Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,20 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/bn/clean_scraped_3.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/bn/merge_bn.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/bn/ocr_newspaper.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/ch/raw_corpus.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/hi/clean_scraped.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/hi/clean_scraped_2.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/hi/clean_scraped_3.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/hi/merge_hi.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/hi/ocr_newspapers.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/kn/clean_scraped_2.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/kn/merge_kn.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/mag/raw_corpus_mag.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/mr/clean_scraped.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/mr/clean_scraped_3.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/mr/merge_mr.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/te/clean_scraped.txt filter=lfs diff=lfs merge=lfs -text
+train_data/train_data/te/ocr_newspaper.txt filter=lfs diff=lfs merge=lfs -text

train_data/.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

train_data/.gitignore

@@ -0,0 +1 @@
+.cache/

train_data/README.md

@@ -0,0 +1,101 @@
+---
+language:
+- hi
+- bn
+- te
+- mr
+- kn
+- bho
+- mai
+- mag
+- hne
+tags:
+- muril
+- bert
+- mlm
+- low-resource
+- indic-languages
+license: apache-2.0
+library_name: transformers
+pipeline_tag: fill-mask
+---
+
+# Fine-Tuned MuRIL Model On Indic Languages (RESPIN)
+
+## 1. Abstract
+
+This document details the evaluation and performance metrics of a fine-tuned MuRIL (Multilingual Representations for Indian Languages) model. The primary objective was to adapt the pre-trained MuRIL architecture to the Indic text domain using the RESPIN (REcognizing SPeech in INdian languages) corpus.
+
+Performance was benchmarked against the original pre-trained `google/muril-base-cased` model, the multilingual industry standard `xlm-roberta-base`, and specialized monolingual architectures (e.g., L3Cube series). The evaluation demonstrates that the fine-tuned model establishes a new state-of-the-art for low-resource languages such as Chhattisgarhi and Magahi.
+
+## 2. Methodology
+
+### 2.1 Training Corpus
+
+The model was fine-tuned utilizing the RESPIN dataset. This corpus comprises text data acquired through extensive web crawling, alongside content derived from newspapers and books processed via Optical Character Recognition (OCR). The heterogeneous nature of these sources necessitated domain-specific adaptation to minimize perplexity (PPL) and enhance model robustness across varied linguistic contexts.
+
+### 2.2 Model Architecture
+
+- **Base Architecture:** MuRIL (BERT-based encoder).
+- **Modification:** Fine-tuned on the RESPIN corpus using Masked Language Modeling (MLM) objectives to align the vector space with the target distribution.
+
+### 2.3 Evaluation Baselines
+
+To provide a rigorous assessment of the model's efficacy, the following comparative baselines were utilized:
+
+- **Base MuRIL (`google/muril-base-cased`):** Used to quantify the net improvement (gain) achieved strictly through the fine-tuning process.
+- **XLM-RoBERTa (`xlm-roberta-base`):** Selected as the high-capacity multilingual baseline to evaluate zero-shot performance on low-resource languages.
+- **Specialist Models (`l3cube-pune`):** Monolingual models trained specifically on a single language. These represent the theoretical upper bound for performance in major languages.
+
+## 3. Evaluation Datasets
+
+Testing was conducted on a diverse suite of held-out evaluation sets to ensure validity. The evaluation suite includes:
+
+- **Benchmark Testing (Legal):** A high-complexity dataset comprising legal documents and bail applications (Available for HI, BN, TE, MR).
+- **Samanantar:** A general-domain parallel corpus utilized for translation benchmarks (Available for HI, BN, KN, TE, MR).
+- **RESPIN (Held-Out):** A specific split from the transcription corpus to test retention of the training distribution (Available for all languages).
+- **Rural Women:** A dialect-rich dataset utilized to test robustness in Bhojpuri (BH).
+- **NanoBEIR:** A retrieval benchmark dataset, cleaned for non-Devanagari artifacts, used for Maithili (MT) and Magahi (MAG).
+- **Chhattisgarh TTS:** A transcription dataset used for Chhattisgarhi (CH).
+- **IISc-MILE:** A speech transcription corpus used for Kannada (KN).
+
+## 4. Empirical Results
+
+The following table presents the Average Perplexity (PPL) scores across the test files for each language. Perplexity is defined as the exponential of the cross-entropy loss; lower values indicate superior predictive performance.
+
+### Table 1: Comparative Average Perplexity (Lower is Better)
+
+| Language Code | Language | Base MuRIL | Fine-Tuned MuRIL (Ours) | XLM-RoBERTa | Specialist Model (L3Cube) |
+| :-: | :-: | :-: | :-: | :-: | :-: |
+| CH | Chhattisgarhi | 764.36 | **21.73** | 169.96 | N/A |
+| MAG | Magahi | 279.29 | **34.35** | 67.77 | N/A |
+| BH | Bhojpuri | 444.35 | **116.41** | 122.09 | N/A |
+| MT | Maithili | 598.79 | 141.09 | **90.83** | N/A |
+| HI | Hindi | 27.58 | 15.86 | **10.29** | N/A |
+| TE | Telugu | 62.98 | 15.41 | 12.07 | **7.43** |
+| MR | Marathi | 85.71 | 25.20 | 17.71 | **19.26** |
+| KN | Kannada | 172.59 | 28.87 | 15.85 | **13.23** |
+| BN | Bengali | 92.62 | 207.80 | **15.96** | 33.43 |
+
+## 5. Analysis and Conclusion
+
+### 5.1 Performance on Low-Resource Languages
+
+The most significant observation is the model's performance on extremely low-resource languages (Chhattisgarhi, Magahi, Bhojpuri), where standard multilingual models typically fail.
+
+- **Chhattisgarhi (CH):** The Base MuRIL model exhibited a perplexity of 764.36, indicating a lack of comprehension. The Fine-Tuned model reduced this drastically to 21.73, outperforming the much larger XLM-R (169.96) by an order of magnitude.
+- **Magahi (MAG):** The Fine-Tuned model achieved a perplexity of 34.35, surpassing both the Base MuRIL (279.29) and the XLM-R baseline (67.77).
+- **Bhojpuri (BH):** The model demonstrated superior robustness, achieving a score of 116.41, edging out XLM-R (122.09) and vastly improving upon the Base MuRIL (444.35).
+
+This validates the efficacy of the RESPIN dataset for adapting encoders to under-represented Indic dialects.
+
+### 5.2 Performance on Major Languages
+
+For widely spoken languages (Hindi, Telugu, Marathi), the fine-tuning process yielded substantial improvements over the Base MuRIL architecture.
+
+- **Telugu (TE) & Marathi (MR):** The Fine-Tuned model reduced perplexity by approximately 75% and 70% respectively compared to the Base model. While the specialized monolingual L3Cube models remain the leader in this category (as expected for single-language specialists), the Fine-Tuned MuRIL is a competitive multilingual alternative.
+- **Bengali (BN):** The model exhibited regression in the average perplexity score (207.80). Detailed analysis reveals that while the model performed well on the respin_bn split (68.14), it struggled to generalize to the benchmark_testing and spring_inx datasets.
+
+### 5.3 Summary
+
+The Fine-Tuned MuRIL model establishes a new benchmark for Chhattisgarhi and Magahi text encoding, significantly surpassing existing open-source multilingual alternatives.

train_data/muril_bh_domain/checkpoint-4000/README.md

@@ -0,0 +1,202 @@
+---
+base_model: google/muril-base-cased
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.2

train_data/muril_bh_domain/checkpoint-4000/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "BertForMaskedLM",
+    "parent_library": "transformers.models.bert.modeling_bert"
+  },
+  "base_model_name_or_path": "google/muril-base-cased",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "query",
+    "value"
+  ],
+  "task_type": null,
+  "use_dora": false,
+  "use_rslora": false
+}

train_data/muril_bh_domain/checkpoint-4000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:deb73549679fd91090a64a88d75c687dde7b28b08950efa95d45de74872e7203
+size 2366064

train_data/muril_bh_domain/checkpoint-4000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1daa4eb30785be66449e5d62b8e08a5908afcf2aa7986702f353f32bc8b42d3
+size 4759290

train_data/muril_bh_domain/checkpoint-4000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26e24e2231a33bdad5887754a5b6968219ef12961d04369739f1350efee44b72
+size 14244

train_data/muril_bh_domain/checkpoint-4000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:edbdfc0163382607dc91e6b9699f3f3a2c3a204482ef6360e91ff42dd0f5b83a
+size 1064

train_data/muril_bh_domain/checkpoint-4000/trainer_state.json

@@ -0,0 +1,341 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.530844669408415,
+  "eval_steps": 1000,
+  "global_step": 4000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.06327111673521038,
+      "grad_norm": 4.5707688331604,
+      "learning_rate": 1.0548523206751056e-05,
+      "loss": 6.0536,
+      "step": 100
+    },
+    {
+      "epoch": 0.12654223347042076,
+      "grad_norm": 7.803924083709717,
+      "learning_rate": 2.1097046413502112e-05,
+      "loss": 5.8177,
+      "step": 200
+    },
+    {
+      "epoch": 0.18981335020563114,
+      "grad_norm": 8.050126075744629,
+      "learning_rate": 3.1645569620253167e-05,
+      "loss": 5.0311,
+      "step": 300
+    },
+    {
+      "epoch": 0.2530844669408415,
+      "grad_norm": 8.803975105285645,
+      "learning_rate": 4.2194092827004224e-05,
+      "loss": 4.7351,
+      "step": 400
+    },
+    {
+      "epoch": 0.3163555836760519,
+      "grad_norm": 12.846040725708008,
+      "learning_rate": 4.96952648851383e-05,
+      "loss": 4.5453,
+      "step": 500
+    },
+    {
+      "epoch": 0.3796267004112623,
+      "grad_norm": 15.128095626831055,
+      "learning_rate": 4.852320675105486e-05,
+      "loss": 4.3979,
+      "step": 600
+    },
+    {
+      "epoch": 0.44289781714647264,
+      "grad_norm": 10.33956527709961,
+      "learning_rate": 4.7351148616971405e-05,
+      "loss": 4.4056,
+      "step": 700
+    },
+    {
+      "epoch": 0.506168933881683,
+      "grad_norm": 23.287179946899414,
+      "learning_rate": 4.617909048288795e-05,
+      "loss": 4.2787,
+      "step": 800
+    },
+    {
+      "epoch": 0.5694400506168934,
+      "grad_norm": 12.955565452575684,
+      "learning_rate": 4.50070323488045e-05,
+      "loss": 4.3504,
+      "step": 900
+    },
+    {
+      "epoch": 0.6327111673521038,
+      "grad_norm": 10.07479476928711,
+      "learning_rate": 4.3834974214721055e-05,
+      "loss": 4.2521,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6327111673521038,
+      "eval_runtime": 33.0567,
+      "eval_samples_per_second": 95.593,
+      "eval_steps_per_second": 11.949,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6959822840873141,
+      "grad_norm": 12.32780647277832,
+      "learning_rate": 4.26629160806376e-05,
+      "loss": 4.2314,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7592534008225246,
+      "grad_norm": 15.515801429748535,
+      "learning_rate": 4.149085794655415e-05,
+      "loss": 4.3058,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8225245175577349,
+      "grad_norm": 12.472834587097168,
+      "learning_rate": 4.03187998124707e-05,
+      "loss": 4.2158,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8857956342929453,
+      "grad_norm": 16.903112411499023,
+      "learning_rate": 3.914674167838725e-05,
+      "loss": 4.142,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9490667510281556,
+      "grad_norm": 13.487791061401367,
+      "learning_rate": 3.79746835443038e-05,
+      "loss": 4.0534,
+      "step": 1500
+    },
+    {
+      "epoch": 1.012337867763366,
+      "grad_norm": 14.721494674682617,
+      "learning_rate": 3.680262541022035e-05,
+      "loss": 4.1405,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0756089844985763,
+      "grad_norm": 16.011690139770508,
+      "learning_rate": 3.56305672761369e-05,
+      "loss": 4.1192,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1388801012337868,
+      "grad_norm": 15.692846298217773,
+      "learning_rate": 3.445850914205345e-05,
+      "loss": 4.1407,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2021512179689973,
+      "grad_norm": 13.71811294555664,
+      "learning_rate": 3.328645100797e-05,
+      "loss": 4.1839,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2654223347042075,
+      "grad_norm": 13.474448204040527,
+      "learning_rate": 3.2114392873886545e-05,
+      "loss": 4.0813,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2654223347042075,
+      "eval_runtime": 33.0889,
+      "eval_samples_per_second": 95.5,
+      "eval_steps_per_second": 11.938,
+      "step": 2000
+    },
+    {
+      "epoch": 1.328693451439418,
+      "grad_norm": 15.276654243469238,
+      "learning_rate": 3.09423347398031e-05,
+      "loss": 3.9624,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3919645681746282,
+      "grad_norm": 13.796238899230957,
+      "learning_rate": 2.9770276605719643e-05,
+      "loss": 4.0653,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4552356849098387,
+      "grad_norm": 16.452594757080078,
+      "learning_rate": 2.8598218471636194e-05,
+      "loss": 4.0338,
+      "step": 2300
+    },
+    {
+      "epoch": 1.518506801645049,
+      "grad_norm": 20.27753257751465,
+      "learning_rate": 2.7426160337552742e-05,
+      "loss": 4.0962,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5817779183802594,
+      "grad_norm": 15.492554664611816,
+      "learning_rate": 2.6254102203469293e-05,
+      "loss": 4.057,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6450490351154698,
+      "grad_norm": 15.298516273498535,
+      "learning_rate": 2.508204406938584e-05,
+      "loss": 4.0301,
+      "step": 2600
+    },
+    {
+      "epoch": 1.70832015185068,
+      "grad_norm": 15.670785903930664,
+      "learning_rate": 2.3909985935302392e-05,
+      "loss": 4.0129,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7715912685858906,
+      "grad_norm": 18.541555404663086,
+      "learning_rate": 2.2737927801218943e-05,
+      "loss": 3.9724,
+      "step": 2800
+    },
+    {
+      "epoch": 1.834862385321101,
+      "grad_norm": 19.13411521911621,
+      "learning_rate": 2.156586966713549e-05,
+      "loss": 4.0044,
+      "step": 2900
+    },
+    {
+      "epoch": 1.8981335020563113,
+      "grad_norm": 14.532624244689941,
+      "learning_rate": 2.039381153305204e-05,
+      "loss": 3.9882,
+      "step": 3000
+    },
+    {
+      "epoch": 1.8981335020563113,
+      "eval_runtime": 33.1181,
+      "eval_samples_per_second": 95.416,
+      "eval_steps_per_second": 11.927,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9614046187915217,
+      "grad_norm": 15.767202377319336,
+      "learning_rate": 1.922175339896859e-05,
+      "loss": 3.9372,
+      "step": 3100
+    },
+    {
+      "epoch": 2.024675735526732,
+      "grad_norm": 17.210546493530273,
+      "learning_rate": 1.804969526488514e-05,
+      "loss": 3.9757,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0879468522619424,
+      "grad_norm": 15.209254264831543,
+      "learning_rate": 1.6877637130801688e-05,
+      "loss": 3.9668,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1512179689971527,
+      "grad_norm": 16.821176528930664,
+      "learning_rate": 1.570557899671824e-05,
+      "loss": 3.9732,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2144890857323634,
+      "grad_norm": 15.914960861206055,
+      "learning_rate": 1.4533520862634786e-05,
+      "loss": 3.9375,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2777602024675736,
+      "grad_norm": 16.489627838134766,
+      "learning_rate": 1.3361462728551336e-05,
+      "loss": 3.9993,
+      "step": 3600
+    },
+    {
+      "epoch": 2.341031319202784,
+      "grad_norm": 17.943021774291992,
+      "learning_rate": 1.2189404594467887e-05,
+      "loss": 3.9889,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4043024359379945,
+      "grad_norm": 14.150239944458008,
+      "learning_rate": 1.1017346460384436e-05,
+      "loss": 4.0,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4675735526732048,
+      "grad_norm": 15.843707084655762,
+      "learning_rate": 9.845288326300985e-06,
+      "loss": 3.9527,
+      "step": 3900
+    },
+    {
+      "epoch": 2.530844669408415,
+      "grad_norm": 16.922142028808594,
+      "learning_rate": 8.673230192217533e-06,
+      "loss": 3.8801,
+      "step": 4000
+    },
+    {
+      "epoch": 2.530844669408415,
+      "eval_runtime": 33.0326,
+      "eval_samples_per_second": 95.663,
+      "eval_steps_per_second": 11.958,
+      "step": 4000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 4740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1700301256769430.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

train_data/muril_bh_domain/checkpoint-4000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a22c26ec129497c28db8c0928a06ac51978ce151cffa0ea84f44917313aa4fe
+size 5304

train_data/muril_bh_domain/checkpoint-4740/README.md

@@ -0,0 +1,202 @@
+---
+base_model: google/muril-base-cased
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.2

train_data/muril_bh_domain/checkpoint-4740/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "BertForMaskedLM",
+    "parent_library": "transformers.models.bert.modeling_bert"
+  },
+  "base_model_name_or_path": "google/muril-base-cased",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "query",
+    "value"
+  ],
+  "task_type": null,
+  "use_dora": false,
+  "use_rslora": false
+}

train_data/muril_bh_domain/checkpoint-4740/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc7f62c24737614448fd04808465491ac77f32241a0a2a9cab226b74deb95ad2
+size 2366064

train_data/muril_bh_domain/checkpoint-4740/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e80f2c68afd84173016108d5470b712a09e93d00829013e7b7624072079df805
+size 4759290

train_data/muril_bh_domain/checkpoint-4740/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c9afe1aba4f99f67ff47a476d35ef83b49ca75cc2da61962433d26d445986b7
+size 14244

train_data/muril_bh_domain/checkpoint-4740/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45472c9e15a883df2692bf1340ef4c87defa24d90bfd8aefefaaecfb0aac3aa4
+size 1064

train_data/muril_bh_domain/checkpoint-4740/trainer_state.json

@@ -0,0 +1,390 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.999050933248972,
+  "eval_steps": 1000,
+  "global_step": 4740,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.06327111673521038,
+      "grad_norm": 4.5707688331604,
+      "learning_rate": 1.0548523206751056e-05,
+      "loss": 6.0536,
+      "step": 100
+    },
+    {
+      "epoch": 0.12654223347042076,
+      "grad_norm": 7.803924083709717,
+      "learning_rate": 2.1097046413502112e-05,
+      "loss": 5.8177,
+      "step": 200
+    },
+    {
+      "epoch": 0.18981335020563114,
+      "grad_norm": 8.050126075744629,
+      "learning_rate": 3.1645569620253167e-05,
+      "loss": 5.0311,
+      "step": 300
+    },
+    {
+      "epoch": 0.2530844669408415,
+      "grad_norm": 8.803975105285645,
+      "learning_rate": 4.2194092827004224e-05,
+      "loss": 4.7351,
+      "step": 400
+    },
+    {
+      "epoch": 0.3163555836760519,
+      "grad_norm": 12.846040725708008,
+      "learning_rate": 4.96952648851383e-05,
+      "loss": 4.5453,
+      "step": 500
+    },
+    {
+      "epoch": 0.3796267004112623,
+      "grad_norm": 15.128095626831055,
+      "learning_rate": 4.852320675105486e-05,
+      "loss": 4.3979,
+      "step": 600
+    },
+    {
+      "epoch": 0.44289781714647264,
+      "grad_norm": 10.33956527709961,
+      "learning_rate": 4.7351148616971405e-05,
+      "loss": 4.4056,
+      "step": 700
+    },
+    {
+      "epoch": 0.506168933881683,
+      "grad_norm": 23.287179946899414,
+      "learning_rate": 4.617909048288795e-05,
+      "loss": 4.2787,
+      "step": 800
+    },
+    {
+      "epoch": 0.5694400506168934,
+      "grad_norm": 12.955565452575684,
+      "learning_rate": 4.50070323488045e-05,
+      "loss": 4.3504,
+      "step": 900
+    },
+    {
+      "epoch": 0.6327111673521038,
+      "grad_norm": 10.07479476928711,
+      "learning_rate": 4.3834974214721055e-05,
+      "loss": 4.2521,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6327111673521038,
+      "eval_runtime": 33.0567,
+      "eval_samples_per_second": 95.593,
+      "eval_steps_per_second": 11.949,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6959822840873141,
+      "grad_norm": 12.32780647277832,
+      "learning_rate": 4.26629160806376e-05,
+      "loss": 4.2314,
+      "step": 1100
+    },
+    {
+      "epoch": 0.7592534008225246,
+      "grad_norm": 15.515801429748535,
+      "learning_rate": 4.149085794655415e-05,
+      "loss": 4.3058,
+      "step": 1200
+    },
+    {
+      "epoch": 0.8225245175577349,
+      "grad_norm": 12.472834587097168,
+      "learning_rate": 4.03187998124707e-05,
+      "loss": 4.2158,
+      "step": 1300
+    },
+    {
+      "epoch": 0.8857956342929453,
+      "grad_norm": 16.903112411499023,
+      "learning_rate": 3.914674167838725e-05,
+      "loss": 4.142,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9490667510281556,
+      "grad_norm": 13.487791061401367,
+      "learning_rate": 3.79746835443038e-05,
+      "loss": 4.0534,
+      "step": 1500
+    },
+    {
+      "epoch": 1.012337867763366,
+      "grad_norm": 14.721494674682617,
+      "learning_rate": 3.680262541022035e-05,
+      "loss": 4.1405,
+      "step": 1600
+    },
+    {
+      "epoch": 1.0756089844985763,
+      "grad_norm": 16.011690139770508,
+      "learning_rate": 3.56305672761369e-05,
+      "loss": 4.1192,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1388801012337868,
+      "grad_norm": 15.692846298217773,
+      "learning_rate": 3.445850914205345e-05,
+      "loss": 4.1407,
+      "step": 1800
+    },
+    {
+      "epoch": 1.2021512179689973,
+      "grad_norm": 13.71811294555664,
+      "learning_rate": 3.328645100797e-05,
+      "loss": 4.1839,
+      "step": 1900
+    },
+    {
+      "epoch": 1.2654223347042075,
+      "grad_norm": 13.474448204040527,
+      "learning_rate": 3.2114392873886545e-05,
+      "loss": 4.0813,
+      "step": 2000
+    },
+    {
+      "epoch": 1.2654223347042075,
+      "eval_runtime": 33.0889,
+      "eval_samples_per_second": 95.5,
+      "eval_steps_per_second": 11.938,
+      "step": 2000
+    },
+    {
+      "epoch": 1.328693451439418,
+      "grad_norm": 15.276654243469238,
+      "learning_rate": 3.09423347398031e-05,
+      "loss": 3.9624,
+      "step": 2100
+    },
+    {
+      "epoch": 1.3919645681746282,
+      "grad_norm": 13.796238899230957,
+      "learning_rate": 2.9770276605719643e-05,
+      "loss": 4.0653,
+      "step": 2200
+    },
+    {
+      "epoch": 1.4552356849098387,
+      "grad_norm": 16.452594757080078,
+      "learning_rate": 2.8598218471636194e-05,
+      "loss": 4.0338,
+      "step": 2300
+    },
+    {
+      "epoch": 1.518506801645049,
+      "grad_norm": 20.27753257751465,
+      "learning_rate": 2.7426160337552742e-05,
+      "loss": 4.0962,
+      "step": 2400
+    },
+    {
+      "epoch": 1.5817779183802594,
+      "grad_norm": 15.492554664611816,
+      "learning_rate": 2.6254102203469293e-05,
+      "loss": 4.057,
+      "step": 2500
+    },
+    {
+      "epoch": 1.6450490351154698,
+      "grad_norm": 15.298516273498535,
+      "learning_rate": 2.508204406938584e-05,
+      "loss": 4.0301,
+      "step": 2600
+    },
+    {
+      "epoch": 1.70832015185068,
+      "grad_norm": 15.670785903930664,
+      "learning_rate": 2.3909985935302392e-05,
+      "loss": 4.0129,
+      "step": 2700
+    },
+    {
+      "epoch": 1.7715912685858906,
+      "grad_norm": 18.541555404663086,
+      "learning_rate": 2.2737927801218943e-05,
+      "loss": 3.9724,
+      "step": 2800
+    },
+    {
+      "epoch": 1.834862385321101,
+      "grad_norm": 19.13411521911621,
+      "learning_rate": 2.156586966713549e-05,
+      "loss": 4.0044,
+      "step": 2900
+    },
+    {
+      "epoch": 1.8981335020563113,
+      "grad_norm": 14.532624244689941,
+      "learning_rate": 2.039381153305204e-05,
+      "loss": 3.9882,
+      "step": 3000
+    },
+    {
+      "epoch": 1.8981335020563113,
+      "eval_runtime": 33.1181,
+      "eval_samples_per_second": 95.416,
+      "eval_steps_per_second": 11.927,
+      "step": 3000
+    },
+    {
+      "epoch": 1.9614046187915217,
+      "grad_norm": 15.767202377319336,
+      "learning_rate": 1.922175339896859e-05,
+      "loss": 3.9372,
+      "step": 3100
+    },
+    {
+      "epoch": 2.024675735526732,
+      "grad_norm": 17.210546493530273,
+      "learning_rate": 1.804969526488514e-05,
+      "loss": 3.9757,
+      "step": 3200
+    },
+    {
+      "epoch": 2.0879468522619424,
+      "grad_norm": 15.209254264831543,
+      "learning_rate": 1.6877637130801688e-05,
+      "loss": 3.9668,
+      "step": 3300
+    },
+    {
+      "epoch": 2.1512179689971527,
+      "grad_norm": 16.821176528930664,
+      "learning_rate": 1.570557899671824e-05,
+      "loss": 3.9732,
+      "step": 3400
+    },
+    {
+      "epoch": 2.2144890857323634,
+      "grad_norm": 15.914960861206055,
+      "learning_rate": 1.4533520862634786e-05,
+      "loss": 3.9375,
+      "step": 3500
+    },
+    {
+      "epoch": 2.2777602024675736,
+      "grad_norm": 16.489627838134766,
+      "learning_rate": 1.3361462728551336e-05,
+      "loss": 3.9993,
+      "step": 3600
+    },
+    {
+      "epoch": 2.341031319202784,
+      "grad_norm": 17.943021774291992,
+      "learning_rate": 1.2189404594467887e-05,
+      "loss": 3.9889,
+      "step": 3700
+    },
+    {
+      "epoch": 2.4043024359379945,
+      "grad_norm": 14.150239944458008,
+      "learning_rate": 1.1017346460384436e-05,
+      "loss": 4.0,
+      "step": 3800
+    },
+    {
+      "epoch": 2.4675735526732048,
+      "grad_norm": 15.843707084655762,
+      "learning_rate": 9.845288326300985e-06,
+      "loss": 3.9527,
+      "step": 3900
+    },
+    {
+      "epoch": 2.530844669408415,
+      "grad_norm": 16.922142028808594,
+      "learning_rate": 8.673230192217533e-06,
+      "loss": 3.8801,
+      "step": 4000
+    },
+    {
+      "epoch": 2.530844669408415,
+      "eval_runtime": 33.0326,
+      "eval_samples_per_second": 95.663,
+      "eval_steps_per_second": 11.958,
+      "step": 4000
+    },
+    {
+      "epoch": 2.5941157861436253,
+      "grad_norm": 19.046831130981445,
+      "learning_rate": 7.501172058134085e-06,
+      "loss": 3.9705,
+      "step": 4100
+    },
+    {
+      "epoch": 2.657386902878836,
+      "grad_norm": 15.516547203063965,
+      "learning_rate": 6.329113924050633e-06,
+      "loss": 3.9296,
+      "step": 4200
+    },
+    {
+      "epoch": 2.720658019614046,
+      "grad_norm": 16.0513916015625,
+      "learning_rate": 5.157055789967183e-06,
+      "loss": 4.0594,
+      "step": 4300
+    },
+    {
+      "epoch": 2.7839291363492564,
+      "grad_norm": 13.820748329162598,
+      "learning_rate": 3.984997655883732e-06,
+      "loss": 3.9353,
+      "step": 4400
+    },
+    {
+      "epoch": 2.847200253084467,
+      "grad_norm": 14.607758522033691,
+      "learning_rate": 2.8129395218002813e-06,
+      "loss": 4.0605,
+      "step": 4500
+    },
+    {
+      "epoch": 2.9104713698196774,
+      "grad_norm": 18.928266525268555,
+      "learning_rate": 1.6408813877168308e-06,
+      "loss": 3.9663,
+      "step": 4600
+    },
+    {
+      "epoch": 2.9737424865548876,
+      "grad_norm": 17.247835159301758,
+      "learning_rate": 4.688232536333803e-07,
+      "loss": 3.9057,
+      "step": 4700
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 4740,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2016724755342822.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

train_data/muril_bh_domain/checkpoint-4740/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a22c26ec129497c28db8c0928a06ac51978ce151cffa0ea84f44917313aa4fe
+size 5304

train_data/muril_bh_domain/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "google/muril-base-cased",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 768,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 197285
+}

train_data/muril_bh_domain/generation_config.json

@@ -0,0 +1,5 @@
+{
+  "_from_model_config": true,
+  "pad_token_id": 0,
+  "transformers_version": "4.46.3"
+}

train_data/muril_bh_domain/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df8f3e88e0c7d9391d346c9abb7e6b7b0729bf480dc131d713e7822db8e5b2a1
+size 951043900

train_data/muril_bh_domain/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

train_data/muril_bh_domain/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "lowercase": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": false,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_data/muril_bn_domain/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "google/muril-base-cased",
+  "architectures": [
+    "BertForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 768,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.3",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 197285
+}

train_data/muril_bn_domain/generation_config.json

@@ -0,0 +1,5 @@
+{
+  "_from_model_config": true,
+  "pad_token_id": 0,
+  "transformers_version": "4.46.3"
+}

train_data/muril_bn_domain/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec9f38111bd7b18e2536f6a68716b77d056876facaf70df2e15d69a6e775c40b
+size 951043900

train_data/muril_bn_domain/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

train_data/muril_bn_domain/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "104": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "105": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "lowercase": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": false,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_data/muril_bn_domain/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7045f6bb9f738a9d8ede00a3453edc500b8eb579edea5aaeff971f5a4baffd29
+size 5304

train_data/muril_ch_domain/checkpoint-30500/README.md

@@ -0,0 +1,202 @@
+---
+base_model: google/muril-base-cased
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.2

train_data/muril_ch_domain/checkpoint-30500/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "BertForMaskedLM",
+    "parent_library": "transformers.models.bert.modeling_bert"
+  },
+  "base_model_name_or_path": "google/muril-base-cased",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "key",
+    "query",
+    "value",
+    "dense"
+  ],
+  "task_type": null,
+  "use_dora": false,
+  "use_rslora": false
+}

train_data/muril_ch_domain/checkpoint-30500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a355714bc7655686649d935524ea8155c31088d897e38f8e91446e8d3699105
+size 42881168

train_data/muril_ch_domain/checkpoint-30500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e53fc7022d7bc695ba1e3ff7fdd1b3368c75e2346cdc9c402d6558e85d67284b
+size 85843898

train_data/muril_ch_domain/checkpoint-30500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dcba7c160327f0ad67913f451ea37b39145ec96b984b91f6a4a72b0a5056736
+size 14244

train_data/muril_ch_domain/checkpoint-30500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e37c8f6a8cf0a15e09cca408bc24278664576b9301e4288b0b47f7768d94d82b
+size 1064

train_data/muril_ch_domain/checkpoint-30500/trainer_state.json

@@ -0,0 +1,2595 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.9839064716528885,
+  "eval_steps": 500,
+  "global_step": 30500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009783299907058651,
+      "grad_norm": 33.780609130859375,
+      "learning_rate": 1.6302575806977503e-06,
+      "loss": 6.574,
+      "step": 100
+    },
+    {
+      "epoch": 0.019566599814117302,
+      "grad_norm": 33.439022064208984,
+      "learning_rate": 3.2605151613955006e-06,
+      "loss": 6.1653,
+      "step": 200
+    },
+    {
+      "epoch": 0.02934989972117595,
+      "grad_norm": 27.25751304626465,
+      "learning_rate": 4.890772742093251e-06,
+      "loss": 5.5515,
+      "step": 300
+    },
+    {
+      "epoch": 0.039133199628234604,
+      "grad_norm": 38.36979675292969,
+      "learning_rate": 6.521030322791001e-06,
+      "loss": 5.0531,
+      "step": 400
+    },
+    {
+      "epoch": 0.048916499535293256,
+      "grad_norm": 29.350488662719727,
+      "learning_rate": 8.15128790348875e-06,
+      "loss": 4.9225,
+      "step": 500
+    },
+    {
+      "epoch": 0.048916499535293256,
+      "eval_runtime": 181.5812,
+      "eval_samples_per_second": 112.578,
+      "eval_steps_per_second": 14.076,
+      "step": 500
+    },
+    {
+      "epoch": 0.0586997994423519,
+      "grad_norm": 33.02122497558594,
+      "learning_rate": 9.781545484186502e-06,
+      "loss": 4.8186,
+      "step": 600
+    },
+    {
+      "epoch": 0.06848309934941056,
+      "grad_norm": 42.41593933105469,
+      "learning_rate": 1.1411803064884251e-05,
+      "loss": 4.5769,
+      "step": 700
+    },
+    {
+      "epoch": 0.07826639925646921,
+      "grad_norm": 40.29044723510742,
+      "learning_rate": 1.3042060645582003e-05,
+      "loss": 4.3963,
+      "step": 800
+    },
+    {
+      "epoch": 0.08804969916352785,
+      "grad_norm": 38.0811653137207,
+      "learning_rate": 1.4672318226279752e-05,
+      "loss": 4.3393,
+      "step": 900
+    },
+    {
+      "epoch": 0.09783299907058651,
+      "grad_norm": 36.08370590209961,
+      "learning_rate": 1.63025758069775e-05,
+      "loss": 4.2421,
+      "step": 1000
+    },
+    {
+      "epoch": 0.09783299907058651,
+      "eval_runtime": 181.886,
+      "eval_samples_per_second": 112.389,
+      "eval_steps_per_second": 14.053,
+      "step": 1000
+    },
+    {
+      "epoch": 0.10761629897764516,
+      "grad_norm": 37.253684997558594,
+      "learning_rate": 1.7932833387675256e-05,
+      "loss": 4.1156,
+      "step": 1100
+    },
+    {
+      "epoch": 0.1173995988847038,
+      "grad_norm": 33.003475189208984,
+      "learning_rate": 1.9563090968373004e-05,
+      "loss": 4.0112,
+      "step": 1200
+    },
+    {
+      "epoch": 0.12718289879176245,
+      "grad_norm": 30.727867126464844,
+      "learning_rate": 2.1193348549070755e-05,
+      "loss": 3.9969,
+      "step": 1300
+    },
+    {
+      "epoch": 0.13696619869882112,
+      "grad_norm": 37.471092224121094,
+      "learning_rate": 2.2823606129768503e-05,
+      "loss": 3.874,
+      "step": 1400
+    },
+    {
+      "epoch": 0.14674949860587977,
+      "grad_norm": 42.32167434692383,
+      "learning_rate": 2.4453863710466254e-05,
+      "loss": 3.8518,
+      "step": 1500
+    },
+    {
+      "epoch": 0.14674949860587977,
+      "eval_runtime": 181.9332,
+      "eval_samples_per_second": 112.36,
+      "eval_steps_per_second": 14.049,
+      "step": 1500
+    },
+    {
+      "epoch": 0.15653279851293841,
+      "grad_norm": 38.00124740600586,
+      "learning_rate": 2.6084121291164005e-05,
+      "loss": 3.918,
+      "step": 1600
+    },
+    {
+      "epoch": 0.16631609841999706,
+      "grad_norm": 44.637386322021484,
+      "learning_rate": 2.7714378871861756e-05,
+      "loss": 3.9134,
+      "step": 1700
+    },
+    {
+      "epoch": 0.1760993983270557,
+      "grad_norm": 49.578609466552734,
+      "learning_rate": 2.9344636452559504e-05,
+      "loss": 3.7507,
+      "step": 1800
+    },
+    {
+      "epoch": 0.18588269823411438,
+      "grad_norm": 36.65715789794922,
+      "learning_rate": 3.0974894033257255e-05,
+      "loss": 3.7551,
+      "step": 1900
+    },
+    {
+      "epoch": 0.19566599814117303,
+      "grad_norm": 36.873443603515625,
+      "learning_rate": 3.2605151613955e-05,
+      "loss": 3.6951,
+      "step": 2000
+    },
+    {
+      "epoch": 0.19566599814117303,
+      "eval_runtime": 181.8273,
+      "eval_samples_per_second": 112.425,
+      "eval_steps_per_second": 14.057,
+      "step": 2000
+    },
+    {
+      "epoch": 0.20544929804823167,
+      "grad_norm": 33.025413513183594,
+      "learning_rate": 3.423540919465276e-05,
+      "loss": 3.6603,
+      "step": 2100
+    },
+    {
+      "epoch": 0.21523259795529032,
+      "grad_norm": 30.105051040649414,
+      "learning_rate": 3.586566677535051e-05,
+      "loss": 3.525,
+      "step": 2200
+    },
+    {
+      "epoch": 0.22501589786234896,
+      "grad_norm": 34.5129280090332,
+      "learning_rate": 3.749592435604825e-05,
+      "loss": 3.6454,
+      "step": 2300
+    },
+    {
+      "epoch": 0.2347991977694076,
+      "grad_norm": 33.16934585571289,
+      "learning_rate": 3.912618193674601e-05,
+      "loss": 3.6356,
+      "step": 2400
+    },
+    {
+      "epoch": 0.24458249767646628,
+      "grad_norm": 33.5789794921875,
+      "learning_rate": 4.0756439517443756e-05,
+      "loss": 3.5605,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24458249767646628,
+      "eval_runtime": 181.7254,
+      "eval_samples_per_second": 112.488,
+      "eval_steps_per_second": 14.065,
+      "step": 2500
+    },
+    {
+      "epoch": 0.2543657975835249,
+      "grad_norm": 34.30876159667969,
+      "learning_rate": 4.238669709814151e-05,
+      "loss": 3.5447,
+      "step": 2600
+    },
+    {
+      "epoch": 0.2641490974905836,
+      "grad_norm": 29.907989501953125,
+      "learning_rate": 4.401695467883926e-05,
+      "loss": 3.5116,
+      "step": 2700
+    },
+    {
+      "epoch": 0.27393239739764225,
+      "grad_norm": 34.08231735229492,
+      "learning_rate": 4.5647212259537006e-05,
+      "loss": 3.4941,
+      "step": 2800
+    },
+    {
+      "epoch": 0.28371569730470086,
+      "grad_norm": 25.034149169921875,
+      "learning_rate": 4.727746984023476e-05,
+      "loss": 3.4863,
+      "step": 2900
+    },
+    {
+      "epoch": 0.29349899721175954,
+      "grad_norm": 32.21685028076172,
+      "learning_rate": 4.890772742093251e-05,
+      "loss": 3.5096,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29349899721175954,
+      "eval_runtime": 181.6612,
+      "eval_samples_per_second": 112.528,
+      "eval_steps_per_second": 14.07,
+      "step": 3000
+    },
+    {
+      "epoch": 0.30328229711881816,
+      "grad_norm": 24.290380477905273,
+      "learning_rate": 4.9940208725902305e-05,
+      "loss": 3.3867,
+      "step": 3100
+    },
+    {
+      "epoch": 0.31306559702587683,
+      "grad_norm": 22.924575805664062,
+      "learning_rate": 4.975902304681838e-05,
+      "loss": 3.398,
+      "step": 3200
+    },
+    {
+      "epoch": 0.3228488969329355,
+      "grad_norm": 19.540430068969727,
+      "learning_rate": 4.957783736773446e-05,
+      "loss": 3.3727,
+      "step": 3300
+    },
+    {
+      "epoch": 0.3326321968399941,
+      "grad_norm": 22.529376983642578,
+      "learning_rate": 4.939665168865053e-05,
+      "loss": 3.3364,
+      "step": 3400
+    },
+    {
+      "epoch": 0.3424154967470528,
+      "grad_norm": 20.821264266967773,
+      "learning_rate": 4.921546600956661e-05,
+      "loss": 3.3126,
+      "step": 3500
+    },
+    {
+      "epoch": 0.3424154967470528,
+      "eval_runtime": 181.7582,
+      "eval_samples_per_second": 112.468,
+      "eval_steps_per_second": 14.063,
+      "step": 3500
+    },
+    {
+      "epoch": 0.3521987966541114,
+      "grad_norm": 24.346153259277344,
+      "learning_rate": 4.903428033048268e-05,
+      "loss": 3.2678,
+      "step": 3600
+    },
+    {
+      "epoch": 0.3619820965611701,
+      "grad_norm": 19.89035415649414,
+      "learning_rate": 4.8853094651398754e-05,
+      "loss": 3.3233,
+      "step": 3700
+    },
+    {
+      "epoch": 0.37176539646822876,
+      "grad_norm": 17.938880920410156,
+      "learning_rate": 4.8671908972314825e-05,
+      "loss": 3.2822,
+      "step": 3800
+    },
+    {
+      "epoch": 0.3815486963752874,
+      "grad_norm": 16.92071533203125,
+      "learning_rate": 4.84907232932309e-05,
+      "loss": 3.2254,
+      "step": 3900
+    },
+    {
+      "epoch": 0.39133199628234605,
+      "grad_norm": 18.241249084472656,
+      "learning_rate": 4.830953761414698e-05,
+      "loss": 3.2116,
+      "step": 4000
+    },
+    {
+      "epoch": 0.39133199628234605,
+      "eval_runtime": 182.8906,
+      "eval_samples_per_second": 111.772,
+      "eval_steps_per_second": 13.976,
+      "step": 4000
+    },
+    {
+      "epoch": 0.40111529618940467,
+      "grad_norm": 17.56020736694336,
+      "learning_rate": 4.812835193506305e-05,
+      "loss": 3.2232,
+      "step": 4100
+    },
+    {
+      "epoch": 0.41089859609646334,
+      "grad_norm": 17.81117057800293,
+      "learning_rate": 4.794716625597913e-05,
+      "loss": 3.1936,
+      "step": 4200
+    },
+    {
+      "epoch": 0.420681896003522,
+      "grad_norm": 19.89581871032715,
+      "learning_rate": 4.77659805768952e-05,
+      "loss": 3.1443,
+      "step": 4300
+    },
+    {
+      "epoch": 0.43046519591058063,
+      "grad_norm": 22.968582153320312,
+      "learning_rate": 4.758479489781128e-05,
+      "loss": 3.2084,
+      "step": 4400
+    },
+    {
+      "epoch": 0.4402484958176393,
+      "grad_norm": 17.119598388671875,
+      "learning_rate": 4.740360921872735e-05,
+      "loss": 3.1263,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4402484958176393,
+      "eval_runtime": 182.3246,
+      "eval_samples_per_second": 112.119,
+      "eval_steps_per_second": 14.019,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4500317957246979,
+      "grad_norm": 19.294527053833008,
+      "learning_rate": 4.722242353964343e-05,
+      "loss": 3.1327,
+      "step": 4600
+    },
+    {
+      "epoch": 0.4598150956317566,
+      "grad_norm": 16.941057205200195,
+      "learning_rate": 4.704123786055951e-05,
+      "loss": 3.0944,
+      "step": 4700
+    },
+    {
+      "epoch": 0.4695983955388152,
+      "grad_norm": 22.43411636352539,
+      "learning_rate": 4.686005218147558e-05,
+      "loss": 3.1093,
+      "step": 4800
+    },
+    {
+      "epoch": 0.4793816954458739,
+      "grad_norm": 19.64097023010254,
+      "learning_rate": 4.667886650239166e-05,
+      "loss": 3.0597,
+      "step": 4900
+    },
+    {
+      "epoch": 0.48916499535293256,
+      "grad_norm": 19.343788146972656,
+      "learning_rate": 4.649768082330773e-05,
+      "loss": 3.1659,
+      "step": 5000
+    },
+    {
+      "epoch": 0.48916499535293256,
+      "eval_runtime": 181.8771,
+      "eval_samples_per_second": 112.395,
+      "eval_steps_per_second": 14.053,
+      "step": 5000
+    },
+    {
+      "epoch": 0.4989482952599912,
+      "grad_norm": 19.657760620117188,
+      "learning_rate": 4.63164951442238e-05,
+      "loss": 3.0506,
+      "step": 5100
+    },
+    {
+      "epoch": 0.5087315951670498,
+      "grad_norm": 16.2425537109375,
+      "learning_rate": 4.613530946513987e-05,
+      "loss": 3.0524,
+      "step": 5200
+    },
+    {
+      "epoch": 0.5185148950741085,
+      "grad_norm": 19.64779281616211,
+      "learning_rate": 4.595412378605595e-05,
+      "loss": 2.9995,
+      "step": 5300
+    },
+    {
+      "epoch": 0.5282981949811671,
+      "grad_norm": 17.29520606994629,
+      "learning_rate": 4.577293810697203e-05,
+      "loss": 3.0932,
+      "step": 5400
+    },
+    {
+      "epoch": 0.5380814948882258,
+      "grad_norm": 17.694602966308594,
+      "learning_rate": 4.55917524278881e-05,
+      "loss": 3.0309,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5380814948882258,
+      "eval_runtime": 181.7231,
+      "eval_samples_per_second": 112.49,
+      "eval_steps_per_second": 14.065,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5478647947952845,
+      "grad_norm": 21.030174255371094,
+      "learning_rate": 4.541056674880418e-05,
+      "loss": 3.0313,
+      "step": 5600
+    },
+    {
+      "epoch": 0.5576480947023431,
+      "grad_norm": 12.339129447937012,
+      "learning_rate": 4.522938106972025e-05,
+      "loss": 3.047,
+      "step": 5700
+    },
+    {
+      "epoch": 0.5674313946094017,
+      "grad_norm": 16.496389389038086,
+      "learning_rate": 4.504819539063633e-05,
+      "loss": 2.9961,
+      "step": 5800
+    },
+    {
+      "epoch": 0.5772146945164603,
+      "grad_norm": 15.456297874450684,
+      "learning_rate": 4.48670097115524e-05,
+      "loss": 2.9821,
+      "step": 5900
+    },
+    {
+      "epoch": 0.5869979944235191,
+      "grad_norm": 17.8603572845459,
+      "learning_rate": 4.468582403246848e-05,
+      "loss": 2.9294,
+      "step": 6000
+    },
+    {
+      "epoch": 0.5869979944235191,
+      "eval_runtime": 181.8258,
+      "eval_samples_per_second": 112.426,
+      "eval_steps_per_second": 14.057,
+      "step": 6000
+    },
+    {
+      "epoch": 0.5967812943305777,
+      "grad_norm": 18.85349464416504,
+      "learning_rate": 4.450463835338455e-05,
+      "loss": 2.9929,
+      "step": 6100
+    },
+    {
+      "epoch": 0.6065645942376363,
+      "grad_norm": 22.971813201904297,
+      "learning_rate": 4.432345267430063e-05,
+      "loss": 2.9684,
+      "step": 6200
+    },
+    {
+      "epoch": 0.616347894144695,
+      "grad_norm": 15.877230644226074,
+      "learning_rate": 4.4142266995216706e-05,
+      "loss": 2.9399,
+      "step": 6300
+    },
+    {
+      "epoch": 0.6261311940517537,
+      "grad_norm": 19.847482681274414,
+      "learning_rate": 4.396108131613278e-05,
+      "loss": 2.88,
+      "step": 6400
+    },
+    {
+      "epoch": 0.6359144939588123,
+      "grad_norm": 15.004170417785645,
+      "learning_rate": 4.377989563704885e-05,
+      "loss": 2.9719,
+      "step": 6500
+    },
+    {
+      "epoch": 0.6359144939588123,
+      "eval_runtime": 182.6045,
+      "eval_samples_per_second": 111.947,
+      "eval_steps_per_second": 13.997,
+      "step": 6500
+    },
+    {
+      "epoch": 0.645697793865871,
+      "grad_norm": 19.473665237426758,
+      "learning_rate": 4.359870995796492e-05,
+      "loss": 2.9246,
+      "step": 6600
+    },
+    {
+      "epoch": 0.6554810937729296,
+      "grad_norm": 18.071683883666992,
+      "learning_rate": 4.3417524278881e-05,
+      "loss": 2.9031,
+      "step": 6700
+    },
+    {
+      "epoch": 0.6652643936799882,
+      "grad_norm": 17.544504165649414,
+      "learning_rate": 4.323633859979707e-05,
+      "loss": 2.8313,
+      "step": 6800
+    },
+    {
+      "epoch": 0.6750476935870469,
+      "grad_norm": 18.936140060424805,
+      "learning_rate": 4.305515292071315e-05,
+      "loss": 2.8536,
+      "step": 6900
+    },
+    {
+      "epoch": 0.6848309934941056,
+      "grad_norm": 14.77696418762207,
+      "learning_rate": 4.2873967241629226e-05,
+      "loss": 2.9104,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6848309934941056,
+      "eval_runtime": 181.938,
+      "eval_samples_per_second": 112.357,
+      "eval_steps_per_second": 14.049,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6946142934011642,
+      "grad_norm": 14.303226470947266,
+      "learning_rate": 4.26927815625453e-05,
+      "loss": 2.8386,
+      "step": 7100
+    },
+    {
+      "epoch": 0.7043975933082228,
+      "grad_norm": 17.11782455444336,
+      "learning_rate": 4.2511595883461376e-05,
+      "loss": 2.9013,
+      "step": 7200
+    },
+    {
+      "epoch": 0.7141808932152816,
+      "grad_norm": 18.661100387573242,
+      "learning_rate": 4.233041020437745e-05,
+      "loss": 2.9428,
+      "step": 7300
+    },
+    {
+      "epoch": 0.7239641931223402,
+      "grad_norm": 15.535719871520996,
+      "learning_rate": 4.2149224525293525e-05,
+      "loss": 2.8582,
+      "step": 7400
+    },
+    {
+      "epoch": 0.7337474930293988,
+      "grad_norm": 15.3306303024292,
+      "learning_rate": 4.19680388462096e-05,
+      "loss": 2.8896,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7337474930293988,
+      "eval_runtime": 181.8938,
+      "eval_samples_per_second": 112.384,
+      "eval_steps_per_second": 14.052,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7435307929364575,
+      "grad_norm": 16.730344772338867,
+      "learning_rate": 4.1786853167125675e-05,
+      "loss": 2.9097,
+      "step": 7600
+    },
+    {
+      "epoch": 0.7533140928435161,
+      "grad_norm": 18.755483627319336,
+      "learning_rate": 4.1605667488041746e-05,
+      "loss": 2.8815,
+      "step": 7700
+    },
+    {
+      "epoch": 0.7630973927505748,
+      "grad_norm": 18.737581253051758,
+      "learning_rate": 4.1424481808957824e-05,
+      "loss": 2.9202,
+      "step": 7800
+    },
+    {
+      "epoch": 0.7728806926576334,
+      "grad_norm": 14.711681365966797,
+      "learning_rate": 4.1243296129873896e-05,
+      "loss": 2.806,
+      "step": 7900
+    },
+    {
+      "epoch": 0.7826639925646921,
+      "grad_norm": 17.5069580078125,
+      "learning_rate": 4.106211045078997e-05,
+      "loss": 2.8576,
+      "step": 8000
+    },
+    {
+      "epoch": 0.7826639925646921,
+      "eval_runtime": 181.9442,
+      "eval_samples_per_second": 112.353,
+      "eval_steps_per_second": 14.048,
+      "step": 8000
+    },
+    {
+      "epoch": 0.7924472924717507,
+      "grad_norm": 17.678852081298828,
+      "learning_rate": 4.0880924771706046e-05,
+      "loss": 2.8035,
+      "step": 8100
+    },
+    {
+      "epoch": 0.8022305923788093,
+      "grad_norm": 17.644638061523438,
+      "learning_rate": 4.069973909262212e-05,
+      "loss": 2.7958,
+      "step": 8200
+    },
+    {
+      "epoch": 0.8120138922858681,
+      "grad_norm": 18.377134323120117,
+      "learning_rate": 4.0518553413538195e-05,
+      "loss": 2.8055,
+      "step": 8300
+    },
+    {
+      "epoch": 0.8217971921929267,
+      "grad_norm": 18.026033401489258,
+      "learning_rate": 4.0337367734454273e-05,
+      "loss": 2.7334,
+      "step": 8400
+    },
+    {
+      "epoch": 0.8315804920999853,
+      "grad_norm": 14.77315616607666,
+      "learning_rate": 4.0156182055370345e-05,
+      "loss": 2.8082,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8315804920999853,
+      "eval_runtime": 182.4176,
+      "eval_samples_per_second": 112.062,
+      "eval_steps_per_second": 14.012,
+      "step": 8500
+    },
+    {
+      "epoch": 0.841363792007044,
+      "grad_norm": 13.729479789733887,
+      "learning_rate": 3.997499637628642e-05,
+      "loss": 2.7939,
+      "step": 8600
+    },
+    {
+      "epoch": 0.8511470919141026,
+      "grad_norm": 16.34333610534668,
+      "learning_rate": 3.9793810697202494e-05,
+      "loss": 2.8517,
+      "step": 8700
+    },
+    {
+      "epoch": 0.8609303918211613,
+      "grad_norm": 22.484411239624023,
+      "learning_rate": 3.961262501811857e-05,
+      "loss": 2.776,
+      "step": 8800
+    },
+    {
+      "epoch": 0.8707136917282199,
+      "grad_norm": 15.922870635986328,
+      "learning_rate": 3.9431439339034644e-05,
+      "loss": 2.7909,
+      "step": 8900
+    },
+    {
+      "epoch": 0.8804969916352786,
+      "grad_norm": 15.06955623626709,
+      "learning_rate": 3.925025365995072e-05,
+      "loss": 2.8416,
+      "step": 9000
+    },
+    {
+      "epoch": 0.8804969916352786,
+      "eval_runtime": 181.9314,
+      "eval_samples_per_second": 112.361,
+      "eval_steps_per_second": 14.049,
+      "step": 9000
+    },
+    {
+      "epoch": 0.8902802915423372,
+      "grad_norm": 16.060428619384766,
+      "learning_rate": 3.9069067980866794e-05,
+      "loss": 2.7803,
+      "step": 9100
+    },
+    {
+      "epoch": 0.9000635914493959,
+      "grad_norm": 16.80124855041504,
+      "learning_rate": 3.888788230178287e-05,
+      "loss": 2.7548,
+      "step": 9200
+    },
+    {
+      "epoch": 0.9098468913564546,
+      "grad_norm": 16.608434677124023,
+      "learning_rate": 3.870669662269894e-05,
+      "loss": 2.8606,
+      "step": 9300
+    },
+    {
+      "epoch": 0.9196301912635132,
+      "grad_norm": 14.83870792388916,
+      "learning_rate": 3.8525510943615015e-05,
+      "loss": 2.7833,
+      "step": 9400
+    },
+    {
+      "epoch": 0.9294134911705718,
+      "grad_norm": 25.778181076049805,
+      "learning_rate": 3.834432526453109e-05,
+      "loss": 2.7434,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9294134911705718,
+      "eval_runtime": 181.99,
+      "eval_samples_per_second": 112.325,
+      "eval_steps_per_second": 14.045,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9391967910776304,
+      "grad_norm": 17.374011993408203,
+      "learning_rate": 3.8163139585447164e-05,
+      "loss": 2.7258,
+      "step": 9600
+    },
+    {
+      "epoch": 0.9489800909846892,
+      "grad_norm": 17.551128387451172,
+      "learning_rate": 3.798195390636324e-05,
+      "loss": 2.824,
+      "step": 9700
+    },
+    {
+      "epoch": 0.9587633908917478,
+      "grad_norm": 14.35797119140625,
+      "learning_rate": 3.7800768227279314e-05,
+      "loss": 2.745,
+      "step": 9800
+    },
+    {
+      "epoch": 0.9685466907988064,
+      "grad_norm": 20.098552703857422,
+      "learning_rate": 3.761958254819539e-05,
+      "loss": 2.7025,
+      "step": 9900
+    },
+    {
+      "epoch": 0.9783299907058651,
+      "grad_norm": 16.218109130859375,
+      "learning_rate": 3.743839686911147e-05,
+      "loss": 2.8093,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9783299907058651,
+      "eval_runtime": 181.8987,
+      "eval_samples_per_second": 112.381,
+      "eval_steps_per_second": 14.052,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9881132906129237,
+      "grad_norm": 17.198423385620117,
+      "learning_rate": 3.725721119002754e-05,
+      "loss": 2.7124,
+      "step": 10100
+    },
+    {
+      "epoch": 0.9978965905199824,
+      "grad_norm": 18.021198272705078,
+      "learning_rate": 3.707602551094362e-05,
+      "loss": 2.6922,
+      "step": 10200
+    },
+    {
+      "epoch": 1.007679890427041,
+      "grad_norm": 15.27678108215332,
+      "learning_rate": 3.689483983185969e-05,
+      "loss": 2.6743,
+      "step": 10300
+    },
+    {
+      "epoch": 1.0174631903340996,
+      "grad_norm": 16.770511627197266,
+      "learning_rate": 3.671365415277577e-05,
+      "loss": 2.857,
+      "step": 10400
+    },
+    {
+      "epoch": 1.0272464902411584,
+      "grad_norm": 18.810932159423828,
+      "learning_rate": 3.653246847369184e-05,
+      "loss": 2.7269,
+      "step": 10500
+    },
+    {
+      "epoch": 1.0272464902411584,
+      "eval_runtime": 181.8537,
+      "eval_samples_per_second": 112.409,
+      "eval_steps_per_second": 14.055,
+      "step": 10500
+    },
+    {
+      "epoch": 1.037029790148217,
+      "grad_norm": 18.56201171875,
+      "learning_rate": 3.635128279460791e-05,
+      "loss": 2.7325,
+      "step": 10600
+    },
+    {
+      "epoch": 1.0468130900552757,
+      "grad_norm": 15.063011169433594,
+      "learning_rate": 3.617009711552399e-05,
+      "loss": 2.7827,
+      "step": 10700
+    },
+    {
+      "epoch": 1.0565963899623343,
+      "grad_norm": 15.339439392089844,
+      "learning_rate": 3.598891143644006e-05,
+      "loss": 2.7472,
+      "step": 10800
+    },
+    {
+      "epoch": 1.066379689869393,
+      "grad_norm": 17.466033935546875,
+      "learning_rate": 3.580772575735614e-05,
+      "loss": 2.7859,
+      "step": 10900
+    },
+    {
+      "epoch": 1.0761629897764515,
+      "grad_norm": 20.727872848510742,
+      "learning_rate": 3.562654007827221e-05,
+      "loss": 2.7278,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0761629897764515,
+      "eval_runtime": 181.8566,
+      "eval_samples_per_second": 112.407,
+      "eval_steps_per_second": 14.055,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0859462896835101,
+      "grad_norm": 16.02055549621582,
+      "learning_rate": 3.544535439918829e-05,
+      "loss": 2.6307,
+      "step": 11100
+    },
+    {
+      "epoch": 1.095729589590569,
+      "grad_norm": 20.069686889648438,
+      "learning_rate": 3.526416872010436e-05,
+      "loss": 2.711,
+      "step": 11200
+    },
+    {
+      "epoch": 1.1055128894976276,
+      "grad_norm": 14.833261489868164,
+      "learning_rate": 3.508298304102044e-05,
+      "loss": 2.6141,
+      "step": 11300
+    },
+    {
+      "epoch": 1.1152961894046862,
+      "grad_norm": 14.86436653137207,
+      "learning_rate": 3.490179736193652e-05,
+      "loss": 2.6816,
+      "step": 11400
+    },
+    {
+      "epoch": 1.1250794893117448,
+      "grad_norm": 17.955862045288086,
+      "learning_rate": 3.472061168285259e-05,
+      "loss": 2.6924,
+      "step": 11500
+    },
+    {
+      "epoch": 1.1250794893117448,
+      "eval_runtime": 181.8085,
+      "eval_samples_per_second": 112.437,
+      "eval_steps_per_second": 14.059,
+      "step": 11500
+    },
+    {
+      "epoch": 1.1348627892188035,
+      "grad_norm": 18.360109329223633,
+      "learning_rate": 3.453942600376867e-05,
+      "loss": 2.6181,
+      "step": 11600
+    },
+    {
+      "epoch": 1.144646089125862,
+      "grad_norm": 17.547542572021484,
+      "learning_rate": 3.435824032468474e-05,
+      "loss": 2.6394,
+      "step": 11700
+    },
+    {
+      "epoch": 1.154429389032921,
+      "grad_norm": 12.194833755493164,
+      "learning_rate": 3.417705464560082e-05,
+      "loss": 2.6684,
+      "step": 11800
+    },
+    {
+      "epoch": 1.1642126889399795,
+      "grad_norm": 17.095104217529297,
+      "learning_rate": 3.399586896651689e-05,
+      "loss": 2.6129,
+      "step": 11900
+    },
+    {
+      "epoch": 1.1739959888470382,
+      "grad_norm": 20.788406372070312,
+      "learning_rate": 3.381468328743296e-05,
+      "loss": 2.5663,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1739959888470382,
+      "eval_runtime": 181.8035,
+      "eval_samples_per_second": 112.44,
+      "eval_steps_per_second": 14.059,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1837792887540968,
+      "grad_norm": 14.261167526245117,
+      "learning_rate": 3.363349760834904e-05,
+      "loss": 2.6544,
+      "step": 12100
+    },
+    {
+      "epoch": 1.1935625886611554,
+      "grad_norm": 24.68012046813965,
+      "learning_rate": 3.345231192926511e-05,
+      "loss": 2.6632,
+      "step": 12200
+    },
+    {
+      "epoch": 1.203345888568214,
+      "grad_norm": 16.10886573791504,
+      "learning_rate": 3.327112625018119e-05,
+      "loss": 2.6366,
+      "step": 12300
+    },
+    {
+      "epoch": 1.2131291884752726,
+      "grad_norm": 18.038848876953125,
+      "learning_rate": 3.308994057109726e-05,
+      "loss": 2.6563,
+      "step": 12400
+    },
+    {
+      "epoch": 1.2229124883823315,
+      "grad_norm": 17.40920639038086,
+      "learning_rate": 3.290875489201334e-05,
+      "loss": 2.718,
+      "step": 12500
+    },
+    {
+      "epoch": 1.2229124883823315,
+      "eval_runtime": 181.9491,
+      "eval_samples_per_second": 112.35,
+      "eval_steps_per_second": 14.048,
+      "step": 12500
+    },
+    {
+      "epoch": 1.23269578828939,
+      "grad_norm": 15.097307205200195,
+      "learning_rate": 3.272756921292941e-05,
+      "loss": 2.7282,
+      "step": 12600
+    },
+    {
+      "epoch": 1.2424790881964487,
+      "grad_norm": 17.63008689880371,
+      "learning_rate": 3.254638353384549e-05,
+      "loss": 2.7104,
+      "step": 12700
+    },
+    {
+      "epoch": 1.2522623881035073,
+      "grad_norm": 16.161130905151367,
+      "learning_rate": 3.236519785476156e-05,
+      "loss": 2.6427,
+      "step": 12800
+    },
+    {
+      "epoch": 1.262045688010566,
+      "grad_norm": 18.786882400512695,
+      "learning_rate": 3.218401217567764e-05,
+      "loss": 2.6105,
+      "step": 12900
+    },
+    {
+      "epoch": 1.2718289879176246,
+      "grad_norm": 24.145421981811523,
+      "learning_rate": 3.2002826496593715e-05,
+      "loss": 2.6322,
+      "step": 13000
+    },
+    {
+      "epoch": 1.2718289879176246,
+      "eval_runtime": 182.5613,
+      "eval_samples_per_second": 111.973,
+      "eval_steps_per_second": 14.001,
+      "step": 13000
+    },
+    {
+      "epoch": 1.2816122878246832,
+      "grad_norm": 15.286133766174316,
+      "learning_rate": 3.1821640817509786e-05,
+      "loss": 2.6465,
+      "step": 13100
+    },
+    {
+      "epoch": 1.291395587731742,
+      "grad_norm": 21.22935676574707,
+      "learning_rate": 3.1640455138425865e-05,
+      "loss": 2.6691,
+      "step": 13200
+    },
+    {
+      "epoch": 1.3011788876388006,
+      "grad_norm": 18.064428329467773,
+      "learning_rate": 3.1459269459341936e-05,
+      "loss": 2.5904,
+      "step": 13300
+    },
+    {
+      "epoch": 1.3109621875458592,
+      "grad_norm": 14.45976448059082,
+      "learning_rate": 3.127808378025801e-05,
+      "loss": 2.6602,
+      "step": 13400
+    },
+    {
+      "epoch": 1.3207454874529179,
+      "grad_norm": 19.72386360168457,
+      "learning_rate": 3.109689810117408e-05,
+      "loss": 2.6337,
+      "step": 13500
+    },
+    {
+      "epoch": 1.3207454874529179,
+      "eval_runtime": 182.4053,
+      "eval_samples_per_second": 112.069,
+      "eval_steps_per_second": 14.013,
+      "step": 13500
+    },
+    {
+      "epoch": 1.3305287873599765,
+      "grad_norm": 17.639583587646484,
+      "learning_rate": 3.091571242209016e-05,
+      "loss": 2.6135,
+      "step": 13600
+    },
+    {
+      "epoch": 1.340312087267035,
+      "grad_norm": 19.71700096130371,
+      "learning_rate": 3.0734526743006235e-05,
+      "loss": 2.6252,
+      "step": 13700
+    },
+    {
+      "epoch": 1.3500953871740937,
+      "grad_norm": 16.715856552124023,
+      "learning_rate": 3.055334106392231e-05,
+      "loss": 2.6475,
+      "step": 13800
+    },
+    {
+      "epoch": 1.3598786870811526,
+      "grad_norm": 12.645075798034668,
+      "learning_rate": 3.0372155384838385e-05,
+      "loss": 2.6199,
+      "step": 13900
+    },
+    {
+      "epoch": 1.3696619869882112,
+      "grad_norm": 20.150625228881836,
+      "learning_rate": 3.0190969705754456e-05,
+      "loss": 2.5567,
+      "step": 14000
+    },
+    {
+      "epoch": 1.3696619869882112,
+      "eval_runtime": 181.9086,
+      "eval_samples_per_second": 112.375,
+      "eval_steps_per_second": 14.051,
+      "step": 14000
+    },
+    {
+      "epoch": 1.3794452868952698,
+      "grad_norm": 19.111286163330078,
+      "learning_rate": 3.0009784026670535e-05,
+      "loss": 2.59,
+      "step": 14100
+    },
+    {
+      "epoch": 1.3892285868023284,
+      "grad_norm": 17.12226104736328,
+      "learning_rate": 2.9828598347586606e-05,
+      "loss": 2.5913,
+      "step": 14200
+    },
+    {
+      "epoch": 1.399011886709387,
+      "grad_norm": 19.741445541381836,
+      "learning_rate": 2.9647412668502684e-05,
+      "loss": 2.5617,
+      "step": 14300
+    },
+    {
+      "epoch": 1.4087951866164456,
+      "grad_norm": 17.605525970458984,
+      "learning_rate": 2.946622698941876e-05,
+      "loss": 2.6077,
+      "step": 14400
+    },
+    {
+      "epoch": 1.4185784865235043,
+      "grad_norm": 17.433218002319336,
+      "learning_rate": 2.928504131033483e-05,
+      "loss": 2.5713,
+      "step": 14500
+    },
+    {
+      "epoch": 1.4185784865235043,
+      "eval_runtime": 181.9305,
+      "eval_samples_per_second": 112.362,
+      "eval_steps_per_second": 14.049,
+      "step": 14500
+    },
+    {
+      "epoch": 1.428361786430563,
+      "grad_norm": 15.442538261413574,
+      "learning_rate": 2.910385563125091e-05,
+      "loss": 2.6499,
+      "step": 14600
+    },
+    {
+      "epoch": 1.4381450863376217,
+      "grad_norm": 15.078730583190918,
+      "learning_rate": 2.892266995216698e-05,
+      "loss": 2.6517,
+      "step": 14700
+    },
+    {
+      "epoch": 1.4479283862446803,
+      "grad_norm": 23.07891273498535,
+      "learning_rate": 2.874148427308306e-05,
+      "loss": 2.594,
+      "step": 14800
+    },
+    {
+      "epoch": 1.457711686151739,
+      "grad_norm": 16.707923889160156,
+      "learning_rate": 2.856029859399913e-05,
+      "loss": 2.6613,
+      "step": 14900
+    },
+    {
+      "epoch": 1.4674949860587976,
+      "grad_norm": 16.731164932250977,
+      "learning_rate": 2.8379112914915208e-05,
+      "loss": 2.5927,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4674949860587976,
+      "eval_runtime": 181.9649,
+      "eval_samples_per_second": 112.34,
+      "eval_steps_per_second": 14.047,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4772782859658564,
+      "grad_norm": 16.020864486694336,
+      "learning_rate": 2.819792723583128e-05,
+      "loss": 2.6464,
+      "step": 15100
+    },
+    {
+      "epoch": 1.4870615858729148,
+      "grad_norm": 16.674760818481445,
+      "learning_rate": 2.8016741556747354e-05,
+      "loss": 2.5853,
+      "step": 15200
+    },
+    {
+      "epoch": 1.4968448857799737,
+      "grad_norm": 16.890748977661133,
+      "learning_rate": 2.7835555877663432e-05,
+      "loss": 2.5748,
+      "step": 15300
+    },
+    {
+      "epoch": 1.5066281856870323,
+      "grad_norm": 20.217845916748047,
+      "learning_rate": 2.7654370198579504e-05,
+      "loss": 2.6204,
+      "step": 15400
+    },
+    {
+      "epoch": 1.516411485594091,
+      "grad_norm": 20.459087371826172,
+      "learning_rate": 2.7473184519495582e-05,
+      "loss": 2.6103,
+      "step": 15500
+    },
+    {
+      "epoch": 1.516411485594091,
+      "eval_runtime": 181.9454,
+      "eval_samples_per_second": 112.352,
+      "eval_steps_per_second": 14.048,
+      "step": 15500
+    },
+    {
+      "epoch": 1.5261947855011495,
+      "grad_norm": 18.207612991333008,
+      "learning_rate": 2.7291998840411654e-05,
+      "loss": 2.5786,
+      "step": 15600
+    },
+    {
+      "epoch": 1.5359780854082081,
+      "grad_norm": 18.084758758544922,
+      "learning_rate": 2.7110813161327732e-05,
+      "loss": 2.6535,
+      "step": 15700
+    },
+    {
+      "epoch": 1.545761385315267,
+      "grad_norm": 15.03881549835205,
+      "learning_rate": 2.6929627482243803e-05,
+      "loss": 2.6061,
+      "step": 15800
+    },
+    {
+      "epoch": 1.5555446852223254,
+      "grad_norm": 16.99995231628418,
+      "learning_rate": 2.6748441803159878e-05,
+      "loss": 2.6151,
+      "step": 15900
+    },
+    {
+      "epoch": 1.5653279851293842,
+      "grad_norm": 15.581089973449707,
+      "learning_rate": 2.6567256124075956e-05,
+      "loss": 2.6163,
+      "step": 16000
+    },
+    {
+      "epoch": 1.5653279851293842,
+      "eval_runtime": 181.8152,
+      "eval_samples_per_second": 112.433,
+      "eval_steps_per_second": 14.058,
+      "step": 16000
+    },
+    {
+      "epoch": 1.5751112850364428,
+      "grad_norm": 21.4382266998291,
+      "learning_rate": 2.6386070444992028e-05,
+      "loss": 2.5975,
+      "step": 16100
+    },
+    {
+      "epoch": 1.5848945849435014,
+      "grad_norm": 15.874536514282227,
+      "learning_rate": 2.6204884765908106e-05,
+      "loss": 2.5851,
+      "step": 16200
+    },
+    {
+      "epoch": 1.59467788485056,
+      "grad_norm": 17.902137756347656,
+      "learning_rate": 2.6023699086824177e-05,
+      "loss": 2.6027,
+      "step": 16300
+    },
+    {
+      "epoch": 1.6044611847576187,
+      "grad_norm": 17.04872703552246,
+      "learning_rate": 2.5842513407740255e-05,
+      "loss": 2.5854,
+      "step": 16400
+    },
+    {
+      "epoch": 1.6142444846646775,
+      "grad_norm": 15.406013488769531,
+      "learning_rate": 2.5661327728656327e-05,
+      "loss": 2.5158,
+      "step": 16500
+    },
+    {
+      "epoch": 1.6142444846646775,
+      "eval_runtime": 181.8647,
+      "eval_samples_per_second": 112.402,
+      "eval_steps_per_second": 14.054,
+      "step": 16500
+    },
+    {
+      "epoch": 1.624027784571736,
+      "grad_norm": 19.62627601623535,
+      "learning_rate": 2.5480142049572402e-05,
+      "loss": 2.5378,
+      "step": 16600
+    },
+    {
+      "epoch": 1.6338110844787948,
+      "grad_norm": 17.825178146362305,
+      "learning_rate": 2.529895637048848e-05,
+      "loss": 2.6162,
+      "step": 16700
+    },
+    {
+      "epoch": 1.6435943843858534,
+      "grad_norm": 15.442023277282715,
+      "learning_rate": 2.511777069140455e-05,
+      "loss": 2.5802,
+      "step": 16800
+    },
+    {
+      "epoch": 1.653377684292912,
+      "grad_norm": 18.695241928100586,
+      "learning_rate": 2.4936585012320626e-05,
+      "loss": 2.585,
+      "step": 16900
+    },
+    {
+      "epoch": 1.6631609841999706,
+      "grad_norm": 18.992969512939453,
+      "learning_rate": 2.4755399333236704e-05,
+      "loss": 2.5448,
+      "step": 17000
+    },
+    {
+      "epoch": 1.6631609841999706,
+      "eval_runtime": 181.91,
+      "eval_samples_per_second": 112.374,
+      "eval_steps_per_second": 14.051,
+      "step": 17000
+    },
+    {
+      "epoch": 1.6729442841070292,
+      "grad_norm": 19.065349578857422,
+      "learning_rate": 2.457421365415278e-05,
+      "loss": 2.6565,
+      "step": 17100
+    },
+    {
+      "epoch": 1.682727584014088,
+      "grad_norm": 20.110734939575195,
+      "learning_rate": 2.439302797506885e-05,
+      "loss": 2.5519,
+      "step": 17200
+    },
+    {
+      "epoch": 1.6925108839211465,
+      "grad_norm": 15.886931419372559,
+      "learning_rate": 2.4211842295984925e-05,
+      "loss": 2.5589,
+      "step": 17300
+    },
+    {
+      "epoch": 1.7022941838282053,
+      "grad_norm": 19.213207244873047,
+      "learning_rate": 2.4030656616901e-05,
+      "loss": 2.5714,
+      "step": 17400
+    },
+    {
+      "epoch": 1.712077483735264,
+      "grad_norm": 17.117481231689453,
+      "learning_rate": 2.3849470937817075e-05,
+      "loss": 2.6682,
+      "step": 17500
+    },
+    {
+      "epoch": 1.712077483735264,
+      "eval_runtime": 181.766,
+      "eval_samples_per_second": 112.463,
+      "eval_steps_per_second": 14.062,
+      "step": 17500
+    },
+    {
+      "epoch": 1.7218607836423225,
+      "grad_norm": 17.19162940979004,
+      "learning_rate": 2.366828525873315e-05,
+      "loss": 2.5591,
+      "step": 17600
+    },
+    {
+      "epoch": 1.7316440835493812,
+      "grad_norm": 15.454411506652832,
+      "learning_rate": 2.3487099579649225e-05,
+      "loss": 2.469,
+      "step": 17700
+    },
+    {
+      "epoch": 1.7414273834564398,
+      "grad_norm": 15.227791786193848,
+      "learning_rate": 2.3305913900565303e-05,
+      "loss": 2.664,
+      "step": 17800
+    },
+    {
+      "epoch": 1.7512106833634986,
+      "grad_norm": 18.5739688873291,
+      "learning_rate": 2.3124728221481374e-05,
+      "loss": 2.5991,
+      "step": 17900
+    },
+    {
+      "epoch": 1.760993983270557,
+      "grad_norm": 12.589066505432129,
+      "learning_rate": 2.294354254239745e-05,
+      "loss": 2.6593,
+      "step": 18000
+    },
+    {
+      "epoch": 1.760993983270557,
+      "eval_runtime": 181.9699,
+      "eval_samples_per_second": 112.337,
+      "eval_steps_per_second": 14.046,
+      "step": 18000
+    },
+    {
+      "epoch": 1.7707772831776158,
+      "grad_norm": 20.695772171020508,
+      "learning_rate": 2.2762356863313524e-05,
+      "loss": 2.5555,
+      "step": 18100
+    },
+    {
+      "epoch": 1.7805605830846745,
+      "grad_norm": 12.731703758239746,
+      "learning_rate": 2.25811711842296e-05,
+      "loss": 2.4617,
+      "step": 18200
+    },
+    {
+      "epoch": 1.790343882991733,
+      "grad_norm": 18.506074905395508,
+      "learning_rate": 2.2399985505145674e-05,
+      "loss": 2.6061,
+      "step": 18300
+    },
+    {
+      "epoch": 1.800127182898792,
+      "grad_norm": 14.8694486618042,
+      "learning_rate": 2.221879982606175e-05,
+      "loss": 2.5779,
+      "step": 18400
+    },
+    {
+      "epoch": 1.8099104828058503,
+      "grad_norm": 22.47985076904297,
+      "learning_rate": 2.2037614146977827e-05,
+      "loss": 2.5012,
+      "step": 18500
+    },
+    {
+      "epoch": 1.8099104828058503,
+      "eval_runtime": 182.3919,
+      "eval_samples_per_second": 112.077,
+      "eval_steps_per_second": 14.014,
+      "step": 18500
+    },
+    {
+      "epoch": 1.8196937827129092,
+      "grad_norm": 25.74334144592285,
+      "learning_rate": 2.1856428467893898e-05,
+      "loss": 2.5265,
+      "step": 18600
+    },
+    {
+      "epoch": 1.8294770826199676,
+      "grad_norm": 18.477630615234375,
+      "learning_rate": 2.1675242788809973e-05,
+      "loss": 2.5555,
+      "step": 18700
+    },
+    {
+      "epoch": 1.8392603825270264,
+      "grad_norm": 14.832316398620605,
+      "learning_rate": 2.1494057109726048e-05,
+      "loss": 2.4609,
+      "step": 18800
+    },
+    {
+      "epoch": 1.849043682434085,
+      "grad_norm": 17.025096893310547,
+      "learning_rate": 2.1312871430642123e-05,
+      "loss": 2.5119,
+      "step": 18900
+    },
+    {
+      "epoch": 1.8588269823411436,
+      "grad_norm": 16.852436065673828,
+      "learning_rate": 2.1131685751558197e-05,
+      "loss": 2.5369,
+      "step": 19000
+    },
+    {
+      "epoch": 1.8588269823411436,
+      "eval_runtime": 181.7443,
+      "eval_samples_per_second": 112.477,
+      "eval_steps_per_second": 14.064,
+      "step": 19000
+    },
+    {
+      "epoch": 1.8686102822482025,
+      "grad_norm": 15.160259246826172,
+      "learning_rate": 2.0950500072474272e-05,
+      "loss": 2.6297,
+      "step": 19100
+    },
+    {
+      "epoch": 1.8783935821552609,
+      "grad_norm": 15.909671783447266,
+      "learning_rate": 2.0769314393390347e-05,
+      "loss": 2.4696,
+      "step": 19200
+    },
+    {
+      "epoch": 1.8881768820623197,
+      "grad_norm": 14.201844215393066,
+      "learning_rate": 2.0588128714306422e-05,
+      "loss": 2.5653,
+      "step": 19300
+    },
+    {
+      "epoch": 1.8979601819693783,
+      "grad_norm": 16.351415634155273,
+      "learning_rate": 2.0406943035222497e-05,
+      "loss": 2.4962,
+      "step": 19400
+    },
+    {
+      "epoch": 1.907743481876437,
+      "grad_norm": 16.943771362304688,
+      "learning_rate": 2.022575735613857e-05,
+      "loss": 2.5091,
+      "step": 19500
+    },
+    {
+      "epoch": 1.907743481876437,
+      "eval_runtime": 181.6486,
+      "eval_samples_per_second": 112.536,
+      "eval_steps_per_second": 14.071,
+      "step": 19500
+    },
+    {
+      "epoch": 1.9175267817834956,
+      "grad_norm": 15.006349563598633,
+      "learning_rate": 2.0044571677054646e-05,
+      "loss": 2.5214,
+      "step": 19600
+    },
+    {
+      "epoch": 1.9273100816905542,
+      "grad_norm": 17.305580139160156,
+      "learning_rate": 1.986338599797072e-05,
+      "loss": 2.4989,
+      "step": 19700
+    },
+    {
+      "epoch": 1.937093381597613,
+      "grad_norm": 17.28044891357422,
+      "learning_rate": 1.9682200318886796e-05,
+      "loss": 2.4008,
+      "step": 19800
+    },
+    {
+      "epoch": 1.9468766815046714,
+      "grad_norm": 18.25079917907715,
+      "learning_rate": 1.950101463980287e-05,
+      "loss": 2.6015,
+      "step": 19900
+    },
+    {
+      "epoch": 1.9566599814117303,
+      "grad_norm": 20.741668701171875,
+      "learning_rate": 1.9319828960718946e-05,
+      "loss": 2.4081,
+      "step": 20000
+    },
+    {
+      "epoch": 1.9566599814117303,
+      "eval_runtime": 181.7745,
+      "eval_samples_per_second": 112.458,
+      "eval_steps_per_second": 14.061,
+      "step": 20000
+    },
+    {
+      "epoch": 1.9664432813187889,
+      "grad_norm": 16.1226863861084,
+      "learning_rate": 1.913864328163502e-05,
+      "loss": 2.5418,
+      "step": 20100
+    },
+    {
+      "epoch": 1.9762265812258475,
+      "grad_norm": 13.914982795715332,
+      "learning_rate": 1.8957457602551095e-05,
+      "loss": 2.5248,
+      "step": 20200
+    },
+    {
+      "epoch": 1.986009881132906,
+      "grad_norm": 15.072690963745117,
+      "learning_rate": 1.877627192346717e-05,
+      "loss": 2.5488,
+      "step": 20300
+    },
+    {
+      "epoch": 1.9957931810399647,
+      "grad_norm": 15.510763168334961,
+      "learning_rate": 1.8595086244383245e-05,
+      "loss": 2.4605,
+      "step": 20400
+    },
+    {
+      "epoch": 2.0055764809470236,
+      "grad_norm": 18.463842391967773,
+      "learning_rate": 1.841390056529932e-05,
+      "loss": 2.522,
+      "step": 20500
+    },
+    {
+      "epoch": 2.0055764809470236,
+      "eval_runtime": 182.07,
+      "eval_samples_per_second": 112.276,
+      "eval_steps_per_second": 14.039,
+      "step": 20500
+    },
+    {
+      "epoch": 2.015359780854082,
+      "grad_norm": 16.670269012451172,
+      "learning_rate": 1.8232714886215394e-05,
+      "loss": 2.5585,
+      "step": 20600
+    },
+    {
+      "epoch": 2.025143080761141,
+      "grad_norm": 20.60368537902832,
+      "learning_rate": 1.805152920713147e-05,
+      "loss": 2.5381,
+      "step": 20700
+    },
+    {
+      "epoch": 2.034926380668199,
+      "grad_norm": 15.686981201171875,
+      "learning_rate": 1.7870343528047544e-05,
+      "loss": 2.5721,
+      "step": 20800
+    },
+    {
+      "epoch": 2.044709680575258,
+      "grad_norm": 14.691718101501465,
+      "learning_rate": 1.768915784896362e-05,
+      "loss": 2.5187,
+      "step": 20900
+    },
+    {
+      "epoch": 2.054492980482317,
+      "grad_norm": 16.31734848022461,
+      "learning_rate": 1.7507972169879694e-05,
+      "loss": 2.5202,
+      "step": 21000
+    },
+    {
+      "epoch": 2.054492980482317,
+      "eval_runtime": 181.9896,
+      "eval_samples_per_second": 112.325,
+      "eval_steps_per_second": 14.045,
+      "step": 21000
+    },
+    {
+      "epoch": 2.0642762803893753,
+      "grad_norm": 12.698554992675781,
+      "learning_rate": 1.732678649079577e-05,
+      "loss": 2.4228,
+      "step": 21100
+    },
+    {
+      "epoch": 2.074059580296434,
+      "grad_norm": 16.34201431274414,
+      "learning_rate": 1.7145600811711843e-05,
+      "loss": 2.3963,
+      "step": 21200
+    },
+    {
+      "epoch": 2.0838428802034925,
+      "grad_norm": 16.52840232849121,
+      "learning_rate": 1.6964415132627918e-05,
+      "loss": 2.4759,
+      "step": 21300
+    },
+    {
+      "epoch": 2.0936261801105513,
+      "grad_norm": 14.856452941894531,
+      "learning_rate": 1.6783229453543993e-05,
+      "loss": 2.4675,
+      "step": 21400
+    },
+    {
+      "epoch": 2.1034094800176097,
+      "grad_norm": 19.68895721435547,
+      "learning_rate": 1.6602043774460068e-05,
+      "loss": 2.5324,
+      "step": 21500
+    },
+    {
+      "epoch": 2.1034094800176097,
+      "eval_runtime": 182.1877,
+      "eval_samples_per_second": 112.203,
+      "eval_steps_per_second": 14.029,
+      "step": 21500
+    },
+    {
+      "epoch": 2.1131927799246686,
+      "grad_norm": 23.248056411743164,
+      "learning_rate": 1.6420858095376143e-05,
+      "loss": 2.5231,
+      "step": 21600
+    },
+    {
+      "epoch": 2.1229760798317274,
+      "grad_norm": 25.471004486083984,
+      "learning_rate": 1.6239672416292217e-05,
+      "loss": 2.5871,
+      "step": 21700
+    },
+    {
+      "epoch": 2.132759379738786,
+      "grad_norm": 17.794851303100586,
+      "learning_rate": 1.6058486737208292e-05,
+      "loss": 2.5008,
+      "step": 21800
+    },
+    {
+      "epoch": 2.1425426796458447,
+      "grad_norm": 15.450346946716309,
+      "learning_rate": 1.5877301058124367e-05,
+      "loss": 2.4194,
+      "step": 21900
+    },
+    {
+      "epoch": 2.152325979552903,
+      "grad_norm": 13.243645668029785,
+      "learning_rate": 1.5696115379040442e-05,
+      "loss": 2.5018,
+      "step": 22000
+    },
+    {
+      "epoch": 2.152325979552903,
+      "eval_runtime": 181.9841,
+      "eval_samples_per_second": 112.328,
+      "eval_steps_per_second": 14.045,
+      "step": 22000
+    },
+    {
+      "epoch": 2.162109279459962,
+      "grad_norm": 16.996198654174805,
+      "learning_rate": 1.5514929699956517e-05,
+      "loss": 2.4492,
+      "step": 22100
+    },
+    {
+      "epoch": 2.1718925793670203,
+      "grad_norm": 20.05558967590332,
+      "learning_rate": 1.5333744020872588e-05,
+      "loss": 2.489,
+      "step": 22200
+    },
+    {
+      "epoch": 2.181675879274079,
+      "grad_norm": 15.66326904296875,
+      "learning_rate": 1.5152558341788666e-05,
+      "loss": 2.5089,
+      "step": 22300
+    },
+    {
+      "epoch": 2.191459179181138,
+      "grad_norm": 17.83564567565918,
+      "learning_rate": 1.4971372662704741e-05,
+      "loss": 2.4945,
+      "step": 22400
+    },
+    {
+      "epoch": 2.2012424790881964,
+      "grad_norm": 21.466899871826172,
+      "learning_rate": 1.4790186983620816e-05,
+      "loss": 2.5467,
+      "step": 22500
+    },
+    {
+      "epoch": 2.2012424790881964,
+      "eval_runtime": 182.8328,
+      "eval_samples_per_second": 111.807,
+      "eval_steps_per_second": 13.98,
+      "step": 22500
+    },
+    {
+      "epoch": 2.211025778995255,
+      "grad_norm": 17.91064453125,
+      "learning_rate": 1.4609001304536891e-05,
+      "loss": 2.5144,
+      "step": 22600
+    },
+    {
+      "epoch": 2.2208090789023136,
+      "grad_norm": 17.678396224975586,
+      "learning_rate": 1.4427815625452964e-05,
+      "loss": 2.5018,
+      "step": 22700
+    },
+    {
+      "epoch": 2.2305923788093724,
+      "grad_norm": 17.510461807250977,
+      "learning_rate": 1.4246629946369039e-05,
+      "loss": 2.4228,
+      "step": 22800
+    },
+    {
+      "epoch": 2.240375678716431,
+      "grad_norm": 24.923967361450195,
+      "learning_rate": 1.4065444267285114e-05,
+      "loss": 2.5249,
+      "step": 22900
+    },
+    {
+      "epoch": 2.2501589786234897,
+      "grad_norm": 17.82384490966797,
+      "learning_rate": 1.388425858820119e-05,
+      "loss": 2.4282,
+      "step": 23000
+    },
+    {
+      "epoch": 2.2501589786234897,
+      "eval_runtime": 182.0459,
+      "eval_samples_per_second": 112.29,
+      "eval_steps_per_second": 14.04,
+      "step": 23000
+    },
+    {
+      "epoch": 2.2599422785305485,
+      "grad_norm": 16.13028335571289,
+      "learning_rate": 1.3703072909117265e-05,
+      "loss": 2.4472,
+      "step": 23100
+    },
+    {
+      "epoch": 2.269725578437607,
+      "grad_norm": 15.137242317199707,
+      "learning_rate": 1.352188723003334e-05,
+      "loss": 2.5985,
+      "step": 23200
+    },
+    {
+      "epoch": 2.2795088783446658,
+      "grad_norm": 16.187530517578125,
+      "learning_rate": 1.3340701550949415e-05,
+      "loss": 2.4862,
+      "step": 23300
+    },
+    {
+      "epoch": 2.289292178251724,
+      "grad_norm": 18.84433937072754,
+      "learning_rate": 1.3159515871865488e-05,
+      "loss": 2.516,
+      "step": 23400
+    },
+    {
+      "epoch": 2.299075478158783,
+      "grad_norm": 20.209121704101562,
+      "learning_rate": 1.2978330192781563e-05,
+      "loss": 2.5031,
+      "step": 23500
+    },
+    {
+      "epoch": 2.299075478158783,
+      "eval_runtime": 181.9806,
+      "eval_samples_per_second": 112.331,
+      "eval_steps_per_second": 14.045,
+      "step": 23500
+    },
+    {
+      "epoch": 2.308858778065842,
+      "grad_norm": 67.4502182006836,
+      "learning_rate": 1.2797144513697637e-05,
+      "loss": 2.4491,
+      "step": 23600
+    },
+    {
+      "epoch": 2.3186420779729002,
+      "grad_norm": 14.940401077270508,
+      "learning_rate": 1.2615958834613712e-05,
+      "loss": 2.5669,
+      "step": 23700
+    },
+    {
+      "epoch": 2.328425377879959,
+      "grad_norm": 16.591793060302734,
+      "learning_rate": 1.2434773155529787e-05,
+      "loss": 2.4565,
+      "step": 23800
+    },
+    {
+      "epoch": 2.3382086777870175,
+      "grad_norm": 16.798791885375977,
+      "learning_rate": 1.2253587476445862e-05,
+      "loss": 2.4046,
+      "step": 23900
+    },
+    {
+      "epoch": 2.3479919776940763,
+      "grad_norm": 17.712255477905273,
+      "learning_rate": 1.2072401797361937e-05,
+      "loss": 2.4453,
+      "step": 24000
+    },
+    {
+      "epoch": 2.3479919776940763,
+      "eval_runtime": 182.0401,
+      "eval_samples_per_second": 112.294,
+      "eval_steps_per_second": 14.041,
+      "step": 24000
+    },
+    {
+      "epoch": 2.3577752776011347,
+      "grad_norm": 18.64284324645996,
+      "learning_rate": 1.1891216118278011e-05,
+      "loss": 2.3973,
+      "step": 24100
+    },
+    {
+      "epoch": 2.3675585775081935,
+      "grad_norm": 18.185895919799805,
+      "learning_rate": 1.1710030439194086e-05,
+      "loss": 2.5045,
+      "step": 24200
+    },
+    {
+      "epoch": 2.377341877415252,
+      "grad_norm": 23.201522827148438,
+      "learning_rate": 1.1528844760110163e-05,
+      "loss": 2.5402,
+      "step": 24300
+    },
+    {
+      "epoch": 2.3871251773223108,
+      "grad_norm": 21.606412887573242,
+      "learning_rate": 1.1347659081026236e-05,
+      "loss": 2.4285,
+      "step": 24400
+    },
+    {
+      "epoch": 2.3969084772293696,
+      "grad_norm": 16.318761825561523,
+      "learning_rate": 1.116647340194231e-05,
+      "loss": 2.5509,
+      "step": 24500
+    },
+    {
+      "epoch": 2.3969084772293696,
+      "eval_runtime": 182.0431,
+      "eval_samples_per_second": 112.292,
+      "eval_steps_per_second": 14.041,
+      "step": 24500
+    },
+    {
+      "epoch": 2.406691777136428,
+      "grad_norm": 17.779014587402344,
+      "learning_rate": 1.0985287722858386e-05,
+      "loss": 2.4245,
+      "step": 24600
+    },
+    {
+      "epoch": 2.416475077043487,
+      "grad_norm": 18.44321060180664,
+      "learning_rate": 1.080410204377446e-05,
+      "loss": 2.5223,
+      "step": 24700
+    },
+    {
+      "epoch": 2.4262583769505452,
+      "grad_norm": 24.017047882080078,
+      "learning_rate": 1.0622916364690535e-05,
+      "loss": 2.4846,
+      "step": 24800
+    },
+    {
+      "epoch": 2.436041676857604,
+      "grad_norm": 14.89560604095459,
+      "learning_rate": 1.044173068560661e-05,
+      "loss": 2.5922,
+      "step": 24900
+    },
+    {
+      "epoch": 2.445824976764663,
+      "grad_norm": 15.532561302185059,
+      "learning_rate": 1.0260545006522685e-05,
+      "loss": 2.3976,
+      "step": 25000
+    },
+    {
+      "epoch": 2.445824976764663,
+      "eval_runtime": 182.1033,
+      "eval_samples_per_second": 112.255,
+      "eval_steps_per_second": 14.036,
+      "step": 25000
+    },
+    {
+      "epoch": 2.4556082766717213,
+      "grad_norm": 18.041282653808594,
+      "learning_rate": 1.007935932743876e-05,
+      "loss": 2.4731,
+      "step": 25100
+    },
+    {
+      "epoch": 2.46539157657878,
+      "grad_norm": 13.40858268737793,
+      "learning_rate": 9.898173648354834e-06,
+      "loss": 2.4838,
+      "step": 25200
+    },
+    {
+      "epoch": 2.4751748764858386,
+      "grad_norm": 17.450841903686523,
+      "learning_rate": 9.71698796927091e-06,
+      "loss": 2.3999,
+      "step": 25300
+    },
+    {
+      "epoch": 2.4849581763928974,
+      "grad_norm": 17.556467056274414,
+      "learning_rate": 9.535802290186984e-06,
+      "loss": 2.3867,
+      "step": 25400
+    },
+    {
+      "epoch": 2.494741476299956,
+      "grad_norm": 18.578310012817383,
+      "learning_rate": 9.354616611103059e-06,
+      "loss": 2.4546,
+      "step": 25500
+    },
+    {
+      "epoch": 2.494741476299956,
+      "eval_runtime": 182.0338,
+      "eval_samples_per_second": 112.298,
+      "eval_steps_per_second": 14.041,
+      "step": 25500
+    },
+    {
+      "epoch": 2.5045247762070146,
+      "grad_norm": 14.936469078063965,
+      "learning_rate": 9.173430932019134e-06,
+      "loss": 2.5562,
+      "step": 25600
+    },
+    {
+      "epoch": 2.514308076114073,
+      "grad_norm": 17.527040481567383,
+      "learning_rate": 8.992245252935209e-06,
+      "loss": 2.4008,
+      "step": 25700
+    },
+    {
+      "epoch": 2.524091376021132,
+      "grad_norm": 12.91336727142334,
+      "learning_rate": 8.811059573851283e-06,
+      "loss": 2.4655,
+      "step": 25800
+    },
+    {
+      "epoch": 2.5338746759281907,
+      "grad_norm": 15.168461799621582,
+      "learning_rate": 8.629873894767358e-06,
+      "loss": 2.4468,
+      "step": 25900
+    },
+    {
+      "epoch": 2.543657975835249,
+      "grad_norm": 17.5390682220459,
+      "learning_rate": 8.448688215683433e-06,
+      "loss": 2.4836,
+      "step": 26000
+    },
+    {
+      "epoch": 2.543657975835249,
+      "eval_runtime": 182.1148,
+      "eval_samples_per_second": 112.248,
+      "eval_steps_per_second": 14.035,
+      "step": 26000
+    },
+    {
+      "epoch": 2.553441275742308,
+      "grad_norm": 15.126510620117188,
+      "learning_rate": 8.267502536599508e-06,
+      "loss": 2.387,
+      "step": 26100
+    },
+    {
+      "epoch": 2.5632245756493663,
+      "grad_norm": 15.374293327331543,
+      "learning_rate": 8.086316857515583e-06,
+      "loss": 2.3652,
+      "step": 26200
+    },
+    {
+      "epoch": 2.573007875556425,
+      "grad_norm": 15.498108863830566,
+      "learning_rate": 7.905131178431657e-06,
+      "loss": 2.4749,
+      "step": 26300
+    },
+    {
+      "epoch": 2.582791175463484,
+      "grad_norm": 16.221315383911133,
+      "learning_rate": 7.723945499347732e-06,
+      "loss": 2.4567,
+      "step": 26400
+    },
+    {
+      "epoch": 2.5925744753705424,
+      "grad_norm": 18.839122772216797,
+      "learning_rate": 7.542759820263806e-06,
+      "loss": 2.3554,
+      "step": 26500
+    },
+    {
+      "epoch": 2.5925744753705424,
+      "eval_runtime": 181.9597,
+      "eval_samples_per_second": 112.344,
+      "eval_steps_per_second": 14.047,
+      "step": 26500
+    },
+    {
+      "epoch": 2.6023577752776013,
+      "grad_norm": 22.626708984375,
+      "learning_rate": 7.361574141179882e-06,
+      "loss": 2.502,
+      "step": 26600
+    },
+    {
+      "epoch": 2.6121410751846597,
+      "grad_norm": 16.519880294799805,
+      "learning_rate": 7.180388462095957e-06,
+      "loss": 2.5034,
+      "step": 26700
+    },
+    {
+      "epoch": 2.6219243750917185,
+      "grad_norm": 27.421489715576172,
+      "learning_rate": 6.999202783012031e-06,
+      "loss": 2.5276,
+      "step": 26800
+    },
+    {
+      "epoch": 2.6317076749987773,
+      "grad_norm": 15.274630546569824,
+      "learning_rate": 6.8180171039281055e-06,
+      "loss": 2.4121,
+      "step": 26900
+    },
+    {
+      "epoch": 2.6414909749058357,
+      "grad_norm": 15.751582145690918,
+      "learning_rate": 6.636831424844181e-06,
+      "loss": 2.5799,
+      "step": 27000
+    },
+    {
+      "epoch": 2.6414909749058357,
+      "eval_runtime": 182.0873,
+      "eval_samples_per_second": 112.265,
+      "eval_steps_per_second": 14.037,
+      "step": 27000
+    },
+    {
+      "epoch": 2.651274274812894,
+      "grad_norm": 16.674850463867188,
+      "learning_rate": 6.455645745760255e-06,
+      "loss": 2.3872,
+      "step": 27100
+    },
+    {
+      "epoch": 2.661057574719953,
+      "grad_norm": 12.62803840637207,
+      "learning_rate": 6.27446006667633e-06,
+      "loss": 2.4,
+      "step": 27200
+    },
+    {
+      "epoch": 2.670840874627012,
+      "grad_norm": 18.055158615112305,
+      "learning_rate": 6.093274387592405e-06,
+      "loss": 2.4681,
+      "step": 27300
+    },
+    {
+      "epoch": 2.68062417453407,
+      "grad_norm": 17.21278190612793,
+      "learning_rate": 5.91208870850848e-06,
+      "loss": 2.5441,
+      "step": 27400
+    },
+    {
+      "epoch": 2.690407474441129,
+      "grad_norm": 20.945236206054688,
+      "learning_rate": 5.7309030294245544e-06,
+      "loss": 2.4388,
+      "step": 27500
+    },
+    {
+      "epoch": 2.690407474441129,
+      "eval_runtime": 182.1279,
+      "eval_samples_per_second": 112.24,
+      "eval_steps_per_second": 14.034,
+      "step": 27500
+    },
+    {
+      "epoch": 2.7001907743481874,
+      "grad_norm": 23.483661651611328,
+      "learning_rate": 5.549717350340629e-06,
+      "loss": 2.4589,
+      "step": 27600
+    },
+    {
+      "epoch": 2.7099740742552463,
+      "grad_norm": 17.954036712646484,
+      "learning_rate": 5.368531671256704e-06,
+      "loss": 2.4477,
+      "step": 27700
+    },
+    {
+      "epoch": 2.719757374162305,
+      "grad_norm": 16.187314987182617,
+      "learning_rate": 5.187345992172779e-06,
+      "loss": 2.4967,
+      "step": 27800
+    },
+    {
+      "epoch": 2.7295406740693635,
+      "grad_norm": 14.324910163879395,
+      "learning_rate": 5.006160313088854e-06,
+      "loss": 2.3921,
+      "step": 27900
+    },
+    {
+      "epoch": 2.7393239739764224,
+      "grad_norm": 20.81557846069336,
+      "learning_rate": 4.8249746340049285e-06,
+      "loss": 2.5201,
+      "step": 28000
+    },
+    {
+      "epoch": 2.7393239739764224,
+      "eval_runtime": 182.146,
+      "eval_samples_per_second": 112.229,
+      "eval_steps_per_second": 14.033,
+      "step": 28000
+    },
+    {
+      "epoch": 2.7491072738834808,
+      "grad_norm": 18.682844161987305,
+      "learning_rate": 4.643788954921003e-06,
+      "loss": 2.4325,
+      "step": 28100
+    },
+    {
+      "epoch": 2.7588905737905396,
+      "grad_norm": 16.227272033691406,
+      "learning_rate": 4.462603275837078e-06,
+      "loss": 2.3864,
+      "step": 28200
+    },
+    {
+      "epoch": 2.7686738736975984,
+      "grad_norm": 16.20302963256836,
+      "learning_rate": 4.281417596753152e-06,
+      "loss": 2.5296,
+      "step": 28300
+    },
+    {
+      "epoch": 2.778457173604657,
+      "grad_norm": 18.634096145629883,
+      "learning_rate": 4.100231917669228e-06,
+      "loss": 2.4514,
+      "step": 28400
+    },
+    {
+      "epoch": 2.7882404735117157,
+      "grad_norm": 13.040008544921875,
+      "learning_rate": 3.919046238585303e-06,
+      "loss": 2.3661,
+      "step": 28500
+    },
+    {
+      "epoch": 2.7882404735117157,
+      "eval_runtime": 181.9164,
+      "eval_samples_per_second": 112.37,
+      "eval_steps_per_second": 14.05,
+      "step": 28500
+    },
+    {
+      "epoch": 2.798023773418774,
+      "grad_norm": 14.142943382263184,
+      "learning_rate": 3.737860559501377e-06,
+      "loss": 2.5074,
+      "step": 28600
+    },
+    {
+      "epoch": 2.807807073325833,
+      "grad_norm": 17.934324264526367,
+      "learning_rate": 3.5566748804174523e-06,
+      "loss": 2.4224,
+      "step": 28700
+    },
+    {
+      "epoch": 2.8175903732328913,
+      "grad_norm": 14.450194358825684,
+      "learning_rate": 3.3754892013335267e-06,
+      "loss": 2.4949,
+      "step": 28800
+    },
+    {
+      "epoch": 2.82737367313995,
+      "grad_norm": 17.746837615966797,
+      "learning_rate": 3.194303522249602e-06,
+      "loss": 2.4153,
+      "step": 28900
+    },
+    {
+      "epoch": 2.8371569730470085,
+      "grad_norm": 13.962541580200195,
+      "learning_rate": 3.0131178431656763e-06,
+      "loss": 2.4804,
+      "step": 29000
+    },
+    {
+      "epoch": 2.8371569730470085,
+      "eval_runtime": 182.0262,
+      "eval_samples_per_second": 112.303,
+      "eval_steps_per_second": 14.042,
+      "step": 29000
+    },
+    {
+      "epoch": 2.8469402729540674,
+      "grad_norm": 16.669286727905273,
+      "learning_rate": 2.831932164081751e-06,
+      "loss": 2.5397,
+      "step": 29100
+    },
+    {
+      "epoch": 2.856723572861126,
+      "grad_norm": 15.421733856201172,
+      "learning_rate": 2.650746484997826e-06,
+      "loss": 2.4175,
+      "step": 29200
+    },
+    {
+      "epoch": 2.8665068727681846,
+      "grad_norm": 14.135702133178711,
+      "learning_rate": 2.4695608059139007e-06,
+      "loss": 2.5069,
+      "step": 29300
+    },
+    {
+      "epoch": 2.8762901726752435,
+      "grad_norm": 17.41412925720215,
+      "learning_rate": 2.2883751268299756e-06,
+      "loss": 2.3997,
+      "step": 29400
+    },
+    {
+      "epoch": 2.886073472582302,
+      "grad_norm": 14.824533462524414,
+      "learning_rate": 2.1071894477460504e-06,
+      "loss": 2.3945,
+      "step": 29500
+    },
+    {
+      "epoch": 2.886073472582302,
+      "eval_runtime": 181.9299,
+      "eval_samples_per_second": 112.362,
+      "eval_steps_per_second": 14.049,
+      "step": 29500
+    },
+    {
+      "epoch": 2.8958567724893607,
+      "grad_norm": 27.31865119934082,
+      "learning_rate": 1.926003768662125e-06,
+      "loss": 2.45,
+      "step": 29600
+    },
+    {
+      "epoch": 2.9056400723964195,
+      "grad_norm": 18.966655731201172,
+      "learning_rate": 1.7448180895781998e-06,
+      "loss": 2.3916,
+      "step": 29700
+    },
+    {
+      "epoch": 2.915423372303478,
+      "grad_norm": 18.538440704345703,
+      "learning_rate": 1.5636324104942746e-06,
+      "loss": 2.4625,
+      "step": 29800
+    },
+    {
+      "epoch": 2.9252066722105368,
+      "grad_norm": 21.757272720336914,
+      "learning_rate": 1.3824467314103494e-06,
+      "loss": 2.3722,
+      "step": 29900
+    },
+    {
+      "epoch": 2.934989972117595,
+      "grad_norm": 16.907358169555664,
+      "learning_rate": 1.201261052326424e-06,
+      "loss": 2.464,
+      "step": 30000
+    },
+    {
+      "epoch": 2.934989972117595,
+      "eval_runtime": 181.9148,
+      "eval_samples_per_second": 112.371,
+      "eval_steps_per_second": 14.051,
+      "step": 30000
+    },
+    {
+      "epoch": 2.944773272024654,
+      "grad_norm": 13.88399600982666,
+      "learning_rate": 1.0200753732424989e-06,
+      "loss": 2.5005,
+      "step": 30100
+    },
+    {
+      "epoch": 2.954556571931713,
+      "grad_norm": 19.77507781982422,
+      "learning_rate": 8.388896941585737e-07,
+      "loss": 2.3829,
+      "step": 30200
+    },
+    {
+      "epoch": 2.9643398718387712,
+      "grad_norm": 16.535932540893555,
+      "learning_rate": 6.577040150746485e-07,
+      "loss": 2.4788,
+      "step": 30300
+    },
+    {
+      "epoch": 2.9741231717458296,
+      "grad_norm": 15.027000427246094,
+      "learning_rate": 4.765183359907233e-07,
+      "loss": 2.5007,
+      "step": 30400
+    },
+    {
+      "epoch": 2.9839064716528885,
+      "grad_norm": 14.9392671585083,
+      "learning_rate": 2.953326569067981e-07,
+      "loss": 2.4847,
+      "step": 30500
+    },
+    {
+      "epoch": 2.9839064716528885,
+      "eval_runtime": 181.9853,
+      "eval_samples_per_second": 112.328,
+      "eval_steps_per_second": 14.045,
+      "step": 30500
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 30663,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.0644527086729788e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

train_data/muril_ch_domain/checkpoint-30500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c39b3134e7e5628432a425a5873f74f59d631bf591a12adab52f3ba906ae6906
+size 5304

train_data/muril_ch_domain/checkpoint-30663/README.md

@@ -0,0 +1,202 @@
+---
+base_model: google/muril-base-cased
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.13.2

train_data/muril_ch_domain/checkpoint-30663/adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": {
+    "base_model_class": "BertForMaskedLM",
+    "parent_library": "transformers.models.bert.modeling_bert"
+  },
+  "base_model_name_or_path": "google/muril-base-cased",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "key",
+    "query",
+    "value",
+    "dense"
+  ],
+  "task_type": null,
+  "use_dora": false,
+  "use_rslora": false
+}

train_data/muril_ch_domain/checkpoint-30663/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d84ccb91fb20f58aa9778952c060eea2273c69c0e40d164972562ff8c0ead9d
+size 42881168

train_data/muril_ch_domain/checkpoint-30663/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ed5ec095ef64893bcddcfeca72b34c0cfd61506702bf2fcf85933f888ec0e1d
+size 85843898

train_data/muril_ch_domain/checkpoint-30663/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c845563310a221ad0236804a9577e46a5928023e42425fd9e7226668ec75f0f
+size 14244

train_data/muril_ch_domain/checkpoint-30663/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5cbeb707e99493c97d2cffe0b096a573843cb8f847ed6fbc608d5da54abe9076
+size 1064

train_data/muril_ch_domain/checkpoint-30663/trainer_state.json

@@ -0,0 +1,2602 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.999853250501394,
+  "eval_steps": 500,
+  "global_step": 30663,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.009783299907058651,
+      "grad_norm": 33.780609130859375,
+      "learning_rate": 1.6302575806977503e-06,
+      "loss": 6.574,
+      "step": 100
+    },
+    {
+      "epoch": 0.019566599814117302,
+      "grad_norm": 33.439022064208984,
+      "learning_rate": 3.2605151613955006e-06,
+      "loss": 6.1653,
+      "step": 200
+    },
+    {
+      "epoch": 0.02934989972117595,
+      "grad_norm": 27.25751304626465,
+      "learning_rate": 4.890772742093251e-06,
+      "loss": 5.5515,
+      "step": 300
+    },
+    {
+      "epoch": 0.039133199628234604,
+      "grad_norm": 38.36979675292969,
+      "learning_rate": 6.521030322791001e-06,
+      "loss": 5.0531,
+      "step": 400
+    },
+    {
+      "epoch": 0.048916499535293256,
+      "grad_norm": 29.350488662719727,
+      "learning_rate": 8.15128790348875e-06,
+      "loss": 4.9225,
+      "step": 500
+    },
+    {
+      "epoch": 0.048916499535293256,
+      "eval_runtime": 181.5812,
+      "eval_samples_per_second": 112.578,
+      "eval_steps_per_second": 14.076,
+      "step": 500
+    },
+    {
+      "epoch": 0.0586997994423519,
+      "grad_norm": 33.02122497558594,
+      "learning_rate": 9.781545484186502e-06,
+      "loss": 4.8186,
+      "step": 600
+    },
+    {
+      "epoch": 0.06848309934941056,
+      "grad_norm": 42.41593933105469,
+      "learning_rate": 1.1411803064884251e-05,
+      "loss": 4.5769,
+      "step": 700
+    },
+    {
+      "epoch": 0.07826639925646921,
+      "grad_norm": 40.29044723510742,
+      "learning_rate": 1.3042060645582003e-05,
+      "loss": 4.3963,
+      "step": 800
+    },
+    {
+      "epoch": 0.08804969916352785,
+      "grad_norm": 38.0811653137207,
+      "learning_rate": 1.4672318226279752e-05,
+      "loss": 4.3393,
+      "step": 900
+    },
+    {
+      "epoch": 0.09783299907058651,
+      "grad_norm": 36.08370590209961,
+      "learning_rate": 1.63025758069775e-05,
+      "loss": 4.2421,
+      "step": 1000
+    },
+    {
+      "epoch": 0.09783299907058651,
+      "eval_runtime": 181.886,
+      "eval_samples_per_second": 112.389,
+      "eval_steps_per_second": 14.053,
+      "step": 1000
+    },
+    {
+      "epoch": 0.10761629897764516,
+      "grad_norm": 37.253684997558594,
+      "learning_rate": 1.7932833387675256e-05,
+      "loss": 4.1156,
+      "step": 1100
+    },
+    {
+      "epoch": 0.1173995988847038,
+      "grad_norm": 33.003475189208984,
+      "learning_rate": 1.9563090968373004e-05,
+      "loss": 4.0112,
+      "step": 1200
+    },
+    {
+      "epoch": 0.12718289879176245,
+      "grad_norm": 30.727867126464844,
+      "learning_rate": 2.1193348549070755e-05,
+      "loss": 3.9969,
+      "step": 1300
+    },
+    {
+      "epoch": 0.13696619869882112,
+      "grad_norm": 37.471092224121094,
+      "learning_rate": 2.2823606129768503e-05,
+      "loss": 3.874,
+      "step": 1400
+    },
+    {
+      "epoch": 0.14674949860587977,
+      "grad_norm": 42.32167434692383,
+      "learning_rate": 2.4453863710466254e-05,
+      "loss": 3.8518,
+      "step": 1500
+    },
+    {
+      "epoch": 0.14674949860587977,
+      "eval_runtime": 181.9332,
+      "eval_samples_per_second": 112.36,
+      "eval_steps_per_second": 14.049,
+      "step": 1500
+    },
+    {
+      "epoch": 0.15653279851293841,
+      "grad_norm": 38.00124740600586,
+      "learning_rate": 2.6084121291164005e-05,
+      "loss": 3.918,
+      "step": 1600
+    },
+    {
+      "epoch": 0.16631609841999706,
+      "grad_norm": 44.637386322021484,
+      "learning_rate": 2.7714378871861756e-05,
+      "loss": 3.9134,
+      "step": 1700
+    },
+    {
+      "epoch": 0.1760993983270557,
+      "grad_norm": 49.578609466552734,
+      "learning_rate": 2.9344636452559504e-05,
+      "loss": 3.7507,
+      "step": 1800
+    },
+    {
+      "epoch": 0.18588269823411438,
+      "grad_norm": 36.65715789794922,
+      "learning_rate": 3.0974894033257255e-05,
+      "loss": 3.7551,
+      "step": 1900
+    },
+    {
+      "epoch": 0.19566599814117303,
+      "grad_norm": 36.873443603515625,
+      "learning_rate": 3.2605151613955e-05,
+      "loss": 3.6951,
+      "step": 2000
+    },
+    {
+      "epoch": 0.19566599814117303,
+      "eval_runtime": 181.8273,
+      "eval_samples_per_second": 112.425,
+      "eval_steps_per_second": 14.057,
+      "step": 2000
+    },
+    {
+      "epoch": 0.20544929804823167,
+      "grad_norm": 33.025413513183594,
+      "learning_rate": 3.423540919465276e-05,
+      "loss": 3.6603,
+      "step": 2100
+    },
+    {
+      "epoch": 0.21523259795529032,
+      "grad_norm": 30.105051040649414,
+      "learning_rate": 3.586566677535051e-05,
+      "loss": 3.525,
+      "step": 2200
+    },
+    {
+      "epoch": 0.22501589786234896,
+      "grad_norm": 34.5129280090332,
+      "learning_rate": 3.749592435604825e-05,
+      "loss": 3.6454,
+      "step": 2300
+    },
+    {
+      "epoch": 0.2347991977694076,
+      "grad_norm": 33.16934585571289,
+      "learning_rate": 3.912618193674601e-05,
+      "loss": 3.6356,
+      "step": 2400
+    },
+    {
+      "epoch": 0.24458249767646628,
+      "grad_norm": 33.5789794921875,
+      "learning_rate": 4.0756439517443756e-05,
+      "loss": 3.5605,
+      "step": 2500
+    },
+    {
+      "epoch": 0.24458249767646628,
+      "eval_runtime": 181.7254,
+      "eval_samples_per_second": 112.488,
+      "eval_steps_per_second": 14.065,
+      "step": 2500
+    },
+    {
+      "epoch": 0.2543657975835249,
+      "grad_norm": 34.30876159667969,
+      "learning_rate": 4.238669709814151e-05,
+      "loss": 3.5447,
+      "step": 2600
+    },
+    {
+      "epoch": 0.2641490974905836,
+      "grad_norm": 29.907989501953125,
+      "learning_rate": 4.401695467883926e-05,
+      "loss": 3.5116,
+      "step": 2700
+    },
+    {
+      "epoch": 0.27393239739764225,
+      "grad_norm": 34.08231735229492,
+      "learning_rate": 4.5647212259537006e-05,
+      "loss": 3.4941,
+      "step": 2800
+    },
+    {
+      "epoch": 0.28371569730470086,
+      "grad_norm": 25.034149169921875,
+      "learning_rate": 4.727746984023476e-05,
+      "loss": 3.4863,
+      "step": 2900
+    },
+    {
+      "epoch": 0.29349899721175954,
+      "grad_norm": 32.21685028076172,
+      "learning_rate": 4.890772742093251e-05,
+      "loss": 3.5096,
+      "step": 3000
+    },
+    {
+      "epoch": 0.29349899721175954,
+      "eval_runtime": 181.6612,
+      "eval_samples_per_second": 112.528,
+      "eval_steps_per_second": 14.07,
+      "step": 3000
+    },
+    {
+      "epoch": 0.30328229711881816,
+      "grad_norm": 24.290380477905273,
+      "learning_rate": 4.9940208725902305e-05,
+      "loss": 3.3867,
+      "step": 3100
+    },
+    {
+      "epoch": 0.31306559702587683,
+      "grad_norm": 22.924575805664062,
+      "learning_rate": 4.975902304681838e-05,
+      "loss": 3.398,
+      "step": 3200
+    },
+    {
+      "epoch": 0.3228488969329355,
+      "grad_norm": 19.540430068969727,
+      "learning_rate": 4.957783736773446e-05,
+      "loss": 3.3727,
+      "step": 3300
+    },
+    {
+      "epoch": 0.3326321968399941,
+      "grad_norm": 22.529376983642578,
+      "learning_rate": 4.939665168865053e-05,
+      "loss": 3.3364,
+      "step": 3400
+    },
+    {
+      "epoch": 0.3424154967470528,
+      "grad_norm": 20.821264266967773,
+      "learning_rate": 4.921546600956661e-05,
+      "loss": 3.3126,
+      "step": 3500
+    },
+    {
+      "epoch": 0.3424154967470528,
+      "eval_runtime": 181.7582,
+      "eval_samples_per_second": 112.468,
+      "eval_steps_per_second": 14.063,
+      "step": 3500
+    },
+    {
+      "epoch": 0.3521987966541114,
+      "grad_norm": 24.346153259277344,
+      "learning_rate": 4.903428033048268e-05,
+      "loss": 3.2678,
+      "step": 3600
+    },
+    {
+      "epoch": 0.3619820965611701,
+      "grad_norm": 19.89035415649414,
+      "learning_rate": 4.8853094651398754e-05,
+      "loss": 3.3233,
+      "step": 3700
+    },
+    {
+      "epoch": 0.37176539646822876,
+      "grad_norm": 17.938880920410156,
+      "learning_rate": 4.8671908972314825e-05,
+      "loss": 3.2822,
+      "step": 3800
+    },
+    {
+      "epoch": 0.3815486963752874,
+      "grad_norm": 16.92071533203125,
+      "learning_rate": 4.84907232932309e-05,
+      "loss": 3.2254,
+      "step": 3900
+    },
+    {
+      "epoch": 0.39133199628234605,
+      "grad_norm": 18.241249084472656,
+      "learning_rate": 4.830953761414698e-05,
+      "loss": 3.2116,
+      "step": 4000
+    },
+    {
+      "epoch": 0.39133199628234605,
+      "eval_runtime": 182.8906,
+      "eval_samples_per_second": 111.772,
+      "eval_steps_per_second": 13.976,
+      "step": 4000
+    },
+    {
+      "epoch": 0.40111529618940467,
+      "grad_norm": 17.56020736694336,
+      "learning_rate": 4.812835193506305e-05,
+      "loss": 3.2232,
+      "step": 4100
+    },
+    {
+      "epoch": 0.41089859609646334,
+      "grad_norm": 17.81117057800293,
+      "learning_rate": 4.794716625597913e-05,
+      "loss": 3.1936,
+      "step": 4200
+    },
+    {
+      "epoch": 0.420681896003522,
+      "grad_norm": 19.89581871032715,
+      "learning_rate": 4.77659805768952e-05,
+      "loss": 3.1443,
+      "step": 4300
+    },
+    {
+      "epoch": 0.43046519591058063,
+      "grad_norm": 22.968582153320312,
+      "learning_rate": 4.758479489781128e-05,
+      "loss": 3.2084,
+      "step": 4400
+    },
+    {
+      "epoch": 0.4402484958176393,
+      "grad_norm": 17.119598388671875,
+      "learning_rate": 4.740360921872735e-05,
+      "loss": 3.1263,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4402484958176393,
+      "eval_runtime": 182.3246,
+      "eval_samples_per_second": 112.119,
+      "eval_steps_per_second": 14.019,
+      "step": 4500
+    },
+    {
+      "epoch": 0.4500317957246979,
+      "grad_norm": 19.294527053833008,
+      "learning_rate": 4.722242353964343e-05,
+      "loss": 3.1327,
+      "step": 4600
+    },
+    {
+      "epoch": 0.4598150956317566,
+      "grad_norm": 16.941057205200195,
+      "learning_rate": 4.704123786055951e-05,
+      "loss": 3.0944,
+      "step": 4700
+    },
+    {
+      "epoch": 0.4695983955388152,
+      "grad_norm": 22.43411636352539,
+      "learning_rate": 4.686005218147558e-05,
+      "loss": 3.1093,
+      "step": 4800
+    },
+    {
+      "epoch": 0.4793816954458739,
+      "grad_norm": 19.64097023010254,
+      "learning_rate": 4.667886650239166e-05,
+      "loss": 3.0597,
+      "step": 4900
+    },
+    {
+      "epoch": 0.48916499535293256,
+      "grad_norm": 19.343788146972656,
+      "learning_rate": 4.649768082330773e-05,
+      "loss": 3.1659,
+      "step": 5000
+    },
+    {
+      "epoch": 0.48916499535293256,
+      "eval_runtime": 181.8771,
+      "eval_samples_per_second": 112.395,
+      "eval_steps_per_second": 14.053,
+      "step": 5000
+    },
+    {
+      "epoch": 0.4989482952599912,
+      "grad_norm": 19.657760620117188,
+      "learning_rate": 4.63164951442238e-05,
+      "loss": 3.0506,
+      "step": 5100
+    },
+    {
+      "epoch": 0.5087315951670498,
+      "grad_norm": 16.2425537109375,
+      "learning_rate": 4.613530946513987e-05,
+      "loss": 3.0524,
+      "step": 5200
+    },
+    {
+      "epoch": 0.5185148950741085,
+      "grad_norm": 19.64779281616211,
+      "learning_rate": 4.595412378605595e-05,
+      "loss": 2.9995,
+      "step": 5300
+    },
+    {
+      "epoch": 0.5282981949811671,
+      "grad_norm": 17.29520606994629,
+      "learning_rate": 4.577293810697203e-05,
+      "loss": 3.0932,
+      "step": 5400
+    },
+    {
+      "epoch": 0.5380814948882258,
+      "grad_norm": 17.694602966308594,
+      "learning_rate": 4.55917524278881e-05,
+      "loss": 3.0309,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5380814948882258,
+      "eval_runtime": 181.7231,
+      "eval_samples_per_second": 112.49,
+      "eval_steps_per_second": 14.065,
+      "step": 5500
+    },
+    {
+      "epoch": 0.5478647947952845,
+      "grad_norm": 21.030174255371094,
+      "learning_rate": 4.541056674880418e-05,
+      "loss": 3.0313,
+      "step": 5600
+    },
+    {
+      "epoch": 0.5576480947023431,
+      "grad_norm": 12.339129447937012,
+      "learning_rate": 4.522938106972025e-05,
+      "loss": 3.047,
+      "step": 5700
+    },
+    {
+      "epoch": 0.5674313946094017,
+      "grad_norm": 16.496389389038086,
+      "learning_rate": 4.504819539063633e-05,
+      "loss": 2.9961,
+      "step": 5800
+    },
+    {
+      "epoch": 0.5772146945164603,
+      "grad_norm": 15.456297874450684,
+      "learning_rate": 4.48670097115524e-05,
+      "loss": 2.9821,
+      "step": 5900
+    },
+    {
+      "epoch": 0.5869979944235191,
+      "grad_norm": 17.8603572845459,
+      "learning_rate": 4.468582403246848e-05,
+      "loss": 2.9294,
+      "step": 6000
+    },
+    {
+      "epoch": 0.5869979944235191,
+      "eval_runtime": 181.8258,
+      "eval_samples_per_second": 112.426,
+      "eval_steps_per_second": 14.057,
+      "step": 6000
+    },
+    {
+      "epoch": 0.5967812943305777,
+      "grad_norm": 18.85349464416504,
+      "learning_rate": 4.450463835338455e-05,
+      "loss": 2.9929,
+      "step": 6100
+    },
+    {
+      "epoch": 0.6065645942376363,
+      "grad_norm": 22.971813201904297,
+      "learning_rate": 4.432345267430063e-05,
+      "loss": 2.9684,
+      "step": 6200
+    },
+    {
+      "epoch": 0.616347894144695,
+      "grad_norm": 15.877230644226074,
+      "learning_rate": 4.4142266995216706e-05,
+      "loss": 2.9399,
+      "step": 6300
+    },
+    {
+      "epoch": 0.6261311940517537,
+      "grad_norm": 19.847482681274414,
+      "learning_rate": 4.396108131613278e-05,
+      "loss": 2.88,
+      "step": 6400
+    },
+    {
+      "epoch": 0.6359144939588123,
+      "grad_norm": 15.004170417785645,
+      "learning_rate": 4.377989563704885e-05,
+      "loss": 2.9719,
+      "step": 6500
+    },
+    {
+      "epoch": 0.6359144939588123,
+      "eval_runtime": 182.6045,
+      "eval_samples_per_second": 111.947,
+      "eval_steps_per_second": 13.997,
+      "step": 6500
+    },
+    {
+      "epoch": 0.645697793865871,
+      "grad_norm": 19.473665237426758,
+      "learning_rate": 4.359870995796492e-05,
+      "loss": 2.9246,
+      "step": 6600
+    },
+    {
+      "epoch": 0.6554810937729296,
+      "grad_norm": 18.071683883666992,
+      "learning_rate": 4.3417524278881e-05,
+      "loss": 2.9031,
+      "step": 6700
+    },
+    {
+      "epoch": 0.6652643936799882,
+      "grad_norm": 17.544504165649414,
+      "learning_rate": 4.323633859979707e-05,
+      "loss": 2.8313,
+      "step": 6800
+    },
+    {
+      "epoch": 0.6750476935870469,
+      "grad_norm": 18.936140060424805,
+      "learning_rate": 4.305515292071315e-05,
+      "loss": 2.8536,
+      "step": 6900
+    },
+    {
+      "epoch": 0.6848309934941056,
+      "grad_norm": 14.77696418762207,
+      "learning_rate": 4.2873967241629226e-05,
+      "loss": 2.9104,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6848309934941056,
+      "eval_runtime": 181.938,
+      "eval_samples_per_second": 112.357,
+      "eval_steps_per_second": 14.049,
+      "step": 7000
+    },
+    {
+      "epoch": 0.6946142934011642,
+      "grad_norm": 14.303226470947266,
+      "learning_rate": 4.26927815625453e-05,
+      "loss": 2.8386,
+      "step": 7100
+    },
+    {
+      "epoch": 0.7043975933082228,
+      "grad_norm": 17.11782455444336,
+      "learning_rate": 4.2511595883461376e-05,
+      "loss": 2.9013,
+      "step": 7200
+    },
+    {
+      "epoch": 0.7141808932152816,
+      "grad_norm": 18.661100387573242,
+      "learning_rate": 4.233041020437745e-05,
+      "loss": 2.9428,
+      "step": 7300
+    },
+    {
+      "epoch": 0.7239641931223402,
+      "grad_norm": 15.535719871520996,
+      "learning_rate": 4.2149224525293525e-05,
+      "loss": 2.8582,
+      "step": 7400
+    },
+    {
+      "epoch": 0.7337474930293988,
+      "grad_norm": 15.3306303024292,
+      "learning_rate": 4.19680388462096e-05,
+      "loss": 2.8896,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7337474930293988,
+      "eval_runtime": 181.8938,
+      "eval_samples_per_second": 112.384,
+      "eval_steps_per_second": 14.052,
+      "step": 7500
+    },
+    {
+      "epoch": 0.7435307929364575,
+      "grad_norm": 16.730344772338867,
+      "learning_rate": 4.1786853167125675e-05,
+      "loss": 2.9097,
+      "step": 7600
+    },
+    {
+      "epoch": 0.7533140928435161,
+      "grad_norm": 18.755483627319336,
+      "learning_rate": 4.1605667488041746e-05,
+      "loss": 2.8815,
+      "step": 7700
+    },
+    {
+      "epoch": 0.7630973927505748,
+      "grad_norm": 18.737581253051758,
+      "learning_rate": 4.1424481808957824e-05,
+      "loss": 2.9202,
+      "step": 7800
+    },
+    {
+      "epoch": 0.7728806926576334,
+      "grad_norm": 14.711681365966797,
+      "learning_rate": 4.1243296129873896e-05,
+      "loss": 2.806,
+      "step": 7900
+    },
+    {
+      "epoch": 0.7826639925646921,
+      "grad_norm": 17.5069580078125,
+      "learning_rate": 4.106211045078997e-05,
+      "loss": 2.8576,
+      "step": 8000
+    },
+    {
+      "epoch": 0.7826639925646921,
+      "eval_runtime": 181.9442,
+      "eval_samples_per_second": 112.353,
+      "eval_steps_per_second": 14.048,
+      "step": 8000
+    },
+    {
+      "epoch": 0.7924472924717507,
+      "grad_norm": 17.678852081298828,
+      "learning_rate": 4.0880924771706046e-05,
+      "loss": 2.8035,
+      "step": 8100
+    },
+    {
+      "epoch": 0.8022305923788093,
+      "grad_norm": 17.644638061523438,
+      "learning_rate": 4.069973909262212e-05,
+      "loss": 2.7958,
+      "step": 8200
+    },
+    {
+      "epoch": 0.8120138922858681,
+      "grad_norm": 18.377134323120117,
+      "learning_rate": 4.0518553413538195e-05,
+      "loss": 2.8055,
+      "step": 8300
+    },
+    {
+      "epoch": 0.8217971921929267,
+      "grad_norm": 18.026033401489258,
+      "learning_rate": 4.0337367734454273e-05,
+      "loss": 2.7334,
+      "step": 8400
+    },
+    {
+      "epoch": 0.8315804920999853,
+      "grad_norm": 14.77315616607666,
+      "learning_rate": 4.0156182055370345e-05,
+      "loss": 2.8082,
+      "step": 8500
+    },
+    {
+      "epoch": 0.8315804920999853,
+      "eval_runtime": 182.4176,
+      "eval_samples_per_second": 112.062,
+      "eval_steps_per_second": 14.012,
+      "step": 8500
+    },
+    {
+      "epoch": 0.841363792007044,
+      "grad_norm": 13.729479789733887,
+      "learning_rate": 3.997499637628642e-05,
+      "loss": 2.7939,
+      "step": 8600
+    },
+    {
+      "epoch": 0.8511470919141026,
+      "grad_norm": 16.34333610534668,
+      "learning_rate": 3.9793810697202494e-05,
+      "loss": 2.8517,
+      "step": 8700
+    },
+    {
+      "epoch": 0.8609303918211613,
+      "grad_norm": 22.484411239624023,
+      "learning_rate": 3.961262501811857e-05,
+      "loss": 2.776,
+      "step": 8800
+    },
+    {
+      "epoch": 0.8707136917282199,
+      "grad_norm": 15.922870635986328,
+      "learning_rate": 3.9431439339034644e-05,
+      "loss": 2.7909,
+      "step": 8900
+    },
+    {
+      "epoch": 0.8804969916352786,
+      "grad_norm": 15.06955623626709,
+      "learning_rate": 3.925025365995072e-05,
+      "loss": 2.8416,
+      "step": 9000
+    },
+    {
+      "epoch": 0.8804969916352786,
+      "eval_runtime": 181.9314,
+      "eval_samples_per_second": 112.361,
+      "eval_steps_per_second": 14.049,
+      "step": 9000
+    },
+    {
+      "epoch": 0.8902802915423372,
+      "grad_norm": 16.060428619384766,
+      "learning_rate": 3.9069067980866794e-05,
+      "loss": 2.7803,
+      "step": 9100
+    },
+    {
+      "epoch": 0.9000635914493959,
+      "grad_norm": 16.80124855041504,
+      "learning_rate": 3.888788230178287e-05,
+      "loss": 2.7548,
+      "step": 9200
+    },
+    {
+      "epoch": 0.9098468913564546,
+      "grad_norm": 16.608434677124023,
+      "learning_rate": 3.870669662269894e-05,
+      "loss": 2.8606,
+      "step": 9300
+    },
+    {
+      "epoch": 0.9196301912635132,
+      "grad_norm": 14.83870792388916,
+      "learning_rate": 3.8525510943615015e-05,
+      "loss": 2.7833,
+      "step": 9400
+    },
+    {
+      "epoch": 0.9294134911705718,
+      "grad_norm": 25.778181076049805,
+      "learning_rate": 3.834432526453109e-05,
+      "loss": 2.7434,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9294134911705718,
+      "eval_runtime": 181.99,
+      "eval_samples_per_second": 112.325,
+      "eval_steps_per_second": 14.045,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9391967910776304,
+      "grad_norm": 17.374011993408203,
+      "learning_rate": 3.8163139585447164e-05,
+      "loss": 2.7258,
+      "step": 9600
+    },
+    {
+      "epoch": 0.9489800909846892,
+      "grad_norm": 17.551128387451172,
+      "learning_rate": 3.798195390636324e-05,
+      "loss": 2.824,
+      "step": 9700
+    },
+    {
+      "epoch": 0.9587633908917478,
+      "grad_norm": 14.35797119140625,
+      "learning_rate": 3.7800768227279314e-05,
+      "loss": 2.745,
+      "step": 9800
+    },
+    {
+      "epoch": 0.9685466907988064,
+      "grad_norm": 20.098552703857422,
+      "learning_rate": 3.761958254819539e-05,
+      "loss": 2.7025,
+      "step": 9900
+    },
+    {
+      "epoch": 0.9783299907058651,
+      "grad_norm": 16.218109130859375,
+      "learning_rate": 3.743839686911147e-05,
+      "loss": 2.8093,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9783299907058651,
+      "eval_runtime": 181.8987,
+      "eval_samples_per_second": 112.381,
+      "eval_steps_per_second": 14.052,
+      "step": 10000
+    },
+    {
+      "epoch": 0.9881132906129237,
+      "grad_norm": 17.198423385620117,
+      "learning_rate": 3.725721119002754e-05,
+      "loss": 2.7124,
+      "step": 10100
+    },
+    {
+      "epoch": 0.9978965905199824,
+      "grad_norm": 18.021198272705078,
+      "learning_rate": 3.707602551094362e-05,
+      "loss": 2.6922,
+      "step": 10200
+    },
+    {
+      "epoch": 1.007679890427041,
+      "grad_norm": 15.27678108215332,
+      "learning_rate": 3.689483983185969e-05,
+      "loss": 2.6743,
+      "step": 10300
+    },
+    {
+      "epoch": 1.0174631903340996,
+      "grad_norm": 16.770511627197266,
+      "learning_rate": 3.671365415277577e-05,
+      "loss": 2.857,
+      "step": 10400
+    },
+    {
+      "epoch": 1.0272464902411584,
+      "grad_norm": 18.810932159423828,
+      "learning_rate": 3.653246847369184e-05,
+      "loss": 2.7269,
+      "step": 10500
+    },
+    {
+      "epoch": 1.0272464902411584,
+      "eval_runtime": 181.8537,
+      "eval_samples_per_second": 112.409,
+      "eval_steps_per_second": 14.055,
+      "step": 10500
+    },
+    {
+      "epoch": 1.037029790148217,
+      "grad_norm": 18.56201171875,
+      "learning_rate": 3.635128279460791e-05,
+      "loss": 2.7325,
+      "step": 10600
+    },
+    {
+      "epoch": 1.0468130900552757,
+      "grad_norm": 15.063011169433594,
+      "learning_rate": 3.617009711552399e-05,
+      "loss": 2.7827,
+      "step": 10700
+    },
+    {
+      "epoch": 1.0565963899623343,
+      "grad_norm": 15.339439392089844,
+      "learning_rate": 3.598891143644006e-05,
+      "loss": 2.7472,
+      "step": 10800
+    },
+    {
+      "epoch": 1.066379689869393,
+      "grad_norm": 17.466033935546875,
+      "learning_rate": 3.580772575735614e-05,
+      "loss": 2.7859,
+      "step": 10900
+    },
+    {
+      "epoch": 1.0761629897764515,
+      "grad_norm": 20.727872848510742,
+      "learning_rate": 3.562654007827221e-05,
+      "loss": 2.7278,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0761629897764515,
+      "eval_runtime": 181.8566,
+      "eval_samples_per_second": 112.407,
+      "eval_steps_per_second": 14.055,
+      "step": 11000
+    },
+    {
+      "epoch": 1.0859462896835101,
+      "grad_norm": 16.02055549621582,
+      "learning_rate": 3.544535439918829e-05,
+      "loss": 2.6307,
+      "step": 11100
+    },
+    {
+      "epoch": 1.095729589590569,
+      "grad_norm": 20.069686889648438,
+      "learning_rate": 3.526416872010436e-05,
+      "loss": 2.711,
+      "step": 11200
+    },
+    {
+      "epoch": 1.1055128894976276,
+      "grad_norm": 14.833261489868164,
+      "learning_rate": 3.508298304102044e-05,
+      "loss": 2.6141,
+      "step": 11300
+    },
+    {
+      "epoch": 1.1152961894046862,
+      "grad_norm": 14.86436653137207,
+      "learning_rate": 3.490179736193652e-05,
+      "loss": 2.6816,
+      "step": 11400
+    },
+    {
+      "epoch": 1.1250794893117448,
+      "grad_norm": 17.955862045288086,
+      "learning_rate": 3.472061168285259e-05,
+      "loss": 2.6924,
+      "step": 11500
+    },
+    {
+      "epoch": 1.1250794893117448,
+      "eval_runtime": 181.8085,
+      "eval_samples_per_second": 112.437,
+      "eval_steps_per_second": 14.059,
+      "step": 11500
+    },
+    {
+      "epoch": 1.1348627892188035,
+      "grad_norm": 18.360109329223633,
+      "learning_rate": 3.453942600376867e-05,
+      "loss": 2.6181,
+      "step": 11600
+    },
+    {
+      "epoch": 1.144646089125862,
+      "grad_norm": 17.547542572021484,
+      "learning_rate": 3.435824032468474e-05,
+      "loss": 2.6394,
+      "step": 11700
+    },
+    {
+      "epoch": 1.154429389032921,
+      "grad_norm": 12.194833755493164,
+      "learning_rate": 3.417705464560082e-05,
+      "loss": 2.6684,
+      "step": 11800
+    },
+    {
+      "epoch": 1.1642126889399795,
+      "grad_norm": 17.095104217529297,
+      "learning_rate": 3.399586896651689e-05,
+      "loss": 2.6129,
+      "step": 11900
+    },
+    {
+      "epoch": 1.1739959888470382,
+      "grad_norm": 20.788406372070312,
+      "learning_rate": 3.381468328743296e-05,
+      "loss": 2.5663,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1739959888470382,
+      "eval_runtime": 181.8035,
+      "eval_samples_per_second": 112.44,
+      "eval_steps_per_second": 14.059,
+      "step": 12000
+    },
+    {
+      "epoch": 1.1837792887540968,
+      "grad_norm": 14.261167526245117,
+      "learning_rate": 3.363349760834904e-05,
+      "loss": 2.6544,
+      "step": 12100
+    },
+    {
+      "epoch": 1.1935625886611554,
+      "grad_norm": 24.68012046813965,
+      "learning_rate": 3.345231192926511e-05,
+      "loss": 2.6632,
+      "step": 12200
+    },
+    {
+      "epoch": 1.203345888568214,
+      "grad_norm": 16.10886573791504,
+      "learning_rate": 3.327112625018119e-05,
+      "loss": 2.6366,
+      "step": 12300
+    },
+    {
+      "epoch": 1.2131291884752726,
+      "grad_norm": 18.038848876953125,
+      "learning_rate": 3.308994057109726e-05,
+      "loss": 2.6563,
+      "step": 12400
+    },
+    {
+      "epoch": 1.2229124883823315,
+      "grad_norm": 17.40920639038086,
+      "learning_rate": 3.290875489201334e-05,
+      "loss": 2.718,
+      "step": 12500
+    },
+    {
+      "epoch": 1.2229124883823315,
+      "eval_runtime": 181.9491,
+      "eval_samples_per_second": 112.35,
+      "eval_steps_per_second": 14.048,
+      "step": 12500
+    },
+    {
+      "epoch": 1.23269578828939,
+      "grad_norm": 15.097307205200195,
+      "learning_rate": 3.272756921292941e-05,
+      "loss": 2.7282,
+      "step": 12600
+    },
+    {
+      "epoch": 1.2424790881964487,
+      "grad_norm": 17.63008689880371,
+      "learning_rate": 3.254638353384549e-05,
+      "loss": 2.7104,
+      "step": 12700
+    },
+    {
+      "epoch": 1.2522623881035073,
+      "grad_norm": 16.161130905151367,
+      "learning_rate": 3.236519785476156e-05,
+      "loss": 2.6427,
+      "step": 12800
+    },
+    {
+      "epoch": 1.262045688010566,
+      "grad_norm": 18.786882400512695,
+      "learning_rate": 3.218401217567764e-05,
+      "loss": 2.6105,
+      "step": 12900
+    },
+    {
+      "epoch": 1.2718289879176246,
+      "grad_norm": 24.145421981811523,
+      "learning_rate": 3.2002826496593715e-05,
+      "loss": 2.6322,
+      "step": 13000
+    },
+    {
+      "epoch": 1.2718289879176246,
+      "eval_runtime": 182.5613,
+      "eval_samples_per_second": 111.973,
+      "eval_steps_per_second": 14.001,
+      "step": 13000
+    },
+    {
+      "epoch": 1.2816122878246832,
+      "grad_norm": 15.286133766174316,
+      "learning_rate": 3.1821640817509786e-05,
+      "loss": 2.6465,
+      "step": 13100
+    },
+    {
+      "epoch": 1.291395587731742,
+      "grad_norm": 21.22935676574707,
+      "learning_rate": 3.1640455138425865e-05,
+      "loss": 2.6691,
+      "step": 13200
+    },
+    {
+      "epoch": 1.3011788876388006,
+      "grad_norm": 18.064428329467773,
+      "learning_rate": 3.1459269459341936e-05,
+      "loss": 2.5904,
+      "step": 13300
+    },
+    {
+      "epoch": 1.3109621875458592,
+      "grad_norm": 14.45976448059082,
+      "learning_rate": 3.127808378025801e-05,
+      "loss": 2.6602,
+      "step": 13400
+    },
+    {
+      "epoch": 1.3207454874529179,
+      "grad_norm": 19.72386360168457,
+      "learning_rate": 3.109689810117408e-05,
+      "loss": 2.6337,
+      "step": 13500
+    },
+    {
+      "epoch": 1.3207454874529179,
+      "eval_runtime": 182.4053,
+      "eval_samples_per_second": 112.069,
+      "eval_steps_per_second": 14.013,
+      "step": 13500
+    },
+    {
+      "epoch": 1.3305287873599765,
+      "grad_norm": 17.639583587646484,
+      "learning_rate": 3.091571242209016e-05,
+      "loss": 2.6135,
+      "step": 13600
+    },
+    {
+      "epoch": 1.340312087267035,
+      "grad_norm": 19.71700096130371,
+      "learning_rate": 3.0734526743006235e-05,
+      "loss": 2.6252,
+      "step": 13700
+    },
+    {
+      "epoch": 1.3500953871740937,
+      "grad_norm": 16.715856552124023,
+      "learning_rate": 3.055334106392231e-05,
+      "loss": 2.6475,
+      "step": 13800
+    },
+    {
+      "epoch": 1.3598786870811526,
+      "grad_norm": 12.645075798034668,
+      "learning_rate": 3.0372155384838385e-05,
+      "loss": 2.6199,
+      "step": 13900
+    },
+    {
+      "epoch": 1.3696619869882112,
+      "grad_norm": 20.150625228881836,
+      "learning_rate": 3.0190969705754456e-05,
+      "loss": 2.5567,
+      "step": 14000
+    },
+    {
+      "epoch": 1.3696619869882112,
+      "eval_runtime": 181.9086,
+      "eval_samples_per_second": 112.375,
+      "eval_steps_per_second": 14.051,
+      "step": 14000
+    },
+    {
+      "epoch": 1.3794452868952698,
+      "grad_norm": 19.111286163330078,
+      "learning_rate": 3.0009784026670535e-05,
+      "loss": 2.59,
+      "step": 14100
+    },
+    {
+      "epoch": 1.3892285868023284,
+      "grad_norm": 17.12226104736328,
+      "learning_rate": 2.9828598347586606e-05,
+      "loss": 2.5913,
+      "step": 14200
+    },
+    {
+      "epoch": 1.399011886709387,
+      "grad_norm": 19.741445541381836,
+      "learning_rate": 2.9647412668502684e-05,
+      "loss": 2.5617,
+      "step": 14300
+    },
+    {
+      "epoch": 1.4087951866164456,
+      "grad_norm": 17.605525970458984,
+      "learning_rate": 2.946622698941876e-05,
+      "loss": 2.6077,
+      "step": 14400
+    },
+    {
+      "epoch": 1.4185784865235043,
+      "grad_norm": 17.433218002319336,
+      "learning_rate": 2.928504131033483e-05,
+      "loss": 2.5713,
+      "step": 14500
+    },
+    {
+      "epoch": 1.4185784865235043,
+      "eval_runtime": 181.9305,
+      "eval_samples_per_second": 112.362,
+      "eval_steps_per_second": 14.049,
+      "step": 14500
+    },
+    {
+      "epoch": 1.428361786430563,
+      "grad_norm": 15.442538261413574,
+      "learning_rate": 2.910385563125091e-05,
+      "loss": 2.6499,
+      "step": 14600
+    },
+    {
+      "epoch": 1.4381450863376217,
+      "grad_norm": 15.078730583190918,
+      "learning_rate": 2.892266995216698e-05,
+      "loss": 2.6517,
+      "step": 14700
+    },
+    {
+      "epoch": 1.4479283862446803,
+      "grad_norm": 23.07891273498535,
+      "learning_rate": 2.874148427308306e-05,
+      "loss": 2.594,
+      "step": 14800
+    },
+    {
+      "epoch": 1.457711686151739,
+      "grad_norm": 16.707923889160156,
+      "learning_rate": 2.856029859399913e-05,
+      "loss": 2.6613,
+      "step": 14900
+    },
+    {
+      "epoch": 1.4674949860587976,
+      "grad_norm": 16.731164932250977,
+      "learning_rate": 2.8379112914915208e-05,
+      "loss": 2.5927,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4674949860587976,
+      "eval_runtime": 181.9649,
+      "eval_samples_per_second": 112.34,
+      "eval_steps_per_second": 14.047,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4772782859658564,
+      "grad_norm": 16.020864486694336,
+      "learning_rate": 2.819792723583128e-05,
+      "loss": 2.6464,
+      "step": 15100
+    },
+    {
+      "epoch": 1.4870615858729148,
+      "grad_norm": 16.674760818481445,
+      "learning_rate": 2.8016741556747354e-05,
+      "loss": 2.5853,
+      "step": 15200
+    },
+    {
+      "epoch": 1.4968448857799737,
+      "grad_norm": 16.890748977661133,
+      "learning_rate": 2.7835555877663432e-05,
+      "loss": 2.5748,
+      "step": 15300
+    },
+    {
+      "epoch": 1.5066281856870323,
+      "grad_norm": 20.217845916748047,
+      "learning_rate": 2.7654370198579504e-05,
+      "loss": 2.6204,
+      "step": 15400
+    },
+    {
+      "epoch": 1.516411485594091,
+      "grad_norm": 20.459087371826172,
+      "learning_rate": 2.7473184519495582e-05,
+      "loss": 2.6103,
+      "step": 15500
+    },
+    {
+      "epoch": 1.516411485594091,
+      "eval_runtime": 181.9454,
+      "eval_samples_per_second": 112.352,
+      "eval_steps_per_second": 14.048,
+      "step": 15500
+    },
+    {
+      "epoch": 1.5261947855011495,
+      "grad_norm": 18.207612991333008,
+      "learning_rate": 2.7291998840411654e-05,
+      "loss": 2.5786,
+      "step": 15600
+    },
+    {
+      "epoch": 1.5359780854082081,
+      "grad_norm": 18.084758758544922,
+      "learning_rate": 2.7110813161327732e-05,
+      "loss": 2.6535,
+      "step": 15700
+    },
+    {
+      "epoch": 1.545761385315267,
+      "grad_norm": 15.03881549835205,
+      "learning_rate": 2.6929627482243803e-05,
+      "loss": 2.6061,
+      "step": 15800
+    },
+    {
+      "epoch": 1.5555446852223254,
+      "grad_norm": 16.99995231628418,
+      "learning_rate": 2.6748441803159878e-05,
+      "loss": 2.6151,
+      "step": 15900
+    },
+    {
+      "epoch": 1.5653279851293842,
+      "grad_norm": 15.581089973449707,
+      "learning_rate": 2.6567256124075956e-05,
+      "loss": 2.6163,
+      "step": 16000
+    },
+    {
+      "epoch": 1.5653279851293842,
+      "eval_runtime": 181.8152,
+      "eval_samples_per_second": 112.433,
+      "eval_steps_per_second": 14.058,
+      "step": 16000
+    },
+    {
+      "epoch": 1.5751112850364428,
+      "grad_norm": 21.4382266998291,
+      "learning_rate": 2.6386070444992028e-05,
+      "loss": 2.5975,
+      "step": 16100
+    },
+    {
+      "epoch": 1.5848945849435014,
+      "grad_norm": 15.874536514282227,
+      "learning_rate": 2.6204884765908106e-05,
+      "loss": 2.5851,
+      "step": 16200
+    },
+    {
+      "epoch": 1.59467788485056,
+      "grad_norm": 17.902137756347656,
+      "learning_rate": 2.6023699086824177e-05,
+      "loss": 2.6027,
+      "step": 16300
+    },
+    {
+      "epoch": 1.6044611847576187,
+      "grad_norm": 17.04872703552246,
+      "learning_rate": 2.5842513407740255e-05,
+      "loss": 2.5854,
+      "step": 16400
+    },
+    {
+      "epoch": 1.6142444846646775,
+      "grad_norm": 15.406013488769531,
+      "learning_rate": 2.5661327728656327e-05,
+      "loss": 2.5158,
+      "step": 16500
+    },
+    {
+      "epoch": 1.6142444846646775,
+      "eval_runtime": 181.8647,
+      "eval_samples_per_second": 112.402,
+      "eval_steps_per_second": 14.054,
+      "step": 16500
+    },
+    {
+      "epoch": 1.624027784571736,
+      "grad_norm": 19.62627601623535,
+      "learning_rate": 2.5480142049572402e-05,
+      "loss": 2.5378,
+      "step": 16600
+    },
+    {
+      "epoch": 1.6338110844787948,
+      "grad_norm": 17.825178146362305,
+      "learning_rate": 2.529895637048848e-05,
+      "loss": 2.6162,
+      "step": 16700
+    },
+    {
+      "epoch": 1.6435943843858534,
+      "grad_norm": 15.442023277282715,
+      "learning_rate": 2.511777069140455e-05,
+      "loss": 2.5802,
+      "step": 16800
+    },
+    {
+      "epoch": 1.653377684292912,
+      "grad_norm": 18.695241928100586,
+      "learning_rate": 2.4936585012320626e-05,
+      "loss": 2.585,
+      "step": 16900
+    },
+    {
+      "epoch": 1.6631609841999706,
+      "grad_norm": 18.992969512939453,
+      "learning_rate": 2.4755399333236704e-05,
+      "loss": 2.5448,
+      "step": 17000
+    },
+    {
+      "epoch": 1.6631609841999706,
+      "eval_runtime": 181.91,
+      "eval_samples_per_second": 112.374,
+      "eval_steps_per_second": 14.051,
+      "step": 17000
+    },
+    {
+      "epoch": 1.6729442841070292,
+      "grad_norm": 19.065349578857422,
+      "learning_rate": 2.457421365415278e-05,
+      "loss": 2.6565,
+      "step": 17100
+    },
+    {
+      "epoch": 1.682727584014088,
+      "grad_norm": 20.110734939575195,
+      "learning_rate": 2.439302797506885e-05,
+      "loss": 2.5519,
+      "step": 17200
+    },
+    {
+      "epoch": 1.6925108839211465,
+      "grad_norm": 15.886931419372559,
+      "learning_rate": 2.4211842295984925e-05,
+      "loss": 2.5589,
+      "step": 17300
+    },
+    {
+      "epoch": 1.7022941838282053,
+      "grad_norm": 19.213207244873047,
+      "learning_rate": 2.4030656616901e-05,
+      "loss": 2.5714,
+      "step": 17400
+    },
+    {
+      "epoch": 1.712077483735264,
+      "grad_norm": 17.117481231689453,
+      "learning_rate": 2.3849470937817075e-05,
+      "loss": 2.6682,
+      "step": 17500
+    },
+    {
+      "epoch": 1.712077483735264,
+      "eval_runtime": 181.766,
+      "eval_samples_per_second": 112.463,
+      "eval_steps_per_second": 14.062,
+      "step": 17500
+    },
+    {
+      "epoch": 1.7218607836423225,
+      "grad_norm": 17.19162940979004,
+      "learning_rate": 2.366828525873315e-05,
+      "loss": 2.5591,
+      "step": 17600
+    },
+    {
+      "epoch": 1.7316440835493812,
+      "grad_norm": 15.454411506652832,
+      "learning_rate": 2.3487099579649225e-05,
+      "loss": 2.469,
+      "step": 17700
+    },
+    {
+      "epoch": 1.7414273834564398,
+      "grad_norm": 15.227791786193848,
+      "learning_rate": 2.3305913900565303e-05,
+      "loss": 2.664,
+      "step": 17800
+    },
+    {
+      "epoch": 1.7512106833634986,
+      "grad_norm": 18.5739688873291,
+      "learning_rate": 2.3124728221481374e-05,
+      "loss": 2.5991,
+      "step": 17900
+    },
+    {
+      "epoch": 1.760993983270557,
+      "grad_norm": 12.589066505432129,
+      "learning_rate": 2.294354254239745e-05,
+      "loss": 2.6593,
+      "step": 18000
+    },
+    {
+      "epoch": 1.760993983270557,
+      "eval_runtime": 181.9699,
+      "eval_samples_per_second": 112.337,
+      "eval_steps_per_second": 14.046,
+      "step": 18000
+    },
+    {
+      "epoch": 1.7707772831776158,
+      "grad_norm": 20.695772171020508,
+      "learning_rate": 2.2762356863313524e-05,
+      "loss": 2.5555,
+      "step": 18100
+    },
+    {
+      "epoch": 1.7805605830846745,
+      "grad_norm": 12.731703758239746,
+      "learning_rate": 2.25811711842296e-05,
+      "loss": 2.4617,
+      "step": 18200
+    },
+    {
+      "epoch": 1.790343882991733,
+      "grad_norm": 18.506074905395508,
+      "learning_rate": 2.2399985505145674e-05,
+      "loss": 2.6061,
+      "step": 18300
+    },
+    {
+      "epoch": 1.800127182898792,
+      "grad_norm": 14.8694486618042,
+      "learning_rate": 2.221879982606175e-05,
+      "loss": 2.5779,
+      "step": 18400
+    },
+    {
+      "epoch": 1.8099104828058503,
+      "grad_norm": 22.47985076904297,
+      "learning_rate": 2.2037614146977827e-05,
+      "loss": 2.5012,
+      "step": 18500
+    },
+    {
+      "epoch": 1.8099104828058503,
+      "eval_runtime": 182.3919,
+      "eval_samples_per_second": 112.077,
+      "eval_steps_per_second": 14.014,
+      "step": 18500
+    },
+    {
+      "epoch": 1.8196937827129092,
+      "grad_norm": 25.74334144592285,
+      "learning_rate": 2.1856428467893898e-05,
+      "loss": 2.5265,
+      "step": 18600
+    },
+    {
+      "epoch": 1.8294770826199676,
+      "grad_norm": 18.477630615234375,
+      "learning_rate": 2.1675242788809973e-05,
+      "loss": 2.5555,
+      "step": 18700
+    },
+    {
+      "epoch": 1.8392603825270264,
+      "grad_norm": 14.832316398620605,
+      "learning_rate": 2.1494057109726048e-05,
+      "loss": 2.4609,
+      "step": 18800
+    },
+    {
+      "epoch": 1.849043682434085,
+      "grad_norm": 17.025096893310547,
+      "learning_rate": 2.1312871430642123e-05,
+      "loss": 2.5119,
+      "step": 18900
+    },
+    {
+      "epoch": 1.8588269823411436,
+      "grad_norm": 16.852436065673828,
+      "learning_rate": 2.1131685751558197e-05,
+      "loss": 2.5369,
+      "step": 19000
+    },
+    {
+      "epoch": 1.8588269823411436,
+      "eval_runtime": 181.7443,
+      "eval_samples_per_second": 112.477,
+      "eval_steps_per_second": 14.064,
+      "step": 19000
+    },
+    {
+      "epoch": 1.8686102822482025,
+      "grad_norm": 15.160259246826172,
+      "learning_rate": 2.0950500072474272e-05,
+      "loss": 2.6297,
+      "step": 19100
+    },
+    {
+      "epoch": 1.8783935821552609,
+      "grad_norm": 15.909671783447266,
+      "learning_rate": 2.0769314393390347e-05,
+      "loss": 2.4696,
+      "step": 19200
+    },
+    {
+      "epoch": 1.8881768820623197,
+      "grad_norm": 14.201844215393066,
+      "learning_rate": 2.0588128714306422e-05,
+      "loss": 2.5653,
+      "step": 19300
+    },
+    {
+      "epoch": 1.8979601819693783,
+      "grad_norm": 16.351415634155273,
+      "learning_rate": 2.0406943035222497e-05,
+      "loss": 2.4962,
+      "step": 19400
+    },
+    {
+      "epoch": 1.907743481876437,
+      "grad_norm": 16.943771362304688,
+      "learning_rate": 2.022575735613857e-05,
+      "loss": 2.5091,
+      "step": 19500
+    },
+    {
+      "epoch": 1.907743481876437,
+      "eval_runtime": 181.6486,
+      "eval_samples_per_second": 112.536,
+      "eval_steps_per_second": 14.071,
+      "step": 19500
+    },
+    {
+      "epoch": 1.9175267817834956,
+      "grad_norm": 15.006349563598633,
+      "learning_rate": 2.0044571677054646e-05,
+      "loss": 2.5214,
+      "step": 19600
+    },
+    {
+      "epoch": 1.9273100816905542,
+      "grad_norm": 17.305580139160156,
+      "learning_rate": 1.986338599797072e-05,
+      "loss": 2.4989,
+      "step": 19700
+    },
+    {
+      "epoch": 1.937093381597613,
+      "grad_norm": 17.28044891357422,
+      "learning_rate": 1.9682200318886796e-05,
+      "loss": 2.4008,
+      "step": 19800
+    },
+    {
+      "epoch": 1.9468766815046714,
+      "grad_norm": 18.25079917907715,
+      "learning_rate": 1.950101463980287e-05,
+      "loss": 2.6015,
+      "step": 19900
+    },
+    {
+      "epoch": 1.9566599814117303,
+      "grad_norm": 20.741668701171875,
+      "learning_rate": 1.9319828960718946e-05,
+      "loss": 2.4081,
+      "step": 20000
+    },
+    {
+      "epoch": 1.9566599814117303,
+      "eval_runtime": 181.7745,
+      "eval_samples_per_second": 112.458,
+      "eval_steps_per_second": 14.061,
+      "step": 20000
+    },
+    {
+      "epoch": 1.9664432813187889,
+      "grad_norm": 16.1226863861084,
+      "learning_rate": 1.913864328163502e-05,
+      "loss": 2.5418,
+      "step": 20100
+    },
+    {
+      "epoch": 1.9762265812258475,
+      "grad_norm": 13.914982795715332,
+      "learning_rate": 1.8957457602551095e-05,
+      "loss": 2.5248,
+      "step": 20200
+    },
+    {
+      "epoch": 1.986009881132906,
+      "grad_norm": 15.072690963745117,
+      "learning_rate": 1.877627192346717e-05,
+      "loss": 2.5488,
+      "step": 20300
+    },
+    {
+      "epoch": 1.9957931810399647,
+      "grad_norm": 15.510763168334961,
+      "learning_rate": 1.8595086244383245e-05,
+      "loss": 2.4605,
+      "step": 20400
+    },
+    {
+      "epoch": 2.0055764809470236,
+      "grad_norm": 18.463842391967773,
+      "learning_rate": 1.841390056529932e-05,
+      "loss": 2.522,
+      "step": 20500
+    },
+    {
+      "epoch": 2.0055764809470236,
+      "eval_runtime": 182.07,
+      "eval_samples_per_second": 112.276,
+      "eval_steps_per_second": 14.039,
+      "step": 20500
+    },
+    {
+      "epoch": 2.015359780854082,
+      "grad_norm": 16.670269012451172,
+      "learning_rate": 1.8232714886215394e-05,
+      "loss": 2.5585,
+      "step": 20600
+    },
+    {
+      "epoch": 2.025143080761141,
+      "grad_norm": 20.60368537902832,
+      "learning_rate": 1.805152920713147e-05,
+      "loss": 2.5381,
+      "step": 20700
+    },
+    {
+      "epoch": 2.034926380668199,
+      "grad_norm": 15.686981201171875,
+      "learning_rate": 1.7870343528047544e-05,
+      "loss": 2.5721,
+      "step": 20800
+    },
+    {
+      "epoch": 2.044709680575258,
+      "grad_norm": 14.691718101501465,
+      "learning_rate": 1.768915784896362e-05,
+      "loss": 2.5187,
+      "step": 20900
+    },
+    {
+      "epoch": 2.054492980482317,
+      "grad_norm": 16.31734848022461,
+      "learning_rate": 1.7507972169879694e-05,
+      "loss": 2.5202,
+      "step": 21000
+    },
+    {
+      "epoch": 2.054492980482317,
+      "eval_runtime": 181.9896,
+      "eval_samples_per_second": 112.325,
+      "eval_steps_per_second": 14.045,
+      "step": 21000
+    },
+    {
+      "epoch": 2.0642762803893753,
+      "grad_norm": 12.698554992675781,
+      "learning_rate": 1.732678649079577e-05,
+      "loss": 2.4228,
+      "step": 21100
+    },
+    {
+      "epoch": 2.074059580296434,
+      "grad_norm": 16.34201431274414,
+      "learning_rate": 1.7145600811711843e-05,
+      "loss": 2.3963,
+      "step": 21200
+    },
+    {
+      "epoch": 2.0838428802034925,
+      "grad_norm": 16.52840232849121,
+      "learning_rate": 1.6964415132627918e-05,
+      "loss": 2.4759,
+      "step": 21300
+    },
+    {
+      "epoch": 2.0936261801105513,
+      "grad_norm": 14.856452941894531,
+      "learning_rate": 1.6783229453543993e-05,
+      "loss": 2.4675,
+      "step": 21400
+    },
+    {
+      "epoch": 2.1034094800176097,
+      "grad_norm": 19.68895721435547,
+      "learning_rate": 1.6602043774460068e-05,
+      "loss": 2.5324,
+      "step": 21500
+    },
+    {
+      "epoch": 2.1034094800176097,
+      "eval_runtime": 182.1877,
+      "eval_samples_per_second": 112.203,
+      "eval_steps_per_second": 14.029,
+      "step": 21500
+    },
+    {
+      "epoch": 2.1131927799246686,
+      "grad_norm": 23.248056411743164,
+      "learning_rate": 1.6420858095376143e-05,
+      "loss": 2.5231,
+      "step": 21600
+    },
+    {
+      "epoch": 2.1229760798317274,
+      "grad_norm": 25.471004486083984,
+      "learning_rate": 1.6239672416292217e-05,
+      "loss": 2.5871,
+      "step": 21700
+    },
+    {
+      "epoch": 2.132759379738786,
+      "grad_norm": 17.794851303100586,
+      "learning_rate": 1.6058486737208292e-05,
+      "loss": 2.5008,
+      "step": 21800
+    },
+    {
+      "epoch": 2.1425426796458447,
+      "grad_norm": 15.450346946716309,
+      "learning_rate": 1.5877301058124367e-05,
+      "loss": 2.4194,
+      "step": 21900
+    },
+    {
+      "epoch": 2.152325979552903,
+      "grad_norm": 13.243645668029785,
+      "learning_rate": 1.5696115379040442e-05,
+      "loss": 2.5018,
+      "step": 22000
+    },
+    {
+      "epoch": 2.152325979552903,
+      "eval_runtime": 181.9841,
+      "eval_samples_per_second": 112.328,
+      "eval_steps_per_second": 14.045,
+      "step": 22000
+    },
+    {
+      "epoch": 2.162109279459962,
+      "grad_norm": 16.996198654174805,
+      "learning_rate": 1.5514929699956517e-05,
+      "loss": 2.4492,
+      "step": 22100
+    },
+    {
+      "epoch": 2.1718925793670203,
+      "grad_norm": 20.05558967590332,
+      "learning_rate": 1.5333744020872588e-05,
+      "loss": 2.489,
+      "step": 22200
+    },
+    {
+      "epoch": 2.181675879274079,
+      "grad_norm": 15.66326904296875,
+      "learning_rate": 1.5152558341788666e-05,
+      "loss": 2.5089,
+      "step": 22300
+    },
+    {
+      "epoch": 2.191459179181138,
+      "grad_norm": 17.83564567565918,
+      "learning_rate": 1.4971372662704741e-05,
+      "loss": 2.4945,
+      "step": 22400
+    },
+    {
+      "epoch": 2.2012424790881964,
+      "grad_norm": 21.466899871826172,
+      "learning_rate": 1.4790186983620816e-05,
+      "loss": 2.5467,
+      "step": 22500
+    },
+    {
+      "epoch": 2.2012424790881964,
+      "eval_runtime": 182.8328,
+      "eval_samples_per_second": 111.807,
+      "eval_steps_per_second": 13.98,
+      "step": 22500
+    },
+    {
+      "epoch": 2.211025778995255,
+      "grad_norm": 17.91064453125,
+      "learning_rate": 1.4609001304536891e-05,
+      "loss": 2.5144,
+      "step": 22600
+    },
+    {
+      "epoch": 2.2208090789023136,
+      "grad_norm": 17.678396224975586,
+      "learning_rate": 1.4427815625452964e-05,
+      "loss": 2.5018,
+      "step": 22700
+    },
+    {
+      "epoch": 2.2305923788093724,
+      "grad_norm": 17.510461807250977,
+      "learning_rate": 1.4246629946369039e-05,
+      "loss": 2.4228,
+      "step": 22800
+    },
+    {
+      "epoch": 2.240375678716431,
+      "grad_norm": 24.923967361450195,
+      "learning_rate": 1.4065444267285114e-05,
+      "loss": 2.5249,
+      "step": 22900
+    },
+    {
+      "epoch": 2.2501589786234897,
+      "grad_norm": 17.82384490966797,
+      "learning_rate": 1.388425858820119e-05,
+      "loss": 2.4282,
+      "step": 23000
+    },
+    {
+      "epoch": 2.2501589786234897,
+      "eval_runtime": 182.0459,
+      "eval_samples_per_second": 112.29,
+      "eval_steps_per_second": 14.04,
+      "step": 23000
+    },
+    {
+      "epoch": 2.2599422785305485,
+      "grad_norm": 16.13028335571289,
+      "learning_rate": 1.3703072909117265e-05,
+      "loss": 2.4472,
+      "step": 23100
+    },
+    {
+      "epoch": 2.269725578437607,
+      "grad_norm": 15.137242317199707,
+      "learning_rate": 1.352188723003334e-05,
+      "loss": 2.5985,
+      "step": 23200
+    },
+    {
+      "epoch": 2.2795088783446658,
+      "grad_norm": 16.187530517578125,
+      "learning_rate": 1.3340701550949415e-05,
+      "loss": 2.4862,
+      "step": 23300
+    },
+    {
+      "epoch": 2.289292178251724,
+      "grad_norm": 18.84433937072754,
+      "learning_rate": 1.3159515871865488e-05,
+      "loss": 2.516,
+      "step": 23400
+    },
+    {
+      "epoch": 2.299075478158783,
+      "grad_norm": 20.209121704101562,
+      "learning_rate": 1.2978330192781563e-05,
+      "loss": 2.5031,
+      "step": 23500
+    },
+    {
+      "epoch": 2.299075478158783,
+      "eval_runtime": 181.9806,
+      "eval_samples_per_second": 112.331,
+      "eval_steps_per_second": 14.045,
+      "step": 23500
+    },
+    {
+      "epoch": 2.308858778065842,
+      "grad_norm": 67.4502182006836,
+      "learning_rate": 1.2797144513697637e-05,
+      "loss": 2.4491,
+      "step": 23600
+    },
+    {
+      "epoch": 2.3186420779729002,
+      "grad_norm": 14.940401077270508,
+      "learning_rate": 1.2615958834613712e-05,
+      "loss": 2.5669,
+      "step": 23700
+    },
+    {
+      "epoch": 2.328425377879959,
+      "grad_norm": 16.591793060302734,
+      "learning_rate": 1.2434773155529787e-05,
+      "loss": 2.4565,
+      "step": 23800
+    },
+    {
+      "epoch": 2.3382086777870175,
+      "grad_norm": 16.798791885375977,
+      "learning_rate": 1.2253587476445862e-05,
+      "loss": 2.4046,
+      "step": 23900
+    },
+    {
+      "epoch": 2.3479919776940763,
+      "grad_norm": 17.712255477905273,
+      "learning_rate": 1.2072401797361937e-05,
+      "loss": 2.4453,
+      "step": 24000
+    },
+    {
+      "epoch": 2.3479919776940763,
+      "eval_runtime": 182.0401,
+      "eval_samples_per_second": 112.294,
+      "eval_steps_per_second": 14.041,
+      "step": 24000
+    },
+    {
+      "epoch": 2.3577752776011347,
+      "grad_norm": 18.64284324645996,
+      "learning_rate": 1.1891216118278011e-05,
+      "loss": 2.3973,
+      "step": 24100
+    },
+    {
+      "epoch": 2.3675585775081935,
+      "grad_norm": 18.185895919799805,
+      "learning_rate": 1.1710030439194086e-05,
+      "loss": 2.5045,
+      "step": 24200
+    },
+    {
+      "epoch": 2.377341877415252,
+      "grad_norm": 23.201522827148438,
+      "learning_rate": 1.1528844760110163e-05,
+      "loss": 2.5402,
+      "step": 24300
+    },
+    {
+      "epoch": 2.3871251773223108,
+      "grad_norm": 21.606412887573242,
+      "learning_rate": 1.1347659081026236e-05,
+      "loss": 2.4285,
+      "step": 24400
+    },
+    {
+      "epoch": 2.3969084772293696,
+      "grad_norm": 16.318761825561523,
+      "learning_rate": 1.116647340194231e-05,
+      "loss": 2.5509,
+      "step": 24500
+    },
+    {
+      "epoch": 2.3969084772293696,
+      "eval_runtime": 182.0431,
+      "eval_samples_per_second": 112.292,
+      "eval_steps_per_second": 14.041,
+      "step": 24500
+    },
+    {
+      "epoch": 2.406691777136428,
+      "grad_norm": 17.779014587402344,
+      "learning_rate": 1.0985287722858386e-05,
+      "loss": 2.4245,
+      "step": 24600
+    },
+    {
+      "epoch": 2.416475077043487,
+      "grad_norm": 18.44321060180664,
+      "learning_rate": 1.080410204377446e-05,
+      "loss": 2.5223,
+      "step": 24700
+    },
+    {
+      "epoch": 2.4262583769505452,
+      "grad_norm": 24.017047882080078,
+      "learning_rate": 1.0622916364690535e-05,
+      "loss": 2.4846,
+      "step": 24800
+    },
+    {
+      "epoch": 2.436041676857604,
+      "grad_norm": 14.89560604095459,
+      "learning_rate": 1.044173068560661e-05,
+      "loss": 2.5922,
+      "step": 24900
+    },
+    {
+      "epoch": 2.445824976764663,
+      "grad_norm": 15.532561302185059,
+      "learning_rate": 1.0260545006522685e-05,
+      "loss": 2.3976,
+      "step": 25000
+    },
+    {
+      "epoch": 2.445824976764663,
+      "eval_runtime": 182.1033,
+      "eval_samples_per_second": 112.255,
+      "eval_steps_per_second": 14.036,
+      "step": 25000
+    },
+    {
+      "epoch": 2.4556082766717213,
+      "grad_norm": 18.041282653808594,
+      "learning_rate": 1.007935932743876e-05,
+      "loss": 2.4731,
+      "step": 25100
+    },
+    {
+      "epoch": 2.46539157657878,
+      "grad_norm": 13.40858268737793,
+      "learning_rate": 9.898173648354834e-06,
+      "loss": 2.4838,
+      "step": 25200
+    },
+    {
+      "epoch": 2.4751748764858386,
+      "grad_norm": 17.450841903686523,
+      "learning_rate": 9.71698796927091e-06,
+      "loss": 2.3999,
+      "step": 25300
+    },
+    {
+      "epoch": 2.4849581763928974,
+      "grad_norm": 17.556467056274414,
+      "learning_rate": 9.535802290186984e-06,
+      "loss": 2.3867,
+      "step": 25400
+    },
+    {
+      "epoch": 2.494741476299956,
+      "grad_norm": 18.578310012817383,
+      "learning_rate": 9.354616611103059e-06,
+      "loss": 2.4546,
+      "step": 25500
+    },
+    {
+      "epoch": 2.494741476299956,
+      "eval_runtime": 182.0338,
+      "eval_samples_per_second": 112.298,
+      "eval_steps_per_second": 14.041,
+      "step": 25500
+    },
+    {
+      "epoch": 2.5045247762070146,
+      "grad_norm": 14.936469078063965,
+      "learning_rate": 9.173430932019134e-06,
+      "loss": 2.5562,
+      "step": 25600
+    },
+    {
+      "epoch": 2.514308076114073,
+      "grad_norm": 17.527040481567383,
+      "learning_rate": 8.992245252935209e-06,
+      "loss": 2.4008,
+      "step": 25700
+    },
+    {
+      "epoch": 2.524091376021132,
+      "grad_norm": 12.91336727142334,
+      "learning_rate": 8.811059573851283e-06,
+      "loss": 2.4655,
+      "step": 25800
+    },
+    {
+      "epoch": 2.5338746759281907,
+      "grad_norm": 15.168461799621582,
+      "learning_rate": 8.629873894767358e-06,
+      "loss": 2.4468,
+      "step": 25900
+    },
+    {
+      "epoch": 2.543657975835249,
+      "grad_norm": 17.5390682220459,
+      "learning_rate": 8.448688215683433e-06,
+      "loss": 2.4836,
+      "step": 26000
+    },
+    {
+      "epoch": 2.543657975835249,
+      "eval_runtime": 182.1148,
+      "eval_samples_per_second": 112.248,
+      "eval_steps_per_second": 14.035,
+      "step": 26000
+    },
+    {
+      "epoch": 2.553441275742308,
+      "grad_norm": 15.126510620117188,
+      "learning_rate": 8.267502536599508e-06,
+      "loss": 2.387,
+      "step": 26100
+    },
+    {
+      "epoch": 2.5632245756493663,
+      "grad_norm": 15.374293327331543,
+      "learning_rate": 8.086316857515583e-06,
+      "loss": 2.3652,
+      "step": 26200
+    },
+    {
+      "epoch": 2.573007875556425,
+      "grad_norm": 15.498108863830566,
+      "learning_rate": 7.905131178431657e-06,
+      "loss": 2.4749,
+      "step": 26300
+    },
+    {
+      "epoch": 2.582791175463484,
+      "grad_norm": 16.221315383911133,
+      "learning_rate": 7.723945499347732e-06,
+      "loss": 2.4567,
+      "step": 26400
+    },
+    {
+      "epoch": 2.5925744753705424,
+      "grad_norm": 18.839122772216797,
+      "learning_rate": 7.542759820263806e-06,
+      "loss": 2.3554,
+      "step": 26500
+    },
+    {
+      "epoch": 2.5925744753705424,
+      "eval_runtime": 181.9597,
+      "eval_samples_per_second": 112.344,
+      "eval_steps_per_second": 14.047,
+      "step": 26500
+    },
+    {
+      "epoch": 2.6023577752776013,
+      "grad_norm": 22.626708984375,
+      "learning_rate": 7.361574141179882e-06,
+      "loss": 2.502,
+      "step": 26600
+    },
+    {
+      "epoch": 2.6121410751846597,
+      "grad_norm": 16.519880294799805,
+      "learning_rate": 7.180388462095957e-06,
+      "loss": 2.5034,
+      "step": 26700
+    },
+    {
+      "epoch": 2.6219243750917185,
+      "grad_norm": 27.421489715576172,
+      "learning_rate": 6.999202783012031e-06,
+      "loss": 2.5276,
+      "step": 26800
+    },
+    {
+      "epoch": 2.6317076749987773,
+      "grad_norm": 15.274630546569824,
+      "learning_rate": 6.8180171039281055e-06,
+      "loss": 2.4121,
+      "step": 26900
+    },
+    {
+      "epoch": 2.6414909749058357,
+      "grad_norm": 15.751582145690918,
+      "learning_rate": 6.636831424844181e-06,
+      "loss": 2.5799,
+      "step": 27000
+    },
+    {
+      "epoch": 2.6414909749058357,
+      "eval_runtime": 182.0873,
+      "eval_samples_per_second": 112.265,
+      "eval_steps_per_second": 14.037,
+      "step": 27000
+    },
+    {
+      "epoch": 2.651274274812894,
+      "grad_norm": 16.674850463867188,
+      "learning_rate": 6.455645745760255e-06,
+      "loss": 2.3872,
+      "step": 27100
+    },
+    {
+      "epoch": 2.661057574719953,
+      "grad_norm": 12.62803840637207,
+      "learning_rate": 6.27446006667633e-06,
+      "loss": 2.4,
+      "step": 27200
+    },
+    {
+      "epoch": 2.670840874627012,
+      "grad_norm": 18.055158615112305,
+      "learning_rate": 6.093274387592405e-06,
+      "loss": 2.4681,
+      "step": 27300
+    },
+    {
+      "epoch": 2.68062417453407,
+      "grad_norm": 17.21278190612793,
+      "learning_rate": 5.91208870850848e-06,
+      "loss": 2.5441,
+      "step": 27400
+    },
+    {
+      "epoch": 2.690407474441129,
+      "grad_norm": 20.945236206054688,
+      "learning_rate": 5.7309030294245544e-06,
+      "loss": 2.4388,
+      "step": 27500
+    },
+    {
+      "epoch": 2.690407474441129,
+      "eval_runtime": 182.1279,
+      "eval_samples_per_second": 112.24,
+      "eval_steps_per_second": 14.034,
+      "step": 27500
+    },
+    {
+      "epoch": 2.7001907743481874,
+      "grad_norm": 23.483661651611328,
+      "learning_rate": 5.549717350340629e-06,
+      "loss": 2.4589,
+      "step": 27600
+    },
+    {
+      "epoch": 2.7099740742552463,
+      "grad_norm": 17.954036712646484,
+      "learning_rate": 5.368531671256704e-06,
+      "loss": 2.4477,
+      "step": 27700
+    },
+    {
+      "epoch": 2.719757374162305,
+      "grad_norm": 16.187314987182617,
+      "learning_rate": 5.187345992172779e-06,
+      "loss": 2.4967,
+      "step": 27800
+    },
+    {
+      "epoch": 2.7295406740693635,
+      "grad_norm": 14.324910163879395,
+      "learning_rate": 5.006160313088854e-06,
+      "loss": 2.3921,
+      "step": 27900
+    },
+    {
+      "epoch": 2.7393239739764224,
+      "grad_norm": 20.81557846069336,
+      "learning_rate": 4.8249746340049285e-06,
+      "loss": 2.5201,
+      "step": 28000
+    },
+    {
+      "epoch": 2.7393239739764224,
+      "eval_runtime": 182.146,
+      "eval_samples_per_second": 112.229,
+      "eval_steps_per_second": 14.033,
+      "step": 28000
+    },
+    {
+      "epoch": 2.7491072738834808,
+      "grad_norm": 18.682844161987305,
+      "learning_rate": 4.643788954921003e-06,
+      "loss": 2.4325,
+      "step": 28100
+    },
+    {
+      "epoch": 2.7588905737905396,
+      "grad_norm": 16.227272033691406,
+      "learning_rate": 4.462603275837078e-06,
+      "loss": 2.3864,
+      "step": 28200
+    },
+    {
+      "epoch": 2.7686738736975984,
+      "grad_norm": 16.20302963256836,
+      "learning_rate": 4.281417596753152e-06,
+      "loss": 2.5296,
+      "step": 28300
+    },
+    {
+      "epoch": 2.778457173604657,
+      "grad_norm": 18.634096145629883,
+      "learning_rate": 4.100231917669228e-06,
+      "loss": 2.4514,
+      "step": 28400
+    },
+    {
+      "epoch": 2.7882404735117157,
+      "grad_norm": 13.040008544921875,
+      "learning_rate": 3.919046238585303e-06,
+      "loss": 2.3661,
+      "step": 28500
+    },
+    {
+      "epoch": 2.7882404735117157,
+      "eval_runtime": 181.9164,
+      "eval_samples_per_second": 112.37,
+      "eval_steps_per_second": 14.05,
+      "step": 28500
+    },
+    {
+      "epoch": 2.798023773418774,
+      "grad_norm": 14.142943382263184,
+      "learning_rate": 3.737860559501377e-06,
+      "loss": 2.5074,
+      "step": 28600
+    },
+    {
+      "epoch": 2.807807073325833,
+      "grad_norm": 17.934324264526367,
+      "learning_rate": 3.5566748804174523e-06,
+      "loss": 2.4224,
+      "step": 28700
+    },
+    {
+      "epoch": 2.8175903732328913,
+      "grad_norm": 14.450194358825684,
+      "learning_rate": 3.3754892013335267e-06,
+      "loss": 2.4949,
+      "step": 28800
+    },
+    {
+      "epoch": 2.82737367313995,
+      "grad_norm": 17.746837615966797,
+      "learning_rate": 3.194303522249602e-06,
+      "loss": 2.4153,
+      "step": 28900
+    },
+    {
+      "epoch": 2.8371569730470085,
+      "grad_norm": 13.962541580200195,
+      "learning_rate": 3.0131178431656763e-06,
+      "loss": 2.4804,
+      "step": 29000
+    },
+    {
+      "epoch": 2.8371569730470085,
+      "eval_runtime": 182.0262,
+      "eval_samples_per_second": 112.303,
+      "eval_steps_per_second": 14.042,
+      "step": 29000
+    },
+    {
+      "epoch": 2.8469402729540674,
+      "grad_norm": 16.669286727905273,
+      "learning_rate": 2.831932164081751e-06,
+      "loss": 2.5397,
+      "step": 29100
+    },
+    {
+      "epoch": 2.856723572861126,
+      "grad_norm": 15.421733856201172,
+      "learning_rate": 2.650746484997826e-06,
+      "loss": 2.4175,
+      "step": 29200
+    },
+    {
+      "epoch": 2.8665068727681846,
+      "grad_norm": 14.135702133178711,
+      "learning_rate": 2.4695608059139007e-06,
+      "loss": 2.5069,
+      "step": 29300
+    },
+    {
+      "epoch": 2.8762901726752435,
+      "grad_norm": 17.41412925720215,
+      "learning_rate": 2.2883751268299756e-06,
+      "loss": 2.3997,
+      "step": 29400
+    },
+    {
+      "epoch": 2.886073472582302,
+      "grad_norm": 14.824533462524414,
+      "learning_rate": 2.1071894477460504e-06,
+      "loss": 2.3945,
+      "step": 29500
+    },
+    {
+      "epoch": 2.886073472582302,
+      "eval_runtime": 181.9299,
+      "eval_samples_per_second": 112.362,
+      "eval_steps_per_second": 14.049,
+      "step": 29500
+    },
+    {
+      "epoch": 2.8958567724893607,
+      "grad_norm": 27.31865119934082,
+      "learning_rate": 1.926003768662125e-06,
+      "loss": 2.45,
+      "step": 29600
+    },
+    {
+      "epoch": 2.9056400723964195,
+      "grad_norm": 18.966655731201172,
+      "learning_rate": 1.7448180895781998e-06,
+      "loss": 2.3916,
+      "step": 29700
+    },
+    {
+      "epoch": 2.915423372303478,
+      "grad_norm": 18.538440704345703,
+      "learning_rate": 1.5636324104942746e-06,
+      "loss": 2.4625,
+      "step": 29800
+    },
+    {
+      "epoch": 2.9252066722105368,
+      "grad_norm": 21.757272720336914,
+      "learning_rate": 1.3824467314103494e-06,
+      "loss": 2.3722,
+      "step": 29900
+    },
+    {
+      "epoch": 2.934989972117595,
+      "grad_norm": 16.907358169555664,
+      "learning_rate": 1.201261052326424e-06,
+      "loss": 2.464,
+      "step": 30000
+    },
+    {
+      "epoch": 2.934989972117595,
+      "eval_runtime": 181.9148,
+      "eval_samples_per_second": 112.371,
+      "eval_steps_per_second": 14.051,
+      "step": 30000
+    },
+    {
+      "epoch": 2.944773272024654,
+      "grad_norm": 13.88399600982666,
+      "learning_rate": 1.0200753732424989e-06,
+      "loss": 2.5005,
+      "step": 30100
+    },
+    {
+      "epoch": 2.954556571931713,
+      "grad_norm": 19.77507781982422,
+      "learning_rate": 8.388896941585737e-07,
+      "loss": 2.3829,
+      "step": 30200
+    },
+    {
+      "epoch": 2.9643398718387712,
+      "grad_norm": 16.535932540893555,
+      "learning_rate": 6.577040150746485e-07,
+      "loss": 2.4788,
+      "step": 30300
+    },
+    {
+      "epoch": 2.9741231717458296,
+      "grad_norm": 15.027000427246094,
+      "learning_rate": 4.765183359907233e-07,
+      "loss": 2.5007,
+      "step": 30400
+    },
+    {
+      "epoch": 2.9839064716528885,
+      "grad_norm": 14.9392671585083,
+      "learning_rate": 2.953326569067981e-07,
+      "loss": 2.4847,
+      "step": 30500
+    },
+    {
+      "epoch": 2.9839064716528885,
+      "eval_runtime": 181.9853,
+      "eval_samples_per_second": 112.328,
+      "eval_steps_per_second": 14.045,
+      "step": 30500
+    },
+    {
+      "epoch": 2.9936897715599473,
+      "grad_norm": 15.128337860107422,
+      "learning_rate": 1.1414697782287289e-07,
+      "loss": 2.4209,
+      "step": 30600
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 30663,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.0701267610290972e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}