Text Generation
PEFT
Safetensors
English
jumplander
jx
qwen2.5
qwen2.5-coder
coding-agent
agentic-ai
software-engineering
repository-understanding
goal-grounding
tool-use
behavioral-policy
qlora
lora
conversational
Instructions to use jumplander/JX-Coder-7B-Agent-Behavior with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use jumplander/JX-Coder-7B-Agent-Behavior with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-Coder-7B-Instruct") model = PeftModel.from_pretrained(base_model, "jumplander/JX-Coder-7B-Agent-Behavior") - Notebooks
- Google Colab
- Kaggle
| { | |
| "name": "JX Coder 7B Agent", | |
| "developer": "JumpLander", | |
| "base_model": "Qwen/Qwen2.5-Coder-7B-Instruct", | |
| "training_method": "4-bit QLoRA", | |
| "mode": "full", | |
| "train_examples": 14265, | |
| "eval_examples": 751, | |
| "max_length": 1024, | |
| "lora_r": 16, | |
| "learning_rate": 0.0001, | |
| "metrics": { | |
| "train_runtime": 49003.1816, | |
| "train_samples_per_second": 0.291, | |
| "train_steps_per_second": 0.018, | |
| "total_flos": 5.5325230506554266e+17, | |
| "train_loss": 0.0687537745764738, | |
| "epoch": 1.0 | |
| }, | |
| "dataset_files": [ | |
| "goal_grounding_sample.jsonl", | |
| "identity_data.jsonl" | |
| ] | |
| } |