SamChat

SamChat is a merge of the following models

• Intel/neural-chat-7b-v3-3
• cognitivecomputations/samantha-mistral-7b

🧩 Configuration

models:
  - model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
# no params for base model
  - model: Intel/neural-chat-7b-v3-3
    parameters:
      weight: 0.55
      density: 0.46
  - model: cognitivecomputations/samantha-mistral-7b
    parameters:
      weight: 0.64
      density: 0.55
merge_method: dare_ties
base_model: macadeliccc/WestLake-7B-v2-laser-truthy-dpo
parameters:
  normalize: true
  int8_mask: true
dtype: float16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "fhai50032/SamChat"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	61.68
AI2 Reasoning Challenge (25-Shot)	62.20
HellaSwag (10-Shot)	81.88
MMLU (5-Shot)	59.70
TruthfulQA (0-shot)	52.89
Winogrande (5-shot)	72.14
GSM8k (5-shot)	41.24