metadata
license: mit
language:
- en
Model Card for Model ID
Finetuned mistralai/Mistral-7B-Instruct-v0.2 on TESTtm7873/ChatCat dataset
Model Description
Developed of our VCC project Finetuned with QLoRA
To use it:
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
base_model_id = "mistralai/Mistral-7B-Instruct-v0.2"
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16
)
base_model = AutoModelForCausalLM.from_pretrained(
base_model_id, # Mistral, same as before
quantization_config=bnb_config, # Same quantization config as before
device_map="auto",
trust_remote_code=True,
)
eval_tokenizer = AutoTokenizer.from_pretrained(base_model_id, add_bos_token=True, trust_remote_code=True)
from peft import PeftModel
ft_model = PeftModel.from_pretrained(base_model, "mistral-journal-finetune/checkpoint-150")
eval_prompt = "You have the softest fur."
model_input = eval_tokenizer(eval_prompt, return_tensors="pt").to("cuda")
ft_model.eval()
with torch.no_grad():
print(eval_tokenizer.decode(ft_model.generate(**model_input, max_new_tokens=100, repetition_penalty=1.15)[0], skip_special_tokens=True))```
- **Developed by:** testtm
- **Funded by:** testtm
- **Model type:** Mistral
- **Language:** English
- **Finetuned from model:** mistralai/Mistral-7B-Instruct-v0.2