Views
No views yet
traffic_law_round1（初版小規模資料集）對
yentinglin/Llama-3-Taiwan-8B-Instruct-rc2
進行微調，主要用於驗證 LLaMA-Factory 訓練流程。後續 Round 2 的兩個模型
（round2-sonnet46-lora / round2-gpt4o-baseline-lora）已取代本模型作為正式比較對象。yentinglin/Llama-3-Taiwan-8B-Instruct-rc2traffic_law_round1lora_target=all, rank=16, alpha=32, dropout=0.051from transformers import AutoModelForCausalLM, AutoTokenizer
2from peft import PeftModel
3
4base_model_id = "yentinglin/Llama-3-Taiwan-8B-Instruct-rc2"
5adapter_id = "traffic-legal-lm/round1-lora"
6
7tokenizer = AutoTokenizer.from_pretrained(base_model_id)
8model = AutoModelForCausalLM.from_pretrained(base_model_id, torch_dtype="bfloat16", device_map="auto")
9model = PeftModel.from_pretrained(model, adapter_id)