Shin-YAM/Agent_try01

1---2base_model: Qwen/Qwen3-4B-Instruct-25073datasets:4- u-10bei/sft_alfworld_trajectory_dataset_v55- u-10bei/dbbench_sft_dataset_react_v46language:7- en8license: apache-2.09library_name: peft10pipeline_tag: text-generation11tags:12- lora13- agent14- tool-use15- alfworld16- dbbench17---18 19# ＜Qwen/Qwen3-4B-Instruct-2507/LoRA-combined_datasets＞20 21This repository provides a **LoRA adapter** fine-tuned from22**Qwen/Qwen3-4B-Instruct-2507** using **LoRA + Unsloth**.23 24This repository contains **LoRA adapter weights only**.25The base model must be loaded separately.26 27## Training Objective28 29This adapter is trained to improve **multi-turn agent task performance**30on ALFWorld (household tasks) and DBBench (database operations).31 32Loss is applied to **all assistant turns** in the multi-turn trajectory,33enabling the model to learn environment observation, action selection,34tool use, and recovery from errors.35 36## Training Configuration37 38- Base model: Qwen/Qwen3-4B-Instruct-250739- Method: LoRA (full precision base)40- Max sequence length: 204841- Epochs: 242- Learning rate: 2e-0643- LoRA: r=64, alpha=12844 45## Usage46 47```python48from transformers import AutoModelForCausalLM, AutoTokenizer49from peft import PeftModel50import torch51 52base = "Qwen/Qwen3-4B-Instruct-2507"53adapter = "your_id/your-repo"54 55tokenizer = AutoTokenizer.from_pretrained(base)56model = AutoModelForCausalLM.from_pretrained(57    base,58    torch_dtype=torch.float16,59    device_map="auto",60)61model = PeftModel.from_pretrained(model, adapter)62```63 64## Sources & Terms (IMPORTANT)65 66Training data: u-10bei/dbbench_sft_dataset_react_v467 68Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.69Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.70