README.md

---
base_model: Qwen/Qwen3-4B-Instruct-2507
datasets:
- u-10bei/sft_alfworld_trajectory_dataset_v5
- u-10bei/dbbench_sft_dataset_react_v4

language:
- en
license: apache-2.0
library_name: peft
pipeline_tag: text-generation
tags:
- lora
- agent
- tool-use
- alfworld
- dbbench
---

# qwen3-4b-agent-trajectory-lora

This repository provides a **LoRA adapter** fine-tuned from
**Qwen/Qwen3-4B-Instruct-2507** using **LoRA + Unsloth**.

This repository contains **LoRA adapter weights only**.
The base model must be loaded separately.

## Training Objective

This adapter is trained on **mixed multi-turn agent trajectories**
from:

- ALFWorld datasets
- DBBench datasets

The datasets are mixed with:

- ALF ratio: 0.7
- DB ratio: 0.3

Data preprocessing includes:

- Marker normalization (THOUGHT / ACTION / ANSWER)
- Removal of unnecessary task-completion lines
- Optional ACTION-only training
- Optional CoT masking
- Assistant-only loss

---

## Training Configuration

- Base model: Qwen/Qwen3-4B-Instruct-2507
- Method: LoRA (full precision base)
- Max sequence length: 2048
- Epochs: 2
- Learning rate: 1e-05
- LoRA: r=128, alpha=256

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = "takayosh/mix"

tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
    base,
    torch_dtype=torch.float16,
    device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)
```

## Sources & Terms (IMPORTANT)

Training data: - u-10bei/sft_alfworld_trajectory_dataset_v5
- u-10bei/dbbench_sft_dataset_react_v4


Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.
Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.