Upload README.md with huggingface_hub

README.md

@@ -1,21 +1,75 @@
 ---
 base_model: Qwen/Qwen3-4B-Instruct-2507
-tags:
-- text-generation-inference
-- transformers
-- unsloth
-- qwen3
-license: apache-2.0
+datasets:
+- u-10bei/sft_alfworld_trajectory_dataset_v3
+- u-10bei/dbbench_sft_dataset_react_v4
 language:
 - en
+license: apache-2.0
+pipeline_tag: text-generation
+tags:
+- unsloth
+- agent
+- tool-use
+- alfworld
+- dbbench
 ---
 
-# Uploaded finetuned  model
+# Qwen3-4B Agent Trajectory (v4)
+
+This repository provides a **fully merged model** fine-tuned from **Qwen/Qwen3-4B-Instruct-2507** using Unsloth.
+
+Unlike standard adapter repositories, this repository contains the **merged weights**, meaning you do not need to load the base model separately.
+
+## Training Objective
+
+This model is trained to improve **multi-turn agent task performance**
+on ALFWorld (household tasks) and DBBench (database operations).
+
+Loss is applied to **all assistant turns** in the multi-turn trajectory,
+enabling the model to learn environment observation, action selection,
+tool use, and recovery from errors.
+
+## Data Processing
+
+- Train/Validation Split: 95% / 5%
+- Random Seed: 3407 (used for shuffling and initialization)
+- Loss Masking: Loss was computed only on the assistant's responses. User prompts and observations were masked during training (`train_on_responses_only` was applied to `<|im_start|>assistant\n`).
+
+## Training Configuration
+- **Base model**: Qwen/Qwen3-4B-Instruct-2507
+- **Method**: LoRA + Unsloth (Merged in 16-bit)
+- **Max sequence length**: 8192
+- **Epochs**: 1
+- **Learning rate**: 1e-05
+- **LoRA**: r=16, alpha=32
+- **PER_DEVICE_TRAIN_BATCH_SIZE** = 8
+- **GRAD_ACCUM** = 2
+- **WARMUP_RATIO** = 0.1
+- **WEIGHT_DECAY** = 0.05
+- **NEFTUNE_NOISE_ALPHA** = 5.0
+- **VAL_RATIO** = 0.05
+
+## Usage
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+model_id = "choco800/qwen3-4b-agent-v4"
 
-- **Developed by:** choco800
-- **License:** apache-2.0
-- **Finetuned from model :** Qwen/Qwen3-4B-Instruct-2507
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+```
+## Sources & Terms (IMPORTANT)
 
-This qwen3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
+Training data:
+- u-10bei/sft_alfworld_trajectory_dataset_v3
+- u-10bei/dbbench_sft_dataset_react_v4
 
-[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
+Dataset License: MIT License. These datasets are used and distributed under the terms of the MIT License.
+Compliance: Users must comply with the dataset licenses and the base model's original terms of use (Apache 2.0).