A full model fine-tuned from melon1891/agentbench-qwen3-4b-2stage-reasoning-20260228 using LoRA + Unsloth, with the adapter merged into the base model.

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.

Training Configuration

Base model: melon1891/agentbench-qwen3-4b-2stage-reasoning-20260228
Method: LoRA (merged into base)
Max sequence length: 8192
Epochs: 3
Learning rate: 1e-06
LoRA: r=16, alpha=32

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("melon1891/agentbench-qwen3-4b-3stage-db-20260301-lr1e6")
tokenizer = AutoTokenizer.from_pretrained("melon1891/agentbench-qwen3-4b-3stage-db-20260301-lr1e6")

Sources & Terms (IMPORTANT)

Training data: melon1891/alfworld-correction-sft-20260301-45

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.