qwen25-7b-agent-exp08-B_v3_synthv1_dbv4

This model is a fine-tuned version of Qwen/Qwen2.5-7B-Instruct using LoRA + Unsloth.

This repository contains the full merged weights. No adapter loading is required.

Training Objective

This adapter 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.

Training Configuration

• Base model: Qwen/Qwen2.5-7B-Instruct
• Data sources: ../../03_data/prepared/alfworld_v3_fixed, ../../05_alfworld_sim/generated/alfworld_synth_v1, ../../03_data/prepared/dbbench_v4
• Method: LoRA (Unsloth)
• Max sequence length: 2048
• Epochs: 2
• Learning rate: 2e-06
• LoRA: r=64, alpha=128

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "curio184/qwen25-7b-agent-exp08-B_v3_synthv1_dbv4"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

Sources & Terms (IMPORTANT)

Training data: ../../03_data/prepared/alfworld_v3_fixed, ../../05_alfworld_sim/generated/alfworld_synth_v1, ../../03_data/prepared/dbbench_v4

Dataset License: MIT License. Compliance: Users must comply with the MIT license and the base model's original terms of use.