Overview

DCAgent/a1-stack_bash_withtests_gpt5mini is an 8 billion parameter language model, fine-tuned from the Qwen/Qwen3-8B architecture. This model has been specialized through fine-tuning on the /e/scratch/jureap59/raoof1/sft_data/hf_hub/datasets--DCAgent--exp_rpt_stack-bash-withtests-gpt5mini_glm_4.7_traces_jupiter/snapshots/169e04b53c79e7edaebf1f46b3ef5ccf609fbfd5_thinking_preprocessed dataset.

Training Details

The model was trained with the following key hyperparameters:

• Learning Rate: 4e-05
• Optimizer: ADAMW_TORCH_FUSED with betas=(0.9, 0.98) and epsilon=1e-08
• Epochs: 7.0
• Batch Size: 1 (train), 8 (eval) with a total train batch size of 16 across 16 GPUs.
• Scheduler: Cosine learning rate scheduler with a warmup ratio of 0.1.

Intended Use

While specific intended uses and limitations are not detailed in the provided README, the fine-tuning dataset suggests an optimization for tasks involving bash scripting and test generation. Developers working on automated scripting or testing environments may find this model particularly useful.