t2ance/atlas-sft_qwen3_8b_pattern_teacher_gpqa_firstcall_noconf

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

sftqwen38bpatternteachergpqafirstcall_noconf

This model is a fine-tuned version of Qwen/Qwen3-8B on the patternteachergpqa_firstcall dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 0.0002
• —trainbatchsize: 2
• —evalbatchsize: 8
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 3
• —totaltrainbatch_size: 6
• —totalevalbatch_size: 24
• —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_ratio: 0.05
• —num_epochs: 1

Training results

Framework versions

• —PEFT 0.18.1
• —Transformers 4.57.6
• —Pytorch 2.10.0+cu128
• —Datasets 4.0.0
• —Tokenizers 0.22.2