graf/Qwen3-4B-SFT-science-1e-5

<!-- 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. -->

Qwen3-4B-SFT-science-1e-5

This model is a fine-tuned version of Qwen/Qwen3-4B on the dolcisciencetrain dataset. It achieves the following results on the evaluation set:

• —Loss: 0.6778

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: 1e-05
• —trainbatchsize: 2
• —evalbatchsize: 2
• —seed: 42
• —distributed_type: multi-GPU
• —num_devices: 4
• —gradientaccumulationsteps: 16
• —totaltrainbatch_size: 128
• —totalevalbatch_size: 8
• —optimizer: Use OptimizerNames.ADAMWTORCHFUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• —lrschedulertype: cosine
• —lrschedulerwarmup_steps: 0.05
• —num_epochs: 3.0

Training results

Framework versions

• —Transformers 5.0.0
• —Pytorch 2.10.0+cu128
• —Datasets 4.0.0
• —Tokenizers 0.22.2