Harshatheeswar/babylama-attention_heads12

<!-- 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. -->

babylama-attention_heads12

This model is a fine-tuned version of Harshatheeswar/babylama-hidden_sizes768 on the None dataset. It achieves the following results on the evaluation set:

• —Loss: 3.8103

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: 5e-05
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 8
• —totaltrainbatch_size: 64
• —optimizer: Use adamwtorch with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: cosine
• —num_epochs: 5
• —mixedprecisiontraining: Native AMP

Training results

Framework versions

• —Transformers 4.46.2
• —Pytorch 2.5.0+cu121
• —Datasets 3.1.0
• —Tokenizers 0.20.3