Meta-Llama-3-8B-Instruct-enzyme-prediction

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 3.8412

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: 0.0001
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
4.0817	0.1214	200	3.9220
3.9957	0.2427	400	3.8707
3.9832	0.3641	600	3.8652
4.0226	0.4854	800	3.8589
3.8973	0.6068	1000	3.8551
4.0458	0.7281	1200	3.8491
3.9373	0.8495	1400	3.8448
3.9891	0.9708	1600	3.8412

Framework versions

• PEFT 0.17.1
• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1