Llama-3.2-1B-Instruct_MED_NLI

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on the zero_shot dataset. It achieves the following results on the evaluation set:

• Loss: 0.0173

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
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.0416	0.1176	1000	0.0616
0.0369	0.2352	2000	0.0368
0.0268	0.3528	3000	0.0269
0.0269	0.4704	4000	0.0256
0.026	0.5880	5000	0.0216
0.018	0.7056	6000	0.0206
0.0174	0.8232	7000	0.0185
0.0172	0.9408	8000	0.0174

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Datasets 3.4.1
• Tokenizers 0.21.0