llama2-finetuned-squad

This model was trained from scratch on the squad dataset. It achieves the following results on the evaluation set:

• Loss: 1.1801

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: 2e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
1.4105	0.96	18	1.2821
1.7789	1.92	36	1.2163
1.7876	2.88	54	1.1932
1.6394	3.84	72	1.1824
1.2972	4.8	90	1.1801

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.0
• Datasets 2.1.0
• Tokenizers 0.13.3