V0410MP2_neo

This model is a fine-tuned version of EleutherAI/gpt-neo-1.3B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 5.8587

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: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_steps: 20
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
14.6876	0.18	20	14.8238
15.6833	0.36	40	15.4468
12.87	0.54	60	10.7786
5.4953	0.73	80	4.9276
8.9421	0.91	100	8.9857
9.9807	1.09	120	9.7099
9.1809	1.27	140	9.2579
6.9349	1.45	160	5.9443
5.8015	1.63	180	5.9521
5.6365	1.81	200	5.7121
5.6957	1.99	220	5.8587

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.14.6
• Tokenizers 0.14.1