gpt2_m050_tiny-stories_1024

This model is a fine-tuned version of on the roneneldan/TinyStories dataset. It achieves the following results on the evaluation set:

• Loss: 1.2035
• Accuracy: 0.6795

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
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.9083	0.0525	1000	2.4374	0.4486
1.9644	0.1049	2000	1.7837	0.5703
1.7149	0.1574	3000	1.5991	0.6031
1.5979	0.2099	4000	1.5038	0.6204
1.5248	0.2623	5000	1.4431	0.6322
1.4723	0.3148	6000	1.3973	0.6411
1.4339	0.3672	7000	1.3621	0.6475
1.406	0.4197	8000	1.3340	0.6530
1.3764	0.4722	9000	1.3089	0.6579
1.3561	0.5246	10000	1.2903	0.6618
1.3357	0.5771	11000	1.2739	0.6649
1.3213	0.6296	12000	1.2586	0.6680
1.3081	0.6820	13000	1.2466	0.6704
1.2962	0.7345	14000	1.2362	0.6726
1.2867	0.7869	15000	1.2277	0.6744
1.2755	0.8394	16000	1.2186	0.6762
1.2709	0.8919	17000	1.2117	0.6776
1.2611	0.9443	18000	1.2070	0.6787
1.2628	0.9968	19000	1.2035	0.6795

Framework versions

• Transformers 4.42.3
• Pytorch 2.2.2+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1