gpt2_m080_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: 1.2099
• Accuracy: 0.6788

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
2.8882	0.0522	1000	2.4481	0.4477
1.9734	0.1043	2000	1.7975	0.5687
1.7272	0.1565	3000	1.6134	0.6017
1.6087	0.2086	4000	1.5135	0.6195
1.5337	0.2608	5000	1.4512	0.6313
1.4808	0.3129	6000	1.4058	0.6399
1.444	0.3651	7000	1.3705	0.6466
1.4094	0.4173	8000	1.3408	0.6524
1.385	0.4694	9000	1.3191	0.6566
1.364	0.5216	10000	1.2988	0.6608
1.3413	0.5737	11000	1.2813	0.6643
1.3267	0.6259	12000	1.2677	0.6669
1.3161	0.6780	13000	1.2534	0.6697
1.3083	0.7302	14000	1.2439	0.6717
1.2955	0.7824	15000	1.2366	0.6731
1.285	0.8345	16000	1.2262	0.6754
1.2796	0.8867	17000	1.2194	0.6767
1.271	0.9388	18000	1.2133	0.6780
1.2678	0.9910	19000	1.2101	0.6787

Framework versions

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