test_1M_1-2025-02-20-22-04

This model is a fine-tuned version of roneneldan/TinyStories-1M on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1219

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: 2.5e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
2.8112	0.5	297	2.7469
2.6839	1.0	594	2.5763
2.4208	1.5	891	2.4286
2.2249	2.0	1188	2.3169
2.2298	2.5	1485	2.2436
2.3641	3.0	1782	2.1996
2.1558	3.5	2079	2.1787
2.119	4.0	2376	2.1618
2.0972	4.5	2673	2.1469
2.0033	5.0	2970	2.1349
2.0667	5.5	3267	2.1275
2.081	6.0	3564	2.1219

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0