llama-3.2-350M-fourier

This model is a fine-tuned version of llama_small_config.json on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 3.7984

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: 0.0005
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 256
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 1000
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
6.2583	0.0754	1000	5.1105
4.7487	0.1508	2000	4.5762
4.4098	0.2262	3000	4.3661
4.2531	0.3016	4000	4.2368
4.1448	0.3770	5000	4.1446
4.0569	0.4524	6000	4.0733
3.9844	0.5278	7000	4.0040
3.9199	0.6032	8000	3.9439
3.8622	0.6786	9000	3.8956
3.8158	0.7541	10000	3.8521
3.7792	0.8295	11000	3.8213
3.7546	0.9049	12000	3.8034
3.7425	0.9803	13000	3.7984

Framework versions

• Transformers 4.46.2
• Pytorch 2.1.2+cu121
• Datasets 3.1.0
• Tokenizers 0.20.1