rationale_model_e3_save5000_rp_f1

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.9372

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
1.7719	0.1908	1000	1.9372
1.3947	0.3816	2000	1.9960
1.0148	0.5724	3000	2.1311
0.6554	0.7632	4000	2.4018
0.382	0.9540	5000	2.7497
0.1972	1.1448	6000	2.9048
0.1697	1.3356	7000	3.0675
0.1468	1.5264	8000	3.1610
0.1288	1.7172	9000	3.3265
0.1163	1.9080	10000	3.4370
0.0983	2.0988	11000	3.5843
0.097	2.2896	12000	3.6571
0.0932	2.4804	13000	3.7338
0.0892	2.6712	14000	3.8644
0.0867	2.8620	15000	3.9677

Framework versions

• Transformers 4.45.0
• Pytorch 2.3.0
• Datasets 2.14.4
• Tokenizers 0.20.3