train_rte_123_1760637669

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B-Instruct on the rte dataset. It achieves the following results on the evaluation set:

• Loss: 0.8256
• Num Input Tokens Seen: 6158144

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: 4
• eval_batch_size: 4
• seed: 123
• 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_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.1621	2.0	996	0.1556	615712
0.1128	4.0	1992	0.1588	1229280
0.1232	6.0	2988	0.1473	1843168
0.0777	8.0	3984	0.2131	2459808
0.1752	10.0	4980	0.3243	3076512
0.0029	12.0	5976	0.5350	3691744
0.0001	14.0	6972	0.7473	4306432
0.0001	16.0	7968	0.7903	4926336
0.0	18.0	8964	0.8237	5542368
0.0	20.0	9960	0.8256	6158144

Framework versions

• PEFT 0.17.1
• Transformers 4.51.3
• Pytorch 2.9.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4