train_copa_101112_1760637986

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

• Loss: 1.1292
• Num Input Tokens Seen: 501344

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: 101112
• 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
0.2347	2.0	160	0.2371	50016
0.2247	4.0	320	0.2412	100256
0.2513	6.0	480	0.2325	150368
0.2255	8.0	640	0.2417	200480
0.086	10.0	800	0.3139	250560
0.0119	12.0	960	0.5935	300928
0.0092	14.0	1120	0.8717	350944
0.0002	16.0	1280	1.1099	401088
0.0001	18.0	1440	1.1466	451200
0.0001	20.0	1600	1.1292	501344

Framework versions

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