train_copa_789_1760637873

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: 0.2308
• Num Input Tokens Seen: 563744

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: 0.03
• train_batch_size: 4
• eval_batch_size: 4
• seed: 789
• 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
0.2715	1.0	90	0.2760	28192
0.2409	2.0	180	0.2308	56192
0.2493	3.0	270	0.2417	84192
0.2453	4.0	360	0.2346	112544
0.2262	5.0	450	0.2422	140960
0.2318	6.0	540	0.2319	169216
0.2352	7.0	630	0.2362	197248
0.2421	8.0	720	0.2336	225440
0.2379	9.0	810	0.2355	253632
0.2244	10.0	900	0.2314	281984
0.2442	11.0	990	0.2330	310176
0.2306	12.0	1080	0.2360	338400
0.2275	13.0	1170	0.2338	366688
0.2299	14.0	1260	0.2351	394752
0.2361	15.0	1350	0.2367	422912
0.236	16.0	1440	0.2350	451008
0.2338	17.0	1530	0.2350	479104
0.2308	18.0	1620	0.2360	507392
0.2297	19.0	1710	0.2361	535584
0.2245	20.0	1800	0.2376	563744

Framework versions

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