train_mnli_42_1773148412

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

• Loss: 0.1012
• Num Input Tokens Seen: 191491960

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.1876	0.2500	11045	0.1526	9565376
0.0607	0.5000	22090	0.1448	19168640
0.1762	0.7500	33135	0.1204	28714304
0.1059	1.0000	44180	0.1136	38289824
0.0964	1.2500	55225	0.1168	47877216
0.0757	1.5000	66270	0.1079	57416032
0.069	1.7500	77315	0.1052	66982176
0.0256	2.0000	88360	0.1062	76602496
0.1102	2.2501	99405	0.1081	86154496
0.0977	2.5001	110450	0.1032	95709312
0.0778	2.7501	121495	0.1012	105304960
0.1157	3.0001	132540	0.1026	114898176
0.0476	3.2501	143585	0.1056	124468928
0.023	3.5001	154630	0.1043	134028992
0.0194	3.7501	165675	0.1036	143607232
0.1094	4.0001	176720	0.1035	153206432
0.0685	4.2501	187765	0.1057	162770528
0.0223	4.5001	198810	0.1050	172345120
0.1275	4.7501	209855	0.1057	181948192

Framework versions

• PEFT 0.17.1
• Transformers 4.51.3
• Pytorch 2.10.0+cu128
• Datasets 4.0.0
• Tokenizers 0.21.4