train_apps_1756979624

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

• Loss: 0.6325
• Num Input Tokens Seen: 698437632

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: 2
• eval_batch_size: 2
• 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: 10.0

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.7169	0.5	26377	0.6975	35040112
0.6416	1.0	52754	0.6713	69817408
0.5317	1.5	79131	0.6545	104642736
0.5538	2.0	105508	0.6427	139606336
0.5162	2.5	131885	0.6355	174505936
0.6337	3.0	158262	0.6286	209511968
0.634	3.5	184639	0.6265	244473968
0.5812	4.0	211016	0.6209	279435760
0.5908	4.5	237393	0.6223	314390208
0.5967	5.0	263770	0.6176	349245872
0.5625	5.5	290147	0.6208	384116416
0.8306	6.0	316524	0.6158	419077696
0.4858	6.5	342901	0.6193	453993952
0.6255	7.0	369278	0.6160	488971616
0.5292	7.5	395655	0.6242	523844752
0.6756	8.0	422032	0.6216	558784736
0.4672	8.5	448409	0.6299	593674416
0.4037	9.0	474786	0.6275	628591376
0.7077	9.5	501163	0.6330	663526640
0.3427	10.0	527540	0.6325	698437632

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1