train_svamp_789_1760637886

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

• Loss: 0.6550
• Num Input Tokens Seen: 1271360

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: 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
0.5975	2.0	280	0.6352	126944
0.658	4.0	560	0.5462	253696
0.3964	6.0	840	0.4913	380768
0.3941	8.0	1120	0.5003	508032
0.2802	10.0	1400	0.5018	635616
0.1987	12.0	1680	0.5148	762944
0.1323	14.0	1960	0.5540	889952
0.1726	16.0	2240	0.6144	1016736
0.1144	18.0	2520	0.6350	1144352
0.1615	20.0	2800	0.6550	1271360

Framework versions

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