train_cola_456_1760637816

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

• Loss: 0.2583
• Num Input Tokens Seen: 7334376

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: 456
• 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
0.2719	1.0	1924	0.2611	366712
0.2655	2.0	3848	0.2600	734016
0.2503	3.0	5772	0.2597	1100824
0.199	4.0	7696	0.2613	1467248
0.2499	5.0	9620	0.2612	1834568
0.2738	6.0	11544	0.2595	2201464
0.2477	7.0	13468	0.2589	2568040
0.1865	8.0	15392	0.2600	2934360
0.3004	9.0	17316	0.2593	3301448
0.2937	10.0	19240	0.2602	3668312
0.265	11.0	21164	0.2593	4034856
0.2789	12.0	23088	0.2589	4401344
0.2578	13.0	25012	0.2591	4767736
0.3012	14.0	26936	0.2584	5134344
0.2325	15.0	28860	0.2585	5501408
0.2727	16.0	30784	0.2583	5867920
0.2254	17.0	32708	0.2586	6234920
0.2709	18.0	34632	0.2587	6601944
0.2282	19.0	36556	0.2587	6968096
0.2455	20.0	38480	0.2586	7334376

Framework versions

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