train_cola_789_1760637929

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: 0.4779
• Num Input Tokens Seen: 6517120

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
0.2744	2.0	3420	0.2695	651168
0.1754	4.0	6840	0.2735	1302560
0.2969	6.0	10260	0.2615	1954368
0.3627	8.0	13680	0.2671	2606240
0.3037	10.0	17100	0.2798	3257824
0.1524	12.0	20520	0.2980	3909216
0.151	14.0	23940	0.3477	4561024
0.1468	16.0	27360	0.4295	5213728
0.1024	18.0	30780	0.4624	5864864
0.1269	20.0	34200	0.4779	6517120

Framework versions

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