train_wic_456_1760637803

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

• Loss: 1.1970
• Num Input Tokens Seen: 7492784

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: 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
0.3472	2.0	2172	0.3455	748208
0.3119	4.0	4344	0.3360	1497968
0.3486	6.0	6516	0.3234	2247488
0.2924	8.0	8688	0.3515	2997264
0.3148	10.0	10860	0.4060	3745952
0.1728	12.0	13032	0.5276	4495360
0.0991	14.0	15204	0.7679	5244688
0.1849	16.0	17376	1.0419	5993648
0.4071	18.0	19548	1.1777	6743280
0.0315	20.0	21720	1.1970	7492784

Framework versions

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