Meta-Llama-3-8B-Instruct-Verifier-logging

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

• Loss: 0.0768

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: 0.0003
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 32
• total_train_batch_size: 512
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
0.3094	0.2708	100	0.1257
0.099	0.5415	200	0.0854
0.0865	0.8123	300	0.0823
0.0842	1.0830	400	0.0813
0.082	1.3538	500	0.0808
0.0797	1.6245	600	0.0816
0.077	1.8953	700	0.0820
0.0771	2.1660	800	0.0828
0.0752	2.4368	900	0.0773
0.0749	2.7075	1000	0.0768
0.0713	2.9783	1100	0.0768

Framework versions

• PEFT 0.10.0
• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.14.7
• Tokenizers 0.19.1