llama3-sudo-sanity

This model is a fine-tuned version of meta-llama/Meta-Llama-3-8B on the HuggingFaceH4/ultrachat_200k dataset. It achieves the following results on the evaluation set:

• Loss: 1.1030

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
1.8735	0.9899	49	1.8325
1.8231	2.0	99	1.7239
1.7516	2.9899	148	1.6330
1.6586	4.0	198	1.5280
1.5571	4.9899	247	1.4166
1.4677	6.0	297	1.3068
1.3422	6.9899	346	1.2082
1.2609	8.0	396	1.1378
1.1647	8.9899	445	1.1074
1.1571	9.8990	490	1.1030

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.1.2
• Datasets 2.20.0
• Tokenizers 0.19.1