llama3-8b-classification-gpt4o-100k

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

• Loss: 2.0032

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
0.9183	1.0	237	1.6517
0.8583	2.0	474	1.6295
0.8179	3.0	711	1.6559
0.7533	4.0	948	1.6894
0.716	5.0	1185	1.7251
0.6876	6.0	1422	1.7830
0.6344	7.0	1659	1.8557
0.591	8.0	1896	1.9240
0.5677	9.0	2133	1.9842
0.5648	10.0	2370	2.0032

Framework versions

• PEFT 0.12.0
• Transformers 4.44.0
• Pytorch 2.4.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1