llama-7b-irony

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7877
• Accuracy: 0.6607
• F1: 0.4527

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	30	1.1347	0.5204	0.5253
No log	2.0	60	0.8200	0.6480	0.5385
No log	3.0	90	0.8212	0.6390	0.4505
No log	4.0	120	0.7801	0.6543	0.4634
No log	5.0	150	0.8322	0.6454	0.4160
No log	6.0	180	0.8362	0.6454	0.4135
No log	7.0	210	0.8376	0.6492	0.3956
No log	8.0	240	0.7727	0.6620	0.4646
No log	9.0	270	0.7935	0.6620	0.4536
No log	10.0	300	0.7877	0.6607	0.4527

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0