Output_llama70B_70-15-15

This model is a fine-tuned version of meta-llama/Llama-3.3-70B-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5734
• Balanced Accuracy: 0.7044
• Accuracy: 0.6923

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Balanced Accuracy	Accuracy
No log	1.0	46	0.7989	0.5634	0.5449
No log	2.0	92	0.6930	0.5440	0.5449
No log	3.0	138	0.6273	0.6321	0.6026
No log	4.0	184	0.6713	0.5862	0.5833
No log	5.0	230	0.6085	0.6298	0.6218
No log	6.0	276	0.6010	0.6623	0.6538
No log	7.0	322	0.5909	0.6800	0.6731
No log	8.0	368	0.5874	0.6646	0.6603
No log	9.0	414	0.5819	0.6722	0.6667
No log	10.0	460	0.5734	0.7044	0.6923

Framework versions

• PEFT 0.10.0
• Transformers 4.45.2
• Pytorch 2.5.1+cu124
• Datasets 2.18.0
• Tokenizers 0.20.3