saiga13summary_v21_FULL

This model is a fine-tuned version of meta-llama/Llama-2-13b-hf on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8578

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: 2
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 10
• total_train_batch_size: 20
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 200

Training results

Training Loss	Epoch	Step	Validation Loss
1.4321	0.7168	20	1.2582
1.2013	1.4337	40	1.1311
1.0033	2.1505	60	1.0325
0.8827	2.8674	80	0.9661
0.8191	3.5842	100	0.9235
0.762	4.3011	120	0.8912
0.7698	5.0179	140	0.8706
0.7386	5.7348	160	0.8638
0.7092	6.4516	180	0.8607
0.6884	7.1685	200	0.8578

Framework versions

• PEFT 0.12.1.dev0
• Transformers 4.45.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.17.1
• Tokenizers 0.19.1