vicuna-adv-robust-u20-sft-lora

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.2191

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 16
• total_train_batch_size: 256
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
2.5536	0.12	1	2.4988
2.5536	1.1	2	2.5688
2.5536	2.08	3	2.5704
2.5536	3.05	4	2.6317
2.5982	4.03	5	2.5004
2.5982	5.13	7	2.5414
2.5982	6.11	8	2.4437
2.5982	7.08	9	2.4887
2.5575	8.06	10	2.5658
2.5575	9.04	11	2.5603
2.5575	10.14	13	2.4061
2.5575	11.11	14	2.4262
2.5037	12.09	15	2.4319
2.5037	13.07	16	2.3978
2.5037	14.05	17	2.3657
2.5037	15.14	19	2.3580
2.4117	16.12	20	2.3231
2.4117	17.1	21	2.3328
2.4117	18.08	22	2.3051
2.4117	19.05	23	2.2653

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0a0+32f93b1
• Datasets 2.14.6
• Tokenizers 0.14.1