Finetuned-llama-11B-VI-Vqav2-v-0-1

This model is a fine-tuned version of meta-llama/Llama-3.2-11B-Vision-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5026

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: 2e-05
• train_batch_size: 1
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• 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: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.4838	0.0667	100	0.5338
0.5618	0.1333	200	0.5331
0.5285	0.2	300	0.5164
0.5217	0.2667	400	0.5143
0.5076	0.3333	500	0.5129
0.525	0.4	600	0.5107
0.5047	0.4667	700	0.5093
0.4999	0.5333	800	0.5082
0.5287	0.6	900	0.5069
0.5153	0.6667	1000	0.5060
0.5074	0.7333	1100	0.5062
0.5095	0.8	1200	0.5050
0.5237	0.8667	1300	0.5040
0.5024	0.9333	1400	0.5027
0.5076	1.0	1500	0.5026

Framework versions

• PEFT 0.14.0
• Transformers 4.49.0
• Pytorch 2.4.0+cu118
• Datasets 3.2.0
• Tokenizers 0.21.0