htlou/mm-interp-RLAIF-V-Cosi-q0_50

RLAIF-V-Cosi-q0_50

This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the RLAIF-V-Cosi-q0_50 dataset. It achieves the following results on the evaluation set:

• Loss: 1.0423

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-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
1.1387	0.3478	50	1.1294
1.0986	0.6957	100	1.0838
0.9413	1.0435	150	1.0630
0.9147	1.3913	200	1.0510
0.9097	1.7391	250	1.0377
0.8125	2.0870	300	1.0442
0.8092	2.4348	350	1.0428
0.8049	2.7826	400	1.0423

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.20.3