htlou/mm-interp-RLAIF-V_L0-q0_25

RLAIF-V_L0-q0_25

This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the RLAIF-V_L0-q0_25 dataset. It achieves the following results on the evaluation set:

• Loss: 1.0725

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
1.1353	0.6944	50	1.1090
0.9543	1.3889	100	1.0778
0.8627	2.0833	150	1.0669
0.8551	2.7778	200	1.0724

Framework versions

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