Downloads · 30 days
15
33% of all-time downloads
backup_0202_llamafactory_AA_preference_l0_0_75-llava-mistral
htlou/backup_0202_llamafactory_AA_preference_l0_0_75-llava-mistral
backup_0202_llamafactory_AA_preference_l0_0_75-llava-mistral is a image-text-to-text model from htlou. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as other.
should probably proofread and complete it, then remove this comment. --
All-time downloads
46
Public
Parameters
7.6B
15.1 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors15.1 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- License
- other
- Model type
- llava_next
- Access
- Public
- Created
- Feb 3, 2025
- Updated
- Feb 3, 2025
- SHA
- 4f1eb59b
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
AA_preference_l0_0_75
This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the AA_preference_l0_0_75 dataset. It achieves the following results on the evaluation set:
- Loss: 0.5643
- Rewards/chosen: 1.3942
- Rewards/rejected: -0.7947
- Rewards/accuracies: 0.8000
- Rewards/margins: 2.1889
- Logps/rejected: -224.1609
- Logps/chosen: -246.0838
- Logits/rejected: -2.3264
- Logits/chosen: -2.3596
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 | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.5946 | 0.7463 | 50 | 0.5850 | 1.2899 | -0.0802 | 0.7333 | 1.3701 | -217.0163 | -247.1266 | -2.3881 | -2.4191 |
| 0.255 | 1.4925 | 100 | 0.5857 | 1.4526 | -0.5760 | 0.7958 | 2.0286 | -221.9741 | -245.4996 | -2.4099 | -2.4342 |
| 0.1492 | 2.2388 | 150 | 0.5706 | 1.4938 | -0.5466 | 0.7917 | 2.0403 | -221.6795 | -245.0877 | -2.3284 | -2.3634 |
| 0.1536 | 2.9851 | 200 | 0.5642 | 1.3949 | -0.7954 | 0.7917 | 2.1903 | -224.1678 | -246.0766 | -2.3262 | -2.3595 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- llava_next
- License
- other
- Created
- Feb 3, 2025
- Updated
- Feb 3, 2025