Downloads · 30 days
12
1% of all-time downloads
gemma-4-e4b-kinetics54K-SQ_FFT
THChou1220/gemma-4-e4b-kinetics54K-SQ_FFT
gemma-4-e4b-kinetics54K-SQ_FFT is a image-text-to-text model from THChou1220. 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.
Full fine-tune of google/gemma-4-e4b-it on AI-generated video data derived from Kinetics. - Dataset: bear7011/gemma-4-e4b-kinetics54K - Samples: 54,618 video instruction examples (80% for training, 10% for validation,…
All-time downloads
1.3K
Public
Parameters
8B
31.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors16 GB · 100%
At a glance
- Task
- Image-Text-to-Text
- Library
- transformers
- Model type
- gemma4
- Access
- Public
- Created
- Jun 27, 2026
- Updated
- Aug 24, 2026
- SHA
- e2f898ea
Try a prompt
Base models
Model card
From the Hugging Face model README
gemma4-e4b-kinetics54K_FFT
Full fine-tune of google/gemma-4-e4b-it on AI-generated video data derived from Kinetics.
Training
- Dataset:
bear7011/gemma-4-e4b-kinetics_54K - Samples: 54,618 video instruction examples (80% for training, 10% for validation, 10% for test)
- Method: full fine-tuning, no LoRA
- Precision: bfloat16
- GPUs: 4
- DeepSpeed: ZeRO-3 with CPU optimizer and parameter offload
- Epochs: 1
- Global steps: 1366
- Per-device batch size: 1
- Gradient accumulation steps: 8
- Optimizer: AdamW
- Learning rate: 5e-6
- Projector learning rate: 5e-6
- Image encoder learning rate: 5e-6
- Weight decay: 0.01
- Warmup ratio: 0.03
- LR scheduler: cosine
- Gradient checkpointing: enabled
- Max sequence length: 3072
- Final training loss: 0.3852
- Task
- Image-Text-to-Text
- Library
- transformers
- Type
- gemma4
- Created
- Jun 27, 2026
- Updated
- Aug 24, 2026