Feature Extraction
Transformers
Safetensors
sensenova_loop
text-generation
vision
vision-encoder
multimodal
qwen3_5
custom_code
Instructions to use gatilin/sensenova-loop-9b-vision-encoder with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use gatilin/sensenova-loop-9b-vision-encoder with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="gatilin/sensenova-loop-9b-vision-encoder", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("gatilin/sensenova-loop-9b-vision-encoder", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download validation.json from gatilin/sensenova-loop-9b-vision-encoder: direct link, hf CLI and curl.
- Browser
- Download file 2.53 kB
-
https://huggingface.co/gatilin/sensenova-loop-9b-vision-encoder/resolve/main/validation.json
- Command line
-
hf download hf://gatilin/sensenova-loop-9b-vision-encoder/validation.json
-
curl -L -o validation.json https://huggingface.co/gatilin/sensenova-loop-9b-vision-encoder/resolve/main/validation.json
2.53 kB
| { | |
| "passed": true, | |
| "strict_load": true, | |
| "model_sha256": "a657c406083ff2cb08b562e4eb819206498b51401e2554d01146405848a309a7", | |
| "device": "cpu", | |
| "dtype": "float32", | |
| "torch": "2.14.1", | |
| "transformers": "5.21.1", | |
| "processor": "Qwen2VLImageProcessor with source normalization and patch settings; testing capped at 224x224", | |
| "attention_method": "Compare explicit softmax(QK^T/sqrt(d))V to native SDPA after rotary embeddings, layers 0,13,26", | |
| "images": [ | |
| { | |
| "image": "000000000009.png", | |
| "image_sha256": "ec671f039c1e2b16ebad3cef3412bc61fdf4286a9b8b98c7c59f8bceb6ce5fbb", | |
| "grid_thw": [ | |
| [ | |
| 1, | |
| 14, | |
| 14 | |
| ] | |
| ], | |
| "last_hidden_state_shape": [ | |
| 196, | |
| 1152 | |
| ], | |
| "pooler_output_shape": [ | |
| 49, | |
| 4096 | |
| ], | |
| "pooler_mean": 0.005065567791461945, | |
| "attention_metrics": { | |
| "0": { | |
| "sdpa_max_absolute_error": 1.1920928955078125e-06, | |
| "row_sum_max_error": 5.364418029785156e-07, | |
| "mean_normalized_entropy": 0.7140987515449524 | |
| }, | |
| "13": { | |
| "sdpa_max_absolute_error": 2.384185791015625e-06, | |
| "row_sum_max_error": 3.5762786865234375e-07, | |
| "mean_normalized_entropy": 0.6647812128067017 | |
| }, | |
| "26": { | |
| "sdpa_max_absolute_error": 4.76837158203125e-06, | |
| "row_sum_max_error": 4.76837158203125e-07, | |
| "mean_normalized_entropy": 0.5053613781929016 | |
| } | |
| } | |
| }, | |
| { | |
| "image": "000000000030.png", | |
| "image_sha256": "1fbe2305d87d89ebd049c9d37e6795b76c10d7b0de5fab8c0cb58313cd1e4e11", | |
| "grid_thw": [ | |
| [ | |
| 1, | |
| 14, | |
| 14 | |
| ] | |
| ], | |
| "last_hidden_state_shape": [ | |
| 196, | |
| 1152 | |
| ], | |
| "pooler_output_shape": [ | |
| 49, | |
| 4096 | |
| ], | |
| "pooler_mean": 0.005192181095480919, | |
| "attention_metrics": { | |
| "0": { | |
| "sdpa_max_absolute_error": 1.430511474609375e-06, | |
| "row_sum_max_error": 4.76837158203125e-07, | |
| "mean_normalized_entropy": 0.7456417083740234 | |
| }, | |
| "13": { | |
| "sdpa_max_absolute_error": 3.814697265625e-06, | |
| "row_sum_max_error": 4.172325134277344e-07, | |
| "mean_normalized_entropy": 0.6580504179000854 | |
| }, | |
| "26": { | |
| "sdpa_max_absolute_error": 5.7220458984375e-06, | |
| "row_sum_max_error": 8.344650268554688e-07, | |
| "mean_normalized_entropy": 0.47674980759620667 | |
| } | |
| } | |
| } | |
| ] | |
| } | |