CLIP-SAE-ViT-L-14

zer0int/CLIP-SAE-ViT-L-14

Zero-Shot Image Classification428M parameterstransformersLicense: mit
Tracked by OpenModelStats since November 8, 2026View on Hugging Face ↗
Downloads 30D
92
Total downloads
1,515
Gained 7D
—
Gained 30D
—
Likes
33
Rank · tracked models
#75,486
411(down 411 places)this week
Parameters
428M
Last source update
Jan 25, 2026

Downloads over time

Daily gain between consecutive observations

Likes over time

Rank among tracked models

Lower is better

Overview

CLIP-SAE-ViT-L-14 is a zero-shot image classification model published by zer0int. OpenModelStats has tracked the model since Nov 8, 2026, most recently observing it Nov 15, 2026.

Published
Dec 8, 2024
Last updated
Jan 25, 2026
Library
transformers
License
mit
Task
Zero-Shot Image Classification
Parameters
427,616,512
Spaces
—
Derivative models
—
Velocity 7D/day
—
transformerssafetensorsclipzero-shot-image-classificationdataset:zer0int/CLIP-adversarial-typographic-attack_text-imagedataset:SPRIGHT-T2I/spright_cocobase_model:openai/clip-vit-large-patch14base_model:finetune:openai/clip-vit-large-patch14license:mitendpoints_compatibleregion:us

Publisher

zer0intView publisher statistics →

Derived from openai/clip-vit-large-patch14 (reported by the source; parent not yet tracked).

Similar models

zer0int/CLIP-SAE-ViT-L-14 — Downloads, Growth & Stats | OpenModelStats