Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_8-mlp-98

Model

0

stars

5

commits

5

linked in READMEs

Apr 18, 2025

updated

Browse cluster: CLIP Transcoder Variants →

README

CLIP Transcoder Checkpoint

This model is a Transcoder trained on CLIP's internal representations.

Model Details

Architecture

Layer: 8
Layer Type: ln2.hook_normalized
Model: laion/CLIP-ViT-B-32-DataComp.XL-s13B-b90K
Dictionary Size: 49152
Input Dimension: 768
Expansion Factor: 64
CLS Token Only: False

Training

Learning Rate: 0.0005932154848461442
Batch Size: 4096
Context Size: 50

Sparsity

L0 (Active Features): 1024

Additional Information

Wandb Run: https://wandb.ai/perceptual-alignment/openclip-transcoders/runs/zh2kpkp6/
Random Seed: 42

Contributors

5 commits

Similar repos

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_9-mlp-98

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_2-mlp-93

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_10-mlp-97

1

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_6-mlp-94

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_5-mlp-91

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_3-mlp-90

0

Prisma-Multimodal/CLIP-transcoder-topk-768-x64-all_patches_0-mlp-96

0

Prisma-Multimodal/CLIP-transcoder-topk-256-x64-all_patches_1-mlp-94

0

Linked in READMEs

Prisma-Multimodal/ViT-Prisma

ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).

388

soniajoseph/ViT-Prisma

ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).

388

leeloolee/df-prisma

0

natihash/my_prisma

0

alexlav17/ViT-Prisma-fix

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_8-mlp-98

Model

0

stars

5

commits

5

linked in READMEs

Apr 18, 2025

updated

Browse cluster: CLIP Transcoder Variants →

README

CLIP Transcoder Checkpoint

This model is a Transcoder trained on CLIP's internal representations.

Model Details

Architecture

Layer: 8
Layer Type: ln2.hook_normalized
Model: laion/CLIP-ViT-B-32-DataComp.XL-s13B-b90K
Dictionary Size: 49152
Input Dimension: 768
Expansion Factor: 64
CLS Token Only: False

Training

Learning Rate: 0.0005932154848461442
Batch Size: 4096
Context Size: 50

Sparsity

L0 (Active Features): 1024

Additional Information

Wandb Run: https://wandb.ai/perceptual-alignment/openclip-transcoders/runs/zh2kpkp6/
Random Seed: 42

Similar repos

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_9-mlp-98

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_2-mlp-93

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_10-mlp-97

1

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_6-mlp-94

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_5-mlp-91

0

Prisma-Multimodal/CLIP-transcoder-topk-1024-x64-all_patches_3-mlp-90

0

Prisma-Multimodal/CLIP-transcoder-topk-768-x64-all_patches_0-mlp-96

0

Prisma-Multimodal/CLIP-transcoder-topk-256-x64-all_patches_1-mlp-94

0

Linked in READMEs

Prisma-Multimodal/ViT-Prisma

ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).

388

soniajoseph/ViT-Prisma

ViT Prisma is a mechanistic interpretability library for Vision and Video Transformers (ViTs).

388

leeloolee/df-prisma

0

natihash/my_prisma

0

alexlav17/ViT-Prisma-fix

0

Contributors

5 commits