lewington
/

CLIP-ViT-L-scope

Model card Files Files and versions Community

lewington commited on 19 days ago

Commit

12ae9d2

•

1 Parent(s): 242c1e5

100 million

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -36,7 +36,7 @@ Training logs are available [via wandb](https://wandb.ai/lewington/ViT-L-14-laio
 - Training Dataset: [Laion-2b](https://huggingface.co/datasets/laion/laion2B-multi-joined-translated-to-en)
 - SAE Architecture: topk with k=32
 - Layer Location: always the residual stream
-- Training Checkpoints: every ~25 million tokens
 - Number of features per autoencoder: 65536 (expansion factor 16)
 ## Usage

 - Training Dataset: [Laion-2b](https://huggingface.co/datasets/laion/laion2B-multi-joined-translated-to-en)
 - SAE Architecture: topk with k=32
 - Layer Location: always the residual stream
+- Training Checkpoints: every ~100 million tokens
 - Number of features per autoencoder: 65536 (expansion factor 16)
 ## Usage