INDEX

Explanations

captions found in image galleries or news articles

oai_token-act-pair · gpt-3.5-turbo

instances of captions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abol

-0.87

bye

-0.84

ught

-0.80

axis

-0.77

ought

-0.74

ggies

-0.72

road

-0.70

omics

-0.70

nder

-0.70

wed

-0.68

POSITIVE LOGITS

 Caption

1.74

 Thumbnails

1.05

 Close

0.96

 Shutdown

0.94

 largeDownload

0.88

Thumbnail

0.86

 Shooting

0.85

 Transcript

0.84

 Solitaire

0.81

ï

0.80

Activations Density 0.009%

captions found in image galleries or news articles

instances of captions

No Comments

No Known Activations