INDEX

Explanations

photo captions and attributions

oai_token-act-pair · gpt-3.5-turbo

references to photo credits or attributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

iqueness

-0.81

oler

-0.81

odox

-0.76

 inver

-0.76

merce

-0.74

rency

-0.70

ngth

-0.70

ixel

-0.70

eele

-0.70

 pest

-0.68

POSITIVE LOGITS

 Provided

0.90

 Â©

0.83

 Courtesy

0.80

 Unknown

0.78

 COURT

0.74

 Getty

0.74

 Thumbnails

0.73

Anonymous

0.72

???

0.71

 Tens

0.70

Activations Density 0.029%

photo captions and attributions

references to photo credits or attributions

No Comments

No Known Activations