INDEX

Explanations

mentions of revealing hidden information or identities

oai_token-act-pair · gpt-3.5-turbo

references to personal identity and significant revelations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

oslav

-0.75

trak

-0.72

Reviewed

-0.70

 upkeep

-0.69

chairs

-0.68

HAM

-0.63

uated

-0.63

ucci

-0.62

assisted

-0.61

sta

-0.60

POSITIVE LOGITS

 secrets

0.94

 vulnerabilities

0.83

 clues

0.82

 whereabouts

0.78

 trove

0.77

 incrim

0.76

 WikiLeaks

0.73

 truths

0.73

 details

0.72

 Hidden

0.72

Activations Density 0.238%

mentions of revealing hidden information or identities

references to personal identity and significant revelations

No Comments

No Known Activations