INDEX

Explanations

phrases or words related to confirming, reiterating, or reassuring a statement or fact

oai_token-act-pair · gpt-3.5-turbo

assertions of confirmation or reaffirmation in statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

isy

-0.71

oided

-0.70

enium

-0.68

cture

-0.68

bang

-0.66

gins

-0.66

iae

-0.66

quer

-0.63

oping

-0.62

rug

-0.61

POSITIVE LOGITS

 suspicions

0.92

 longstanding

0.80

 anecdotal

0.78

 assertions

0.74

 misconceptions

0.73

 assumptions

0.73

 rumours

0.72

avorite

0.71

 unequivocally

0.70

 observations

0.68

Activations Density 0.143%

phrases or words related to confirming, reiterating, or reassuring a statement or fact

assertions of confirmation or reaffirmation in statements

No Comments

No Known Activations