INDEX

Explanations

mentions of individuals who are knowledgeable or aware of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "know" and its variations in the text

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

interstitial

-0.83

otion

-0.82

oples

-0.78

ermanent

-0.77

phrine

-0.77

isco

-0.76

ItemTracker

-0.76

anmar

-0.73

plet

-0.71

phabet

-0.70

POSITIVE LOGITS

ledged

1.25

ledge

1.14

how

0.87

lege

0.83

LED

0.80

 beforehand

0.78

ingly

0.77

 firsthand

0.77

 exactly

0.76

how

0.76

Activations Density 0.070%

mentions of individuals who are knowledgeable or aware of something

instances of the word "know" and its variations in the text

No Comments

No Known Activations