INDEX

Explanations

phrases related to looking at or examining something

oai_token-act-pair · gpt-3.5-turbo

phrases containing the word "look" followed by various prepositions or objects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

trap

-0.72

externalActionCode

-0.72

widget

-0.70

Cry

-0.67

enthal

-0.66

Skill

-0.66

enegger

-0.65

IER

-0.65

bur

-0.65

WARNING

-0.65

POSITIVE LOGITS

 them

0.77

 pictures

0.75

how

0.74

the

0.71

 these

0.71

hetically

0.70

awed

0.69

 things

0.68

 what

0.67

 screens

0.66

Activations Density 0.071%

phrases related to looking at or examining something

phrases containing the word "look" followed by various prepositions or objects

No Comments

No Known Activations