INDEX

Explanations

terms related to popular culture, specifically focusing on pop music

oai_token-act-pair · gpt-3.5-turbo

references to pop culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

RAW

-0.79

APH

-0.74

IVES

-0.72

BILITIES

-0.71

 ACTIONS

-0.71

perse

-0.66

 horrible

-0.63

adr

-0.63

 captcha

-0.62

 LORD

-0.62

POSITIVE LOGITS

ulates

1.00

ulating

0.96

corn

0.95

lar

0.89

ularity

0.87

ulated

0.87

 quiz

0.87

ulations

0.85

pop

0.84

pop

0.84

Activations Density 0.007%

terms related to popular culture, specifically focusing on pop music

references to pop culture

No Comments

No Known Activations