INDEX

Explanations

mentions of public appearances or appearances in events

oai_token-act-pair · gpt-3.5-turbo

instances of the word "appearance" in various contexts

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

cise

-0.86

rals

-0.83

EStreamFrame

-0.75

regate

-0.71

lethal

-0.71

ced

-0.70

brace

-0.69

omb

-0.69

HQ

-0.66

zb

-0.65

POSITIVE LOGITS

 appearance

1.17

 appearances

1.10

 Appearance

1.07

 flaws

0.75

earance

0.74

lihood

0.72

 Discrimination

0.72

 resemblance

0.70

iP

0.70

 contestant

0.70

Activations Density 0.019%

mentions of public appearances or appearances in events

instances of the word "appearance" in various contexts

No Comments

No Known Activations