INDEX

Explanations

words related to problem-solving or fixing issues

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "fixing" problems or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ographies

-0.81

NetMessage

-0.78

ographic

-0.75

ammy

-0.71

 Schwar

-0.66

ography

-0.64

 Flavoring

-0.64

ographical

-0.63

ographically

-0.62

 unprotected

-0.62

POSITIVE LOGITS

ysis

0.92

esty

0.84

ated

0.83

ayne

0.76

edo

0.76

lear

0.71

 fixing

0.71

ate

0.70

ating

0.70

hift

0.70

Activations Density 0.012%

words related to problem-solving or fixing issues

references to the concept of "fixing" problems or issues

No Comments

No Known Activations