INDEX

Explanations

adjectives expressing strong opinions or beliefs

oai_token-act-pair · gpt-3.5-turbo

words that denote strong ideological commitment or resolve

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

hops

-1.02

ammy

-0.89

hazard

-0.77

nesota

-0.76

anders

-0.73

NetMessage

-0.72

 Crew

-0.69

 Disorder

-0.68

dule

-0.68

 Fields

-0.68

POSITIVE LOGITS

ly

0.96

wart

0.94

 staunch

0.93

SourceFile

0.79

 exponent

0.77

 pacif

0.76

ness

0.75

nesses

0.74

ãĥ¼ãĤ¯

0.73

 adherent

0.73

Activations Density 0.012%

adjectives expressing strong opinions or beliefs

words that denote strong ideological commitment or resolve

No Comments

No Known Activations