INDEX

Explanations

proper nouns related to public figures

oai_token-act-pair · gpt-3.5-turbo

references to leaders or prominent figures in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

sav

-0.73

alien

-0.72

ships

-0.71

sburgh

-0.68

Ī

-0.66

race

-0.64

âĺħ

-0.63

=-=-=-=-

-0.63

drawn

-0.62

selected

-0.62

POSITIVE LOGITS

asar

0.92

utical

0.92

utic

0.92

utics

0.84

illance

0.82

estro

0.81

ema

0.80

emon

0.79

ovic

0.77

verend

0.74

Activations Density 0.020%

proper nouns related to public figures

references to leaders or prominent figures in the text

No Comments

No Known Activations