INDEX

Explanations

official titles or positions

oai_token-act-pair · gpt-3.5-turbo

references to officials or official statements

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

ï¸

-1.00

ĸļ

-0.91

esville

-0.84

xual

-0.73

âķĲ

-0.72

âķĲâķĲ

-0.72

oppers

-0.72

avery

-0.72

 Bengal

-0.70

Beg

-0.69

POSITIVE LOGITS

 official

0.84

dom

0.81

ially

0.80

 sanctioned

0.80

 announcement

0.74

 fiat

0.73

iating

0.71

wcs

0.70

 pronounce

0.70

 documentation

0.68

Activations Density 0.021%

official titles or positions

references to officials or official statements

No Comments

No Known Activations