INDEX

Explanations

sequences related to highlighting or calling attention to specific information within a text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "marks" related to various contexts or meanings

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

 Garner

-0.68

ADS

-0.68

rouse

-0.66

OTE

-0.66

erest

-0.65

 disproportionate

-0.64

lux

-0.63

 Huang

-0.63

GAN

-0.63

ernand

-0.63

POSITIVE LOGITS

manship

1.48

 marks

1.26

marks

1.22

 mark

0.99

mark

0.96

emark

0.93

 marking

0.91

terday

0.86

marked

0.85

 markings

0.84

Activations Density 0.008%

sequences related to highlighting or calling attention to specific information within a text

instances of the word "marks" related to various contexts or meanings

No Comments

No Known Activations