INDEX

Explanations

phrases related to instructions or guidelines

oai_token-act-pair · gpt-3.5-turbo

references to community engagement and participation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

edia

-0.75

ahime

-0.75

ocious

-0.65

eka

-0.61

ews

-0.61

eez

-0.60

neck

-0.59

android

-0.58

lace

-0.58

iframe

-0.58

POSITIVE LOGITS

 except

1.54

except

1.27

 imaginable

1.01

 alike

0.92

 irrespective

0.92

 equally

0.86

 whatsoever

0.84

 regardless

0.83

soever

0.83

excluding

0.81

Activations Density 0.391%

phrases related to instructions or guidelines

references to community engagement and participation

No Comments

No Known Activations