INDEX

Explanations

strings with letters followed by a number indicating significance

oai_token-act-pair · gpt-3.5-turbo

words or phrases that indicate specific regulations or prohibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

enhagen

-0.80

 Keller

-0.78

Kop

-0.77

å§«

-0.76

 Shepherd

-0.72

 Grayson

-0.71

schild

-0.70

 Forth

-0.70

 Chronicle

-0.70

kamp

-0.69

POSITIVE LOGITS

anc

0.93

arching

0.84

itting

0.81

arc

0.79

ounded

0.79

usterity

0.78

withstanding

0.77

abor

0.75

imm

0.74

arr

0.73

Activations Density 0.234%

strings with letters followed by a number indicating significance

words or phrases that indicate specific regulations or prohibitions

No Comments

No Known Activations