INDEX

Explanations

the word "rewritten" in various contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "rewrite" or its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ello

-0.86

ells

-0.82

ulhu

-0.79

waukee

-0.73

acqu

-0.70

aren

-0.70

arijuana

-0.69

 Charity

-0.69

alach

-0.68

uay

-0.68

POSITIVE LOGITS

 rewritten

1.32

 rewrite

1.18

 rewriting

1.00

é»Ĵ

0.86

dated

0.82

RFC

0.75

 typed

0.73

 restruct

0.72

 vengeance

0.67

 rethink

0.66

Activations Density 0.014%

the word "rewritten" in various contexts

instances of the word "rewrite" or its variations

No Comments

No Known Activations