INDEX

Explanations

references to the word "Pet"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Pet" and related branding or products

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

xual

-0.92

éĹĺ

-0.86

IDER

-0.75

hower

-0.72

 PUBLIC

-0.72

 Reson

-0.71

hirt

-0.70

sers

-0.66

 seeded

-0.65

Îº

-0.64

POSITIVE LOGITS

ertodd

1.36

pet

1.03

rov

0.96

abyte

0.95

itions

0.94

roleum

0.93

terson

0.93

unia

0.93

lyak

0.92

rol

0.89

Activations Density 0.007%

references to the word "Pet"

mentions of the word "Pet" and related branding or products

No Comments

No Known Activations