INDEX

Explanations

mentions of the word "Cord"

oai_token-act-pair · gpt-3.5-turbo

the word "Cord" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

geant

-0.78

 IMAGES

-0.71

veyard

-0.67

ccording

-0.63

iesta

-0.63

hh

-0.61

ĨĴ

-0.61

Ô

-0.61

 midterm

-0.61

ixels

-0.61

POSITIVE LOGITS

 cord

1.18

 Cord

1.03

ovan

0.94

lessly

0.94

ially

0.82

ages

0.80

oba

0.79

rier

0.79

 cords

0.79

inarily

0.78

Activations Density 0.015%

mentions of the word "Cord"

the word "Cord" and its variations

No Comments

No Known Activations