INDEX

Explanations

the specific name "Ko" with various numerical ratings

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Ko"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Creed

-0.79

ingham

-0.74

senal

-0.70

Interstitial

-0.66

 Defin

-0.66

 Redux

-0.65

 narrator

-0.63

å¤©

-0.63

nces

-0.63

 à¨

-0.62

POSITIVE LOGITS

osta

1.04

zzi

1.03

essler

0.97

eln

0.95

Å¡

0.94

unin

0.94

zy

0.94

lyak

0.93

etter

0.93

ogle

0.89

Activations Density 0.035%

the specific name "Ko" with various numerical ratings

mentions of the name "Ko"

No Comments

No Known Activations