INDEX

Explanations

the word "episode"

oai_token-act-pair · gpt-3.5-turbo

phrases indicating episodes or segments of media content

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

imei

-0.88

chwitz

-0.82

ghan

-0.78

anchester

-0.75

jri

-0.73

nod

-0.72

 hint

-0.70

uckland

-0.70

tymology

-0.69

ÃŁ

-0.69

POSITIVE LOGITS

 RELE

0.74

 Forbes

0.73

icial

0.70

 course

0.66

 Fortune

0.66

 Notting

0.65

 scrimmage

0.65

 Celebrity

0.65

 Nightmares

0.65

 installments

0.64

Activations Density 0.174%

the word "episode"

phrases indicating episodes or segments of media content

No Comments

No Known Activations