INDEX

Explanations

words related to a specific actress, possibly analyzing news articles or social media posts

oai_token-act-pair · gpt-3.5-turbo

adjectives related to artistic expression

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

lessly

-0.82

lessness

-0.82

enegger

-0.79

ãĥĨ

-0.76

cloth

-0.75

shire

-0.74

hig

-0.74

fman

-0.71

plin

-0.68

patrick

-0.67

POSITIVE LOGITS

andum

0.90

ity

0.86

onduct

0.80

henko

0.78

hes

0.75

 acid

0.74

acia

0.69

onite

0.67

atic

0.67

anwhile

0.67

Activations Density 0.042%

words related to a specific actress, possibly analyzing news articles or social media posts

adjectives related to artistic expression

No Comments

No Known Activations