INDEX

Explanations

instances where something is taken away or removed from someone or something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "stripped" 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

rious

-1.00

enegger

-0.79

é

-0.76

VERTIS

-0.76

asio

-0.72

lov

-0.67

SEE

-0.67

abul

-0.65

lished

-0.64

 Admir

-0.64

POSITIVE LOGITS

 stripping

0.91

 malls

0.84

utic

0.84

 stripped

0.80

cloth

0.75

jit

0.74

 naked

0.71

 nudity

0.71

hya

0.71

leted

0.71

Activations Density 0.016%

instances where something is taken away or removed from someone or something

instances of the word "stripped" and its variations

No Comments

No Known Activations