INDEX

Explanations

phrases related to significant or prominent figures or entities

oai_token-act-pair · gpt-3.5-turbo

the term "leading" in various contexts related to prominence or influence

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

ascade

-0.79

adas

-0.73

rehens

-0.71

yang

-0.70

aves

-0.70

ombies

-0.70

atto

-0.70

nery

-0.67

arthy

-0.66

tera

-0.63

POSITIVE LOGITS

 Leading

0.83

better

0.81

boards

0.81

quartered

0.79

 scorer

0.79

wcs

0.77

 leading

0.77

lines

0.76

TAIN

0.70

ership

0.68

Activations Density 0.019%

phrases related to significant or prominent figures or entities

the term "leading" in various contexts related to prominence or influence

No Comments

No Known Activations