INDEX

Explanations

prepositions indicating a specific point in time or location

oai_token-act-pair · gpt-3.5-turbo

the phrase "before" indicating sequences of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

geist

-0.76

hack

-0.75

aez

-0.74

ische

-0.72

ophe

-0.71

arna

-0.71

Force

-0.71

ISM

-0.69

rone

-0.68

WIND

-0.68

POSITIVE LOGITS

 settling

1.11

 returning

1.02

 realizing

1.01

 retiring

1.01

 disappearing

1.00

 abruptly

0.95

 concluding

0.95

 committing

0.94

 heading

0.94

 fading

0.94

Activations Density 0.064%

prepositions indicating a specific point in time or location

the phrase "before" indicating sequences of events

No Comments

No Known Activations