INDEX

Explanations

Tab-related terms or actions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Tab" in various formats and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĨĴ

-0.76

 Lauder

-0.72

IVERS

-0.68

 Worst

-0.65

Yel

-0.62

Downloadha

-0.62

 CHRIST

-0.60

 Flavoring

-0.60

 Gina

-0.58

 antid

-0.58

POSITIVE LOGITS

ular

1.36

ulated

1.31

bed

1.30

ulation

1.29

lished

1.20

ulations

1.20

oola

1.18

ulator

1.17

bing

1.13

ula

1.12

Activations Density 0.022%

Tab-related terms or actions

occurrences of the word "Tab" in various formats and contexts

No Comments

No Known Activations