INDEX

Explanations

mentions of specific people and organizations, potentially related to biographies or political statements

oai_token-act-pair · gpt-3.5-turbo

the names of people or entities associated with specific contexts or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 occasion

-0.80

 Cher

-0.73

 Atlas

-0.71

 direction

-0.71

 Gran

-0.70

Aux

-0.68

Mus

-0.68

 Evangel

-0.68

 reserves

-0.68

 Civil

-0.67

POSITIVE LOGITS

Bi

2.28

Po

2.13

Ho

2.11

Va

1.96

Ma

1.85

Lo

1.66

Li

1.64

Ha

1.63

Hu

1.63

Sa

1.56

Activations Density 0.043%

mentions of specific people and organizations, potentially related to biographies or political statements

the names of people or entities associated with specific contexts or events

No Comments

No Known Activations