INDEX

Explanations

phrases related to government actions or involvement

oai_token-act-pair · gpt-3.5-turbo

references to government entities and actions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etter

-0.74

Myr

-0.70

 Clicker

-0.70

âĦ¢:

-0.68

+,

-0.66

 Ogre

-0.66

 Purg

-0.65

hots

-0.65

inth

-0.65

igmatic

-0.63

POSITIVE LOGITS

 government

3.38

 Government

2.79

government

2.76

Government

2.71

 governments

2.36

 governmental

2.27

 Governments

2.00

 federal

1.76

vernment

1.74

governmental

1.57

Activations Density 0.057%

phrases related to government actions or involvement

references to government entities and actions

No Comments

No Known Activations