INDEX

Explanations

terms related to public health organizations or government health departments

oai_token-act-pair · gpt-3.5-turbo

mentions of health-related entities and departments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

uble

-0.79

 tongues

-0.72

ãĥ£

-0.71

coni

-0.71

Äĩ

-0.70

aunder

-0.68

ãĤ©

-0.68

ptions

-0.68

 goose

-0.67

xual

-0.67

POSITIVE LOGITS

 Care

1.14

Care

1.05

care

1.03

Health

0.98

 Insurance

0.95

 Sciences

0.91

 Health

0.90

 Affairs

0.86

 Advis

0.85

 Resources

0.84

Activations Density 0.014%

terms related to public health organizations or government health departments

mentions of health-related entities and departments

No Comments

No Known Activations