INDEX

Explanations

phrases mentioning the "Justice Department."

oai_token-act-pair · gpt-3.5-turbo

references to the Justice Department

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

hest

-0.79

iple

-0.78

 livest

-0.75

uers

-0.74

ulum

-0.72

hetically

-0.70

ength

-0.69

otin

-0.67

vre

-0.67

irst

-0.66

POSITIVE LOGITS

 Dept

1.00

 Department

0.99

 Scalia

0.90

 League

0.85

 Corps

0.80

 Assistance

0.78

 Justice

0.77

 Anton

0.76

 Holmes

0.74

 Centers

0.71

Activations Density 0.022%

phrases mentioning the "Justice Department."

references to the Justice Department

No Comments

No Known Activations