INDEX

Explanations

mentions of "HUD" (Department of Housing and Urban Development) related terms in the text

oai_token-act-pair · gpt-3.5-turbo

references to the Department of Housing and Urban Development (HUD)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Belg

-0.81

Koz

-0.68

 Thames

-0.67

Oro

-0.67

AES

-0.65

 Prev

-0.65

ãĥīãĥ©ãĤ´ãĥ³

-0.64

 Glou

-0.64

Wit

-0.62

 Shannon

-0.62

POSITIVE LOGITS

HUD

1.14

HUD

1.11

eller

0.94

ilitating

0.91

assad

0.87

 subsidized

0.79

ailed

0.79

uilt

0.79

arre

0.79

ndra

0.77

Activations Density 0.016%

mentions of "HUD" (Department of Housing and Urban Development) related terms in the text

references to the Department of Housing and Urban Development (HUD)

No Comments

No Known Activations