INDEX

Explanations

terms related to specific geographical locations or institutions

oai_token-act-pair · gpt-3.5-turbo

references to Georgetown and its affiliated locations or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

acca

-0.79

emic

-0.72

userc

-0.72

astical

-0.71

wana

-0.70

gypt

-0.66

ramid

-0.66

usal

-0.66

uscript

-0.64

usp

-0.64

POSITIVE LOGITS

orgetown

1.01

Hoy

0.92

 Georgetown

0.84

 Morg

0.79

heit

0.79

icago

0.76

 Tennis

0.75

 University

0.73

 Germ

0.72

 Marriott

0.71

Activations Density 0.013%

terms related to specific geographical locations or institutions

references to Georgetown and its affiliated locations or contexts

No Comments

No Known Activations