INDEX

Explanations

locations or places, especially in the context of businesses, universities, and cities

oai_token-act-pair · gpt-3.5-turbo

locations or references associated with "in" followed by specific geographic areas

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

0000000000000000

-0.75

tre

-0.75

potion

-0.67

sle

-0.66

raf

-0.65

romeda

-0.65

trump

-0.64

tal

-0.64

pad

-0.63

izoph

-0.62

POSITIVE LOGITS

 lieu

1.13

 conjunction

1.12

 accordance

1.08

 rural

1.02

 northeastern

1.01

 suburban

0.97

effic

0.97

roads

0.97

clusions

0.96

 Pennsylvania

0.96

Activations Density 0.312%

locations or places, especially in the context of businesses, universities, and cities

locations or references associated with "in" followed by specific geographic areas

No Comments

No Known Activations