INDEX

Explanations

names of specific places, including cities and businesses

oai_token-act-pair · gpt-3.5-turbo

names of locations, brands, and franchises

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

mble

-0.90

berra

-0.78

interstitial

-0.77

aird

-0.75

orate

-0.72

resent

-0.70

hor

-0.68

unal

-0.68

unic

-0.67

glomer

-0.66

POSITIVE LOGITS

 Wars

0.83

 Islanders

0.77

 City

0.74

 Suit

0.74

 Point

0.74

 Lantern

0.74

 Fighters

0.73

 Mages

0.73

 Fans

0.73

 Island

0.71

Activations Density 0.316%

names of specific places, including cities and businesses

names of locations, brands, and franchises

No Comments

No Known Activations