INDEX

Explanations

proper nouns, specifically names of individuals and characters

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

names of people or entities

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 hypoc

-0.63

 Rebels

-0.62

 hereafter

-0.60

 Skydragon

-0.59

 dracon

-0.58

 dungeons

-0.57

 blow

-0.57

 recapt

-0.56

tc

-0.56

CPC

-0.56

POSITIVE LOGITS

berto

1.09

lene

1.05

endra

1.03

isha

1.03

oslav

1.02

ilyn

1.01

issa

1.01

anya

1.00

leen

0.98

antha

0.98

Activations Density 0.395%

proper nouns, specifically names of individuals and characters

names of people or entities

No Comments

No Known Activations