INDEX

Explanations

mentions of individuals holding leadership positions

oai_token-act-pair · gpt-3.5-turbo

references to leadership positions or titles

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

 Parenthood

-0.68

ĸļ

-0.65

 Guests

-0.62

âĨĳ

-0.60

 Memories

-0.60

 IMAGES

-0.58

-0.57

 Niet

-0.56

-0.55

Constructed

-0.54

POSITIVE LOGITS

quartered

1.21

quarter

1.13

master

1.13

hunter

1.11

quarters

1.10

liner

1.09

lining

1.08

mist

1.07

hun

1.03

liners

1.00

Activations Density 0.022%

mentions of individuals holding leadership positions

references to leadership positions or titles

No Comments

No Known Activations