INDEX

Explanations

people's surnames

oai_token-act-pair · gpt-3.5-turbo

references to the name "Mull" or variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ec

-0.87

ar

-0.75

ESA

-0.72

zy

-0.69

sci

-0.68

ivable

-0.68

opia

-0.65

 bearing

-0.64

 cape

-0.62

 dystopian

-0.60

POSITIVE LOGITS

 Mull

4.02

 mull

1.95

 Maul

1.38

 Mush

1.24

 Dill

1.19

Haf

1.19

ullah

1.17

 Mahar

1.10

 Cumm

1.09

 Kund

1.08

Activations Density 0.012%

people's surnames

references to the name "Mull" or variations of it

No Comments

No Known Activations