INDEX

Explanations

references to the word "food" or words related to it

oai_token-act-pair · gpt-3.5-turbo

instances of the word "food" and its variations

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

ps

-0.74

 contested

-0.68

 juggling

-0.66

 delegates

-0.66

 divor

-0.64

 Europa

-0.63

 prerequisite

-0.62

 levers

-0.62

 compar

-0.61

 dreaded

-0.60

POSITIVE LOGITS

ood

4.63

oodle

1.98

OOD

1.79

oud

1.21

oor

1.15

ORK

1.13

ook

1.11

ork

1.03

orks

1.01

oot

1.01

Activations Density 0.012%

references to the word "food" or words related to it

instances of the word "food" and its variations

No Comments

No Known Activations