INDEX

Explanations

names starting with "Fr"

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual named "Fr."

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

 Izan

-0.71

 writ

-0.66

 pleas

-0.65

 thumbs

-0.64

eers

-0.63

 odds

-0.59

 tones

-0.58

eer

-0.58

 thumb

-0.56

 forth

-0.56

POSITIVE LOGITS

aternity

1.27

inge

1.26

ustration

1.25

antz

1.23

aternal

1.23

ictional

1.22

annie

1.18

isco

1.17

antic

1.17

idge

1.16

Activations Density 0.020%

names starting with "Fr"

references to a specific individual named "Fr."

No Comments

No Known Activations