INDEX

Explanations

questions or inquiries in the form of requests or queries

oai_token-act-pair · gpt-3.5-turbo

questions beginning with "Can."

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

é¾įå¥ĳå£«

-0.65

 whence

-0.60

eering

-0.57

 feared

-0.56

 Powered

-0.54

 valued

-0.53

 dispersed

-0.53

 seeded

-0.53

 starring

-0.52

aiden

-0.52

POSITIVE LOGITS

't

1.60

berra

1.26

vas

1.13

adian

1.08

 anyone

0.97

 anybody

0.96

opy

0.94

you

0.93

elo

0.93

nery

0.89

Activations Density 0.041%

questions or inquiries in the form of requests or queries

questions beginning with "Can."

No Comments

No Known Activations