INDEX

Explanations

words related to being satisfied or approved of, as well as words indicating support or backing

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of justification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

thur

-0.72

dimension

-0.72

fall

-0.70

 Bees

-0.68

kers

-0.65

ACH

-0.65

 Archdemon

-0.64

FTWARE

-0.64

ker

-0.64

alter

-0.63

POSITIVE LOGITS

ified

1.15

ification

0.92

ifies

0.87

ify

0.86

ourgeois

0.84

urally

0.81

IFIED

0.81

ifix

0.79

ibaba

0.79

ific

0.78

Activations Density 0.013%

words related to being satisfied or approved of, as well as words indicating support or backing

words related to the concept of justification

No Comments

No Known Activations