INDEX

Explanations

proper names and identifiers related to authors and researchers

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_131k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.33.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Brainz

-0.35

 snippetHide

-0.34

endregion

-0.32

cloths

-0.31

ramid

-0.30

`{

-0.30

 twij

-0.30

`/

-0.30

<![

-0.29

BUF

-0.29

POSITIVE LOGITS

 Hong

1.16

 Zhang

1.06

 Cantonese

1.05

 Wong

1.03

 Wang

1.02

Hong

1.00

 Zhou

0.99

 Taiwanese

0.97

 Huang

0.96

 cinese

0.96

Activations Density 0.232%

proper names and identifiers related to authors and researchers

No Comments

No Known Activations