INDEX

Explanations

proper nouns or identifiers related to people and entities

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

 BrowserModule

-0.62

setVerticalGroup

-0.61

 hyö

-0.58

 çerçe

-0.56

 kağı

-0.55

 cappuccio

-0.54

Билгалдахарш

-0.54

 bileklik

-0.54

 kaplama

-0.54

 yüzük

-0.54

POSITIVE LOGITS

devamını

0.49

勝

0.42

 Considering

0.42

بوابة

0.41

 researches

0.40

getJSONArray

0.40

ehe

0.40

 bete

0.39

 Adnan

0.38

 Koval

0.38

Activations Density 0.001%

proper nouns or identifiers related to people and entities

No Comments

No Known Activations