INDEX

Explanations

phrases that denote a quantity or number of individuals or entities

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

featureID

-0.57

autorest

-0.50

arrangement

-0.43

 unknownFields

-0.42

 hipótesis

-0.42

 aset

-0.40

 Zwiebel

-0.40

 proposons

-0.40

 défendre

-0.40

amaño

-0.39

POSITIVE LOGITS

 EconPapers

0.55

 autorytatywna

0.44

Tikang

0.43

 المعيارى

0.39

 indisponible

0.39

+#+

0.39

 invokingState

0.35

 ujednoznacz

0.35

__(/*!

0.34

 intptr

0.33

Activations Density 0.003%

phrases that denote a quantity or number of individuals or entities

No Comments

No Known Activations