INDEX

Explanations

classifications and categorizations of entities or concepts

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Италијани

-0.46

 Houſe

-0.46

 testSet

-0.45

writeFieldEnd

-0.45

OGND

-0.44

antMatchers

-0.42

 endfor

-0.41

ollectionView

-0.41

 houſe

-0.40

dflare

-0.40

POSITIVE LOGITS

 split

0.57

 divided

0.52

 تقسیم

0.49

 Divided

0.48

 Split

0.46

 classifications

0.45

 Categories

0.45

 types

0.45

 splits

0.45

 categorize

0.45

Activations Density 0.042%

classifications and categorizations of entities or concepts

No Comments

No Known Activations