INDEX

Explanations

keywords and concepts related to computational processes and analysis

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

quantification and entities

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Personendaten

-0.66

исленность

-0.66

 Administrativna

-0.62

autoIncrement

-0.54

 ſche

-0.54

queryInterface

-0.53

TagNumber

-0.53

 ligiloj

-0.52

AsUp

-0.52

 JpaRepository

-0.52

POSITIVE LOGITS

matchCondition

0.46

QMetaType

0.41

bootstrapcdn

0.38

writeFieldEnd

0.36

tagext

0.35

CrossRef

0.32

 dressed

0.30

waitKey

0.29

+:+

0.29

ifflin

0.29

Activations Density 0.129%

keywords and concepts related to computational processes and analysis

quantification and entities

No Comments

No Known Activations