INDEX

Explanations

syntax related to programming and variable assignment

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

code snippets and punctuation

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

setVerticalGroup

-0.54

 houſe

-0.54

 queſta

-0.53

<?

-0.52

 myſelf

-0.49

 Houſe

-0.49

 surla

-0.48

 beſch

-0.48

 Weiſe

-0.48

 ſans

-0.48

POSITIVE LOGITS

titleMargin

0.47

ErrUnexpectedEOF

0.46

 FishBase

0.42

 Италијани

0.41

 defaultstate

0.40

 nonUne

0.40

 invokingState

0.39

Personensuche

0.38

Personendaten

0.38

 Numerade

0.36

Activations Density 0.067%

syntax related to programming and variable assignment

code snippets and punctuation

No Comments

No Known Activations