INDEX

Explanations

tokens associated with assignment operations and declaration in programming contexts

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

`new` keyword, code structure

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

AsUp

-0.69

AISSEE

-0.67

 Administrativna

-0.64

 &___

-0.63

ьаж

-0.62

 kohdetta

-0.61

 Italijanski

-0.61

OGND

-0.61

 Италијани

-0.60

MigrationBuilder

-0.59

POSITIVE LOGITS

 ExecuteAsync

0.39

 anzunehmen

0.31

itário

0.30

</b>

0.29

withstanding

0.29

 inadvertently

0.29

 základ

0.28

 überw

0.28

 gradually

0.28

by

0.28

Activations Density 0.010%

tokens associated with assignment operations and declaration in programming contexts

`new` keyword, code structure

No Comments

No Known Activations