INDEX

Explanations

specific names and relevant identifiers in technical and procedural contexts

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

after numbers

np_acts-logits-general · gemini-2.0-flash

numbers, units, and abbreviations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

ValueStyle

-0.68

 "..\..\..\

-0.66

Personendaten

-0.62

 noDo

-0.59

webElementXpaths

-0.56

 tartalomajánló

-0.56

paramMap

-0.53

 ModelExpression

-0.53

 defaultstate

-0.52

aarrggbb

-0.52

POSITIVE LOGITS

az

0.62

0.59

 Económica

0.58

 PhpStorm

0.50

parsedMessage

0.50

age

0.47

az

0.47

 BORN

0.47

Corollary

0.47

aze

0.45

Activations Density 0.041%

specific names and relevant identifiers in technical and procedural contexts

after numbers

numbers, units, and abbreviations

No Comments

No Known Activations