INDEX

Explanations

certain keywords related to processes or functions in programming and mathematics

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

mathematical expressions or code snippets

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

 للمعارف

-0.46

 beginnetje

-0.37

 vastaan

-0.34

 skóra

-0.33

 satisfecho

-0.32

ValueStyle

-0.31

 člověk

-0.31

 micrófono

-0.31

 domésticos

-0.30

 défaut

-0.30

POSITIVE LOGITS

+#+#

0.65

httphttps

0.64

fromnode

0.63

 MainAxisSize

0.56

keyColumn

0.54

ostar

0.53

 wireType

0.53

EE

0.52

ChildScrollView

0.52

idor

0.52

Activations Density 0.729%

certain keywords related to processes or functions in programming and mathematics

mathematical expressions or code snippets

No Comments

No Known Activations