INDEX

Explanations

mathematical expressions and variables used in scientific contexts

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

numerical units

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

 виправивши

-0.66

rån

-0.57

GraphicsUnit

-0.56

">//

-0.56

providedIn

-0.55

 MainAxisSize

-0.54

OGND

-0.52

CloseOperation

-0.52

ViewFeatures

-0.51



-0.50

POSITIVE LOGITS

 enfans

0.46

 Schmer

0.44

BibitemOpen

0.44

 falsche

0.43

lojik

0.42

pekte

0.42

 sanno

0.42

 Socialista

0.41

뀔

0.41

 giornal

0.40

Activations Density 0.144%

mathematical expressions and variables used in scientific contexts

numerical units

No Comments

No Known Activations