INDEX

Explanations

references to data structures and operations within programming contexts

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

code and math syntax

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

 Poitiers

-0.45

gebob

-0.44

 wikipagina

-0.43



-0.42

mannian

-0.42

 golem

-0.42

 étoient

-0.41

 />';

-0.41

Comté

-0.40

 maxX

-0.40

POSITIVE LOGITS

ंदीखरीदारी

0.43

Autoritní

0.42

delwed

0.41

verifyException

0.40

 uLocal

0.40

 simplifié

0.39

 Italijanski

0.39

 kohdetta

0.38

fillType

0.38

setVerticalGroup

0.37

Activations Density 0.340%

references to data structures and operations within programming contexts

code and math syntax

No Comments

No Known Activations