INDEX

Explanations

mathematical formatting and notation

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

latex math typesetting

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

 transfieras

-0.56



-0.55

tagHelperRunner

-0.52

Diweddarwch

-0.51

principalColumn

-0.50

 Italijani

-0.50

########.

-0.48

❋

-0.44

OGND

-0.43

UnitTesting

-0.43

POSITIVE LOGITS

center

0.45

setVerticalGroup

0.44

centerline

0.43

Center

0.42

endcsname

0.42

 center

0.41

 central

0.40

feira

0.39

Цель

0.38

teu

0.38

Activations Density 0.024%

mathematical formatting and notation

latex math typesetting

No Comments

No Known Activations