INDEX

Explanations

mathematical expressions and symbols

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

latest firmware

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

OrWhiteSpace

-0.42

addPreferredGap

-0.39

/*

-0.34

 rozhod

-0.32

 lámpara

-0.31

nloa

-0.30

 keluarganya

-0.30

อ้าง

-0.30

 beginnetje

-0.30

<0xCE>

-0.30

POSITIVE LOGITS

reportWebVitals

0.61

ItemBackground

0.58

complexContent

0.55

ValueGenerated

0.54

 Numerade

0.53

OGND

0.52

:✨

0.52

 betweenstory

0.51

 baga

0.51

 بيها

0.50

Activations Density 0.111%

mathematical expressions and symbols

latest firmware

No Comments

No Known Activations