INDEX

Explanations

references to equations, sections, and figures within a scientific or mathematical context

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

references to sections, algorithms, or figures

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

kháu

-0.61

 ModelExpression

-0.57

ंदीखरीदारी

-0.56

 propOrder

-0.52

ChildScrollView

-0.50

Personendaten

-0.48

ParallelGroup

-0.48

Jeografia

-0.47

AccessorTable

-0.46

MigrationBuilder

-0.45

POSITIVE LOGITS

Pref

0.42

tab

0.39

 NEAR

0.39

敢

0.39

 fileName

0.36

elemField

0.35

csolódó

0.35

openModal

0.35

Tit

0.34

bootstrapcdn

0.34

Activations Density 0.469%

references to equations, sections, and figures within a scientific or mathematical context

references to sections, algorithms, or figures

No Comments

No Known Activations