INDEX

Explanations

references to programming or technical content, especially in coding syntax or structures

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

mathematical expressions

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

 defaultstate

-0.57

 Италијани

-0.51

LookAnd

-0.50

olyb

-0.50

 Numerade

-0.49

期刊论文

-0.49

 cherchés

-0.48

Derbyniad

-0.46

webElementXpaths

-0.45

 Infórmanos

-0.45

POSITIVE LOGITS

RegressionTest

0.48

 ویکی‌آمباردا

0.41

$_"

0.41

TagMode

0.40

\{\\

0.40

 forskj

0.36

consulté

0.36

 affitto

0.35

 Rajan

0.35

 Option

0.35

Activations Density 0.005%

references to programming or technical content, especially in coding syntax or structures

mathematical expressions

No Comments

No Known Activations