INDEX

Explanations

code-related commands and output functions

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

outputting information

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

UnitTesting

-0.80

MessageTagHelper

-0.79

IntoConstraints

-0.78

出版年

-0.71

principalColumn

-0.68

 propOrder

-0.66

FormTagHelper

-0.66

WebElementEntity

-0.65

 autorytatywna

-0.64

featureID

-0.63

POSITIVE LOGITS

getStringExtra

0.31

 Venez

0.27

Slf

0.27

"?>

0.26

InSeconds

0.25

ầm

0.25

linearGradient

0.25

 démission

0.25

Pinned

0.24

습

0.24

Activations Density 0.003%

code-related commands and output functions

outputting information

No Comments

No Known Activations