INDEX

Explanations

instances of the word "test"

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

unit tests

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

 StringTokenizer

-0.61

 Мексичка

-0.59

탤

-0.58

 arşivlendi

-0.57

MethodManager

-0.55

 ""],

-0.52

 BeautifulSoup

-0.52

 ویکی‌پدی

-0.52

AndEndTag

-0.51

 jScrollPane

-0.51

POSITIVE LOGITS

UnitTesting

0.80

Test

0.68

 Xunit

0.66

TestMethod

0.57

ViewFeatures

0.57

AndroidJUnit

0.56

test

0.55

 Test

0.52

 test

0.52

scalatest

0.51

Activations Density 0.001%

instances of the word "test"

unit tests

No Comments

No Known Activations