INDEX

Explanations

elements related to code tests and assertions in programming

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

test and user context

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

ⓧ

-0.51

AutoModerator

-0.48

 Мексичка

-0.46

gameserver

-0.46

*/].

-0.45

 "..\..\..\

-0.45

bootstrapcdn

-0.45

 Италијани

-0.45

BindingSource

-0.44

 informée

-0.43

POSITIVE LOGITS

 Xunit

0.60

scalatest

0.56

getTest

0.52

 test

0.52

entile

0.50

Test

0.49

 Test

0.48

 tests

0.47

test

0.44

AndroidJUnit

0.44

Activations Density 0.000%

elements related to code tests and assertions in programming

test and user context

No Comments

No Known Activations