INDEX

Explanations

terms related to class and context identification in programming

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

module name loading

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

 Photocase

-0.66

enumii

-0.63

AsUp

-0.63

 CreateTagHelper

-0.59

səhifə

-0.57

ollectionView

-0.57

şiv

-0.54

becher

-0.54

 promoção

-0.54

 oprot

-0.54

POSITIVE LOGITS

MODULE

0.60

ModuleName

0.47

forName

0.43

module

0.41

moduleName

0.37

TestingModule

0.37

subpackage

0.36

AccessorTable

0.36

oa̍t

0.35

 AssemblyTitle

0.35

Activations Density 0.001%

terms related to class and context identification in programming

module name loading

No Comments

No Known Activations