INDEX

Explanations

comment and documentation markers in code

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

Java programming language

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



-0.54

 BrowserModule

-0.53

iNdEx

-0.49

 nakalista

-0.42

 opérateur

-0.42

+#+#

-0.41

rawDesc

-0.41

 ocupación

-0.41

 jsPsych

-0.41

PageRoute

-0.41

POSITIVE LOGITS

setVerticalGroup

0.54

GOTREF

0.50

 propOrder

0.47

complexType

0.44

AndroidJUnit

0.43

 Prologue

0.39

PMC

0.39

mení

0.39

endpush

0.38

 Cooker

0.38

Activations Density 0.001%

comment and documentation markers in code

Java programming language

No Comments

No Known Activations