INDEX

Explanations

occurrences of code-related keywords and structures

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

code structure

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

AnimationsModule

-0.48

bootstrapcdn

-0.42

 Polaroid

-0.40

 Rhode

-0.39

ТО

-0.39

 fail

-0.39

getRuntime

-0.39

 scrap

-0.39

లాలు

-0.38

ValueStyle

-0.38

POSITIVE LOGITS

 يتيمه

0.63

 betweenstory

0.53

期刊论文

0.52

 HFILL

0.50

UnusedPrivate

0.48

contentLoaded

0.48

webElementXpaths

0.47

Autoritní

0.47

RegressionTest

0.47

 enfans

0.47

Activations Density 0.000%

occurrences of code-related keywords and structures

code structure

No Comments

No Known Activations