INDEX

Explanations

references to data structures or components within a programming context

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

math.complex.ComplexUtils test

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

styleType

-0.84

parsedMessage

-0.73

 '\\;'

-0.72

setVerticalGroup

-0.68

contentLoaded

-0.68

ChildScrollView

-0.67

MemoryWarning

-0.66

 CreateTagHelper

-0.65

onViewCreated

-0.65

webElementXpaths

-0.61

POSITIVE LOGITS

 score

0.35

 scoring

0.33

and

0.33

 assets

0.33

 convention

0.32

 census

0.31

 conventions

0.31

 rendel

0.30

 chapters

0.30

 casualties

0.30

Activations Density 0.003%

references to data structures or components within a programming context

math.complex.ComplexUtils test

No Comments

No Known Activations