INDEX

Explanations

coding-related terms and structures, particularly in object-oriented programming and event handling

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

allocating variables in code

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

 arşivlendi

-0.55

HtmlAttribute

-0.48

SequentialGroup

-0.48

StoryboardSegue

-0.47

сылкі

-0.46

 CURIAM

-0.45

 Absch

-0.44

StructEnd

-0.44

 الاطلاع

-0.43

ChildScrollView

-0.43

POSITIVE LOGITS

getDoctrine

0.46

 plufieurs

0.46

 nôtre

0.46

LookAnd

0.44

éndolo

0.43

 normaux

0.43

ándolos

0.41

 médec

0.41

ragalactic

0.40

theoremstyle

0.40

Activations Density 0.009%

coding-related terms and structures, particularly in object-oriented programming and event handling

allocating variables in code

No Comments

No Known Activations