INDEX

Explanations

programming-related terminology and class structures

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

utility code and constants

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits



-0.66

 Paglinawan

-0.65

OGND

-0.65

 IndexPath

-0.64

 وتسجيلات

-0.64

 ब्रेकडाउन

-0.63

 Lokales

-0.62

 pinulongan

-0.62

 يتيمه

-0.60

 otomatig

-0.60

POSITIVE LOGITS

 ujednoznacz

0.48

oredCriteria

0.46

homonymie

0.45

 gynhyrchwyd

0.44

توض

0.43

queryInterface

0.42

webElementXpaths

0.42

 &___

0.42

exitRule

0.40

 resourceCulture

0.34

Activations Density 0.108%

programming-related terminology and class structures

utility code and constants

No Known Activations