INDEX

Explanations

keywords related to programming functions and structures within code

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

Options objects used for tests

np_acts-logits-general · gemini-2.0-flash

parameters and options

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 OMITBAD

-0.47

يكب

-0.47

 dAtA

-0.44

 pinulongan

-0.44

 JpaRepository

-0.43

AsUp

-0.42

ViewInit

-0.42

inSlope

-0.42

InjectMocks

-0.42

featureID

-0.41

POSITIVE LOGITS

 spéciaux

0.47

GEBURTS

0.44

Péri

0.41

 enfans

0.40

 cérami

0.39

 meille

0.39

Internal

0.39

Hentet

0.39

 normaux

0.39

ffilm

0.38

Activations Density 0.198%

keywords related to programming functions and structures within code

Options objects used for tests

parameters and options

No Comments

No Known Activations