INDEX

Explanations

method names used in programming code

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

getter method calls

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

RegressionTest

-0.93

Tikang

-0.91

 betweenstory

-0.88

="@+

-0.83

 zuſammen

-0.81

ロウィン

-0.80

webElementXpaths

-0.79

 queſta

-0.78

 zwiſchen

-0.78

 nakalista

-0.78

POSITIVE LOGITS

 getId

0.33

sbereich

0.24

 electrico

0.24

 OnInit

0.23

 échanges

0.22

 interno

0.22

lindung

0.21

setObjectName

0.21

 abstracto

0.21

AutoSize

0.21

Activations Density 0.002%

method names used in programming code

getter method calls

No Comments

No Known Activations