INDEX

Explanations

programming constructs and data structures

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

code snippets and declarations

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

Negative Logits

دانشنامهٔ

-0.68

writeFieldEnd

-0.62

المناصب

-0.60

restling

-0.57

 TextInputType

-0.56

 فريبيس

-0.55

Obrázky

-0.54

ссора

-0.54

 surla

-0.54

########.

-0.52

POSITIVE LOGITS

">//

0.44

 propOrder

0.40

Viitteet

0.33

 unknownFields

0.31

 pinulongan

0.29

 MonoBehaviour

0.29

tableLayoutPanel

0.29

 nennt

0.29

/**

0.29

 mám

0.28

Activations Density 0.032%

programming constructs and data structures

code snippets and declarations

No Known Activations