INDEX

Explanations

references to mathematical or programming concepts, specifically in technical documentation or code contexts

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

programming or mathematical derivative

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

 nakalista

-0.58

Personensuche

-0.56

 CURIAM

-0.52

 تكبرها

-0.52

存于互联网档案馆

-0.52

 Италијани

-0.45

ंदीखरीदारी

-0.44

 happiest

-0.44

Controllo

-0.44

libur

-0.43

POSITIVE LOGITS

 AssemblyTitle

0.53

TypedDataSet

0.50

 NSCoder

0.41

gdx

0.36

AddHtmlAttribute

0.36

protoimpl

0.35

AnimationsModule

0.35

 ferons

0.35

AnchorStyles

0.34

 BorderRadius

0.33

Activations Density 0.076%

references to mathematical or programming concepts, specifically in technical documentation or code contexts

programming or mathematical derivative

No Comments

No Known Activations