INDEX

Explanations

coding or programming syntax and structures

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

programming constructs and data manipulation

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

verwijspagina

-0.79

 kasarigan

-0.69

 >=",

-0.69

webElementXpaths

-0.67

 ſind

-0.67

$_"

-0.67

 jadx

-0.65

 AssemblyCompany

-0.65

 ſont

-0.65

Personensuche

-0.65

POSITIVE LOGITS

 Johansen

0.36

<eos>

0.36

مصادر

0.34

0.33

 Schulte

0.32

 Schmel

0.31

Ols

0.30

autoIncrement

0.29

 Fleisch

0.29

ansı

0.29

Activations Density 0.154%

coding or programming syntax and structures

programming constructs and data manipulation

No Comments

No Known Activations