INDEX

Explanations

code-related terms and identifiers

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

test handle / user

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

 estekak

-0.71

 GoogleFonts

-0.65

}{@

-0.57

GEBURTS

-0.55

devamını

-0.54

LEncoder

-0.54

VPB

-0.53

styleType

-0.52

bootstrapcdn

-0.52

+#+

-0.52

POSITIVE LOGITS

 AssemblyCompany

0.47

Personensuche

0.44

invokeLater

0.40

tanleria

0.38

 initComponents

0.38

InputTagHelper

0.37

 Usaha

0.31

 mergeFrom

0.31

 illustrationer

0.31

 importancia

0.31

Activations Density 0.000%

code-related terms and identifiers

test handle / user

No Comments

No Known Activations