INDEX

Explanations

technical abbreviations and specific identifiers within a context, likely related to coding or scientific terminologies

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

Words/acronyms followed by uppercase letters

np_acts-logits-general · gemini-2.0-flash

API endpoints and database queries

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

-0.70

iae

-0.59

 estekak

-0.57

jacency

-0.55

 Klik

-0.55

IQUES

-0.54

 tanto

-0.53

 MainAxisSize

-0.53

IATION

-0.53

IKE

-0.52

POSITIVE LOGITS

oredCriteria

0.84

 utafitiHapana

0.70

WebElementEntity

0.69

 Мексичка

0.68

Externé

0.68

MigrationBuilder

0.67

isContained

0.60

tagext

0.59

jspx

0.58

 AspNetCore

0.58

Activations Density 0.372%

technical abbreviations and specific identifiers within a context, likely related to coding or scientific terminologies

Words/acronyms followed by uppercase letters

API endpoints and database queries

No Comments

No Known Activations