INDEX

Explanations

references to scientific methodologies and statistical terms

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

sold to buyers

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

-0.70

DeleteBehavior

-0.65

Autoritní

-0.63

IUrlHelper

-0.60

 ویکی‌پدی

-0.58

 Roskov

-0.58

فایل‌لار

-0.57

ThroughAttribute

-0.55

VYMaps

-0.54

CloseOperation

-0.54

POSITIVE LOGITS

Cons

0.45

0.44

0.42

Consider

0.40

↵↵

0.40

<h1>

0.39

Peny

0.38

 llegamos

0.37

Mes

0.37

 licenses

0.37

Activations Density 0.271%

references to scientific methodologies and statistical terms

sold to buyers

No Comments

No Known Activations