INDEX

Explanations

references to technical terms and specifications related to projects or studies

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

copyright policy rules

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

 MainAxisSize

-0.34

 հղումներ

-0.30

ⓧ

-0.30

 InputDecoration

-0.30

énagement

-0.29

ceğini

-0.28

fication

-0.28

アナ

-0.27

setViewportView

-0.27

VolleyError

-0.26

POSITIVE LOGITS

 surla

0.63

HtmlAttribute

0.60

 verſ

0.60

 ſte

0.59

XmlAccessType

0.59

rungsseite

0.57

 Inſ

0.57

 titolata

0.54

attutto

0.53

 ſou

0.51

Activations Density 0.106%

references to technical terms and specifications related to projects or studies

copyright policy rules

No Comments

No Known Activations