INDEX

Explanations

mathematical and programming expressions related to operations and structures

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

mathematical and code symbols

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

 kasarigan

-0.91

WriteTagHelper

-0.90

 autorytatywna

-0.84

ंदीखरीदारी

-0.73

Personensuche

-0.73

msgTypes

-0.71

oredCriteria

-0.71

ConstraintMaker

-0.71

parsedMessage

-0.71

 surla

-0.68

POSITIVE LOGITS

 properly

0.28

 privately

0.27

poj

0.27

 magnetic

0.25

../

0.24

 separately

0.24

 pasteur

0.23

थ

0.23

 higher

0.23

 averaged

0.23

Activations Density 0.040%

mathematical and programming expressions related to operations and structures

mathematical and code symbols

No Comments

No Known Activations