INDEX

Explanations

technical terms and components related to machinery or programming constructs

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

devices that perform actions

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

исленность

-0.62

 ब्रेकडाउन

-0.59

 autorytatywna

-0.57

 ivelany

-0.56

Datuak

-0.56

CORBA

-0.54

ymce

-0.54

 metav

-0.54

 Photocase

-0.54

ᅠ

-0.52

POSITIVE LOGITS

 cargador

0.51

 oídos

0.46

 downloader

0.46

 amplifier

0.44

 feeder

0.43

generator

0.43

 chargeur

0.42

 reproductor

0.42

evaluator

0.42

feeder

0.42

Activations Density 0.203%

technical terms and components related to machinery or programming constructs

devices that perform actions

No Comments

No Known Activations