INDEX

Explanations

numerical sequences and patterns, often within data or code contexts

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

numbers and technical terms

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

LookAnd

-0.52

 betweenstory

-0.49

UserScript

-0.46

 autorytatywna

-0.44

pexpr

-0.44

 kaarangay

-0.44

currentColor

-0.43

 Chwiliwch

-0.42

 transfieras

-0.41

ürme

-0.40

POSITIVE LOGITS

Diwedd

0.50

WebpackPlugin

0.46

PerformLayout

0.45

basicConfig

0.41

 صوتيه

0.38

rouvez

0.38

tableFuture

0.37

reactivex

0.37

 tadi

0.37

 opérés

0.36

Activations Density 0.339%

numerical sequences and patterns, often within data or code contexts

numbers and technical terms

No Comments

No Known Activations