INDEX

Explanations

specific keywords and phrases related to various commands or functions, particularly in programming or technical contexts

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

various language fragments

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

-0.42

you

-0.41

You

-0.41

 Geduld

-0.40

 Wege

-0.40

if

-0.40

 Sieger

-0.39

 Erbe

-0.38

 Raub

-0.37

POSITIVE LOGITS

iſen

1.14

iſten

1.03

ſſung

0.93

niſſe

0.93

ロウィン

0.92

 ſind

0.92

ſicht

0.91

ðsíða

0.91

ésultats

0.91

ſcher

0.91

Activations Density 0.037%

specific keywords and phrases related to various commands or functions, particularly in programming or technical contexts

various language fragments

No Comments

No Known Activations