INDEX

Explanations

specific names and references to individuals, brands, and chemical compounds

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

after place names or names

np_acts-logits-general · gemini-2.0-flash

names of people and places

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

("/:

-0.30

hadap

-0.29

SuspendLayout

-0.28

 BioLib

-0.28

apimachinery

-0.28

('/:

-0.27

 gemeinsam

-0.27

väg

-0.26

Caractéristiques

-0.25

ufig

-0.24

POSITIVE LOGITS

 kasarigan

1.02

EDEFAULT

0.82

 Normdatei

0.76

enderror

0.74

 CreateTagHelper

0.73

 المعيارى

0.72

Aktualisiert

0.71

 ब्रेकडाउन

0.69

<unused28>

0.69

<unused43>

0.69

Activations Density 1.020%

specific names and references to individuals, brands, and chemical compounds

after place names or names

names of people and places

No Comments

No Known Activations