INDEX

Explanations

expressions of gratitude and appreciation towards individuals and groups involved in various efforts

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

Thanks and acknowledgements

np_acts-logits-general · gemini-2.0-flash

thanking groups

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

✭✭

-0.40

 Anſ

-0.40

 wikipagina

-0.39

qvarna

-0.37

UserScript

-0.36

jillo

-0.35

 univ

-0.35

velte

-0.35

 CreateTagHelper

-0.34

 okuyayım

-0.34

POSITIVE LOGITS

prochen

0.47

EDEFAULT

0.46

 danken

0.44

 AssemblyCulture

0.44

 thanked

0.42

ÊN

0.41

ơn

0.41

 contributed

0.41

pantalones

0.41

labelledby

0.41

Activations Density 0.133%

expressions of gratitude and appreciation towards individuals and groups involved in various efforts

Thanks and acknowledgements

thanking groups

No Comments

No Known Activations