INDEX

Explanations

elements related to server responses and statuses

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

Code and math symbols

np_acts-logits-general · gemini-2.0-flash

code and fractions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

 Normdatei

-1.17

 Administrativna

-1.03

 Italijanski

-1.02

 autorytatywna

-0.98

rungsseite

-0.90

Datuak

-0.89

 مرئيه

-0.86

 فريبيس

-0.84

ⓧ

-0.84

ⓘ

-0.84

POSITIVE LOGITS

json

0.60

 json

0.49

send

0.47

 send

0.43

 land

0.42

unton

0.41

end

0.39

0.38

jung

0.38

Activations Density 0.052%

elements related to server responses and statuses

Code and math symbols

code and fractions

No Comments

No Known Activations