INDEX

Explanations

mentions of Google and related web services

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

"Google" and related terms

np_acts-logits-general · gemini-2.0-flash

google

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

出版年

-0.56

 tarko

-0.42

 ModelExpression

-0.41

 Anspruch

-0.41

kyt

-0.40

 hoort

-0.40

mbals

-0.39

transQ

-0.38

大事な

-0.38

Przyp

-0.36

POSITIVE LOGITS

 Google

1.46

Google

1.45

google

1.42

 google

1.36

 GOOGLE

1.34

GOOGLE

1.18

 goog

0.93

oogle

0.91

goog

0.88

 googled

0.87

Activations Density 0.055%

mentions of Google and related web services

"Google" and related terms

google

No Comments

No Known Activations