INDEX

Explanations

the ampersand symbol '&' used in various contexts

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

B&B and R&D

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

 صوتيه

-0.60

bootstrapcdn

-0.53

impanan

-0.47

gridx

-0.45

 Stroh

-0.44

aternion

-0.42

 doPost

-0.42

 Wisata

-0.42

issau

-0.41

찮

-0.41

POSITIVE LOGITS

1.02

\&

0.89

0.85

>&

0.81

'&

0.81

"&

0.80

;&

0.79

&\

0.73

?&

0.73

+"&

0.72

Activations Density 0.003%

the ampersand symbol '&' used in various contexts

B&B and R&D

No Comments

No Known Activations