INDEX

Explanations

punctuation marks

np_max-act-logits · gemini-2.0-flash

the boundaries between conversational turns in a dialogue

oai_token-act-pair · gemini-2.0-flash

HTML tags or code

np_token-act-pair-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

…

-0.42

zon

-0.41

©

-0.37

...

-0.36

wy

-0.36

人

-0.36

<eos>

-0.35

tem

-0.35

oris

-0.35

sur

-0.35

POSITIVE LOGITS

 ویکی‌پدیای

1.05

Autoritní

1.02

 purpoſe

1.00

 myſelf

0.97

 Italijanski

0.95

 وتسجيلات

0.93

Personendaten

0.92

 تضيفلها

0.91

 himſelf

0.91

 CreateTagHelper

0.91

Activations Density 0.992%

punctuation marks

the boundaries between conversational turns in a dialogue

HTML tags or code

No Comments

No Known Activations