INDEX

Explanations

a confusing mix of first-person pronouns, mathematical formulas, numbered lists, and the word "at"

oai_token-act-pair · gemini-2.0-flash

code and strings

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

Efq

-0.91

 Jefus

-0.80

 chofe

-0.74

 myſelf

-0.73

 itſelf

-0.71

 Monfieur

-0.71

 himſelf

-0.70

 ſeveral

-0.69

 uſed

-0.69

 poffible

-0.68

POSITIVE LOGITS

}{-

0.56

IVEREF

0.54

ред

0.52

 protoimpl

0.51

__*/

0.50

 kasarigan

0.50

\|_{\

0.47

>|</

0.47

\|^{

0.47

és

0.46

Activations Density 2.986%

a confusing mix of first-person pronouns, mathematical formulas, numbered lists, and the word "at"

code and strings

No Comments

No Known Activations