INDEX

Explanations

helping verbs

oai_token-act-pair · gemini-2.0-flash

memories and appreciations

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_24/width_16k/average_l0_37

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

AsUp

-0.76

 виправивши

-0.75

Portály

-0.69

uxxxx

-0.67

 BoxFit

-0.67

 Roskov

-0.66

IntoConstraints

-0.65

 itſelf

-0.65

 incomplète

-0.63

 himſelf

-0.61

POSITIVE LOGITS

RenderAtEndOf

0.51

 cuer

0.47

 réessayer

0.46

Pal

0.46

감

0.45

 Pain

0.44

ാല

0.43

lasse

0.43

odis

0.42

aman

0.42

Activations Density 1.249%

helping verbs

memories and appreciations

No Comments

No Known Activations