INDEX

Explanations

email headers and/or signatures

oai_token-act-pair · gemini-2.0-flash

writing/saying

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 Meksiku

-0.53

GraphicsUnit

-0.52

 VizieR

-0.49

RTDA

-0.49

etine

-0.49

ktı

-0.48

xase

-0.47

✨:

-0.47

 виправивши

-0.46

 ddelweddau

-0.46

POSITIVE LOGITS

 wrote

1.23

 writes

1.16

 said

1.15

writes

1.15

wrote

1.10

said

1.08

Said

1.05

 write

1.03

 écrit

1.02

 schrieb

1.00

Activations Density 2.355%

email headers and/or signatures

writing/saying

No Comments

No Known Activations