INDEX

Explanations

French contractions of the form "l'"

oai_token-act-pair · gemini-2.0-flash

apostrophe

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

di

-0.98

ver

-0.79

-0.77

die

-0.77

-0.76

dat

-0.75

-0.74

POSITIVE LOGITS

 itſelf

1.44

 Monfieur

1.38

 pleaſure

1.38

 Majefty

1.31

 myſelf

1.30

 greateſt

1.27

 Diſ

1.27

 Anſ

1.26

 Jefus

1.25

ſelf

1.23

Activations Density 0.181%

French contractions of the form "l'"

apostrophe

No Comments

No Known Activations