INDEX

Explanations

phrases like "another *"

oai_token-act-pair · gemini-2.0-flash

another

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

Most

-0.56

 referenties

-0.56

 mattino

-0.48

 bonté

-0.47

 tuturor

-0.46

most

-0.46

 accett

-0.45

 most

-0.45

 mogelijkheden

-0.44

POSITIVE LOGITS

SOUNDBITE

0.85

 cref

0.80

IsContent

0.77

numerusform

0.77

 تانيه

0.73

AddTagHelper

0.71

Билгалдахарш

0.71

SuppressMessage

0.66

Itoa

0.66

__":

0.66

Activations Density 2.762%

phrases like "another *"

another

No Comments

No Known Activations