INDEX

Explanations

some variation of the prefix "pro."

oai_token-act-pair · gemini-2.0-flash

pro

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

UnsafeEnabled

-0.66

bootstrapcdn

-0.65

 sén

-0.62

GEBURTSDATUM

-0.59

AddTagHelper

-0.57

MLLoader

-0.56

XmlAccessorType

-0.56

zeuge

-0.55

vangen

-0.54

 smtplib

-0.54

POSITIVE LOGITS

Sucesor

0.43

sted

0.43

AlterField

0.43

ieste

0.42

icordia

0.40

ściej

0.40

 eventdata

0.40

༘

0.39

➞

0.39

itez

0.39

Activations Density 0.148%

some variation of the prefix "pro."

pro

No Comments

No Known Activations