INDEX

Explanations

words used to describe different or unique instances of things

oai_token-act-pair · gemini-2.0-flash

types and descriptions

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

 estekak

-1.20

Efq

-1.11

 الرياضيه

-1.01

UnsafeEnabled

-0.93

 houſe

-0.92

 Houſe

-0.90

 Jefus

-0.89

 autorytatywna

-0.88

 ſtate

-0.88

ſelf

-0.88

POSITIVE LOGITS

 vintage

0.52

vin

0.49

0.47

ial

0.43

 trade

0.42

 stripe

0.41

 extra

0.40

 custom

0.40

0.39

entsch

0.39

Activations Density 3.446%

words used to describe different or unique instances of things

types and descriptions

No Comments

No Known Activations