INDEX

Explanations

and

np_max-act-logits · gemini-2.0-flash

conjunctions, country names, and US state names

oai_token-act-pair · gemini-2.0-flash

and

np_token-act-pair-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

GEBURTSDATUM

-0.68

<bos>

-0.64

 nakalista

-0.54

Архівовано

-0.52

 allAfrica

-0.51

Missi

-0.50

 Roskov

-0.48

RegressionTest

-0.47

minu

-0.47

 Comple

-0.47

POSITIVE LOGITS

DispatchToProps

0.64

Diweddarwch

0.56

SequentialGroup

0.56

 برانيه

0.55

SharedCtor

0.54

们

0.51

CloseOperation

0.48

辈子

0.48

rítica

0.47

 setTotal

0.47

Activations Density 2.415%

and

conjunctions, country names, and US state names

and

No Comments

No Known Activations