INDEX

Explanations

words related to movies and the actors/crew who make them

oai_token-act-pair · gemini-2.0-flash

Unclear, cannot discern pattern from activations or logits

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

foreign languages

np_max-act-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

 estekak

-0.62

ennes

-0.59

withIdentifier

-0.57

 للمعارف

-0.55

manos

-0.52

 externi

-0.52

TagMode

-0.52

issau

-0.50

 Punt

-0.50

tanleria

-0.49

POSITIVE LOGITS

 básica

0.56

oa̍t

0.56

liothèque

0.54

 Beſ

0.54

 Monfieur

0.54

 ſte

0.52

 purpoſe

0.52

 făcut

0.51

 TestBed

0.50

 Inſ

0.50

Activations Density 1.623%

words related to movies and the actors/crew who make them

Unclear, cannot discern pattern from activations or logits

foreign languages

No Comments

No Known Activations