INDEX

Explanations

tokens that come before or after numbers, as well as some other apparently insignificant words ("the," "with")

oai_token-act-pair · gemini-2.0-flash

code/data snippets

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

Negative Logits

RenderAtEndOf

-0.50

PullParser

-0.50

aarrggbb

-0.49

 morire

-0.48

Paj

-0.48

ToBounds

-0.47

 passively

-0.47

 molesta

-0.47

Ligações

-0.46

 StatefulWidget

-0.46

POSITIVE LOGITS

parsedMessage

0.69

-------------</

0.66

GEBURTSDATUM

0.65

addCriterion

0.63

 kaarangay

0.59

urbation

0.57

 free

0.56

abestanden

0.56

RESPOND

0.55

itoba

0.55

Activations Density 1.887%

tokens that come before or after numbers, as well as some other apparently insignificant words ("the," "with")

code/data snippets

No Known Activations