INDEX

Explanations

the word "at" followed by a time

oai_token-act-pair · gemini-2.0-flash

at or from

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_20/width_16k/average_l0_11

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 везе

-0.51

 Care

-0.49

mtable

-0.48

 tutti

-0.47

protoimpl

-0.47

racht

-0.47

 rendre

-0.46

 Zeit

-0.46

ientôt

-0.46

raise

-0.45

POSITIVE LOGITS

 وتسجيلات

0.84

0.73

 afternoon

0.73

 AssemblyCulture

0.69

StackNavigator

0.67

 MORNING

0.66

 morning

0.66

Morning

0.65

RegressionTest

0.63

 noon

0.59

Activations Density 2.423%

the word "at" followed by a time

at or from

No Known Activations