INDEX

Explanations

hexadecimal color codes and legal citations

oai_token-act-pair · gemini-2.0-flash

tokens in a specific format or language

np_token-act-pair-logits · gpt-4o-mini

foreign words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Seven

-0.56

 FetchType

-0.52

Seven

-0.51

 tujuh

-0.50

 seven

-0.50

DrawerToggle

-0.48

EndContext

-0.46

seven

-0.45

 SEVEN

-0.44

bufio

-0.43

POSITIVE LOGITS

 nahilalakip

0.70

ReferenceEquals

0.66

omock

0.62

ThroughAttribute

0.59

Попис

0.59

Izvori

0.58

 فريبيس

0.58

 Besøkt

0.57

şiv

0.57

 оригіналу

0.57

Activations Density 2.072%

hexadecimal color codes and legal citations

tokens in a specific format or language

foreign words

No Comments

No Known Activations