INDEX

Explanations

the word "across" followed by words indicating geographical regions or political entities

oai_token-act-pair · gemini-2.0-flash

across

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

Not in Any Lists

No Comments

Negative Logits

 Across

-1.48

Across

-1.37

 across

-1.35

across

-1.34

 ACROSS

-1.33

 kasarigan

-0.90

 crossing

-0.89

 nakalista

-0.88

 crossed

-0.84

 cross

-0.84

POSITIVE LOGITS

the

0.57

an

0.51

0.47

 Reaktion

0.44

urable

0.43

 reaction

0.42

__(/*!

0.42

Idade

0.41

achuca

0.39

 Negara

0.39

Activations Density 1.080%

the word "across" followed by words indicating geographical regions or political entities

across

No Comments

No Known Activations