INDEX

Explanations

phrases with the word "Act" and URLs

oai_token-act-pair · gemini-2.0-flash

News articles

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

aarrggbb

-0.99

PreferredItem

-0.91

 initComponents

-0.85

 propOrder

-0.84

parsedMessage

-0.80

 CreateTagHelper

-0.79

 resourceCulture

-0.75

@[+][

-0.75

OGND

-0.73

MessageState

-0.73

POSITIVE LOGITS

//

0.44

 actuelles

0.40

 echar

0.39

 sauvages

0.39

 usted

0.39

 alimentaires

0.39

textLabel

0.36

 electrónico

0.36

 blessés

0.36

 dazu

0.36

Activations Density 0.945%

phrases with the word "Act" and URLs

News articles

No Known Activations