INDEX

Explanations

Initialisms like CNN, CNW and NYSE

oai_token-act-pair · gemini-2.0-flash

news abbreviations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WebVitals

-0.58

 насељу

-0.56

mergeFrom

-0.54

存于互联网档案馆

-0.52

frontmatter

-0.52

tonode

-0.50

<bos>

-0.50

 متعلقه

-0.49

⎪

-0.48

 nahilalakip

-0.48

POSITIVE LOGITS

exitRule

0.59

WebElementEntity

0.57

redor

0.51

✢

0.50

ˌ

0.48

 CYTHON

0.48

PMID

0.47

 noires

0.46

 WaitForSeconds

0.46

AutoSize

0.46

Activations Density 0.371%

Initialisms like CNN, CNW and NYSE

news abbreviations

No Comments

No Known Activations