INDEX

Explanations

python transformers tokenizer

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

THUR

2.44

𝑖

2.42

𝑛

2.37

𝑜

2.20

𝗻

2.19

 cosines

2.17

 inquiet

2.17

𝑦

2.14

 columnspan

2.14

𝑎

2.09

POSITIVE LOGITS

gató

1.92

'"

1.86

관

1.86

<bos>

1.79

 निका

1.75

ਾਂ

1.63

================

1.59

puff

1.56

acco

1.51

های

1.50

Activations Density 0.001%