Skip to content
mini-lab
Features of prelude-2

Feature #2940

Fires on “.” (not labeled yet: read the examples)

Fires on
0.64% of tokens
Mostly in
worksheet
Layer
after block 4

Where it fires

  • worksheet 34%
  • story 18%
  • python 17%
  • chat 16%
  • agent 16%

Makes the model say

4 `53 main off cream(

Makes it not say

Fixeddefemoopencodeconfig# happened man

Strongest examples

On text the dictionary never trained on. The underlined token is where it fires most.

  1. story 2.7
    there was a little girl named Lily. She had a toy named Mr. Bear. They liked
  2. story 2.68
    He wanted to talk to the crab. Tim asked, "Mr. Crab,
  3. chat 2.67
    wanted to play with it.⏎Tim said, "Hello, Mr. Crab!
  4. chat 2.65
    wanted to play with it.⏎Tim said, "Hello, Mr. Crab!
  5. story 2.64
    there was a little girl named Lily. Lily had a toy named Mr. Bear. Mr.
  6. story 2.6
    bear and flew near him. The bird said, "Hello, Mr. Bear! What are
  7. python 2.59
    python⏎def only_even(nums):⏎ return [n for n in nums if
  8. story 2.59
    a small, fast car. The car said, "Hello, Mr. Wagon!
  9. python 2.58
    (nums):⏎ return [n for n in nums if
  10. python 2.52
    has `return [n for n in nums if
  11. chat 2.5
    wanted to play with it.⏎Tim said, "Hello, Mr. Crab!
  12. python 2.48
    ⏎⏎```python⏎def evens(nums):⏎ return [n for n in nums if