Skip to content
mini-lab
Features of prelude-2

Feature #986

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

Fires on
0.74% of tokens
Mostly in
story
Layer
after block 4

Where it fires

  • story 45%
  • python 30%
  • agent 17%
  • chat 6%
  • worksheet 2%

Makes the model say

The[ their until Their `[ READMEassert

Makes it not say

5 Now4 show inside with onstrings

Strongest examples

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

  1. python 2.98
    , `combine("tiger", "robot")` returns `"tiger robot"`.⏎⏎`includes(
  2. python 2.97
    , `combine("tiger", "tree")` returns `"tiger tree"`.⏎<|bos|><|system_start|>tools:
  3. python 2.93
    , `combine("apple", "lemon")` returns `"apple lemon"`.⏎<|bos|><|user_start|>I need
  4. python 2.86
    . For example, `excite("hello")` returns `"hello!"`.⏎<|bos|>8 plus
  5. python 2.86
    . For example, `exclaim("python")` returns `"python!"`.⏎⏎```python⏎
  6. python 2.83
    1])` returns `[10, 14, 2]`.⏎⏎`sum_
  7. python 2.82
    3, 6, 5], 1)⏎False⏎<|bos|>Once upon a
  8. python 2.8
    "!"⏎```⏎⏎>>> excite("sun")⏎"sun!"⏎⏎```python⏎
  9. python 2.79
    example, `hello("Mia")` returns `"Hello, Mia!"`.⏎⏎```python⏎
  10. python 2.78
    ```⏎⏎>>> loud("cat")⏎"CAT"⏎⏎```python⏎
  11. python 2.76
    ```⏎⏎>>> quiet("Robot")⏎"robot"⏎⏎```python⏎
  12. python 2.76
    `welcome("Zoe")` returns `"Hello, Zoe!"`.⏎<|bos|><|system_start|>tools: