L28 EVALUATION WRITE-UP
Parts 1 to 3: No AI. Building and running the set, and drafting parts 4, 5 and 7:
AI allowed — keep a record. Part 6 must be your own.

1. Purpose. What is the system for, and for whom? (one or two sentences)

2. What counts as correct. State the rule you applied to every item, before you ran it.

3. The set. How many items, how they were chosen, and how many you expected to fail.

4. Result. Give the count, the settings, and the date. Example form:
   "N of M items passed (CHUNK_WORDS = __, TOP_K = __, run on ____)."
   List every failed item with the stage you attribute it to.

5. What the set did not cover. At least three specific gaps.

6. What I would still not claim. At least one sentence that begins "This result does not show".

7. What would change my mind. What further evidence would raise or lower your confidence?

AI-USE RECORD
Label:            AI allowed — keep a record
Assistant:        (name and version, or "local model")
Date(s):
Prompts:          (paste them, or a link to the conversation)
What I kept:
What I changed:   (and why)
What I verified:  (which claims or results I checked, and against what)
