JesseMarkowitzandClaude Opus 5
fa3769d0fe
Research toward building a content pack from a story corpus, kept on its own branch and independent of the game. Records the selection experiments against blind labels, and settles selection as gate G2 followed by a human review: review.py writes REVIEW.md and a review.json form, apply_review.py checks the filled form and writes situations.json for the next stage. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01C6UDQ9o6L6Ey173U7XVou6
28 lines
2.1 KiB
Plaintext
28 lines
2.1 KiB
Plaintext
qwen2.5:3b-instruct: 250 distinct clusters, 5 already judged
|
|
10/250 11s parse failures 0 last: D15 fit 0.80 "Eviction and housing disputes"
|
|
20/250 24s parse failures 1 last: D25 fit 0.33 "seeking advice or assistance"
|
|
30/250 37s parse failures 1 last: D35 fit 0.73 "romantic interest or potential romance"
|
|
40/250 48s parse failures 1 last: T05 fit 0.87 "con or scam"
|
|
50/250 60s parse failures 1 last: T15 fit 0.40 "manipulation and hidden agenda"
|
|
60/250 71s parse failures 1 last: T25 fit 0.40 "flattery and attention seeking"
|
|
70/250 83s parse failures 1 last: T35 fit 0.20 "Demanding payment/repayment"
|
|
80/250 94s parse failures 1 last: R004 fit 0.47 "confrontation over authority or status"
|
|
90/250 108s parse failures 1 last: R014 fit 0.87 "drinking and socializing"
|
|
100/250 120s parse failures 1 last: X003 fit 1.00 "demand and confrontation"
|
|
110/250 131s parse failures 1 last: X013 fit 0.43 "seeking financial advice"
|
|
120/250 142s parse failures 1 last: X023 fit 0.47 "social gathering or dinner party"
|
|
130/250 153s parse failures 1 last: X033 fit 0.73 "negotiation and payment"
|
|
140/250 165s parse failures 1 last: X043 fit 0.93 "dinner party/hosting"
|
|
150/250 177s parse failures 1 last: X053 fit 0.67 "exploit and exploitation"
|
|
160/250 189s parse failures 1 last: X063 fit 0.67 "confrontation"
|
|
170/250 200s parse failures 1 last: X073 fit 0.60 "coercion and manipulation"
|
|
180/250 214s parse failures 1 last: Y010 fit 0.00 "NONE"
|
|
190/250 226s parse failures 1 last: Y020 fit 0.64 "seeking connections or companionship"
|
|
200/250 236s parse failures 1 last: Y030 fit 0.71 "dining and socializing"
|
|
210/250 248s parse failures 1 last: Y040 fit 1.00 "people seek social status or attention"
|
|
220/250 258s parse failures 1 last: Y050 fit 0.25 "courtship and seduction"
|
|
230/250 268s parse failures 1 last: Y060 fit 0.40 "confrontation over professional issue"
|
|
240/250 280s parse failures 1 last: Y070 fit 0.80 "people pursuing their livelihood or interests"
|
|
done: 245 judged this session in 286s; parse failures 1
|
|
exit 0
|