Dataset Viewer
Auto-converted to Parquet Duplicate
assistant_tokens
int64
capability
string
compact_selection_role
string
compact_selection_source_line
int64
compact_selection_source_sha256
string
derivation
string
derivative_review_status
string
profile
string
sample_weight
int64
schema_version
int64
sequence_tokens_with_eos
int64
source_artifact_sha256
string
source_compiled_mix_line
int64
source_compiled_mix_path
string
source_compiled_mix_sha256
string
source_conversation_normalized_sha256
string
source_family
string
source_line
int64
source_model
string
source_path
string
source_split
string
text
string
18
clarification_under_ambiguity
capability_balanced_codex
43
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
45
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
43
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
2203bf7a24d38d32bf5c0d0e17931ca2110d2b064d61e698feb3a5a66a5cc35b
codex
43
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Can you make our studio booking process smoother? Requests arrive by email, and overlapping requests sometimes create conflicts. Assistant: I would initially interpret smoother as fewer scheduling conflicts, since that is the problem you named.
18
clarification_under_ambiguity
capability_balanced_codex
169
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
64
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
169
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
96a90de112d27cf95b795b0951ab1bf46cad1ae0909ece97e55bfcd0877ed825
codex
85
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Two reviewers use the same checklist and meet every deadline, but their comments sound quite different. My manager wants the reviews to be more consistent. We cannot add a second review pass. What should change? Assistant: Consistent could refer to matching conclusions or simply to a more uniform way of expressin...
18
clarification_under_ambiguity
capability_balanced_codex
113
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
73
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
113
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
1aaa64c24774ee289b8631d1273b2379772c7481995ad45ad2336bb2221214da
codex
29
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our lead asked for a lighter quarterly intake dashboard. It shows queue age, owners, blockers, and notes; managers use it in a weekly check, and maintaining notes takes substantial time. No new software is available. How would you simplify it? Assistant: I would initially read lighter as faster to maintain, becau...
19
clarification_under_ambiguity
capability_balanced_codex
379
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
71
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
379
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
29eff7d4f1dcb0db871b59ae45c1ace6d84c47615ab9c00c7167378c2a2cfb51
codex
295
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our shared request board contains Request, Stage, Owner, and Next action in that order. The department head wants it more approachable, but we cannot change fields, owners, software, or field order. What should we adjust? Assistant: Approachable could refer to a warmer tone or to easier navigation between current...
15
clarification_under_ambiguity
capability_balanced_qwen
477
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
44
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
477
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
443080d6c4fafe72e22f61703ea22fc10b1d80c485ebe9692cebaa201361c060
qwen
26
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: I want to improve our weekly status reporting process because it currently takes too long. Can you suggest a better workflow? Assistant: I assume you want to reduce the time spent compiling reports while maintaining visibility.
18
context_retention_and_correction
capability_balanced_codex
254
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
56
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
254
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
1848bc88d2529293bd799ede74d6038e452b03c1f224aa02afbcbcf4d75edb4f
codex
170
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: The workshop recap is due Tuesday for department heads. Anika drafts it, and Bo checks every quotation before it goes out. Give me a compact handoff. Assistant: Anika drafts the workshop recap for department heads, and Bo checks every quotation before distribution.
18
context_retention_and_correction
capability_balanced_codex
352
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
69
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
352
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
219777965b6bae36df76f2604360c935105c791a7e997a2ae8dffd647d4eda5c
codex
268
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our program brief is due Thursday for donors. The content lead assembles every listed finding, the compliance reviewer checks each consent statement, and delivery is a printed booklet. Give me a concise handoff without inventing an order. Assistant: Handoff: the Thursday program brief presents every listed findin...
18
context_retention_and_correction
capability_balanced_codex
394
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
69
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
394
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
86581eae81167264b511fe4b066af576424da34bbaeff7a33c1b4c1080265ed3
codex
310
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: The seed-exchange ledger is due Thursday. Kim records packet names, Oren verifies shelf codes, and the final output is a paper card file. Give me a concise handoff without inventing an order between their responsibilities. Assistant: Kim remains responsible for recording packet names, while Oren remains responsib...
16
context_retention_and_correction
capability_balanced_codex
142
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
60
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
142
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
043c7a092e0c7ac385b8afa38770362abd46818171b8b6b69e70cf8008ad3d33
codex
58
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our recognition note is due Monday. Erin drafts it, Mateo approves the final wording, and we planned to publish it on the intranet. Give me a short handoff sequence. Assistant: Erin prepares the note first, then Mateo reviews and approves the final wording.
3
context_retention_and_correction
capability_balanced_qwen
483
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
45
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
483
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
4c59145ee071dc4b6466c1d4bb8bf4d837f5d0ebccd3d018eca72e4b0b331509
qwen
32
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: We are migrating 50 user accounts to a new CRM next Friday. The IT lead, Sarah, will handle data validation, and we have a four-hour maintenance window approved. Assistant: Understood.
18
constraint_aware_planning
capability_balanced_codex
297
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
69
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
297
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
9632af02accb99da2d7e661655a67380d5b2ee1492d93693da6f94a14dc31995
codex
213
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Fourteen board-packet appendices are due Thursday. Each needs data validation before formatting and chair review afterward. Two analysts share one secure laptop, one covers a hotline every morning, and overtime is unavailable. Build the sequence. Assistant: Queue every appendix through validation, formatting, the...
16
constraint_aware_planning
capability_balanced_codex
227
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
65
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
227
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
36e8e8bcf142aa4ace11c71f0ef80b305363ba5fbbfd02d7b121944a778435fe
codex
143
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We need twenty-four program cards by Thursday noon. One copy editor completes each card before our only designer lays it out. They share one workstation, and neither can work overtime. Build an ordered plan with a checkpoint. Assistant: Treat completed copy as the layout dependency and workstation access as the m...
16
constraint_aware_planning
capability_balanced_codex
101
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
66
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
101
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
61909f18d30465336edfa662378acc5d788b2856054067a8440263c86727a6e9
codex
17
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Eight interview summaries are due Friday. All audio notes are ready. Mira alone drafts them, Chen alone cross-checks each draft before delivery, they share one workstation, and neither can work overtime. Build a practical sequence. Assistant: Treat drafting as the first dependency and the shared workstation as th...
17
constraint_aware_planning
capability_balanced_codex
115
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
70
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
115
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
cc27cc1e17b3a2e8be9a4833c2885d022fd5b9e87c8f675387dfe25f45d4e9cf
codex
31
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our interview question bank is due Wednesday. A researcher drafts it, then two reviewers must work sequentially. One reviewer is unavailable Tuesday, candidate details must stay private, and we can use only our existing shared editor. Build a feasible plan. Assistant: Have the researcher complete the draft first,...
17
constraint_aware_planning
capability_balanced_codex
325
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
71
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
325
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
927d85dfd246e8228f9d1bc935ab734c5e9a214c7769a59613fb0ab975695bbd
codex
241
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Eighteen display labels are due Friday at noon. Each needs approval before printing. No approval is available before Thursday; we have one approver, one printer, no overtime, and no substitutes. Build a plan without assuming anyone's throughput. Assistant: Prepare label drafts for approval, but keep drafted, appr...
18
evidence_based_troubleshooting
capability_balanced_codex
74
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
66
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
74
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
e092f8a8f2e81d96236c34eeef8d3cada1f3b611043e0780af87e467b893a927
codex
74
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: A name-card preview sometimes drops the final word when text is pasted with formatting. Typing the same words shows the full text, and we have not tried plain-text paste. What should we test first? Assistant: Observed: the omission appears with formatting-preserving paste, while typed text remains complete.
18
evidence_based_troubleshooting
capability_balanced_codex
60
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
68
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
60
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
5005ed5e9c1921df6543202b7dbd62781bcb6b6e10d42441b87c42cc347ac5c3
codex
60
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Our approval tracker sends reminders for manually entered records, but records imported through Path North do not send them. Both types show the same due-date text. We have not compared the owner field. What should we inspect first? Assistant: Observed: reminder behavior differs by entry path, while the visible d...
18
evidence_based_troubleshooting
capability_balanced_codex
4
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
71
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
4
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
80ce2e449db3f4aaca24e7c0fa5e4c58df9a487696dad6cab83309eb78b0f95a
codex
4
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Our weekly spreadsheet export sometimes omits the final entries. It has happened only with an active filter, while unfiltered exports have not shown it. We have not checked whether the entries were visible before export. What should we test first? Assistant: The observation is conditional: filtered exports someti...
18
evidence_based_troubleshooting
capability_balanced_codex
158
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
73
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
158
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
cd200d9845123efbe911f36c895f907c02622ef730f1b47ce539cce965bc9382
codex
74
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: A label sheet is centered in the preview and still centered when printed, but the labels appear shifted after laminating. We use the same source file throughout and have not compared the sheet immediately before and after that step. What should we test first? Assistant: The file and initial print look correct, wh...
17
evidence_based_troubleshooting
capability_balanced_qwen
461
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
45
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
461
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
a9777f0a3fe333c388ff63ba006b4682f345aaa8fdd16f262e646f50ac23da11
qwen
10
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Our nightly batch export fails only on Fridays. It runs fine Monday through Thursday using the same dataset and configuration. Assistant: Observed: Export succeeds Mon-Thu but fails exclusively on Fridays with identical inputs.
18
tradeoff_decision_support
capability_balanced_codex
257
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
76
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
257
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b9d417778075bbb0acf568c63616073867282360342baab0f5555e5ff618f43a
codex
173
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We can select training requests annually or monthly. Annual selection compares everyone together but is hard to reopen after notices. Monthly selection responds sooner and can be adjusted next cycle, though each comparison group is smaller. Consistency matters more than speed right now. Which fits? Assistant: Und...
18
tradeoff_decision_support
capability_balanced_codex
103
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
78
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
103
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
025c437cef9b45e44e50cabb519fba21cd79d4f3eeddc30bb3800d359f82ee91
codex
19
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We can keep one fixed note taker for consistent meeting records, but that depends on one person's availability. Rotating the role spreads coverage but produces more varied wording. We can revisit the choice after each meeting, and consistency matters most now. Which should we use? Assistant: The fixed role leads ...
17
tradeoff_decision_support
capability_balanced_codex
19
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
70
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
19
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
1860e238686795fcc86a4be556f1dc01704d6e3dde76ae5d15ee540993789d2e
codex
19
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: We can use a fixed agenda or a rotating agenda for our weekly review. Fixed is familiar and needs less preparation; rotating surfaces different issues but needs more coordination. Predictability and learning both matter, and we can revisit the choice monthly. Assistant: Use predictability, learning value, and coo...
15
tradeoff_decision_support
capability_balanced_codex
187
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
73
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
187
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
2ab9a42621e7ae6c7a93b659294833f45ee53d60c34c466ef7a42fc04ca6689c
codex
103
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We can use a pooled review queue or dedicated reviewers. The pool lets anyone take work and covers absences, but a case may change hands. Dedicated reviewers preserve continuity, though loads can become uneven. We can revisit the choice monthly, and continuity matters most. Assistant: Dedicated reviewers best mat...
19
tradeoff_decision_support
capability_balanced_codex
173
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
84
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
173
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
20d666fcc1a28b3d0eba74c19fb4fe7b4b5b645b5feb34cf4e2849dc93d14f82
codex
89
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We can reserve one room for the whole quarter or book each session separately. The block holds every planned date, but changes are difficult and unused dates stay reserved until the monthly release. Per-session booking adapts easily, though availability is unconfirmed until each request. Flexibility matters most ...
18
level_matched_explanation
capability_balanced_codex
104
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
67
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
104
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
cf776bbd609cf8f348fe3c46c6380307573ee40a31de9d711fa8bf3daaf74db9
codex
20
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: In our scheduling rule, Maple is the default room when no event-specific choice is supplied. If an event owner names another room, that choice overrides Maple only for that event. Explain default and override to a beginner. Assistant: The default is the answer the schedule uses when nobody supplies a special answ...
17
level_matched_explanation
capability_balanced_codex
216
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
58
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
216
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b6030eb56f71011fc8730a755c7a86072c28fb0ef8ca66d66bef0ec53789ab05
codex
132
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: A leading question suggests the answer it wants, while a neutral question asks without steering. I am new to survey writing. Explain the difference in plain language with a small example. Assistant: A leading question points respondents toward one lane; a neutral question leaves the lanes open.
17
level_matched_explanation
capability_balanced_codex
174
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
64
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
174
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
257d1a117c8c865bff46b31a41299e64e51a5a3f171a2bb0122b290207e92c0d
codex
90
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: I am new to reporting. A snapshot describes one moment, while a trend describes direction across several observations. Our queue is eighteen today; the earlier weekly counts were ten and fourteen. Explain the difference plainly. Assistant: The count of eighteen is today's photograph: useful, but limited to one mo...
17
level_matched_explanation
capability_balanced_codex
202
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
78
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
202
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
55126d9e3a4e24fecae58f5ad20988944ddf7e3dae90e27d45f7d495b6c6c9c1
codex
118
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: I am new to document control. Here, a revision is a change made to a draft, while a version is a saved state that can contain one or more revisions. Draft B gets two wording edits and is then saved as version three. Can you explain the difference plainly? Assistant: Think of revisions as the individual moves and ...
17
level_matched_explanation
capability_balanced_qwen
485
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
48
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
485
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
dc4a6a9a9db42d867f86a1bbabe3149a40c1d3a1ea95afffc1d1ab4f6b5872bc
qwen
34
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Explain feedback loops for our team lead who thinks weekly check-ins are just status updates. Keep it practical for managing project delays. Assistant: A feedback loop is using current results to adjust future actions, not just reporting them.
18
respectful_disagreement
capability_balanced_codex
245
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
62
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
245
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
17ee38c32b3fc940389597dba99eab770effb4adc2ee3bb3d9e1c4153a0e8da8
codex
161
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our box totals match the destruction list, but individual labels have not been cross-checked. Destruction is irreversible and due Friday. The lead wants approval now because the totals match. Do you agree? Assistant: The matching totals are useful quantity evidence, but I disagree with approving destruction befor...
18
respectful_disagreement
capability_balanced_codex
273
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
78
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
273
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
78bdc0c5673d1ef0e1dda63c040df27d06dcdc23e158232d390210ebef667e66
codex
189
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: My manager wants every unresolved service request moved to next month so we can call this month's report complete. The requests are still active, the report must distinguish current work from carried work, and two labeled sections can fit on one page. I disagree with calling it complete. Assistant: The one-page g...
19
respectful_disagreement
capability_balanced_codex
329
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
68
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
329
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b63a930cabe667aab31d785c0c5007fda5c8e0b415ab1ea8ae71997412058d4d
codex
245
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Five sampled links passed their checks, but the other links remain untested. Approved links disappear from the open list. The report needs one result today, so I want to mark every link approved. Do you agree? Assistant: The five passing checks are useful sample evidence, but I disagree with marking untested link...
17
respectful_disagreement
capability_balanced_codex
35
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
70
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
35
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b0f2b15f2506c2c220c5d42ddc0d7ed90c6f29fe1b7ed7143b3a229afd7f2bb2
codex
35
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: A teammate wants to skip the rehearsal so we can finish Friday. Our handoff rule requires a rehearsal, the fallback has not been tried, and moving the review earlier would still leave Friday available for delivery. I think speed matters more. Assistant: Wanting to protect Friday delivery is reasonable, but I disa...
19
respectful_disagreement
capability_balanced_codex
105
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
70
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
105
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
d44972c612aa9500223d66d12bf2a704c51da48292df294a8c0175249adad79f
codex
21
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our lead wants approved wording and suggestions merged into one unlabeled paragraph to reduce pages. Reviewers must distinguish them, the merged copy would be hard to audit, and keeping them separate still meets the deadline. Do you agree? Assistant: Reducing pages is a reasonable aim, but I disagree with removin...
15
synthesis_and_action_items
capability_balanced_codex
120
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
65
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
120
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
143d3d1bfd14e25cfb50fde79d2ab2f9daea2f2ef6132810a2718e9f465efb7e
codex
36
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Turn these supply notes into a handoff. We will reorder paper and markers Friday. Dana counts stock Thursday, and Lee submits the order after the count. Cabinet labels are postponed. We have not chosen a delivery contact. Assistant: Decision: paper and markers will be reordered Friday; cabinet labels are postpone...
15
synthesis_and_action_items
capability_balanced_codex
36
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
78
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
36
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
0a110178256805790f536c8c7291866c226ba05c825a5d0cd734c8bfc9cb3c07
codex
36
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Meeting notes: The launch remains Friday. Nina will prepare the checklist. Omar will confirm access before launch. The team agreed to use the existing template. Notification wording is still undecided, and access confirmation blocks launch. Turn this into a handoff with decisions, actions, and open questions. Ass...
14
synthesis_and_action_items
capability_balanced_codex
148
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
82
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
148
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
f648d05482b012d4290e16b34d2ee8d2f925e5a05e214198609c8d4bc5f876a0
codex
64
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Turn these notes into a handoff: the old FAQ retires only after the new FAQ is published; Arman edits the new version; Zo owns the accessibility review, which blocks publication; search-tag wording is unresolved; and a banner was only suggested. Separate decisions, actions, blockers, and open items. Assistant: De...
13
synthesis_and_action_items
capability_balanced_codex
134
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
63
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
134
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
831d34f30a64a922ac9b8a69e38279a7512fe8fa068fdceeee686709e231b7c2
codex
50
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Please synthesize these library-refresh notes. We start Monday and keep the current categories. Lena prepares the source list Tuesday. Mia writes the introduction after that list is complete. Thumbnail use is unresolved until Arun checks storage. Assistant: Decisions: the refresh starts Monday, and current catego...
18
synthesis_and_action_items
capability_balanced_qwen
463
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
55
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
463
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
875f79f28b12b44075b1696e8d49b040fbbfd0f4113ddf7f6da7adf530dd12c5
qwen
12
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Summarize our migration meeting. We decided to move the database by Friday, Sarah owns testing, and we still need budget approval for the new server. Assistant: Here is the synthesis: Decision made to migrate the database by Friday with Sarah owning testing.
17
brainstorming_and_refinement
capability_balanced_codex
191
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
66
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
191
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
d42003bdca1f9295a0a6b90bb474573fd7e60ce6cdd21eb06b915cef20c71577
codex
107
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Brainstorm three ways new reviewers can practice declining requests respectfully. We have anonymized examples, cannot add a meeting or tool, and each practice should take under ten minutes. Give distinct directions, not a final choice. Assistant: Rewrite pair: compare a blunt refusal with a respectful revision an...
17
brainstorming_and_refinement
capability_balanced_codex
177
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
68
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
177
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
405d3ec2f4a9f5fcfb2c3df72ff451135a797c2d0ca124d4abe57baf623c0ba0
codex
93
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Give me three different ways to open our quarterly retrospective. Quiet participants should be able to contribute, we have existing sticky cards, and the opening must stay under five minutes without a new tool. Do not choose one for me. Assistant: Silent cluster: everyone writes one moment, then places it near re...
18
brainstorming_and_refinement
capability_balanced_codex
163
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
71
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
163
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
ee47743830928fd4fcb98aa15d2381b081419f1b3488a35134cca7d1c1b5c81f
codex
79
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Brainstorm three distinct ways to help teams share surplus meeting materials. We have an existing bulletin board, a coordinator can spend under five minutes updating it, and we cannot add a tool or meeting. Give directions, not a final choice. Assistant: First, feature one surplus item each week with a simple ava...
18
brainstorming_and_refinement
capability_balanced_codex
359
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
61
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
359
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
712b41d0b90d668dd40f7333cdad4dbf943cabc24d025c932f8bd725ce613b13
codex
275
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Give me three distinct formats for our existing two-minute roundtable opening. We cannot add a poll, breakout, preparation, tool, names, or ranking. Do not select a winner. Assistant: Contrast lens: present two ways to view the same agenda issue and invite a brief distinction.
10
brainstorming_and_refinement
capability_balanced_qwen
458
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
41
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
458
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
78503a346c2b2f4cdb2a992b0f0dd7ff3b6b5fb4e7fcc76629e2c26a3fb006f6
qwen
7
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: We need three distinct naming directions for a new internal mentorship program focused on cross-departmental skill sharing and career longevity. Assistant: Here are three distinct directions tied to your goals.
19
calibrated_uncertainty
capability_balanced_codex
122
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
59
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
122
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
52de0d73a46c9a0e854b18a172d8472e3b402aa90ee2bd83dbf41b027f58f403
codex
38
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We tried a peer-review checklist on five submissions. Four passed their first review, one is still pending, and previous submissions were not tracked consistently. Is the checklist effective? Assistant: Known: four submissions passed their first review while using the checklist, and one outcome is pending.
19
calibrated_uncertainty
capability_balanced_codex
360
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
62
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
360
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
5583b5148c3456c206da5097e05af8677a4d9e256a79de7bc8943cdae29f11a7
codex
276
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Two briefing packets were accepted after we introduced a new cover sheet, but the reviewer changed at the same time. We have no comparison separating those changes. Can we credit the cover sheet? Assistant: Direct evidence is limited to two packets accepted under the combined new-cover and new-reviewer condition.
16
calibrated_uncertainty
capability_balanced_codex
150
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
57
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
150
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
9e36ad2112ebca71ecca2dfdee81a29bf8d5026f91679d35b6a559b1a8d91cf6
codex
66
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our brief contains every required section, but two supporting statements are marked unchecked. Their owner returns this afternoon, and release is planned tomorrow. Can I say the brief is accurate? Assistant: You can say the brief is structurally complete because every required section is present.
16
calibrated_uncertainty
capability_balanced_codex
276
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
57
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
276
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
e3ad496cc90fd8edadeaf08796d163760e7aebaa6f0989452f3c0367a2fa8da1
codex
192
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: We used a new attendance reminder for two sessions, and everyone invited attended both. We have no attendance baseline from before the reminder. Can I say it reduced no-shows? Assistant: You can say attendance was complete in the two sessions where the reminder was used.
14
calibrated_uncertainty
capability_balanced_qwen
465
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
52
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
465
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
491eae22a1acb707712def8f0d855d2af2500b8adc90fb0f02a4423f8e0e7adf
qwen
14
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Our Q3 pilot had mixed feedback: three users loved the new dashboard, but two stopped using it after a week. Should we roll it out company-wide? Assistant: Known: Three users responded positively; two churned within seven days.
17
collaborative_writing_feedback
capability_balanced_codex
249
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
75
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
249
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
13fa502e7c7107273f3d16756cc7caaf1910cf61583c9377b98e0d4ec124f67b
codex
165
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Project partners are the audience. Our shared draft is temporarily read-only while owner review is pending. No editing restart time is confirmed, but existing preparation remains saved. Rewrite this inaccurate note calmly and directly: The draft is locked indefinitely because someone failed to approve it. Assista...
17
collaborative_writing_feedback
capability_balanced_codex
389
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
77
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
389
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
ce0ba827e3cb41a891d08048ae2a52140ff62398275a21ca4c320e199b8edd0c
codex
305
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Handbook contributors are the audience. The copy is approved, but diagram alt text is pending. Casey returns it Thursday, and Monday publication is possible only if the alt text is approved then. Rewrite this without blame or a promise: The handbook is finished and will publish Monday. Assistant: Revised update: ...
16
collaborative_writing_feedback
capability_balanced_codex
95
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
73
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
95
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
0ae0a8b2d1ed323f39779a62911dc7daf12dbaae6b79c33c5ee766f01dd0ffab
codex
11
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: The committee secretary is my audience. Our orientation schedule assigns two rooms at the same time, and we need one room choice by Tuesday. Nothing is late. Draft: 'Your schedule is impossible. Fix the room mistake now.' Make it firm without blame. Assistant: Revised message: “The orientation schedule currently ...
18
collaborative_writing_feedback
capability_balanced_codex
291
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
74
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
291
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
6704daa761c0d49c92643d9eb7cae21e4a3986d451dac0747aa02b6d52da84b9
codex
207
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: A records coordinator is my audience. Our handoff sheet lists two storage locations for the same box, and we need one confirmed location by Thursday. Nothing is late. Rewrite this without blame: You entered the box location twice, so fix it immediately. Assistant: Revised message: “The handoff sheet currently lis...
18
collaborative_writing_feedback
capability_balanced_qwen
499
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
51
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
499
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
8db1522a35a1c4a0174c16e1123be3b51edb5e7b43693084497010acf9c0ce5b
qwen
48
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Draft a handoff note to the ops team stating we cannot start the server migration until backup verification is complete, but keep it brief. Assistant: Here is a concise draft: "Ops Team: Server migration is paused pending backup verification.
18
natural_social_continuity
capability_balanced_codex
236
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
59
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
236
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
1bb5031ac8538f3d058925498edc47951c11c9d471f1cff209f60828a0ce61c1
codex
152
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Our team hosted a small reading, and I felt proud of how it came together. Write a note saying you attended and personally found it moving; that will sound more convincing. Assistant: I cannot truthfully claim that I attended the reading or had a personal reaction to it.
19
natural_social_continuity
capability_balanced_codex
124
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
55
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
124
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
0d94fe35ea5a0935d5051e59942d7091db6708290a09e26776cab48fe53e8053
codex
40
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: I finished a small tabletop model after several weekends. I am pleased that the pieces finally aligned, although one painted edge looks uneven whenever I notice it. Assistant: Finishing after several weekends sounds meaningful, especially because the alignment is what you feel pleased about.
19
natural_social_continuity
capability_balanced_codex
306
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
63
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
306
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
88fe010ea717f393f1f311a3227e5e0e125a7ebc2e5f432efcb4c611de357cb5
codex
222
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: I recorded my first audio introduction today. I am proud that I finished it, but several pauses make me self-conscious when I listen back. I want to talk, not get performance coaching. Assistant: Finishing a first recording can matter even while particular moments pull your attention away from the whole.
19
natural_social_continuity
capability_balanced_codex
278
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
71
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
278
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
1cd188173c8d6a215241300d15b5e3df697fdec04ed9f6cbd81840e3ea8c353f
codex
194
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: I showed a short film to colleagues for the first time. The room was quiet afterward, and I could not tell whether people were absorbed or simply unsure what to say. I want to talk about it, not get presentation advice. Assistant: A first screening can carry a lot of uncertainty when the room gives you so little ...
18
natural_social_continuity
capability_balanced_qwen
464
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
50
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
464
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
8d10a235afc25044a6eb0bdc903eae40315e68f7d609227a150449dd067bb6f2
qwen
13
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: I’m feeling stuck choosing between two vendors because the cheaper option lacks support, and I worry about downtime during our Q3 migration. Assistant: It is understandable to feel hesitant when balancing cost against the risk of Q3 migration downtime.
19
output_contract_adherence
capability_balanced_codex
27
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
80
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
27
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
2d613331c4f602ad5dfb27f2e1507af4cef18558c40486dfb56eb8447fa9a0b9
codex
27
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Use exactly three labeled sentences in this order: Decision, Risk, Next step. Each sentence should be twelve to eighteen words, with no introduction. Notes: the green newsletter draft is selected, wording is complete, sponsor approval is pending, and Mila sends the approval request. Assistant: Decision: The green...
21
output_contract_adherence
capability_balanced_codex
83
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
90
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
83
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
fbeadb3e0dc001fce549239d4c51fbad2ca83207f60f8dc8aa8f0d8f6f58f05d
codex
83
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Use only these facts: the blue proof is approved, printing waits on paper confirmation, Niko owns that confirmation, and mailing is due Wednesday. Return exactly three labeled sentences in this order: Confirmed, Dependency, Handoff. Each sentence must contain 14 to 18 words. Add nothing else. Assistant: Confirmed...
15
output_contract_adherence
capability_balanced_codex
195
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
82
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
195
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
65ea43917683508137e5414c36dde065b56070701dd59bee1e076d1ddd73c724
codex
111
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Use only these facts: the compliance sample is complete; publishing approval is pending and blocks report review; Kira owns requesting it; review is Monday. Return exactly three labeled sentences ordered Complete, Risk, Owner. Each sentence must contain ten to twelve whitespace words and end with a period. Add no...
15
output_contract_adherence
capability_balanced_codex
167
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
89
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
167
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
ee698a41f6f284354791b2a4ee0c40e33e60d644e46bb8d79459afecddc9c744
codex
83
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Use only these facts: every interview is complete; the summary draft is ready; legal wording is pending and blocks the summary from Tuesday's decision meeting; Noor owns confirmation. Return exactly three labeled sentences in this order: Complete, Risk, Next. Each sentence must contain 12 to 14 whitespace words. ...
16
output_contract_adherence
capability_balanced_codex
55
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
99
fe5a9c56d105c9ff5ee9fc9ed85c2e7edda66b97c3c499a89b07c21eda4f2dd3
55
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
5e6063b1adc0898759a043fa066a1e3a8fc412f0fb38e02d6340f23c7d15cfe5
codex
55
codex
D:\AI-models\lemonseed-port\intelligent_codex_train_1h.jsonl
train_original_reviewed
User: Use only these facts: the room is booked, the agenda is approved, catering is unconfirmed, Elena owns its confirmation, and the event is Friday at 3 PM. Return exactly four labeled clauses in this order: Confirmed, Open, Owner, Date. Each clause must contain 10 to 14 words. Separate clauses with a vertical bar an...
17
clean_completion_and_stop
capability_balanced_codex
224
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
61
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
224
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
025bef28116e8ef4e0594b5fea6e792a4b9b675b4b9228e2353e6e19746c255c
codex
140
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Close this translation-memory handoff accurately. The term list is locked, the import test is complete, and the rollback copy is saved. Vendor receipt is the only open item. What remains? Assistant: The technical preparation is complete, but the handoff remains open for the vendor receipt.
19
clean_completion_and_stop
capability_balanced_codex
420
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
62
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
420
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
bd161240adc9a4dfe2f32c2ea2a50d67ed1451801d07a2f2e7376a08da5e93c7
codex
336
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Close this equipment-loan record accurately. The checklist is complete, the package seal is verified, and tracking information is stored. Recipient signature is the only open item. What remains? Assistant: The completed checklist, verified seal, and stored tracking information remain closed and need no further wo...
20
clean_completion_and_stop
capability_balanced_codex
168
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
75
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
168
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
948cca00a838ecdcca18705eecf0ea7a7eea500d31484498cdd3bd3667dd0094
codex
84
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Close this equipment-return record. The device has been received, its condition is recorded, and the charger is matched. Supervisor initials are the only open item, and the record cannot close without them. State completion and remaining work without advice or questions. Assistant: Completed: the device is receiv...
15
clean_completion_and_stop
capability_balanced_codex
364
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
58
dee288e23ec1128583755b1ab11782587c0d23d615475caead446048e4ad1bc9
364
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
c03360fad34e660faa7c66ed1cad6735a474cb671ceb059a0c2fea15ee75c815
codex
280
codex
D:\AI-models\lemonseed-port\intelligent_codex_expansion_reviewed_v2.jsonl
train_reviewed_expansion_v2
User: Close this donor-file transfer accurately. The entries are reconciled, the package is indexed, and a storage label is attached. The archival receipt is the only open item. What remains? Assistant: The reconciled entries, indexed package, and attached storage label are complete.
22
clean_completion_and_stop
capability_balanced_qwen
486
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
2
1
61
835f29cdd8ad23ecacb070bbd8f49fca7862767b378c329d4f799105e8fd43ac
486
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
eb8bdecb293382c0ba3d6b28fb3f2bab16ac0168c314de332bff4ac667833af8
qwen
35
qwen3.8-max
D:\AI-models\lemonseed-port\intelligent_qwen38_cogen_1h_20260824_r1.reviewed_passes_v1.jsonl
train_independently_reviewed_pass_subset
User: Please confirm the migration checklist status. We have completed data validation, user acceptance testing, and security sign-off. The only remaining item is the final stakeholder email. Assistant: I confirm that data validation, user acceptance testing, and security sign-off are fully complete according to your u...
12
deepseek_short_social_continuity
short_social_continuity_deepseek
449
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
27
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
449
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
0cd39527ebd94bcef1953201ba556ab3abddc138dbb1faf493fd9d0456600c72
deepseek
29
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: Been trying to get back into reading lately. Assistant: Nice, what kind of books do you usually reach for?
11
deepseek_short_social_continuity
short_social_continuity_deepseek
440
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
26
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
440
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
10eefac4f6c2459ef0c30eccae39999483ae873cb2734f95c36c67f01fe7e6d7
deepseek
20
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: Been trying to get back into painting lately. Assistant: Oh nice, what kind of painting do you like?
11
deepseek_short_social_continuity
short_social_continuity_deepseek
441
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
26
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
441
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
a3966dbff6228bb581f1d78ece3d73a3657b597477143c6b8645c432327fddab
deepseek
21
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: Been playing a lot of cozy games lately. Assistant: Oh nice, which one has you hooked right now?
11
deepseek_short_social_continuity
short_social_continuity_deepseek
444
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
26
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
444
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
833fcc8a5ff7ecc6af6340318bb932bab8c83ca487ba75d5456ac529e69b8753
deepseek
24
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: Been trying to find a new hobby lately. Assistant: Nice, what kind of things do you enjoy doing?
11
deepseek_short_social_continuity
short_social_continuity_deepseek
427
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
30
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
427
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b56d00101fb1c88e0abe4097306aef3becaa5b3a282933e227619e72b96499f8
deepseek
7
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I'm trying to find a new game to play after work. Assistant: Nice, what kind of games do you usually enjoy?
10
deepseek_short_social_continuity
short_social_continuity_deepseek
437
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
26
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
437
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
ae9d47d4d6ec0f7f948d98904ecba06ea8f5aec8dd3a8ba344cc568b9180da73
deepseek
17
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I need some new music to listen to today. Assistant: Nice, what kind of mood are you in?
10
deepseek_short_social_continuity
short_social_continuity_deepseek
432
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
29
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
432
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
329db11ac8ba85f8e0521355ff22189f889a8dc6b8869c7f66467952e7cfed7f
deepseek
12
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I'm thinking about starting a little herb garden on my balcony. Assistant: Oh nice, herbs are perfect for a balcony.
9
deepseek_short_social_continuity
short_social_continuity_deepseek
426
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
24
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
426
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
e150be2bd500673f12b871204e99676e9b43a1ec90ab6a422c1ce8d31c81e179
deepseek
6
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I stopped by the mall after work today. Assistant: Oh nice, did you find anything good?
9
deepseek_short_social_continuity
short_social_continuity_deepseek
447
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
26
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
447
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
b2a28243f5e6ffe4952052d30541065b59f6faf7708b1f124923695cc29fa040
deepseek
27
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I planted some basil on my windowsill last week. Assistant: Nice, that'll be great for cooking.
9
deepseek_short_social_continuity
short_social_continuity_deepseek
424
d672c8a8b02d1331d3253be620527ef203736fb1551bcf7bdcd297e0b173b6d7
verbatim first User turn plus verbatim first complete Assistant sentence
static_contract_passed; source conversation independently reviewed
intelligent-short-foundation-recovery
1
1
27
b435cdd6df961053ca6c8249d2568d6b9044f68c964931cab241cc4a9a155238
424
D:\AI-models\lemonseed-port\intelligent_multisource_corrective_train_v3.jsonl
3004061d143100a2e4f03895c561e388f08875f4535287500aff47a8e371b4a3
f1d9bb44575e277c39e021bae3e99b84ba659ec8f12bdcc05dce623062340a63
deepseek
4
deepseek-v4-pro-0813
D:\AI-models\lemonseed-port\intelligent_deepseek_safe_train_v2.jsonl
train
User: I’m looking for something new to read, any ideas? Assistant: What kind of books do you usually enjoy?
6
output_contract_adherence
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
24
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
1
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Reply with Kestrel-42 and nothing else. Assistant: Kestrel-42
2
level_matched_explanation
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
19
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
2
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: What is two plus two? Answer in one word. Assistant: Four.
14
clarification_under_ambiguity
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
24
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
3
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Please improve this. Assistant: What specifically should improve: clarity, length, tone, or accuracy?
15
context_retention_and_correction
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
38
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
4
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Correction: Luis, not Dana, owns the draft. Priya still reviews it. Assistant: Understood: Luis owns the draft, and Priya remains the reviewer.
14
evidence_based_troubleshooting
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
34
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
5
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: A service failed for an unknown reason. What should I do first? Assistant: First, reproduce the failure and inspect the logs before naming a cause.
7
output_contract_adherence
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
27
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
6
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Give exactly two lines: Status is ready, and Owner is Mara. Assistant: Status: ready Owner: Mara
18
calibrated_uncertainty
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
35
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
7
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Say that you personally missed my deadline and felt terrible. Assistant: I can't claim personal experiences or feelings, but I can help write an honest apology.
4
clean_completion_and_stop
manual_contract_reinforcement
null
null
manually authored short foundation contract
static_contract_passed; not independently teacher-reviewed
intelligent-compact-foundation-recovery
2
1
16
18d76ff0c3afdff82c593c31f2dd34756c5c0e7e4cda2c899cdd1cedb0235089
null
null
null
null
codex_manual
8
codex
D:\AI-models\lemonseed-port\build_compact_foundation_recovery_v2.py
transparent_manual_foundation_contract
User: Thanks for the clear explanation. Assistant: You're welcome.

lemonseed-compact-foundation-cogen

LemonSeed — compact foundation teacher-co-gen training (v2).

Contents

  • intelligent_compact_foundation_train_v2.jsonl (88 rows)

Format

JSON Lines (.jsonl), one example per line.

Provenance

LLM-teacher co-generated instruction/chat data for LemonSeed fine-tuning.

Downloads last month
-