Datasets:
schema_version stringclasses 1
value | example_id stringlengths 64 64 | role stringclasses 2
values | variant stringclasses 1
value | split stringclasses 1
value | reviewer_id stringlengths 29 29 | target_id stringlengths 29 29 | trajectory_id stringlengths 64 64 | turn_index int64 0 0 | criterion_index int64 | candidate_index int64 | loss_mask_type stringclasses 1
value | messages listlengths 2 2 | metadata_json stringlengths 908 1.32k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
amazon-c2-varied-rubrics-sft-row-v1 | 8d285ac4a5d12faa73038c4bfc4f9ec0c90e9a22a2aa122b90f913d81ea06268 | rubric_judge | non-diverse | train | amzd_9c945afb912799ecce4a38a3 | amzt_862dec92f9fa07278df87472 | 5f217d7cd8e59e4bcdb306cc785ef6dea4f54ef2ce4cb10532fb4c7a5c55e048 | 0 | null | null | qwen3_final_response | [
{
"role": "user",
"content": "Score EACH candidate Amazon product review independently against the weighted rubric for what this reviewer would likely write about this product. Do not reward general review quality.\n\n<|The Start of Rubric|>\n<rubric>\n### Opens with a personal, first-person purchase story\... | {"available_quality_snapshots":6,"candidate_count":40,"candidate_order":[11,37,26,5,6,10,12,33,13,25,7,35,30,29,34,17,23,18,38,32,36,8,24,9,31,20,27,22,4,14,15,19,1,28,21,3,16,0,2,39],"candidates_sha256":"73bff04f11c142d3bd13009daba80230b38125385f71779537f9489bc9c9a143","config_name":"non-diverse-varied-k4","criterion_... |
amazon-c2-varied-rubrics-sft-row-v1 | a8e0450d9a512103a3a822a24785daa9b34fcb0fb1a0cb6f8d4e4bad68ceb3cd | rubric_writer | non-diverse | train | amzd_ee4fff5a85a0eec86f7e8734 | amzt_a79cb116d7c36a1fa2d098ef | 7a25db35b8414957301ae3853e1027866576237ba5014cc6a1c41c03cb7dee86 | 0 | null | null | qwen3_final_response | [
{
"role": "user",
"content": "You are an expert at characterising how one specific person writes Amazon product reviews.\n\nYou will be shown 8 reviews written by ONE reviewer, each preceded by the corresponding product metadata. Using only these reviews, write a rubric that captures this reviewer's distinc... | {"available_quality_snapshots":12,"candidate_count":40,"config_name":"non-diverse-varied-k4","criterion_count":20,"min_combined_distance":0.3976081180637724,"min_functional_distance":0.19521623612754474,"min_structural_distance":0.6,"parent_trajectory_id":"ed618628a02d6e16ca8889f88897b8a997dd03d5d2858cb26d7b7fa3007e30b... |
amazon-c2-varied-rubrics-sft-row-v1 | 85a5ea871da3d65063a7df34c368d8b3f9adb29fe135efb7cbe93b71915ae050 | rubric_judge | non-diverse | train | amzd_9c71f6f62480deb45725e8d4 | amzt_222b7b98559ccf632139ac77 | 3dd490f0e88cc285cbcfee3376c457ed0fe22dd56df3b041cfbe4c10b93754df | 0 | null | null | qwen3_final_response | [
{
"role": "user",
"content": "Score EACH candidate Amazon product review independently against the weighted rubric for what this reviewer would likely write about this product. Do not reward general review quality.\n\n<|The Start of Rubric|>\n<rubric>\n### Overall positive verdict on the bedding set\n**Crit... | {"available_quality_snapshots":20,"candidate_count":40,"candidate_order":[16,3,12,7,9,28,13,14,30,25,5,33,20,6,38,27,19,4,22,36,1,24,32,17,23,35,10,31,34,39,37,2,18,11,8,21,15,0,26,29],"candidates_sha256":"c3f926097e088719db60a38884d95dac1a7648380fa1727972041d2b0bb83f33","config_name":"non-diverse-varied-k4","criterion... |
amazon-c2-varied-rubrics-sft-row-v1 | e947a99d2a987a689b2fefaed47ea8ca5f8b76f90edf6eab7d0e98ea47170d7f | rubric_writer | non-diverse | train | amzd_f7909a12617578a04b31d332 | amzt_e90e82961286e7aac7124360 | ee7a44699947701fa0e0034773835b373ba8952d3258e957dd653c9930e0d91c | 0 | null | null | qwen3_final_response | [
{
"role": "user",
"content": "You are an expert at characterising how one specific person writes Amazon product reviews.\n\nYou will be shown 8 reviews written by ONE reviewer, each preceded by the corresponding product metadata. Using only these reviews, write a rubric that captures this reviewer's distinc... | {"available_quality_snapshots":19,"candidate_count":40,"config_name":"non-diverse-varied-k4","criterion_count":7,"min_combined_distance":0.11127817311635474,"min_functional_distance":0.07969920337556657,"min_structural_distance":0.1428571428571429,"parent_trajectory_id":"f6a3d9d23a848c815dd5c2eebd9d1f3e074d5df7ee3c75c0... |
amazon-c2-varied-rubrics-sft-row-v1 | 2bc7ebb9b74546383582d0fc9c79ec20ea0415d1f814019b4488afa784f9c6bd | rubric_judge | non-diverse | train | amzd_9c945afb912799ecce4a38a3 | amzt_862dec92f9fa07278df87472 | 5df808921589f89decf9b24397e6317485d8b9057e3ae779bc82b051a988b12e | 0 | null | null | qwen3_final_response | [
{
"role": "user",
"content": "Score EACH candidate Amazon product review independently against the weighted rubric for what this reviewer would likely write about this product. Do not reward general review quality.\n\n<|The Start of Rubric|>\n<rubric>\n### Opens with a personal, first-person purchase story\... | {"available_quality_snapshots":6,"candidate_count":40,"candidate_order":[11,37,26,5,6,10,12,33,13,25,7,35,30,29,34,17,23,18,38,32,36,8,24,9,31,20,27,22,4,14,15,19,1,28,21,3,16,0,2,39],"candidates_sha256":"73bff04f11c142d3bd13009daba80230b38125385f71779537f9489bc9c9a143","config_name":"non-diverse-varied-k4","criterion_... |
amazon-c2-varied-rubrics-sft-row-v1 | 925830cd9b3181938a0d95bdc5d950d23974f73d39ec72b6d8e2456b2e826b04 | rubric_writer | non-diverse | train | amzd_6ab6a379f71095b6b7c372ce | amzt_9e3186bf05c7b8c50d19f80d | b12d82af80138e9673430f43db73f773a70166ccbbc8d97c2c5833204b777522 | 0 | null | null | qwen3_final_response | [{"role":"user","content":"You are an expert at characterising how one specific person writes Amazon(...TRUNCATED) | "{\"available_quality_snapshots\":15,\"candidate_count\":40,\"config_name\":\"non-diverse-varied-k4\(...TRUNCATED) |
amazon-c2-varied-rubrics-sft-row-v1 | bb95bf422c8203055602e77febb998b4d45f9d71999629f1ec1b99d773e2590b | rubric_judge | non-diverse | train | amzd_fd5e7b739856e220a625ebfd | amzt_3e3d576630bc4d4792546c69 | ee444c943c249150df973213970f528de5d14c2c92b07d6bc289a8b14483ec08 | 0 | null | null | qwen3_final_response | [{"role":"user","content":"Score EACH candidate Amazon product review independently against the weig(...TRUNCATED) | "{\"available_quality_snapshots\":5,\"candidate_count\":40,\"candidate_order\":[2,32,22,29,19,34,23,(...TRUNCATED) |
amazon-c2-varied-rubrics-sft-row-v1 | 9b1e3477b4811bb29aa815586bdb99b7e8eef85609d3fef1ae0bf0a67774b04a | rubric_judge | non-diverse | train | amzd_63e350a72904d81e6fbc44f4 | amzt_fbc4194528031e36f441d441 | 3eaa87c1d31ca5a01be58b4afa00b5b1b604c978e3bba8c95e423604a9676020 | 0 | null | null | qwen3_final_response | [{"role":"user","content":"Score EACH candidate Amazon product review independently against the weig(...TRUNCATED) | "{\"available_quality_snapshots\":9,\"candidate_count\":40,\"candidate_order\":[27,14,38,6,20,29,18,(...TRUNCATED) |
amazon-c2-varied-rubrics-sft-row-v1 | a8d5eedeb563dcc4b6361bddca387ff6b4c17b69f8983eab90604799dd62012f | rubric_writer | non-diverse | train | amzd_0b838abd2ad0e1e6f46283a7 | amzt_9806656e37134ba3ba702941 | e19d9cbbcf069f2d0cc3bb01ca3868de10d860900b1baf765a6eacb452abbacc | 0 | null | null | qwen3_final_response | [{"role":"user","content":"You are an expert at characterising how one specific person writes Amazon(...TRUNCATED) | "{\"available_quality_snapshots\":6,\"candidate_count\":40,\"config_name\":\"non-diverse-varied-k4\"(...TRUNCATED) |
amazon-c2-varied-rubrics-sft-row-v1 | ea458f623a261a1cbaec4fb46b3eba5d39e56ae1c59389d89ea20ab091a50e0b | rubric_judge | non-diverse | train | amzd_2b54411781ec873cc32dfe7d | amzt_4fb2753a12f764298bab22a0 | a6a5868cbfc44e5dd8242da495f860f8647c23a9863c13f1bfc37068f74ca0f0 | 0 | null | null | qwen3_final_response | [{"role":"user","content":"Score EACH candidate Amazon product review independently against the weig(...TRUNCATED) | "{\"available_quality_snapshots\":5,\"candidate_count\":40,\"candidate_order\":[25,10,26,17,28,11,1,(...TRUNCATED) |
End of preview. Expand in Data Studio
Amazon C2 varied-rubric distillation
This release exposes six balanced C2 SFT configurations: latent-state and non-diverse candidate panels at
K=1, K=2, and K=4 rubrics per retained reviewer. Each rubric-writer target is paired with one full-rubric
listwise judge target over the same variant's frozen 40-candidate panel. The K arms within a variant share one
reviewer cohort and are exact nested prefixes.
| Config | Train rows | Validation | Test | Train reviewers |
|---|---|---|---|---|
latent-state-varied-k1 |
9,202 | 98 | 98 | 4,601 |
latent-state-varied-k2 |
18,404 | 196 | 196 | 4,601 |
latent-state-varied-k4 |
36,808 | 392 | 392 | 4,601 |
non-diverse-varied-k1 |
9,202 | 98 | 98 | 4,601 |
non-diverse-varied-k2 |
18,404 | 196 | 196 | 4,601 |
non-diverse-varied-k4 |
36,808 | 392 | 392 | 4,601 |
Selection and provenance
Every chosen rubric is an accepted snapshot from one sequential teacher trajectory, not an independent teacher sample. It must pass the pinned quality policy on both candidate panels. Selection ranks eligible snapshots on the arm's own panel, then greedily maximizes an equal blend of rank disagreement and exact criterion-set Jaccard distance. Additional rubrics must clear the pinned combined-distance floor.
The 102 aggregate selection audits are transported as deterministic gzip under
selections/<split>/part-*.json.gz; their uncompressed hashes remain signed by the block manifests. They contain
rubric quality and distance diagnostics but no private review targets or raw oracle vectors. The companion
selection-audit viewer reads those sidecars at a pinned dataset
commit. selection_transport.json maps every signed source sidecar to its compressed transport path, encoding,
size, and hash so the unchanged block manifests remain machine-resolvable.
- Rubric source:
shared-sequential-trajectory-snapshot - Source collection manifest SHA-256:
866c7bc6801f5fa8813a498c51eaa176ce6f5ff4042197a160f2e6162aa42d45 - Source config SHA-256:
f0c8a1ee29cc3b2ad93d3ad14b55149da73d6920e496e26037ee984749b83c52 - Quality policy SHA-256:
3bafbdac25d0c745d75a44fa420ed647916ebf9f0ef64f26870c8690fe11d60c - Selection policy SHA-256:
60d1678f791fec71809667fff5d895ac0a66a7d60e8c8daa27968be62dc481ff - Dataset catalog SHA-256:
8f6080e56114ac5ea1bfffdb497d11f598e03a552b98b588e6e713487c0d89a7 - Row schema:
amazon-c2-varied-rubrics-sft-row-v1 - Loss mask:
qwen3_final_response
- Downloads last month
- 132