messages listlengths 2 2 | response stringlengths 740 66.6k | thinking stringlengths 118 65.2k | answer stringlengths 444 2.84k | native_messages listlengths 2 2 | metadata dict |
|---|---|---|---|---|---|
[
{
"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 distinctive voice, content ... | <think>
Let me analyze the current state of the rubric file and what needs to be done.
Current criteria in the file:
1. Use-case specificity and fit (+8)
2. Trade-off and caveat acknowledgment (+8)
3. Physical and tactile evaluation (+6)
4. Practical functionality and ease of use (+6)
5. Comparison to alternatives or ... | Let me analyze the current state of the rubric file and what needs to be done.
Current criteria in the file:
1. Use-case specificity and fit (+8)
2. Trade-off and caveat acknowledgment (+8)
3. Physical and tactile evaluation (+6)
4. Practical functionality and ease of use (+6)
5. Comparison to alternatives or expectat... | <patch>
```diff
@@ -64,12 +64,6 @@
**Grading:** Full credit: explicit comparison with a specific advantage or disadvantage (e.g., 'much easier than using regular scissors' or 'doesn't snag like my old scissors'). Partial: implied comparison (e.g., 'better than trying to use scissors') or comparison without a clear eva... | [
{
"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 distinctive voice, content ... | {
"example_id": "000ac3402ec595fabad79dc84641ea63",
"task_id": "ee63e76d9926928e9776800cdea18d87",
"split": "train",
"row_id": "43145d361c2b64d69afcd7ab",
"uid": "amzu_2be3f7cf0dfaee9b5c7550b2",
"target_index": 0,
"target_review_id": "Home_and_Kitchen:B08HWBFKRQ:1663649714565",
"target_category": "Home_... |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nLet me analyze the current state of the rubric and what needs to be added or improved.\n\n(...TRUNCATED) | "Let me analyze the current state of the rubric and what needs to be added or improved.\n\nCurrent c(...TRUNCATED) | "<patch>\n```diff\n@@ -51,3 +51,9 @@\n **Points:** +4\n **Grading:** Full credit if the review names(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"00d9c807af9943f40cc02dc2801f06fd","task_id":"cd73be72e8715165458aea8a926f35da","split(...TRUNCATED) |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nLet me analyze the current state and what needs to be added.\n\nCurrent file has 1 criteri(...TRUNCATED) | "Let me analyze the current state and what needs to be added.\n\nCurrent file has 1 criterion:\n1. D(...TRUNCATED) | "<patch>\n```diff\n@@ -3,3 +3,9 @@\n Points: +8 ###\n Grading: 1.0: Explicitly states the product f(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"00e6b7d5cddf135fb8e5bbbf71d600b0","task_id":"1943a6ac7a77618522b46bd45d1d56a5","split(...TRUNCATED) |
[{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | "<think>\nLet me re-examine the task. I need to score each candidate review against the 5 criteria a(...TRUNCATED) | "Let me re-examine the task. I need to score each candidate review against the 5 criteria and output(...TRUNCATED) | "```json\n{\"1\":[1.0,1.0,1.0,0.0,0.0],\"2\":[0.0,0.0,0.0,0.0,0.0],\"3\":[1.0,1.0,1.0,0.0,0.0],\"4\"(...TRUNCATED) | [{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | {"example_id":"00ea09684a7e6f0fed6b89118fbd0ac0","task_id":"1943a6ac7a77618522b46bd45d1d56a5","split(...TRUNCATED) |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nThe task is to add a new criterion about \"Personal project grounding and hands-on use\" t(...TRUNCATED) | "The task is to add a new criterion about \"Personal project grounding and hands-on use\" to the rub(...TRUNCATED) | "<patch>\n```diff\n@@ -15,3 +15,9 @@\n **Points:** +5\n **Grading:** 1.0 = Review explicitly ties qu(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"014b493f1d5fcad091e45130ce66173e","task_id":"c4ed017a84a5bf03e5d91ed66405b1b7","split(...TRUNCATED) |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nLet me analyze the current state and what to add.\n\nCurrent file has 1 criterion:\n1. Pro(...TRUNCATED) | "Let me analyze the current state and what to add.\n\nCurrent file has 1 criterion:\n1. Progressive (...TRUNCATED) | "<patch>\n```diff\n@@ -3,3 +3,9 @@\n **Points:** +4\n **Grading:** 1.0 = Explicitly states the top i(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"015d328943acf02c1651ce698e40d542","task_id":"859681d094625b044058aa79b141fa53","split(...TRUNCATED) |
[{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | "<think>\nThe user wants me to return the exact JSON schema with activation fractions for all 40 can(...TRUNCATED) | "The user wants me to return the exact JSON schema with activation fractions for all 40 candidates a(...TRUNCATED) | "```json\n{\"1\":[1.0,1.0,0.7,0.7,1.0,1.0,0.7],\"2\":[1.0,1.0,0.4,0.7,0.4,0.7,0.4],\"3\":[0.7,1.0,0.(...TRUNCATED) | [{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | {"example_id":"017b140bef00b85392f149241e287082","task_id":"1eb80495f973cfc0f73a0a4ea0259bb2","split(...TRUNCATED) |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nLet me analyze the current state of the rubric and what needs to be added.\n\nCurrent crit(...TRUNCATED) | "Let me analyze the current state of the rubric and what needs to be added.\n\nCurrent criteria:\n1.(...TRUNCATED) | "<patch>\n```diff\n@@ -15,3 +15,9 @@\n **Points:** +12\n **Grading:** 1.0 = Clear expression of skep(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"01f0c347bb325e260856666638ae89b5","task_id":"80deb7c7f0c06d277c977b0c8d49537b","split(...TRUNCATED) |
[{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | "<think>\nLet me analyze the current state of the rubric file and what needs to be done.\n\nCurrent (...TRUNCATED) | "Let me analyze the current state of the rubric file and what needs to be done.\n\nCurrent file has (...TRUNCATED) | "<patch>\n```diff\n@@ -75,3 +75,9 @@\n **Points:** +7\n **Grading:** Full credit if the review's mai(...TRUNCATED) | [{"content":"You are an expert at characterising how one specific person writes Amazon product revie(...TRUNCATED) | {"example_id":"020ef24dcb04826921f48caf031ecb0b","task_id":"9295873f79ec60da76c2b71730e342e8","split(...TRUNCATED) |
[{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | "<think>\nThe user wants me to return the exact JSON schema with 40 candidates, each having 7 activa(...TRUNCATED) | "The user wants me to return the exact JSON schema with 40 candidates, each having 7 activation frac(...TRUNCATED) | "```json\n{\"1\":[0.7,0.0,0.7,0.7,0.7,0.7,0.0],\"2\":[0.7,1.0,1.0,0.7,1.0,1.0,0.7],\"3\":[0.7,1.0,1.(...TRUNCATED) | [{"content":"Score EACH candidate Amazon product review independently against the weighted rubric fo(...TRUNCATED) | {"example_id":"02305aa983d046deb6dbd0843298df58","task_id":"04a2a012818495d6d503e159ff7d9baa","split(...TRUNCATED) |
End of preview. Expand in Data Studio
Amazon Coevolve Mid-training SFT Preview
This is an accepted-only point-in-time snapshot of amazon-mt-pilot-native-v7 for inspecting and launching initial SFT experiments. Collection is still active, so this repository is deliberately marked incomplete.
Configurations and splits
| Configuration | Train | Validation |
|---|---|---|
single |
540 | 11 |
structured |
1,239 | 24 |
diff |
1,240 | 23 |
merged |
2,561 | 50 |
The 50-row merged validation set holds out one complete reviewer, selected deterministically to closely match the snapshot's target-kind and construction/repair proportions. No reviewer occurs in both SFT splits. The smaller track-specific validation counts are the corresponding rows for that same reviewer.
Each row uses a portable SFT target: messages[-1].content and the top-level response are exactly <think>\n{{reasoning}}\n</think>\n{{answer}}. The component fields remain available as thinking and answer, while native_messages preserves the original Qwen-native reasoning_content representation. Only the final assistant turn is supervised via step_loss_mask=1; prompt turns have step_loss_mask=0.
from datasets import load_dataset
dataset = load_dataset("asingh15/amazon-coevolve-midtrain-pilot-sft-preview", "merged")
Snapshot status
- Snapshot:
20260908T144909Z - Accepted rows: 2,611
- Expected completed pilot rows: 8,192
- Provisional validation: 50 rows, one reviewer, all five target kinds, both workflows
- Candidate:
Qwen/Qwen3.8-27B@1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0 - Teacher:
deepseek-ai/DeepSeek-V4-Flash-0731@7872f01b1d1fe23eabc4c98b48bffcef5a386062
See snapshot_manifest.json for split details and SHA-256 checksums, and audit/accepted_trace_index.jsonl for per-example generation metadata. Do not treat this preview validation split as the final official held-out evaluation set.
- Downloads last month
- 68