Mephisto-MathCode: 2M rows (1M Code + 1M Math, no_think), English-filtered and globally shuffled

.gitattributes

@@ -58,3 +58,23 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+data/train_000.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_001.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_002.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_003.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_004.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_005.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_006.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_007.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_008.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_009.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_010.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_011.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_012.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_013.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_014.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_015.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_016.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_017.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_018.jsonl filter=lfs diff=lfs merge=lfs -text
+data/train_019.jsonl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,108 @@
+---
+license: apache-2.0
+language:
+- en
+task_categories:
+- text-generation
+tags:
+- sft
+- math
+- code
+- reasoning
+- shuffled
+size_categories:
+- 1M<n<10M
+configs:
+- config_name: default
+  data_files:
+  - split: train
+    path: data/train_*.jsonl
+---
+
+# Mephisto-MathCode_2M
+
+**2,000,000** non-thinking SFT examples — an even 1M/1M split of math and code
+— drawn from
+[openbmb/UltraData-SFT-2605](https://huggingface.co/datasets/openbmb/UltraData-SFT-2605),
+filtered to English and globally shuffled.
+
+This is a **curation** pass, not a generation one: no model produced these
+answers for this dataset. All credit for the content belongs to OpenBMB. What
+is added here is language filtering, an exact 1M/1M balance, and a global
+shuffle so the file can be streamed without a shuffle buffer.
+
+| `domain` | rows |
+|---|---:|
+| `Code` | 1,000,000 |
+| `Math` | 1,000,000 |
+
+Source config/split: `Code/no_think` and `Math/no_think`.
+
+## Format
+
+```json
+{
+  "uid": "...",
+  "messages": [
+    {"role": "user", "content": "..."},
+    {"role": "assistant", "content": "..."}
+  ],
+  "source": "UltraData-sft-2605",
+  "domain": "Code",
+  "think_type": "no_think"
+}
+```
+
+The source schema is preserved unchanged. Every row is exactly two turns
+(`user`, `assistant`) and `think_type` is `no_think` throughout — responses are
+direct answers with **no chain-of-thought block**, though math answers do show
+their working as ordinary prose/LaTeX.
+
+`domain` distinguishes the two halves, and `uid` maps back to the source row.
+
+## Filtering
+
+Only one filter was applied: **Chinese removal**. Rows whose prompt or answer
+is more than 5% CJK characters were dropped.
+
+| | scanned | kept | dropped (CJK) |
+|---|---:|---:|---:|
+| Code | 1,000,171 | 1,000,000 | 171 (0.017%) |
+| Math | 1,000,154 | 1,000,000 | 154 (0.015%) |
+
+Verified by re-running the filter over the output: 0 CJK rows and 0 duplicate
+`uid`s remained. **No quality, length, or degeneracy filtering was performed** —
+inspect before training.
+
+## Shuffling
+
+Shuffled globally across both halves with a seeded Fisher–Yates permutation
+(ChaCha8, seed 42), so every shard and every prefix is representative:
+
+| | Code | Math |
+|---|---:|---:|
+| shard 000 | 50.4% | 49.6% |
+| shard 019 | 49.2% | 50.8% |
+
+`take(n)` on a streaming load gives an unbiased, balanced sample without an
+extra shuffle buffer.
+
+The shuffle was done with a small memory-mapped Rust tool that permutes an
+index of `(file_id, offset, length)` — 16 bytes per row, so 2M rows cost 32 MB
+of RAM regardless of the 10 GB of text behind them, and line bytes go straight
+from mmap to the output without ever entering the process heap. This matters
+if you want to re-shuffle it yourself on a modest machine.
+
+## Caveats
+
+- Unfiltered for quality. The source is broadly good but nothing here has been
+  checked for correctness, degeneracy, or answer length.
+- Math rows are long (the 1M math rows are ~7.1 GB vs ~2.7 GB for 1M code
+  rows), so a token-balanced mix is **not** 50/50 by row count.
+- English-only by construction; the source split is bilingual.
+- No deduplication beyond exact `uid` collisions.
+
+## Provenance and license
+
+All content from `openbmb/UltraData-SFT-2605` (Apache-2.0); see the source
+dataset for its terms. Released under Apache-2.0.

data/train_000.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1cd9c3e2d35b202264200387db61dc5810a17ac3270f4ec3afa392688c540ae
+size 529943388

data/train_001.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dcab545e60be456095340e4fce80ad6fcd94ebf895f2cca4aec32f7f6631e12
+size 525704829

data/train_002.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f85558e0fcb9577958aedf95333331fabfea5efb70b3d9ed02d0aa39d870088
+size 524994580

data/train_003.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b2212f7c1622e491c9205388669966dc2a1c239895a71821ffb412bfc99ddd2
+size 517858943

data/train_004.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cee2615a62b044f26d0205a9ca7e65fdac8835ead54ccdc651e109d7fcd2273e
+size 526399656

data/train_005.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2159f27f265d7c0ab81b6388c00ff6a4f8ee2ae450fe298bb7b927ccfdfd69d
+size 519353903

data/train_006.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:313e0a12e3ec9e7a5633d35ed8c85c49b33ddffc46d5e43d37ffcf7dce174bea
+size 524190903

data/train_007.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bb44428420bd63e8a33a872bc539c5296a198634a5b75dc7d1f23209e5c6f0e
+size 518750265

data/train_008.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c208c52fe0eb3c3c99f034a3a7fccd77b42e03d3af3bb833c5050a0646bc514
+size 526335774

data/train_009.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8190f23a97977e6c1b616b359d14aa96668e5f1a07ab12da58bdf5c4781de11
+size 527995533

data/train_010.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:147e04b545d15b5877738136192c74b26eeb7590d81114b8a1edcdb9976ecba6
+size 528271176

data/train_011.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:190a6eb3eadefbcd1e194cea10f8216e2f90b2327794770dd709d6e72305aeea
+size 522933165

data/train_012.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:055ddcd977b5381b7e2d183ea9d7be2daf21f1b3d7fe5afdba99a357164ae74b
+size 525128377

data/train_013.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d163de5fe8e32d1a6e606cf5f0e1d2fccfd06156e6356b6aa140d4dd2984a93
+size 526871496

data/train_014.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7acca022c7c1f0ea0797bb1a691d71b8958558eadfd5b0a8ba84d58ca89321d7
+size 526431146

data/train_015.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:898c03b5259243e975bb29fd1377de38fc047b1791aeab77daa98bdc59d95693
+size 522184098

data/train_016.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc5d4a45fa158c3c340c0bfae41e709d6c0b9f147b287b9416c93da19dcdcd2c
+size 530271594

data/train_017.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5c06e85a1296ea51cfe09c6e774b95e836f60bdd283bf626b116064eda32edc
+size 529867971

data/train_018.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a299aa006a4ce35d00fe0ee4af0f69a3ffcf81be8dff81b8f173cbe0523d721
+size 521550367

data/train_019.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f643e98886a68cd913e6e3c6d1bc31511e097c58f94409041467943cbbd3fb6
+size 525808553