Beam reasoning traces: 53,306 verified SFT rows + 124,945 raw generations

.gitattributes

@@ -58,3 +58,5 @@ 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/sft.jsonl filter=lfs diff=lfs merge=lfs -text
+data/traces_all.jsonl filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,102 @@
+---
+license: cc-by-4.0
+language:
+- en
+task_categories:
+- text-generation
+- question-answering
+tags:
+- reasoning
+- distillation
+- chain-of-thought
+- math
+- science
+- beam
+- synthetic
+size_categories:
+- 100K<n<1M
+pretty_name: Beam Reasoning Traces
+configs:
+- config_name: sft
+  default: true
+  data_files:
+  - split: train
+    path: data/sft.jsonl
+- config_name: all
+  data_files:
+  - split: train
+    path: data/traces_all.jsonl
+---
+
+# Beam Reasoning Traces
+
+Verified math and science reasoning traces from Reflection's
+[Beam-1.0-501B-A23B](https://huggingface.co/reflection/Beam-1.0-501B-A23B) at `reasoning_effort=medium`.
+They were built to distill Beam into a small model, and were used to train
+[ProCreations/Dory-Beam-4B](https://huggingface.co/ProCreations/Dory-Beam-4B), a looped NVIDIA Dory-4B that
+reasons about half as long as the base model at the same accuracy. Made for the Build with Beam challenge.
+
+| Config | Rows | What it is |
+|---|---|---|
+| `sft` (default) | 53,306 | One verified-correct trace per prompt (the shortest correct one), in chat `messages` form |
+| `all` | 124,945 | Every generation over 62,417 prompts, correct or not, with the grader's verdict |
+
+## How it was made
+
+**Prompts.** These are decontaminated, auto-verifiable problems from two sources:
+- [zwhe99/DeepMath-103K](https://huggingface.co/datasets/zwhe99/DeepMath-103K) (MIT): difficulty 5–9, closed-form answers
+- [nvidia/OpenScienceReasoning-2](https://huggingface.co/datasets/nvidia/OpenScienceReasoning-2) (CC-BY-4.0): multiple choice, single-letter answers
+
+Any prompt sharing a 13-gram with MATH-500, AIME 2024/2025, AMC 2023 or GPQA was dropped. A separate check
+against the 198 GPQA-Diamond questions found zero 13-gram overlap. 300 further pool prompts were held out for
+evaluation and are not in this dataset.
+
+**Generation.** Every trace comes from Beam-1.0-501B-A23B with `reasoning_effort=medium`, apart from 30 pilot
+rows at `high`. Most were generated by self-hosting the open weights with SGLang (`lmsysorg/sglang:beam`) at
+temperature 0.7 and top-p 0.99, usually with 2 samples per prompt. The rest came from the Reflection API with
+its default sampling:
+
+| `teacher` | Checkpoint | `sft` rows | Math accuracy | Science accuracy | Median tokens |
+|---|---|---|---|---|---|
+| `selfhost-nvfp4` | Beam NVFP4 on 2× B300 | 48,826 | 85.8% | 76.5% | 1,248 |
+| `selfhost-fp8` | Beam FP8 on 4× B300 | 2,640 | 86.3% | 78.1% | 1,152 |
+| `api` | Reflection API | 1,840 | 83.5% | 78.7% | 987 |
+
+Self-hosted Beam matched the API's accuracy but wrote about 30–37% longer reasoning on token-identical
+prompts. This held for both NVFP4 and FP8, so quantization is unlikely to be the cause; sampling or serving
+differences on Reflection's side are more likely. Keeping the shortest correct of two samples narrows the gap.
+
+**Verification.** Math answers are taken from the last `\boxed{}` and compared with
+[math-verify](https://github.com/huggingface/Math-Verify). Science answers must match the gold letter.
+Truncated generations always count as incorrect.
+
+## Fields
+
+`sft`: `id`, `domain` (`math`/`science`), `source`, `difficulty` (DeepMath only), `teacher`, `reasoning_effort`,
+`messages` (user prompt, then an assistant turn with `reasoning_content` and `content`), `gold`, `n_samples` and
+`n_correct` (Beam's generations and correct ones for this prompt).
+
+`all`: `id`, `domain`, `source`, `topic`, `difficulty`, `teacher`, `reasoning_effort`, `prompt`, `reasoning`,
+`content`, `gold`, `answer_type`, `pred` (the extracted answer), `correct`, `finish_reason`, `completion_tokens`.
+Self-hosted SGLang requests reported only total usage across samples, so their per-sample token counts are
+approximate.
+
+`run_a_ids.txt` lists the 52,660 prompts in the snapshot that Dory-Beam-4B was trained on. Later rows come from
+nightly API top-ups.
+
+## Use
+
+```python
+from datasets import load_dataset
+ds = load_dataset("ProCreations/Beam-Reasoning-Traces", "sft", split="train")
+```
+
+The `messages` format works with any chat template that renders `reasoning_content`, including Dory's and
+Qwen3's. To train on another model, the code in the
+[model repo](https://huggingface.co/ProCreations/Dory-Beam-4B/tree/main/code) rebuilds everything:
+`build_pool.py`, `generate.py`, `verify.py` and `build_dataset.py`.
+
+## License
+
+CC-BY-4.0. Prompts come from DeepMath-103K (MIT) and OpenScienceReasoning-2 (CC-BY-4.0, please credit NVIDIA).
+Responses were generated by Beam-1.0-501B-A23B, which is released under Apache-2.0.

data/sft.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c12a4daf400c1928129f6ed13fc5d12bce11aed3ca76de0611e812853169331
+size 308228420

data/traces_all.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84be386e189cf26a9a10022fff593b094bf36cf41ae4afdb518a7c4be1b4aa5f
+size 921089884