Functional Benchmarks and the Reasoning Gap
active 2024-03-01 → 2025-07-03 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 490 days from 2024-03-01 to 2025-07-03. Pushes: 79 total, peak 14 in a day. Pull requests: 16 total, peak 3 in a day. Issues: 10 total, peak 2 in a day. Comments: 2 total, peak 2 in a day. Stars: 91 total, peak 22 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| saurabh-srivastava | 106 | 79 | 16 | 2 |
| brando90 | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#23brando902024-11-24 06:27is there a MATH variations snapshot that can that can be used and it's original problems companion?
- Issue#22saurabh-srivastava2024-04-26 20:46Compute Phi3 reasoning gap: Add vllm model_api accessor
- Pull request#21saurabh-srivastava2024-04-23 03:45
- Pull request#20saurabh-srivastava2024-04-23 03:42
- Pull request#20saurabh-srivastava2024-04-23 03:18
- Issue#19saurabh-srivastava2024-04-19 19:35Add Llama 3
- Issue#18saurabh-srivastava2024-04-13 17:26Update OAI GPT4 prompt and function to what they made public
- Issue#17saurabh-srivastava2024-03-31 15:53Reward Model'
- Pull request#16saurabh-srivastava2024-03-24 22:55
- Pull request#15saurabh-srivastava2024-03-22 19:15
- Pull request#15saurabh-srivastava2024-03-22 19:11
- Pull request#14saurabh-srivastava2024-03-21 00:02
- Pull request#13saurabh-srivastava2024-03-20 23:59
- Releasesaurabh-srivastava2024-03-20 00:33Reasoning Gap, first publish
- Pull request#12saurabh-srivastava2024-03-20 00:08
- Pull request#13saurabh-srivastava2024-03-16 21:48
- Pull request#12saurabh-srivastava2024-03-15 20:09
- Issue comment#6saurabh-srivastava2024-03-15 02:14Mixtral 8x7B - Replicate Mistral's reported static numbers: Do CoT postprocessing, OR instruction to make it stop doing default CoT?
- Issue#6saurabh-srivastava2024-03-15 02:14Mixtral 8x7B - Replicate Mistral's reported static numbers: Do CoT postprocessing, OR instruction to make it stop doing default CoT?
- Issue comment#6saurabh-srivastava2024-03-15 02:13Mixtral 8x7B - Replicate Mistral's reported static numbers: Do CoT postprocessing, OR instruction to make it stop doing default CoT?
- Issue#5saurabh-srivastava2024-03-15 02:13Claude 3 - Replicate static accuracy of Opus with 0-shot CoT
- Pull request#8saurabh-srivastava2024-03-15 02:13
- Pull request#10saurabh-srivastava2024-03-15 01:51
- Issue#11saurabh-srivastava2024-03-14 20:18Re-evaluation happening for some already cached entries
- Pull request#10saurabh-srivastava2024-03-08 00:36
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 91 stars here means stars gained during the window, not the repo's star count.