Normalized Transformer (nGPT) implementation by NVIDIA
active 2024-11-07 → 2026-04-02 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 512 days from 2024-11-07 to 2026-04-02. Pushes: 2 total, peak 2 in a day. Pull requests: 1 total, peak 1 in a day. Issues: 8 total, peak 2 in a day. Comments: 10 total, peak 4 in a day. Stars: 174 total, peak 16 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 |
|---|---|---|---|---|
| iloshchilov | 8 | 2 | 0 | 4 |
| Thiggel | 3 | 0 | 0 | 2 |
| akhilkedia | 2 | 0 | 0 | 1 |
| shashwat14 | 2 | 0 | 0 | 1 |
| mbrukman | 1 | 0 | 0 | 0 |
| AumGupta | 1 | 0 | 0 | 0 |
| gabbard | 1 | 0 | 0 | 1 |
| alxndrTL | 1 | 0 | 0 | 1 |
| afrozenator | 1 | 0 | 1 | 0 |
| asphytheghoul | 1 | 0 | 0 | 0 |
| Kevinstone-199898 | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#9shashwat142025-03-05 18:57nGPT and small models (100M-300M)
- Issue comment#5shashwat142025-02-21 03:23[Not a Bug] Regarding nGPT and Quantization
- Issue#8asphytheghoul2025-01-19 05:16How was the validation split created?
- Issue comment#4iloshchilov2024-12-12 14:48Checkpoints for best 1B models at 8K context
- Issue comment#4Thiggel2024-12-12 14:43Checkpoints for best 1B models at 8K context
- Issue comment#4iloshchilov2024-12-12 14:39Checkpoints for best 1B models at 8K context
- Issue#6Kevinstone-1998982024-12-12 11:27Question when reproducing the experiment in the paper
- Issue comment#4Thiggel2024-12-12 11:21Checkpoints for best 1B models at 8K context
- Issue comment#1akhilkedia2024-12-09 12:32GPT baseline block computation error
- Issue#5akhilkedia2024-12-06 05:37[Not a Bug] Regarding nGPT and Quantization
- Issue comment#1gabbard2024-12-04 15:40GPT baseline block computation error
- Issue#4Thiggel2024-11-30 11:05Checkpoints for best 1B models at 8K context
- Pull request#3afrozenator2024-11-20 14:57
- Issue comment#1alxndrTL2024-11-19 09:12GPT baseline block computation error
- Issue#2iloshchilov2024-11-19 03:26Please move the license at the top of the README to a LICENSE file
- Issue comment#2iloshchilov2024-11-19 03:26Please move the license at the top of the README to a LICENSE file
- Issue#1iloshchilov2024-11-19 02:49GPT baseline block computation error
- Issue comment#1iloshchilov2024-11-19 02:49GPT baseline block computation error
- Issue#2mbrukman2024-11-18 22:34Please move the license at the top of the README to a LICENSE file
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 174 stars here means stars gained during the window, not the repo's star count.