| !2513 [core-llm][dskv3]master同步2.0.0的deepseek3 A3脚本 Merge pull request !2513 from guozhihua/2.0.0 | 1 年前 |
| !2473 update:2.0.0 弃用老rope算子默认使用新算子 Merge pull request !2473 from yanzhixiao/rope-new-2.0.0 | 1 年前 |
| !2012 gpt-moe-8k 256die预训练脚本 Merge pull request !2012 from zhaowei1936/mybranch | 1 年前 |
| !2399 修改grok1和deepseek3_60b_128die脚本 Merge pull request !2399 from guozhihua/master | 1 年前 |
| !2108 poc脚本修改 Merge pull request !2108 from xiecheng/master | 1 年前 |
| !2638 add swap-attention to poc/405b Merge pull request !2638 from jzh/2.0.0 | 1 年前 |
| !2334 update qwen2moe update for faster Merge pull request !2334 from jzh/master | 1 年前 |