Star180
224
代码介绍
代码
Issues56
Pull Requests53
流水线
Actions
讨论
Wiki
项目成员54
分析
项目设置
Star180
224
  1. MindSpeed-LLM
  2. /
  3. tests
  4. /
  5. poc
  6. /
  7. deepseek3_671b_64p_perf
ascend-robotascend-robotfeat(pytorch): add deepseek3 8p and 64p scripts in A5
c796a1cd创建于 9 天前历史提交
文件最后提交记录最后更新时间
pretrain_deepseek3_16.2b_256k_bf16_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前
pretrain_deepseek3_16.2b_256k_fp8_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前
pretrain_deepseek3_26.1b_8k_bf16_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前
pretrain_deepseek3_26.1b_8k_fp8_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前
pretrain_deepseek3_75.9b_128k_bf16_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前
pretrain_deepseek3_75.9b_128k_fp8_A5_ptd.sh
feat(pytorch): add deepseek3 8p and 64p scripts in A59 天前