2.2.0

ascend-robot[pytorch][doc]doc file update

2fadeded创建于 2025年10月20日历史提交

文件	最后提交记录	最后更新时间
ckpt_convert_mixtral_hf2mcore.sh	!2160 Fix: 一些脚本的错误参数 Merge pull request !2160 from shenjiarun/master	1 年前
ckpt_convert_mixtral_hf2mcore_qlora.sh	!2415 更新 Mixtral-8x22B QLoRA 权重转换&微调脚本 Merge pull request !2415 from CY-Slightwind/qlora	1 年前
ckpt_convert_mixtral_mg2mg_lora.sh	!2295 更新仓库readme链接和脚本 Merge pull request !2295 from xiecheng/master	1 年前
data_convert_mixtral_instruction.sh	!2673 docs readme modify Merge pull request !2673 from jzh/docs_0517	1 年前
data_convert_mixtral_pairwise.sh	!1858 dpo、simpo方案特性支持：支持vpp、dpp、ep、cp、断点续训等 Merge pull request !1858 from glhyy/master	1 年前
data_convert_mixtral_pretrain.sh	!1555 新增 Mixtral 8x22B 预训练、推理和评估 Merge pull request !1555 from shengjy/master	1 年前
data_convert_mixtral_process_reward.sh	!1882 新增过程奖励模型PRM训练框架 Merge pull request !1882 from zhangjianxiang/prm	1 年前
dpo_mixtral_8x7b_full_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
dpo_mixtral_8x7b_lora_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
evaluate_mixtral_8x22b_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
evaluate_mixtral_8x7b_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
generate_mixtral_8x22b_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
generate_mixtral_8x7b_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
pretrain_mixtral_8x22b_32k_A3_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
pretrain_mixtral_8x22b_32k_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
pretrain_mixtral_8x22b_64k_A3_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
pretrain_mixtral_8x7b_ptd.sh	[pytorch][bugfix] fix reapeated allgather in 2.2.0 Co-authored-by: yanzhixiao<yanzhixiao@h-partners.com> # message auto-generated for no-merge-commit merge: !3477 merge bugfix-0928-2.2.0 into 2.2.0 [pytorch][bugfix] fix reapeated allgather in 2.2.0 Created-by: yanzhixiao23 Commit-by: yanzhixiao Merged-by: ascend-robot Description: fix reapeated allgather in 2.2.0 See merge request: Ascend/MindSpeed-LLM!3477	9 个月前
train_orm_mixtral_8x7B.sh	!3129 [pytorch][feature] update cp args Merge pull request !3129 from yanzhixiao/update-cp-args	11 个月前
train_prm_mixtral_8x7B_full_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
tune_mixtral_8x22b_qlora_ptd.sh	[pytorch][doc]doc file update Co-authored-by: guihaowen666<guihaowen@huawei.com> # message auto-generated for no-merge-commit merge: !3535 merge br_220_update_doc_file_1017 into 2.2.0 [pytorch][doc]doc file update Created-by: guihaowen666 Commit-by: guihaowen666 Merged-by: ascend-robot Description: 2.2.0 doc file update See merge request: Ascend/MindSpeed-LLM!3535	9 个月前
tune_mixtral_8x7B_full_ptd.sh	!2712 [pytorch][feature]upgrading Megatron to r0.12.1 Merge pull request !2712 from yanzhixiao/llm-0.12.0-0526	1 年前
tune_mixtral_8x7b_lora_ptd.sh	!3252 [pytorch][bugfix]Fix LoRA finetune bug Merge pull request !3252 from sunjunjie/master	10 个月前