2.2.0

Ii-robot!2971 [pytorch][refactor]MLA module upgrade, parameter alignment with Megatron.

f0d60bfe创建于 2025年7月18日历史提交

文件	最后提交记录	最后更新时间
chat_qwen2_7b_ptd.sh	!2169 core0.8.0 upgrade to mindspeed12.25 Merge pull request !2169 from MeiFei/core0.8.0-1225	1 年前
ckpt_convert_qwen2_hf2mcore.sh	!2828 [pytorch][sh]Add CUDA_DEVICE_MAX_CONNECTIONS Merge pull request !2828 from yanzhixiao/add-environment-variable	1 年前
ckpt_convert_qwen2_mcore2hf.sh	!2828 [pytorch][sh]Add CUDA_DEVICE_MAX_CONNECTIONS Merge pull request !2828 from yanzhixiao/add-environment-variable	1 年前
data_convert_qwen2_instruction.sh	!1759 新增Qwen2-7B/Qwen1.5-4B mcore全参微调脚本 Merge pull request !1759 from 徐源徽/master	1 年前
data_convert_qwen2_pretrain.sh	!1474 适配Qwen2-1.5B模型 Merge pull request !1474 from changlei/master	1 年前
evaluate_qwen2_0point5b_ptd.sh	!1901 添加Qwen2.5-0.5B适配 Merge pull request !1901 from 商元义/qwen25_0point5b	1 年前
evaluate_qwen2_1point5b_ptd.sh	!1901 添加Qwen2.5-0.5B适配 Merge pull request !1901 from 商元义/qwen25_0point5b	1 年前
evaluate_qwen2_72b_ptd.sh	!1477 新模型Qwen2_72B适配mcore分支 Merge pull request !1477 from AresLzk/master	1 年前
evaluate_qwen2_7b_full_ptd.sh	!2169 core0.8.0 upgrade to mindspeed12.25 Merge pull request !2169 from MeiFei/core0.8.0-1225	1 年前
evaluate_qwen2_7b_needle_bench_ptd.sh	!2971 [pytorch][refactor]MLA module upgrade, parameter alignment with Megatron. Merge pull request !2971 from mhh001/master	1 年前
evaluate_qwen2_7b_ptd.sh	!1579 添加Qwen2-7B适配 Merge pull request !1579 from 商元义/qwen2-7b	1 年前
generate_qwen2_0point5b_ptd.sh	!1901 添加Qwen2.5-0.5B适配 Merge pull request !1901 from 商元义/qwen25_0point5b	1 年前
generate_qwen2_1point5b_ptd.sh	!1901 添加Qwen2.5-0.5B适配 Merge pull request !1901 from 商元义/qwen25_0point5b	1 年前
generate_qwen2_72b_ptd.sh	!1477 新模型Qwen2_72B适配mcore分支 Merge pull request !1477 from AresLzk/master	1 年前
generate_qwen2_7b_lora_ptd.sh	!1819 专家以及共享专家支持tp+ep切分&MiniCPM3预训练性能修复&Qwen-LoRA推理bug修复 Merge pull request !1819 from 商元义/master	1 年前
generate_qwen2_7b_ptd.sh	!1579 添加Qwen2-7B适配 Merge pull request !1579 from 商元义/qwen2-7b	1 年前
pretrain_qwen2_0point5b_32k_ptd.sh	!2507 Rename all seq length naming K to k Merge pull request !2507 from shenjiarun/master	1 年前
pretrain_qwen2_0point5b_4k_ptd.sh	!2507 Rename all seq length naming K to k Merge pull request !2507 from shenjiarun/master	1 年前
pretrain_qwen2_1point5b_32k_ptd.sh	!2507 Rename all seq length naming K to k Merge pull request !2507 from shenjiarun/master	1 年前
pretrain_qwen2_1point5b_4k_ptd.sh	!2507 Rename all seq length naming K to k Merge pull request !2507 from shenjiarun/master	1 年前
pretrain_qwen2_72b_32k_ptd.sh	!2929 [pytorch][bugfix]update install_guide Merge pull request !2929 from jzh/master_doc	1 年前
pretrain_qwen2_72b_4k_ptd.sh	!2986 [pytorch][bugfix]qwen2-72b 4k script add HCCL_CONNECT_TIMEOUT Merge pull request !2986 from sunjunjie/master	1 年前
pretrain_qwen2_7b_32k_ptd.sh	!2815 [pytorch][bugfix] fix logs bug Merge pull request !2815 from jzh/master_a	1 年前
pretrain_qwen2_7b_4k_ptd.sh	!2507 Rename all seq length naming K to k Merge pull request !2507 from shenjiarun/master	1 年前
tune_qwen2_72b_lora_pad_4k.sh	!2306 修复新增patch导致patch顺序错乱报错的bug Merge pull request !2306 from shenjiarun/master	1 年前
tune_qwen2_7b_full_ptd.sh	!2169 core0.8.0 upgrade to mindspeed12.25 Merge pull request !2169 from MeiFei/core0.8.0-1225	1 年前
tune_qwen2_7b_lora_ptd.sh	!2169 core0.8.0 upgrade to mindspeed12.25 Merge pull request !2169 from MeiFei/core0.8.0-1225	1 年前