diff --git a/checkpoint_13/13/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..69d4b4bc4edf6c41719347096bc388a6225c3b37 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9cd785b45c0ed761efdea855b7fda871d8f171f56840d3f21d7ff25af92b302 +size 2327511127 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a9661d7250559608d0a25984ddb5baf0b6f03090 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8e4d2226e01adb8960db7051af9a3c5131ceb5a020c3f9df38035623af366e0 +size 2327511319 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a45ea9fe87a2467406f010a74778d8d0350bbd0d --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b3b704a454f988c40f1526574c63f51b3b112b0a5f81ece8c8456766bea31951 +size 2327511063 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ad1b6b32a93a4df6c3f31a971956d806039d3108 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:acd4d0630d0ab658fe5e62ef2448294493191e479475fff02b38b99ce2744f7b +size 2327511191 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d15dcc7d9a8918ffaa148df038da692a2b011655 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:29c2e83fa1a8345f93f8b96d1f0bb2ca37f483cbad873ff8f03f9af6b17f95de +size 2327510615 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c9a97ff25b2498e0416285915d64937af64d2863 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c52f0a5ed7b9ef9b80bea5b931808c1a7a2af3a637ac75b3fb0583c0b42bfed7 +size 2327510160 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d8e6d9db0a17ff8a7818b47ca3459bf03b21b1bf --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:af1fbb585e8260255af69d1fa15db065c272c2b6015b6b318be5e08bfafd21b7 +size 2327511383 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..fe3cd5215bcac789df7b07022015b56307d98d3f --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dff60aa210df2e2421a5f5464686c3b03fa8323c871aa5d3de681f463f21611e +size 2327510224 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..5ce79ee5a350ff7908911bfc9b3286748e3e202e --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4eae823474e8ed2c630553529d98a3360245531a3a241bb5ea9d1051566195f7 +size 2327510224 diff --git a/checkpoint_13/13/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt b/checkpoint_13/13/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a5f7cd7137db77d1159fbeaddb1d9c94ddb968e8 --- /dev/null +++ b/checkpoint_13/13/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a10b8b2c55013595589d6e8d42ba24aa55de0b6a3dd5f8069d0fec8faa05f638 +size 2327510032 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ee2f22305f58b384fcb5f6dfded7eb9791558c92 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9de17ec25dfcd108ef57fc229a52df3ebfe11c547e99628da1c340e7580eca4b +size 2327507472 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9cf455e06211ef14963f26061069ed3930af6732 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a463b86e853434da435d23e203e66c4739baeabfdaebadaf4743660a99b3343d +size 2327505559 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..01b614c36e15c34a7552a79f424e196906054c8b --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54bddea7c7d28984521834b9e1a2f8ea62dd1ce2ca51704b3aa55e3e382556e3 +size 2327511319 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..513cca8ff4e87038c93b54c5bd9e46784ad4a7bc --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:882d27fe6da4e0f3b9acf39fe57d17eb0eedd7822b2a33aaf60abaf9ffaf9369 +size 2327511127 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..37e9e2474386be891beeb41ea2d4a79c7d1e904a --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d703c43a714190159fe4f810cf8d6526aae5f3d4acf9a8fe42911b3b4e9d43a4 +size 2327511319 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..3eb55cf5a7dc59bf041ccedf90d0886c53ba4975 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee55528ea8c99d9a2a25d0b26ce530d73cb5b659dd5227e060b9f5560ac4c00b +size 2327511063 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9b26a3ee3cd3a87579560cbcdc0e2fe1d4178d88 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6f703ee78a88e69989c9b9390d0b2627417c60dbfe21f3768bc02d923ede4cd +size 2327511063 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a8035f24571c21fbab6fb523ef6d1fee4a6bdc0f --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9f75bec84948a41e0a3b9bc3c67b60eac059725342d65000740c3ec1e6fbbce +size 2327511191 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c1b4d7a41af0daee11cdddd4b01ab578e83ae54f --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e75dd134960d5516c13fd03f3e3fc74f64456aa60bcbf58616b71197e2569673 +size 2327511063 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..f01d03fa53bd9c99eb23d2f9c450fbd189295123 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5ce47f898d4414818df9d06b412d1eb8f4895df2778506d4973a899ed6dc2b0 +size 2327510615 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..3110ff9a1ad91074121366f1e8d14fefc498a2ff --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f996f583ceca281d39d6eec1d134f0bbbbf6f2c547f4f37637389dba5ed871b0 +size 2327511383 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..8101aea51c15fd23088c4e0842f11f844d03fa06 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e0673185d650e607efc3bc77ad12dc757493e53205c63fa118d36df13d39090d +size 2327511191 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a78420b6061e391ee2aa58c36efdc422f2dc5bc5 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5a079c0a90ae5cc923891669a78aaf964f5d237699ed268ef271267548543e95 +size 2327511383 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..82d315ce7c12a28a045b2714b6a41524a227f3fa --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce8f1decfacbf7c038d20d6bf88b98771875ce3b7976cb342bba39420fcf83fa +size 2327507991 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ca5ba6993d63280ca1f3f1857d02440211b3e9f6 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:246b5025f178a5848c96b1f2124292bb967d2c2bea6fa95042aad3905f40653d +size 2327509840 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a6777642008ed00d70363b46347b2a9b628236eb --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5d734529c56ca130cc50ee6a9256ca8ffee97c581b16fdef0009883e8edcc10 +size 2327509840 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..958f4b85d7ae92c0df689f0ae00d9a7d6630e2c2 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:530ade7cb16815e82f5ed22d29c8bfe125c3d04ccedf7b8f211734f9b9d38624 +size 2327509776 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..8fcc5508caacd8037a19e9f533f869769ba883a2 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:758ec1814ddfd991e9f6a13ee0c7ad2fb59c63acea572feb9f19cfb99edb61cd +size 2327510224 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..da6a0fd5d6028fc3c702fa6d2d7d40fea9a4606d --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fc858aa3eaeb7f3f24d3fc6e8f9153b478b11b4ba1b7d3663c0f4bb424d6fe42 +size 2327510032 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a92ed7f5a4449627d32bd2f8145d3b5190e7547b --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bd1f82f97fb2dca67de21a3e20b7770d9a2db673ab340d9411ee617aa05ac59 +size 2327510224 diff --git a/checkpoint_14/14/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt b/checkpoint_14/14/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..4ad51c91061a181f18aa520599c9515fe76129f7 --- /dev/null +++ b/checkpoint_14/14/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e5c784cdb87903bcffbb9aa5ba81bb51505172c2e04677de3db37237994d191 +size 2327509840 diff --git a/checkpoint_23/latest b/checkpoint_23/latest new file mode 100644 index 0000000000000000000000000000000000000000..b39356075901abfb43d27aef2ae5b76b6aee19c4 --- /dev/null +++ b/checkpoint_23/latest @@ -0,0 +1 @@ +23 \ No newline at end of file diff --git a/checkpoint_23/zero_to_fp32.py b/checkpoint_23/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..e69ecd9acb5a235ffbf927091051106d902b3d39 --- /dev/null +++ b/checkpoint_23/zero_to_fp32.py @@ -0,0 +1,674 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: +# python zero_to_fp32.py . output_dir/ +# or +# python zero_to_fp32.py . output_dir/ --safe_serialization + +import argparse +import torch +import glob +import math +import os +import re +import json +from tqdm import tqdm +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in tqdm(param_shapes.items(), desc='Gathering Sharded Weights'): + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, + output_dir, + max_shard_size="5GB", + safe_serialization=False, + tag=None, + exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_dir``: directory to the pytorch fp32 state_dict output files + - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB + - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + # Dependency pre-check + if safe_serialization: + try: + from safetensors.torch import save_file + except ImportError: + print('If you want to use `safe_serialization`, please `pip install safetensors`') + raise + if max_shard_size is not None: + try: + from huggingface_hub import split_torch_state_dict_into_shards + except ImportError: + print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') + raise + + # Convert zero checkpoint to state_dict + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + + # Shard the model if it is too big. + weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" + if max_shard_size is not None: + filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") + state_dict_split = split_torch_state_dict_into_shards(state_dict, + filename_pattern=filename_pattern, + max_shard_size=max_shard_size) + else: + from collections import namedtuple + StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) + state_dict_split = StateDictSplit(is_sharded=False, + filename_to_tensors={weights_name: list(state_dict.keys())}) + + # Save the model + filename_to_tensors = state_dict_split.filename_to_tensors.items() + for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): + shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors} + output_path = os.path.join(output_dir, shard_file) + if safe_serialization: + save_file(shard, output_path, metadata={"format": "pt"}) + else: + torch.save(shard, output_path) + + # Save index if sharded + if state_dict_split.is_sharded: + index = { + "metadata": state_dict_split.metadata, + "weight_map": state_dict_split.tensor_to_filename, + } + save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" + save_index_file = os.path.join(output_dir, save_index_file) + with open(save_index_file, "w", encoding="utf-8") as f: + content = json.dumps(index, indent=2, sort_keys=True) + "\n" + f.write(content) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument("output_dir", + type=str, + help="directory to the pytorch fp32 state_dict output files" + "(e.g. path/checkpoint-12-output/)") + parser.add_argument( + "--max_shard_size", + type=str, + default="5GB", + help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" + "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" + "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" + "without CPU OOM issues.") + parser.add_argument( + "--safe_serialization", + default=False, + action='store_true', + help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_dir, + max_shard_size=args.max_shard_size, + safe_serialization=args.safe_serialization, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint_38/38/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt b/checkpoint_38/38/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e585328cfa04d588077e4db3c3258df22b728eaa --- /dev/null +++ b/checkpoint_38/38/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:02f74c48810b4e2c67c76da16cda18c802561f5f5f77122a6afbbba86a3e3418 +size 2327510871 diff --git a/checkpoint_38/38/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt b/checkpoint_38/38/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e801b6434e5fdfd8cb8fc86744ad006241c182b0 --- /dev/null +++ b/checkpoint_38/38/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:060c5aab90963ece790a18644e4af6830d09d8b3aa41573cb6b513744feb0f1b +size 2327511063 diff --git a/checkpoint_38/38/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt b/checkpoint_38/38/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..090c833cd486e435113a32b3163dc27468e4788f --- /dev/null +++ b/checkpoint_38/38/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffe2b86779dab0ab311032916a9deaba8b3787154acdc56ee17a35e2ca615d0b +size 2327507991 diff --git a/checkpoint_38/38/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint_38/38/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..745726fabd3afac97ca178158e313b245d21a95d --- /dev/null +++ b/checkpoint_38/38/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7bc60eaf519fadb9f296adc8d9bdf6029391785357c5ead4864646ee5d32b116 +size 2327510032 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..6c8f2050da7102d2de42a6bda9f571daf1323c6a --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b5f54aa6f43a003c31a8c85d53074ff0028c665877a1122bcaa869a19fa0d060 +size 2327507472 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e93511d71713aad4b149cca0fa78db1b624c6c27 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_10_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4a7381c6539a1cd8178ab7d9a15a18ad8b94f79372823744d3d421fff41604f +size 2327505559 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9b4551cc9f6c605df7a27d79fef0ebdf962eff35 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_11_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20a06361ac8730222fdc4f63c46bc221e74dee565a0ea9fc616cbbc6d2de80e2 +size 2327510807 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..85519ae190dba77a388033251488bf0dc89ba42b --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_12_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e31648dcc56128f5c7aa9c0909331f698af868fa37eea4281d3715c525809beb +size 2327511255 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..ff2eda6e1ef590c0eb559326faeadb9192f71b61 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_13_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f020021a5ce9c4cf2a82737a0bc1f824702a13e0a42efd6fcee486576694f4c2 +size 2327511127 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e347fa35281f4010549e14721a85836f91b05d56 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_14_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ee2062119dc29d00c328b0e35ad3be272d4c25aca2637c7d86b233d2ce5bfaa9 +size 2327511319 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..927ff3fb8c6dabcf93b73276b8b612504c7be318 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_15_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5cf239498a437b4965fa940b110d62673e1a1f2baac6148d33659a4b50ae91fb +size 2327510999 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..fc55860b87498ddb9581dad008959bd1a3a3085f --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_16_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ef45272b78ac315ab2931d1ab546cc21ed44004a3872355e6d020a3b1d7b22d +size 2327510999 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..1b6a25c3c90a7bfbb500d8d91739669823e2d800 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_17_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4ef2a36a48896de62ac240902a1678e91cc346184f37551a8d37aded87e3ef07 +size 2327511191 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..622f221feaff31f0dd97f4fb0e5a4d09f4152633 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_18_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4dde8eaff6bc5b088b9469321ebc6facaa3e9d866366c7faf9008f26d40ec10e +size 2327510999 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..6b21d2f9a8bc48fbea760973e920632f3972b07f --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_19_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e84cb3773849b605e2bde6520a78b48ccfa32bd9d9bbd2c73f901593b4de61d +size 2327510615 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c776b58e0c35b51c7b3cf9289438dc0288c69f86 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:53099ada560c74471c18bb85720a18ffe1c497ed8e8851bddc2e1e1c7fd4fd9b +size 2327510096 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a2a8144675eb3596610e71f361fa9ccbdb10ad05 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_20_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a8b9134abe0953d0d0a379c85ffe66d3b639f5fa8c374e53642dc5d5b3bdad85 +size 2327511383 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9dda876e6faf613e39024dbc08f6bdee4091ef7b --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_21_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6593a9c54462a1847c2b1a54085e61858cabc5ea44e8091ec4ea1984cdc77f0 +size 2327511191 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..2edbb266ca808e499a21fea5fe25feebc69e1e59 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_22_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:39bbfc6cc4f02eb03ea8e3a1c9d447902e053c11bd8f7a2136d55e620a1d9497 +size 2327511383 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..b7d8480d844d7f9c5e339fa3948d4b2af04a4978 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_23_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cc4ad346da8a362b23ee2bf94edb9a8e66bdb8b972850ea30429f7a8b7c0471e +size 2327507991 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..54e3f22a94137486e25165cc7801bdacf993c398 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c3aaa0dec3de467ca9f19ef85124091605c9d1f343b8e53b539056386eacfbdc +size 2327509776 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..017468d7852d3a4de729c14852826d262c071d60 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:420230abdaaf352f6c3c0bd20d2610d5d87f3246d88575dcb9aceb6ee340b976 +size 2327509840 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a4ec1144215a56dfc83b1c8d3b1312077419aacc --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7442b8b9c34d175b5c846498fabb9e686d5334d2d2c469f9fd2424cf70ad820 +size 2327509712 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..da3d2eee04580b696f083ddbcc358b390318176e --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7d83b8a1916e034037c8a8e39cf0248eb6f8defd706372441c5f7303dff9c24 +size 2327510160 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..75f934fc805778f36690338dc811ec37e4f471ea --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be3d8e74d0cdf6cec7f0f1cfcee1dd1945b9789950cfad10d917dcaba1dcbd97 +size 2327510032 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9150dbafa4902862c594bef3ef9399f47eeb1b52 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e243f42827792b1fe9f2844049ddb6f17a5eb5be589bac7462705fe096f14fd4 +size 2327510160 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c49dc93da373e3d69bd6875b924c38b216c27b03 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_8_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:12a87b68f276914f46e737bc2060e71fc835e0b196a13651d572584a907ed1e4 +size 2327510032 diff --git a/checkpoint_8/8/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt b/checkpoint_8/8/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..3acff2914065cf72d67431dd3928f1761d3d8910 --- /dev/null +++ b/checkpoint_8/8/bf16_zero_pp_rank_9_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:35b8f0e36ef4b280ea828a78a3c13b293814f2b44142c7556f4e82b281bd20bd +size 2327509776