Upload folder using huggingface_hub

base/60_128_e3_3e-5/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "ibm-granite/granite-3.3-8b-base",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "down_proj",
+    "up_proj",
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "gate_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

base/60_128_e3_3e-5/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4c1ad9f092012e559d066d33bdfb0fef4f52f1781102acd7a30f7a86597eafe
+size 791751704

base/60_128_e3_3e-5/latest

@@ -0,0 +1 @@
+global_step393

base/60_128_e3_3e-5/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e61478b45110045030b61fc1489fe5587c28af27f833c93177e96f4531096b4f
+size 16389

base/60_128_e3_3e-5/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c1ff3d737bd6b102551ab1f38d592dbdcb07fc812330e8e6e1d92669a500041
+size 16389

base/60_128_e3_3e-5/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42d7e82cb948c9db833c65a1dd63284cc57d63fecb5612e38b1ae4b57bd5d636
+size 16389

base/60_128_e3_3e-5/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6da0ea4df6410c949f6cc3576c64bcc0cd7289610744bbd51f820eb5bde6148c
+size 16389

base/60_128_e3_3e-5/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e434881dbd1addcacefdbbf122ad9471cf9f75d78a8a2e81385d9951c1965b42
+size 16389

base/60_128_e3_3e-5/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee080a970e113880d65c438f5b13724c8ed15b09082bed2734bbd98c2789d3e5
+size 16389

base/60_128_e3_3e-5/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba5c055f7df48cf78ab31cb8c66f35453c526308f97c8c71694de7823b089f4a
+size 16389

base/60_128_e3_3e-5/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0352d0d9388103cba491ea18ec7002251700426f2e85d7a55aee3b6fc6709058
+size 16389

base/60_128_e3_3e-5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:626af754f553d744da743dae2dc8fd8d66efe897c6167fdb66841328c3ab38b0
+size 1401

base/60_128_e3_3e-5/special_tokens_map.json

@@ -0,0 +1,45 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<reponame>",
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

base/60_128_e3_3e-5/tokenizer_config.json

@@ -0,0 +1,188 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<fim_prefix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<fim_middle>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<fim_suffix>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<fim_pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<filename>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<gh_stars>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "7": {
+      "content": "<issue_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "8": {
+      "content": "<issue_comment>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "9": {
+      "content": "<issue_closed>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "10": {
+      "content": "<jupyter_start>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "11": {
+      "content": "<jupyter_text>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "12": {
+      "content": "<jupyter_code>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "13": {
+      "content": "<jupyter_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "14": {
+      "content": "<empty_output>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "15": {
+      "content": "<commit_before>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "16": {
+      "content": "<commit_msg>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "17": {
+      "content": "<commit_after>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "18": {
+      "content": "<reponame>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim_prefix>",
+    "<fim_middle>",
+    "<fim_suffix>",
+    "<fim_pad>",
+    "<filename>",
+    "<gh_stars>",
+    "<issue_start>",
+    "<issue_comment>",
+    "<issue_closed>",
+    "<jupyter_start>",
+    "<jupyter_text>",
+    "<jupyter_code>",
+    "<jupyter_output>",
+    "<empty_output>",
+    "<commit_before>",
+    "<commit_msg>",
+    "<commit_after>",
+    "<reponame>"
+  ],
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<reponame>",
+  "padding_side": "left",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}

base/60_128_e3_3e-5/trainer_state.json

@@ -0,0 +1,580 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 393,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03816793893129771,
+      "grad_norm": 0.7092479467391968,
+      "learning_rate": 6e-06,
+      "loss": 1.2264,
+      "step": 5
+    },
+    {
+      "epoch": 0.07633587786259542,
+      "grad_norm": 0.6547515392303467,
+      "learning_rate": 1.3500000000000001e-05,
+      "loss": 1.2422,
+      "step": 10
+    },
+    {
+      "epoch": 0.11450381679389313,
+      "grad_norm": 0.6427267789840698,
+      "learning_rate": 2.1e-05,
+      "loss": 1.1531,
+      "step": 15
+    },
+    {
+      "epoch": 0.15267175572519084,
+      "grad_norm": 0.576325535774231,
+      "learning_rate": 2.8499999999999998e-05,
+      "loss": 1.1218,
+      "step": 20
+    },
+    {
+      "epoch": 0.19083969465648856,
+      "grad_norm": 0.638084888458252,
+      "learning_rate": 2.9991488183550813e-05,
+      "loss": 1.075,
+      "step": 25
+    },
+    {
+      "epoch": 0.22900763358778625,
+      "grad_norm": 0.7571313977241516,
+      "learning_rate": 2.9956925484750416e-05,
+      "loss": 0.9714,
+      "step": 30
+    },
+    {
+      "epoch": 0.26717557251908397,
+      "grad_norm": 0.8607004284858704,
+      "learning_rate": 2.9895841154498428e-05,
+      "loss": 0.8878,
+      "step": 35
+    },
+    {
+      "epoch": 0.3053435114503817,
+      "grad_norm": 0.7799422740936279,
+      "learning_rate": 2.980834350757492e-05,
+      "loss": 0.8266,
+      "step": 40
+    },
+    {
+      "epoch": 0.3435114503816794,
+      "grad_norm": 0.9758269190788269,
+      "learning_rate": 2.969458769487283e-05,
+      "loss": 0.7162,
+      "step": 45
+    },
+    {
+      "epoch": 0.3816793893129771,
+      "grad_norm": 0.9388594031333923,
+      "learning_rate": 2.9554775428284285e-05,
+      "loss": 0.6496,
+      "step": 50
+    },
+    {
+      "epoch": 0.4198473282442748,
+      "grad_norm": 1.2068301439285278,
+      "learning_rate": 2.9389154623024927e-05,
+      "loss": 0.6274,
+      "step": 55
+    },
+    {
+      "epoch": 0.4580152671755725,
+      "grad_norm": 1.0586284399032593,
+      "learning_rate": 2.919801895803047e-05,
+      "loss": 0.5717,
+      "step": 60
+    },
+    {
+      "epoch": 0.4961832061068702,
+      "grad_norm": 1.2534061670303345,
+      "learning_rate": 2.8981707355204967e-05,
+      "loss": 0.532,
+      "step": 65
+    },
+    {
+      "epoch": 0.5343511450381679,
+      "grad_norm": 1.2999385595321655,
+      "learning_rate": 2.8740603378444267e-05,
+      "loss": 0.4759,
+      "step": 70
+    },
+    {
+      "epoch": 0.5725190839694656,
+      "grad_norm": 1.1170881986618042,
+      "learning_rate": 2.8475134553500226e-05,
+      "loss": 0.4313,
+      "step": 75
+    },
+    {
+      "epoch": 0.6106870229007634,
+      "grad_norm": 1.1361089944839478,
+      "learning_rate": 2.8185771609891696e-05,
+      "loss": 0.3924,
+      "step": 80
+    },
+    {
+      "epoch": 0.648854961832061,
+      "grad_norm": 1.1748210191726685,
+      "learning_rate": 2.7873027646206612e-05,
+      "loss": 0.3838,
+      "step": 85
+    },
+    {
+      "epoch": 0.6870229007633588,
+      "grad_norm": 1.185474157333374,
+      "learning_rate": 2.7537457220275198e-05,
+      "loss": 0.3573,
+      "step": 90
+    },
+    {
+      "epoch": 0.7251908396946565,
+      "grad_norm": 1.2445701360702515,
+      "learning_rate": 2.7179655365827598e-05,
+      "loss": 0.3242,
+      "step": 95
+    },
+    {
+      "epoch": 0.7633587786259542,
+      "grad_norm": 4.9902024269104,
+      "learning_rate": 2.6800256537379607e-05,
+      "loss": 0.3284,
+      "step": 100
+    },
+    {
+      "epoch": 0.8015267175572519,
+      "grad_norm": 1.407232403755188,
+      "learning_rate": 2.639993348521748e-05,
+      "loss": 0.2715,
+      "step": 105
+    },
+    {
+      "epoch": 0.8396946564885496,
+      "grad_norm": 1.2961313724517822,
+      "learning_rate": 2.597939606247659e-05,
+      "loss": 0.2913,
+      "step": 110
+    },
+    {
+      "epoch": 0.8778625954198473,
+      "grad_norm": 1.2154102325439453,
+      "learning_rate": 2.5539389966429358e-05,
+      "loss": 0.2548,
+      "step": 115
+    },
+    {
+      "epoch": 0.916030534351145,
+      "grad_norm": 1.1938343048095703,
+      "learning_rate": 2.5080695416214235e-05,
+      "loss": 0.2379,
+      "step": 120
+    },
+    {
+      "epoch": 0.9541984732824428,
+      "grad_norm": 1.1521358489990234,
+      "learning_rate": 2.4604125769350592e-05,
+      "loss": 0.2201,
+      "step": 125
+    },
+    {
+      "epoch": 0.9923664122137404,
+      "grad_norm": 1.1257578134536743,
+      "learning_rate": 2.4110526079492545e-05,
+      "loss": 0.1997,
+      "step": 130
+    },
+    {
+      "epoch": 1.0305343511450382,
+      "grad_norm": 1.0462419986724854,
+      "learning_rate": 2.360077159797917e-05,
+      "loss": 0.1741,
+      "step": 135
+    },
+    {
+      "epoch": 1.0687022900763359,
+      "grad_norm": 1.1780959367752075,
+      "learning_rate": 2.3075766221838187e-05,
+      "loss": 0.1458,
+      "step": 140
+    },
+    {
+      "epoch": 1.1068702290076335,
+      "grad_norm": 0.957473874092102,
+      "learning_rate": 2.2536440890995068e-05,
+      "loss": 0.1497,
+      "step": 145
+    },
+    {
+      "epoch": 1.1450381679389312,
+      "grad_norm": 1.1523516178131104,
+      "learning_rate": 2.198375193752968e-05,
+      "loss": 0.1261,
+      "step": 150
+    },
+    {
+      "epoch": 1.183206106870229,
+      "grad_norm": 1.0756340026855469,
+      "learning_rate": 2.14186793899075e-05,
+      "loss": 0.1286,
+      "step": 155
+    },
+    {
+      "epoch": 1.2213740458015268,
+      "grad_norm": 1.2219468355178833,
+      "learning_rate": 2.0842225235192427e-05,
+      "loss": 0.1388,
+      "step": 160
+    },
+    {
+      "epoch": 1.2595419847328244,
+      "grad_norm": 0.9866964221000671,
+      "learning_rate": 2.0255411642322573e-05,
+      "loss": 0.1326,
+      "step": 165
+    },
+    {
+      "epoch": 1.297709923664122,
+      "grad_norm": 0.9707256555557251,
+      "learning_rate": 1.9659279149599526e-05,
+      "loss": 0.1109,
+      "step": 170
+    },
+    {
+      "epoch": 1.33587786259542,
+      "grad_norm": 0.9792926907539368,
+      "learning_rate": 1.9054884819605057e-05,
+      "loss": 0.1305,
+      "step": 175
+    },
+    {
+      "epoch": 1.3740458015267176,
+      "grad_norm": 0.935828685760498,
+      "learning_rate": 1.8443300364816972e-05,
+      "loss": 0.1091,
+      "step": 180
+    },
+    {
+      "epoch": 1.4122137404580153,
+      "grad_norm": 1.2550859451293945,
+      "learning_rate": 1.7825610247247726e-05,
+      "loss": 0.1034,
+      "step": 185
+    },
+    {
+      "epoch": 1.450381679389313,
+      "grad_norm": 1.0032892227172852,
+      "learning_rate": 1.720290975547559e-05,
+      "loss": 0.0974,
+      "step": 190
+    },
+    {
+      "epoch": 1.4885496183206106,
+      "grad_norm": 0.9914925694465637,
+      "learning_rate": 1.657630306247812e-05,
+      "loss": 0.0871,
+      "step": 195
+    },
+    {
+      "epoch": 1.5267175572519083,
+      "grad_norm": 1.1812331676483154,
+      "learning_rate": 1.594690126771179e-05,
+      "loss": 0.091,
+      "step": 200
+    },
+    {
+      "epoch": 1.5648854961832062,
+      "grad_norm": 0.799457311630249,
+      "learning_rate": 1.5315820426909635e-05,
+      "loss": 0.1012,
+      "step": 205
+    },
+    {
+      "epoch": 1.6030534351145038,
+      "grad_norm": 0.8296777606010437,
+      "learning_rate": 1.4684179573090368e-05,
+      "loss": 0.0829,
+      "step": 210
+    },
+    {
+      "epoch": 1.6412213740458015,
+      "grad_norm": 0.9046019911766052,
+      "learning_rate": 1.4053098732288213e-05,
+      "loss": 0.0853,
+      "step": 215
+    },
+    {
+      "epoch": 1.6793893129770994,
+      "grad_norm": 0.8362064957618713,
+      "learning_rate": 1.3423696937521884e-05,
+      "loss": 0.0726,
+      "step": 220
+    },
+    {
+      "epoch": 1.717557251908397,
+      "grad_norm": 0.7103888988494873,
+      "learning_rate": 1.2797090244524408e-05,
+      "loss": 0.0787,
+      "step": 225
+    },
+    {
+      "epoch": 1.7557251908396947,
+      "grad_norm": 0.6839487552642822,
+      "learning_rate": 1.2174389752752274e-05,
+      "loss": 0.0697,
+      "step": 230
+    },
+    {
+      "epoch": 1.7938931297709924,
+      "grad_norm": 0.833463191986084,
+      "learning_rate": 1.1556699635183027e-05,
+      "loss": 0.0762,
+      "step": 235
+    },
+    {
+      "epoch": 1.83206106870229,
+      "grad_norm": 0.5735163688659668,
+      "learning_rate": 1.0945115180394947e-05,
+      "loss": 0.0653,
+      "step": 240
+    },
+    {
+      "epoch": 1.8702290076335877,
+      "grad_norm": 0.8502929210662842,
+      "learning_rate": 1.034072085040048e-05,
+      "loss": 0.0666,
+      "step": 245
+    },
+    {
+      "epoch": 1.9083969465648853,
+      "grad_norm": 0.6898216009140015,
+      "learning_rate": 9.744588357677433e-06,
+      "loss": 0.0631,
+      "step": 250
+    },
+    {
+      "epoch": 1.9465648854961832,
+      "grad_norm": 0.6909482479095459,
+      "learning_rate": 9.157774764807577e-06,
+      "loss": 0.0574,
+      "step": 255
+    },
+    {
+      "epoch": 1.984732824427481,
+      "grad_norm": 0.7462769746780396,
+      "learning_rate": 8.581320610092509e-06,
+      "loss": 0.0641,
+      "step": 260
+    },
+    {
+      "epoch": 2.0229007633587788,
+      "grad_norm": 0.700482189655304,
+      "learning_rate": 8.016248062470326e-06,
+      "loss": 0.0577,
+      "step": 265
+    },
+    {
+      "epoch": 2.0610687022900764,
+      "grad_norm": 0.6542515754699707,
+      "learning_rate": 7.463559109004936e-06,
+      "loss": 0.0505,
+      "step": 270
+    },
+    {
+      "epoch": 2.099236641221374,
+      "grad_norm": 0.611177921295166,
+      "learning_rate": 6.9242337781618155e-06,
+      "loss": 0.0456,
+      "step": 275
+    },
+    {
+      "epoch": 2.1374045801526718,
+      "grad_norm": 0.7389485239982605,
+      "learning_rate": 6.3992284020208344e-06,
+      "loss": 0.0434,
+      "step": 280
+    },
+    {
+      "epoch": 2.1755725190839694,
+      "grad_norm": 0.572051465511322,
+      "learning_rate": 5.889473920507455e-06,
+      "loss": 0.0387,
+      "step": 285
+    },
+    {
+      "epoch": 2.213740458015267,
+      "grad_norm": 0.5540604591369629,
+      "learning_rate": 5.395874230649409e-06,
+      "loss": 0.0479,
+      "step": 290
+    },
+    {
+      "epoch": 2.2519083969465647,
+      "grad_norm": 0.6635771989822388,
+      "learning_rate": 4.919304583785768e-06,
+      "loss": 0.0467,
+      "step": 295
+    },
+    {
+      "epoch": 2.2900763358778624,
+      "grad_norm": 0.4915962517261505,
+      "learning_rate": 4.460610033570641e-06,
+      "loss": 0.037,
+      "step": 300
+    },
+    {
+      "epoch": 2.32824427480916,
+      "grad_norm": 0.5220828056335449,
+      "learning_rate": 4.020603937523412e-06,
+      "loss": 0.0452,
+      "step": 305
+    },
+    {
+      "epoch": 2.366412213740458,
+      "grad_norm": 0.7002636194229126,
+      "learning_rate": 3.6000665147825233e-06,
+      "loss": 0.0403,
+      "step": 310
+    },
+    {
+      "epoch": 2.404580152671756,
+      "grad_norm": 0.5554895401000977,
+      "learning_rate": 3.1997434626203965e-06,
+      "loss": 0.0468,
+      "step": 315
+    },
+    {
+      "epoch": 2.4427480916030535,
+      "grad_norm": 0.5509517192840576,
+      "learning_rate": 2.8203446341724045e-06,
+      "loss": 0.0428,
+      "step": 320
+    },
+    {
+      "epoch": 2.480916030534351,
+      "grad_norm": 0.4921354055404663,
+      "learning_rate": 2.4625427797248014e-06,
+      "loss": 0.046,
+      "step": 325
+    },
+    {
+      "epoch": 2.519083969465649,
+      "grad_norm": 0.644952654838562,
+      "learning_rate": 2.1269723537933883e-06,
+      "loss": 0.0379,
+      "step": 330
+    },
+    {
+      "epoch": 2.5572519083969465,
+      "grad_norm": 0.49749132990837097,
+      "learning_rate": 1.814228390108304e-06,
+      "loss": 0.0406,
+      "step": 335
+    },
+    {
+      "epoch": 2.595419847328244,
+      "grad_norm": 0.591105043888092,
+      "learning_rate": 1.5248654464997753e-06,
+      "loss": 0.0444,
+      "step": 340
+    },
+    {
+      "epoch": 2.633587786259542,
+      "grad_norm": 0.4215145409107208,
+      "learning_rate": 1.2593966215557295e-06,
+      "loss": 0.0413,
+      "step": 345
+    },
+    {
+      "epoch": 2.67175572519084,
+      "grad_norm": 0.3882170617580414,
+      "learning_rate": 1.0182926447950364e-06,
+      "loss": 0.0395,
+      "step": 350
+    },
+    {
+      "epoch": 2.7099236641221376,
+      "grad_norm": 0.6219843029975891,
+      "learning_rate": 8.019810419695328e-07,
+      "loss": 0.0444,
+      "step": 355
+    },
+    {
+      "epoch": 2.7480916030534353,
+      "grad_norm": 0.47858303785324097,
+      "learning_rate": 6.108453769750721e-07,
+      "loss": 0.037,
+      "step": 360
+    },
+    {
+      "epoch": 2.786259541984733,
+      "grad_norm": 0.39236730337142944,
+      "learning_rate": 4.452245717157177e-07,
+      "loss": 0.0396,
+      "step": 365
+    },
+    {
+      "epoch": 2.8244274809160306,
+      "grad_norm": 0.4560660123825073,
+      "learning_rate": 3.0541230512717456e-07,
+      "loss": 0.0381,
+      "step": 370
+    },
+    {
+      "epoch": 2.8625954198473282,
+      "grad_norm": 0.51278156042099,
+      "learning_rate": 1.9165649242508244e-07,
+      "loss": 0.0352,
+      "step": 375
+    },
+    {
+      "epoch": 2.900763358778626,
+      "grad_norm": 0.3950483798980713,
+      "learning_rate": 1.0415884550157239e-07,
+      "loss": 0.0411,
+      "step": 380
+    },
+    {
+      "epoch": 2.9389312977099236,
+      "grad_norm": 0.3286290168762207,
+      "learning_rate": 4.307451524958428e-08,
+      "loss": 0.0352,
+      "step": 385
+    },
+    {
+      "epoch": 2.9770992366412212,
+      "grad_norm": 0.6519626379013062,
+      "learning_rate": 8.511816449186526e-09,
+      "loss": 0.043,
+      "step": 390
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 393,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.152156030591304e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

base/60_128_e3_3e-5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b54ff940926eb37640edbec9f00ec8ebdbf9214a81d93044a3244d7dc7115cf
+size 8273

base/60_128_e3_3e-5/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/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 . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+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 param_shapes.items():
+
+        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_file, 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_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``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
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+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_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    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_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)