Upload folder using huggingface_hub

119_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": [
+    "up_proj",
+    "k_proj",
+    "gate_proj",
+    "o_proj",
+    "v_proj",
+    "q_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

119_128_e3_3e-5/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a17f2e594ea8d340278f1bf9839ceffc6e1127fbd73159dda5aa00b8fba00e7
+size 791751704

119_128_e3_3e-5/latest

@@ -0,0 +1 @@
+global_step804

119_128_e3_3e-5/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b6c12d10c919682100e85736f10dd5975f0f345cde5559461625962187dbc08
+size 16389

119_128_e3_3e-5/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:103732f00d5d46011f1b2c0367a52d9cb34650a8421ec42641018a03cd163624
+size 16389

119_128_e3_3e-5/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:714dbd2b4da330d085c96a0c423f6cf5bead2ba01ab7ff39de6e92c10d00d966
+size 16389

119_128_e3_3e-5/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:beed1ab45634d74d923d276d90ac131dfd019edd1bed099a3ce8fc8c4c7ea36a
+size 16389

119_128_e3_3e-5/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2c65d7ddc0ba7a87f599c2468e0587b13ea9b08b00460d99b832629ff0fd8fd
+size 16389

119_128_e3_3e-5/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12ec5d68d07443be57f2d25d5c50f658716e3daeeaf3465bc5506abc4adf5eec
+size 16389

119_128_e3_3e-5/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ba456ef6ccd84225a8dd87d71723609e72b1eab20d0a596d15d4e32bd5fab85
+size 16389

119_128_e3_3e-5/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d58066683e2e3c0d1fe58f377f4b0c0590559a40711c6235468c0dea49b5cec6
+size 16389

119_128_e3_3e-5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be013fe951879cb0c37f4c4c27620140c6313009bb94ba3fae79a4c8f2288d62
+size 1401

119_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
+  }
+}

119_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
+}

119_128_e3_3e-5/trainer_state.json

@@ -0,0 +1,1154 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 804,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.018691588785046728,
+      "grad_norm": 0.7529240250587463,
+      "learning_rate": 2.9268292682926833e-06,
+      "loss": 1.4292,
+      "step": 5
+    },
+    {
+      "epoch": 0.037383177570093455,
+      "grad_norm": 0.5391725897789001,
+      "learning_rate": 6.585365853658537e-06,
+      "loss": 1.529,
+      "step": 10
+    },
+    {
+      "epoch": 0.056074766355140186,
+      "grad_norm": 0.6047134399414062,
+      "learning_rate": 1.024390243902439e-05,
+      "loss": 1.4678,
+      "step": 15
+    },
+    {
+      "epoch": 0.07476635514018691,
+      "grad_norm": 0.5919334292411804,
+      "learning_rate": 1.3902439024390245e-05,
+      "loss": 1.4613,
+      "step": 20
+    },
+    {
+      "epoch": 0.09345794392523364,
+      "grad_norm": 0.5360497236251831,
+      "learning_rate": 1.7560975609756096e-05,
+      "loss": 1.385,
+      "step": 25
+    },
+    {
+      "epoch": 0.11214953271028037,
+      "grad_norm": 0.4371679425239563,
+      "learning_rate": 2.121951219512195e-05,
+      "loss": 1.3415,
+      "step": 30
+    },
+    {
+      "epoch": 0.1308411214953271,
+      "grad_norm": 0.487432599067688,
+      "learning_rate": 2.4878048780487805e-05,
+      "loss": 1.3458,
+      "step": 35
+    },
+    {
+      "epoch": 0.14953271028037382,
+      "grad_norm": 0.47176486253738403,
+      "learning_rate": 2.8536585365853658e-05,
+      "loss": 1.2981,
+      "step": 40
+    },
+    {
+      "epoch": 0.16822429906542055,
+      "grad_norm": 0.5130228400230408,
+      "learning_rate": 2.9998855676227193e-05,
+      "loss": 1.301,
+      "step": 45
+    },
+    {
+      "epoch": 0.18691588785046728,
+      "grad_norm": 0.5486770868301392,
+      "learning_rate": 2.9991863218779613e-05,
+      "loss": 1.2237,
+      "step": 50
+    },
+    {
+      "epoch": 0.205607476635514,
+      "grad_norm": 0.6038556098937988,
+      "learning_rate": 2.997851699922284e-05,
+      "loss": 1.2286,
+      "step": 55
+    },
+    {
+      "epoch": 0.22429906542056074,
+      "grad_norm": 0.6618132591247559,
+      "learning_rate": 2.995882267387209e-05,
+      "loss": 1.2172,
+      "step": 60
+    },
+    {
+      "epoch": 0.24299065420560748,
+      "grad_norm": 0.658949077129364,
+      "learning_rate": 2.9932788589458942e-05,
+      "loss": 1.1229,
+      "step": 65
+    },
+    {
+      "epoch": 0.2616822429906542,
+      "grad_norm": 0.8154000043869019,
+      "learning_rate": 2.9900425779593876e-05,
+      "loss": 1.0453,
+      "step": 70
+    },
+    {
+      "epoch": 0.2803738317757009,
+      "grad_norm": 0.6942453980445862,
+      "learning_rate": 2.9861747960090068e-05,
+      "loss": 1.0478,
+      "step": 75
+    },
+    {
+      "epoch": 0.29906542056074764,
+      "grad_norm": 44.16806411743164,
+      "learning_rate": 2.981677152315044e-05,
+      "loss": 1.046,
+      "step": 80
+    },
+    {
+      "epoch": 0.3177570093457944,
+      "grad_norm": 0.8369907140731812,
+      "learning_rate": 2.9765515530420406e-05,
+      "loss": 0.9808,
+      "step": 85
+    },
+    {
+      "epoch": 0.3364485981308411,
+      "grad_norm": 0.8791264295578003,
+      "learning_rate": 2.9708001704909293e-05,
+      "loss": 0.9998,
+      "step": 90
+    },
+    {
+      "epoch": 0.35514018691588783,
+      "grad_norm": 0.8433237671852112,
+      "learning_rate": 2.9644254421783808e-05,
+      "loss": 0.9373,
+      "step": 95
+    },
+    {
+      "epoch": 0.37383177570093457,
+      "grad_norm": 0.9307782053947449,
+      "learning_rate": 2.9574300698037523e-05,
+      "loss": 0.9556,
+      "step": 100
+    },
+    {
+      "epoch": 0.3925233644859813,
+      "grad_norm": 0.8911663889884949,
+      "learning_rate": 2.949817018104066e-05,
+      "loss": 0.9024,
+      "step": 105
+    },
+    {
+      "epoch": 0.411214953271028,
+      "grad_norm": 0.9401324391365051,
+      "learning_rate": 2.941589513597517e-05,
+      "loss": 0.854,
+      "step": 110
+    },
+    {
+      "epoch": 0.42990654205607476,
+      "grad_norm": 0.9261566400527954,
+      "learning_rate": 2.9327510432160222e-05,
+      "loss": 0.8139,
+      "step": 115
+    },
+    {
+      "epoch": 0.4485981308411215,
+      "grad_norm": 1.0186972618103027,
+      "learning_rate": 2.923305352827418e-05,
+      "loss": 0.8415,
+      "step": 120
+    },
+    {
+      "epoch": 0.4672897196261682,
+      "grad_norm": 1.1378459930419922,
+      "learning_rate": 2.913256445647902e-05,
+      "loss": 0.7822,
+      "step": 125
+    },
+    {
+      "epoch": 0.48598130841121495,
+      "grad_norm": 1.0377293825149536,
+      "learning_rate": 2.9026085805454182e-05,
+      "loss": 0.8111,
+      "step": 130
+    },
+    {
+      "epoch": 0.5046728971962616,
+      "grad_norm": 1.1396936178207397,
+      "learning_rate": 2.8913662702346876e-05,
+      "loss": 0.6871,
+      "step": 135
+    },
+    {
+      "epoch": 0.5233644859813084,
+      "grad_norm": 1.1937830448150635,
+      "learning_rate": 2.879534279364654e-05,
+      "loss": 0.7301,
+      "step": 140
+    },
+    {
+      "epoch": 0.5420560747663551,
+      "grad_norm": 1.1447718143463135,
+      "learning_rate": 2.867117622499164e-05,
+      "loss": 0.7428,
+      "step": 145
+    },
+    {
+      "epoch": 0.5607476635514018,
+      "grad_norm": 1.3902647495269775,
+      "learning_rate": 2.854121561991718e-05,
+      "loss": 0.7281,
+      "step": 150
+    },
+    {
+      "epoch": 0.5794392523364486,
+      "grad_norm": 1.378641128540039,
+      "learning_rate": 2.840551605755218e-05,
+      "loss": 0.7486,
+      "step": 155
+    },
+    {
+      "epoch": 0.5981308411214953,
+      "grad_norm": 1.1851186752319336,
+      "learning_rate": 2.8264135049276306e-05,
+      "loss": 0.6579,
+      "step": 160
+    },
+    {
+      "epoch": 0.616822429906542,
+      "grad_norm": 1.2538806200027466,
+      "learning_rate": 2.811713251434577e-05,
+      "loss": 0.6374,
+      "step": 165
+    },
+    {
+      "epoch": 0.6355140186915887,
+      "grad_norm": 1.273149847984314,
+      "learning_rate": 2.7964570754498716e-05,
+      "loss": 0.6353,
+      "step": 170
+    },
+    {
+      "epoch": 0.6542056074766355,
+      "grad_norm": 1.2391812801361084,
+      "learning_rate": 2.780651442755083e-05,
+      "loss": 0.6358,
+      "step": 175
+    },
+    {
+      "epoch": 0.6728971962616822,
+      "grad_norm": 1.2821063995361328,
+      "learning_rate": 2.7643030519992505e-05,
+      "loss": 0.5896,
+      "step": 180
+    },
+    {
+      "epoch": 0.6915887850467289,
+      "grad_norm": 1.2253949642181396,
+      "learning_rate": 2.7474188318598977e-05,
+      "loss": 0.5656,
+      "step": 185
+    },
+    {
+      "epoch": 0.7102803738317757,
+      "grad_norm": 1.5632805824279785,
+      "learning_rate": 2.7300059381065646e-05,
+      "loss": 0.5457,
+      "step": 190
+    },
+    {
+      "epoch": 0.7289719626168224,
+      "grad_norm": 1.3139008283615112,
+      "learning_rate": 2.7120717505680908e-05,
+      "loss": 0.5503,
+      "step": 195
+    },
+    {
+      "epoch": 0.7476635514018691,
+      "grad_norm": 1.3450722694396973,
+      "learning_rate": 2.693623870004941e-05,
+      "loss": 0.5441,
+      "step": 200
+    },
+    {
+      "epoch": 0.7663551401869159,
+      "grad_norm": 1.4465762376785278,
+      "learning_rate": 2.6746701148878944e-05,
+      "loss": 0.5245,
+      "step": 205
+    },
+    {
+      "epoch": 0.7850467289719626,
+      "grad_norm": 1.2329579591751099,
+      "learning_rate": 2.6552185180844704e-05,
+      "loss": 0.541,
+      "step": 210
+    },
+    {
+      "epoch": 0.8037383177570093,
+      "grad_norm": 1.2821564674377441,
+      "learning_rate": 2.6352773234544835e-05,
+      "loss": 0.4976,
+      "step": 215
+    },
+    {
+      "epoch": 0.822429906542056,
+      "grad_norm": 1.323890209197998,
+      "learning_rate": 2.6148549823561803e-05,
+      "loss": 0.4998,
+      "step": 220
+    },
+    {
+      "epoch": 0.8411214953271028,
+      "grad_norm": 1.3277744054794312,
+      "learning_rate": 2.5939601500644357e-05,
+      "loss": 0.4919,
+      "step": 225
+    },
+    {
+      "epoch": 0.8598130841121495,
+      "grad_norm": 1.456680417060852,
+      "learning_rate": 2.5726016821025273e-05,
+      "loss": 0.4629,
+      "step": 230
+    },
+    {
+      "epoch": 0.8785046728971962,
+      "grad_norm": 1.4164416790008545,
+      "learning_rate": 2.5507886304890365e-05,
+      "loss": 0.4717,
+      "step": 235
+    },
+    {
+      "epoch": 0.897196261682243,
+      "grad_norm": 1.6264890432357788,
+      "learning_rate": 2.5285302399014808e-05,
+      "loss": 0.4337,
+      "step": 240
+    },
+    {
+      "epoch": 0.9158878504672897,
+      "grad_norm": 1.2559916973114014,
+      "learning_rate": 2.5058359437582863e-05,
+      "loss": 0.4143,
+      "step": 245
+    },
+    {
+      "epoch": 0.9345794392523364,
+      "grad_norm": 1.2903261184692383,
+      "learning_rate": 2.4827153602207723e-05,
+      "loss": 0.3816,
+      "step": 250
+    },
+    {
+      "epoch": 0.9532710280373832,
+      "grad_norm": 1.507580041885376,
+      "learning_rate": 2.4591782881168424e-05,
+      "loss": 0.4048,
+      "step": 255
+    },
+    {
+      "epoch": 0.9719626168224299,
+      "grad_norm": 1.4054561853408813,
+      "learning_rate": 2.4352347027881003e-05,
+      "loss": 0.3973,
+      "step": 260
+    },
+    {
+      "epoch": 0.9906542056074766,
+      "grad_norm": 2.1033549308776855,
+      "learning_rate": 2.4108947518621632e-05,
+      "loss": 0.3989,
+      "step": 265
+    },
+    {
+      "epoch": 1.0074766355140188,
+      "grad_norm": 1.3985251188278198,
+      "learning_rate": 2.3861687509519523e-05,
+      "loss": 0.3541,
+      "step": 270
+    },
+    {
+      "epoch": 1.0261682242990655,
+      "grad_norm": 1.3118833303451538,
+      "learning_rate": 2.3610671792837917e-05,
+      "loss": 0.323,
+      "step": 275
+    },
+    {
+      "epoch": 1.0448598130841122,
+      "grad_norm": 1.3135764598846436,
+      "learning_rate": 2.3356006752561656e-05,
+      "loss": 0.3455,
+      "step": 280
+    },
+    {
+      "epoch": 1.063551401869159,
+      "grad_norm": 1.4228386878967285,
+      "learning_rate": 2.3097800319310165e-05,
+      "loss": 0.3504,
+      "step": 285
+    },
+    {
+      "epoch": 1.0822429906542057,
+      "grad_norm": 1.4900109767913818,
+      "learning_rate": 2.2836161924594928e-05,
+      "loss": 0.3338,
+      "step": 290
+    },
+    {
+      "epoch": 1.1009345794392524,
+      "grad_norm": 1.4829893112182617,
+      "learning_rate": 2.2571202454440864e-05,
+      "loss": 0.3284,
+      "step": 295
+    },
+    {
+      "epoch": 1.1196261682242992,
+      "grad_norm": 1.4645193815231323,
+      "learning_rate": 2.2303034202391312e-05,
+      "loss": 0.3061,
+      "step": 300
+    },
+    {
+      "epoch": 1.1383177570093457,
+      "grad_norm": 1.3824594020843506,
+      "learning_rate": 2.2031770821916414e-05,
+      "loss": 0.3307,
+      "step": 305
+    },
+    {
+      "epoch": 1.1570093457943926,
+      "grad_norm": 1.389153003692627,
+      "learning_rate": 2.17575272782452e-05,
+      "loss": 0.292,
+      "step": 310
+    },
+    {
+      "epoch": 1.1757009345794391,
+      "grad_norm": 1.1946080923080444,
+      "learning_rate": 2.1480419799641695e-05,
+      "loss": 0.2734,
+      "step": 315
+    },
+    {
+      "epoch": 1.194392523364486,
+      "grad_norm": 1.427320122718811,
+      "learning_rate": 2.1200565828145762e-05,
+      "loss": 0.2737,
+      "step": 320
+    },
+    {
+      "epoch": 1.2130841121495326,
+      "grad_norm": 1.4717649221420288,
+      "learning_rate": 2.0918083969799503e-05,
+      "loss": 0.2802,
+      "step": 325
+    },
+    {
+      "epoch": 1.2317757009345796,
+      "grad_norm": 1.2406470775604248,
+      "learning_rate": 2.0633093944380375e-05,
+      "loss": 0.2854,
+      "step": 330
+    },
+    {
+      "epoch": 1.250467289719626,
+      "grad_norm": 1.4278013706207275,
+      "learning_rate": 2.0345716534662234e-05,
+      "loss": 0.257,
+      "step": 335
+    },
+    {
+      "epoch": 1.269158878504673,
+      "grad_norm": 1.289311408996582,
+      "learning_rate": 2.0056073535225946e-05,
+      "loss": 0.2812,
+      "step": 340
+    },
+    {
+      "epoch": 1.2878504672897195,
+      "grad_norm": 1.6223173141479492,
+      "learning_rate": 1.976428770084108e-05,
+      "loss": 0.2633,
+      "step": 345
+    },
+    {
+      "epoch": 1.3065420560747665,
+      "grad_norm": 1.2059963941574097,
+      "learning_rate": 1.9470482694440755e-05,
+      "loss": 0.2638,
+      "step": 350
+    },
+    {
+      "epoch": 1.325233644859813,
+      "grad_norm": 1.31736159324646,
+      "learning_rate": 1.9174783034711544e-05,
+      "loss": 0.2543,
+      "step": 355
+    },
+    {
+      "epoch": 1.34392523364486,
+      "grad_norm": 1.3725029230117798,
+      "learning_rate": 1.887731404332068e-05,
+      "loss": 0.2514,
+      "step": 360
+    },
+    {
+      "epoch": 1.3626168224299064,
+      "grad_norm": 1.3406190872192383,
+      "learning_rate": 1.8578201791803015e-05,
+      "loss": 0.2563,
+      "step": 365
+    },
+    {
+      "epoch": 1.3813084112149534,
+      "grad_norm": 1.5269060134887695,
+      "learning_rate": 1.8277573048130105e-05,
+      "loss": 0.2483,
+      "step": 370
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 1.4657032489776611,
+      "learning_rate": 1.7975555222984166e-05,
+      "loss": 0.2233,
+      "step": 375
+    },
+    {
+      "epoch": 1.4186915887850469,
+      "grad_norm": 1.1344116926193237,
+      "learning_rate": 1.7672276315759654e-05,
+      "loss": 0.22,
+      "step": 380
+    },
+    {
+      "epoch": 1.4373831775700934,
+      "grad_norm": 1.611379861831665,
+      "learning_rate": 1.736786486031531e-05,
+      "loss": 0.2634,
+      "step": 385
+    },
+    {
+      "epoch": 1.45607476635514,
+      "grad_norm": 1.1742888689041138,
+      "learning_rate": 1.706244987049972e-05,
+      "loss": 0.2088,
+      "step": 390
+    },
+    {
+      "epoch": 1.4747663551401868,
+      "grad_norm": 1.533969759941101,
+      "learning_rate": 1.6756160785473403e-05,
+      "loss": 0.2168,
+      "step": 395
+    },
+    {
+      "epoch": 1.4934579439252336,
+      "grad_norm": 1.37166166305542,
+      "learning_rate": 1.6449127414850725e-05,
+      "loss": 0.2102,
+      "step": 400
+    },
+    {
+      "epoch": 1.5121495327102803,
+      "grad_norm": 1.3960413932800293,
+      "learning_rate": 1.6141479883684717e-05,
+      "loss": 0.2134,
+      "step": 405
+    },
+    {
+      "epoch": 1.5308411214953273,
+      "grad_norm": 1.1707730293273926,
+      "learning_rate": 1.5833348577318295e-05,
+      "loss": 0.1918,
+      "step": 410
+    },
+    {
+      "epoch": 1.5495327102803738,
+      "grad_norm": 1.404576301574707,
+      "learning_rate": 1.552486408612507e-05,
+      "loss": 0.1725,
+      "step": 415
+    },
+    {
+      "epoch": 1.5682242990654207,
+      "grad_norm": 1.5168726444244385,
+      "learning_rate": 1.5216157150163359e-05,
+      "loss": 0.2034,
+      "step": 420
+    },
+    {
+      "epoch": 1.5869158878504672,
+      "grad_norm": 1.2666687965393066,
+      "learning_rate": 1.4907358603766686e-05,
+      "loss": 0.1751,
+      "step": 425
+    },
+    {
+      "epoch": 1.6056074766355142,
+      "grad_norm": 1.1670969724655151,
+      "learning_rate": 1.4598599320094367e-05,
+      "loss": 0.1629,
+      "step": 430
+    },
+    {
+      "epoch": 1.6242990654205607,
+      "grad_norm": 1.1096911430358887,
+      "learning_rate": 1.4290010155665629e-05,
+      "loss": 0.1564,
+      "step": 435
+    },
+    {
+      "epoch": 1.6429906542056076,
+      "grad_norm": 1.606921672821045,
+      "learning_rate": 1.3981721894900769e-05,
+      "loss": 0.1765,
+      "step": 440
+    },
+    {
+      "epoch": 1.6616822429906541,
+      "grad_norm": 1.4207931756973267,
+      "learning_rate": 1.3673865194692878e-05,
+      "loss": 0.1677,
+      "step": 445
+    },
+    {
+      "epoch": 1.680373831775701,
+      "grad_norm": 1.5100213289260864,
+      "learning_rate": 1.3366570529033646e-05,
+      "loss": 0.176,
+      "step": 450
+    },
+    {
+      "epoch": 1.6990654205607476,
+      "grad_norm": 1.34877347946167,
+      "learning_rate": 1.3059968133716607e-05,
+      "loss": 0.1995,
+      "step": 455
+    },
+    {
+      "epoch": 1.7177570093457943,
+      "grad_norm": 1.095085859298706,
+      "learning_rate": 1.2754187951141428e-05,
+      "loss": 0.1648,
+      "step": 460
+    },
+    {
+      "epoch": 1.736448598130841,
+      "grad_norm": 1.284014344215393,
+      "learning_rate": 1.2449359575242456e-05,
+      "loss": 0.1431,
+      "step": 465
+    },
+    {
+      "epoch": 1.7551401869158878,
+      "grad_norm": 1.5210627317428589,
+      "learning_rate": 1.2145612196565009e-05,
+      "loss": 0.1829,
+      "step": 470
+    },
+    {
+      "epoch": 1.7738317757009345,
+      "grad_norm": 1.471913456916809,
+      "learning_rate": 1.1843074547512611e-05,
+      "loss": 0.1552,
+      "step": 475
+    },
+    {
+      "epoch": 1.7925233644859813,
+      "grad_norm": 1.162643551826477,
+      "learning_rate": 1.1541874847788381e-05,
+      "loss": 0.1287,
+      "step": 480
+    },
+    {
+      "epoch": 1.811214953271028,
+      "grad_norm": 1.4394381046295166,
+      "learning_rate": 1.1242140750053734e-05,
+      "loss": 0.1641,
+      "step": 485
+    },
+    {
+      "epoch": 1.8299065420560747,
+      "grad_norm": 1.2295345067977905,
+      "learning_rate": 1.0943999285827381e-05,
+      "loss": 0.1419,
+      "step": 490
+    },
+    {
+      "epoch": 1.8485981308411215,
+      "grad_norm": 1.2940006256103516,
+      "learning_rate": 1.06475768116476e-05,
+      "loss": 0.1534,
+      "step": 495
+    },
+    {
+      "epoch": 1.8672897196261682,
+      "grad_norm": 1.3033008575439453,
+      "learning_rate": 1.0352998955520538e-05,
+      "loss": 0.1447,
+      "step": 500
+    },
+    {
+      "epoch": 1.885981308411215,
+      "grad_norm": 1.0752450227737427,
+      "learning_rate": 1.006039056367734e-05,
+      "loss": 0.1485,
+      "step": 505
+    },
+    {
+      "epoch": 1.9046728971962616,
+      "grad_norm": 0.9898340702056885,
+      "learning_rate": 9.769875647662504e-06,
+      "loss": 0.1226,
+      "step": 510
+    },
+    {
+      "epoch": 1.9233644859813084,
+      "grad_norm": 1.3222194910049438,
+      "learning_rate": 9.48157733177611e-06,
+      "loss": 0.1348,
+      "step": 515
+    },
+    {
+      "epoch": 1.9420560747663551,
+      "grad_norm": 1.4025565385818481,
+      "learning_rate": 9.195617800891965e-06,
+      "loss": 0.1207,
+      "step": 520
+    },
+    {
+      "epoch": 1.9607476635514018,
+      "grad_norm": 1.1852610111236572,
+      "learning_rate": 8.912118248673967e-06,
+      "loss": 0.1173,
+      "step": 525
+    },
+    {
+      "epoch": 1.9794392523364486,
+      "grad_norm": 1.3928327560424805,
+      "learning_rate": 8.631198826212516e-06,
+      "loss": 0.1304,
+      "step": 530
+    },
+    {
+      "epoch": 1.9981308411214953,
+      "grad_norm": 1.3401037454605103,
+      "learning_rate": 8.352978591102797e-06,
+      "loss": 0.1221,
+      "step": 535
+    },
+    {
+      "epoch": 2.0149532710280376,
+      "grad_norm": 0.9207136034965515,
+      "learning_rate": 8.077575456986505e-06,
+      "loss": 0.1003,
+      "step": 540
+    },
+    {
+      "epoch": 2.033644859813084,
+      "grad_norm": 1.1109579801559448,
+      "learning_rate": 7.805106143578365e-06,
+      "loss": 0.0965,
+      "step": 545
+    },
+    {
+      "epoch": 2.052336448598131,
+      "grad_norm": 1.1716532707214355,
+      "learning_rate": 7.5356861271987265e-06,
+      "loss": 0.0859,
+      "step": 550
+    },
+    {
+      "epoch": 2.0710280373831775,
+      "grad_norm": 1.0723094940185547,
+      "learning_rate": 7.269429591833027e-06,
+      "loss": 0.0799,
+      "step": 555
+    },
+    {
+      "epoch": 2.0897196261682245,
+      "grad_norm": 0.9724428057670593,
+      "learning_rate": 7.00644938073904e-06,
+      "loss": 0.0926,
+      "step": 560
+    },
+    {
+      "epoch": 2.108411214953271,
+      "grad_norm": 1.178525447845459,
+      "learning_rate": 6.746856948622318e-06,
+      "loss": 0.0852,
+      "step": 565
+    },
+    {
+      "epoch": 2.127102803738318,
+      "grad_norm": 1.4153668880462646,
+      "learning_rate": 6.490762314400115e-06,
+      "loss": 0.1139,
+      "step": 570
+    },
+    {
+      "epoch": 2.1457943925233645,
+      "grad_norm": 1.140940546989441,
+      "learning_rate": 6.238274014573849e-06,
+      "loss": 0.1111,
+      "step": 575
+    },
+    {
+      "epoch": 2.1644859813084114,
+      "grad_norm": 1.0335090160369873,
+      "learning_rate": 5.989499057229761e-06,
+      "loss": 0.0954,
+      "step": 580
+    },
+    {
+      "epoch": 2.183177570093458,
+      "grad_norm": 1.104285478591919,
+      "learning_rate": 5.744542876687429e-06,
+      "loss": 0.077,
+      "step": 585
+    },
+    {
+      "epoch": 2.201869158878505,
+      "grad_norm": 0.7969491481781006,
+      "learning_rate": 5.503509288815222e-06,
+      "loss": 0.0764,
+      "step": 590
+    },
+    {
+      "epoch": 2.2205607476635514,
+      "grad_norm": 1.1409050226211548,
+      "learning_rate": 5.266500447031647e-06,
+      "loss": 0.0878,
+      "step": 595
+    },
+    {
+      "epoch": 2.2392523364485983,
+      "grad_norm": 0.9332407712936401,
+      "learning_rate": 5.033616799011316e-06,
+      "loss": 0.0775,
+      "step": 600
+    },
+    {
+      "epoch": 2.257943925233645,
+      "grad_norm": 1.0020838975906372,
+      "learning_rate": 4.804957044113814e-06,
+      "loss": 0.0844,
+      "step": 605
+    },
+    {
+      "epoch": 2.2766355140186914,
+      "grad_norm": 0.8991174101829529,
+      "learning_rate": 4.580618091553511e-06,
+      "loss": 0.0919,
+      "step": 610
+    },
+    {
+      "epoch": 2.2953271028037383,
+      "grad_norm": 0.9004872441291809,
+      "learning_rate": 4.360695019328101e-06,
+      "loss": 0.0824,
+      "step": 615
+    },
+    {
+      "epoch": 2.3140186915887853,
+      "grad_norm": 1.0106133222579956,
+      "learning_rate": 4.145281033923183e-06,
+      "loss": 0.0904,
+      "step": 620
+    },
+    {
+      "epoch": 2.3327102803738318,
+      "grad_norm": 1.2012033462524414,
+      "learning_rate": 3.9344674308100566e-06,
+      "loss": 0.0791,
+      "step": 625
+    },
+    {
+      "epoch": 2.3514018691588783,
+      "grad_norm": 1.0751651525497437,
+      "learning_rate": 3.7283435557534185e-06,
+      "loss": 0.0883,
+      "step": 630
+    },
+    {
+      "epoch": 2.3700934579439252,
+      "grad_norm": 0.8737074136734009,
+      "learning_rate": 3.5269967669453855e-06,
+      "loss": 0.0854,
+      "step": 635
+    },
+    {
+      "epoch": 2.388785046728972,
+      "grad_norm": 1.0530178546905518,
+      "learning_rate": 3.3305123979818477e-06,
+      "loss": 0.0765,
+      "step": 640
+    },
+    {
+      "epoch": 2.4074766355140187,
+      "grad_norm": 0.7321348190307617,
+      "learning_rate": 3.1389737216968906e-06,
+      "loss": 0.0819,
+      "step": 645
+    },
+    {
+      "epoch": 2.426168224299065,
+      "grad_norm": 0.8886009454727173,
+      "learning_rate": 2.9524619148706163e-06,
+      "loss": 0.0733,
+      "step": 650
+    },
+    {
+      "epoch": 2.444859813084112,
+      "grad_norm": 0.8825504183769226,
+      "learning_rate": 2.77105602382528e-06,
+      "loss": 0.0689,
+      "step": 655
+    },
+    {
+      "epoch": 2.463551401869159,
+      "grad_norm": 0.996822714805603,
+      "learning_rate": 2.594832930924342e-06,
+      "loss": 0.0722,
+      "step": 660
+    },
+    {
+      "epoch": 2.4822429906542056,
+      "grad_norm": 0.8003243207931519,
+      "learning_rate": 2.4238673219886388e-06,
+      "loss": 0.0731,
+      "step": 665
+    },
+    {
+      "epoch": 2.500934579439252,
+      "grad_norm": 0.6386939883232117,
+      "learning_rate": 2.2582316546434957e-06,
+      "loss": 0.0814,
+      "step": 670
+    },
+    {
+      "epoch": 2.519626168224299,
+      "grad_norm": 0.6979541778564453,
+      "learning_rate": 2.0979961276101666e-06,
+      "loss": 0.0819,
+      "step": 675
+    },
+    {
+      "epoch": 2.538317757009346,
+      "grad_norm": 0.958196759223938,
+      "learning_rate": 1.9432286509546208e-06,
+      "loss": 0.0825,
+      "step": 680
+    },
+    {
+      "epoch": 2.5570093457943925,
+      "grad_norm": 0.7895479202270508,
+      "learning_rate": 1.7939948173063098e-06,
+      "loss": 0.071,
+      "step": 685
+    },
+    {
+      "epoch": 2.575700934579439,
+      "grad_norm": 0.8437227606773376,
+      "learning_rate": 1.6503578740590903e-06,
+      "loss": 0.0943,
+      "step": 690
+    },
+    {
+      "epoch": 2.594392523364486,
+      "grad_norm": 0.7973619103431702,
+      "learning_rate": 1.5123786965660875e-06,
+      "loss": 0.0679,
+      "step": 695
+    },
+    {
+      "epoch": 2.613084112149533,
+      "grad_norm": 0.7281593680381775,
+      "learning_rate": 1.380115762339877e-06,
+      "loss": 0.0786,
+      "step": 700
+    },
+    {
+      "epoch": 2.6317757009345795,
+      "grad_norm": 0.6109554767608643,
+      "learning_rate": 1.2536251262688765e-06,
+      "loss": 0.0642,
+      "step": 705
+    },
+    {
+      "epoch": 2.650467289719626,
+      "grad_norm": 0.7290362119674683,
+      "learning_rate": 1.1329603968605024e-06,
+      "loss": 0.0688,
+      "step": 710
+    },
+    {
+      "epoch": 2.669158878504673,
+      "grad_norm": 0.9455207586288452,
+      "learning_rate": 1.0181727135211477e-06,
+      "loss": 0.0824,
+      "step": 715
+    },
+    {
+      "epoch": 2.68785046728972,
+      "grad_norm": 0.718041181564331,
+      "learning_rate": 9.093107248825621e-07,
+      "loss": 0.0735,
+      "step": 720
+    },
+    {
+      "epoch": 2.7065420560747664,
+      "grad_norm": 0.8444967269897461,
+      "learning_rate": 8.064205681838899e-07,
+      "loss": 0.079,
+      "step": 725
+    },
+    {
+      "epoch": 2.725233644859813,
+      "grad_norm": 0.6965999603271484,
+      "learning_rate": 7.09545849718059e-07,
+      "loss": 0.0725,
+      "step": 730
+    },
+    {
+      "epoch": 2.74392523364486,
+      "grad_norm": 0.8532543778419495,
+      "learning_rate": 6.187276263508169e-07,
+      "loss": 0.0765,
+      "step": 735
+    },
+    {
+      "epoch": 2.762616822429907,
+      "grad_norm": 0.7777957320213318,
+      "learning_rate": 5.340043881202649e-07,
+      "loss": 0.0728,
+      "step": 740
+    },
+    {
+      "epoch": 2.7813084112149533,
+      "grad_norm": 0.853276789188385,
+      "learning_rate": 4.554120419242302e-07,
+      "loss": 0.081,
+      "step": 745
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.9259778261184692,
+      "learning_rate": 3.8298389630242237e-07,
+      "loss": 0.0705,
+      "step": 750
+    },
+    {
+      "epoch": 2.8186915887850468,
+      "grad_norm": 0.8668875098228455,
+      "learning_rate": 3.167506473198012e-07,
+      "loss": 0.0724,
+      "step": 755
+    },
+    {
+      "epoch": 2.8373831775700937,
+      "grad_norm": 0.6263694167137146,
+      "learning_rate": 2.5674036555716817e-07,
+      "loss": 0.0671,
+      "step": 760
+    },
+    {
+      "epoch": 2.8560747663551402,
+      "grad_norm": 0.6742134690284729,
+      "learning_rate": 2.0297848421443943e-07,
+      "loss": 0.0612,
+      "step": 765
+    },
+    {
+      "epoch": 2.8747663551401867,
+      "grad_norm": 0.8330395221710205,
+      "learning_rate": 1.5548778833171461e-07,
+      "loss": 0.0694,
+      "step": 770
+    },
+    {
+      "epoch": 2.8934579439252337,
+      "grad_norm": 0.720735490322113,
+      "learning_rate": 1.1428840513264982e-07,
+      "loss": 0.0753,
+      "step": 775
+    },
+    {
+      "epoch": 2.91214953271028,
+      "grad_norm": 0.8907384276390076,
+      "learning_rate": 7.939779549426263e-08,
+      "loss": 0.0665,
+      "step": 780
+    },
+    {
+      "epoch": 2.930841121495327,
+      "grad_norm": 0.7753194570541382,
+      "learning_rate": 5.083074654676001e-08,
+      "loss": 0.0718,
+      "step": 785
+    },
+    {
+      "epoch": 2.9495327102803737,
+      "grad_norm": 0.7027153372764587,
+      "learning_rate": 2.8599365406551394e-08,
+      "loss": 0.0662,
+      "step": 790
+    },
+    {
+      "epoch": 2.9682242990654206,
+      "grad_norm": 0.7937668561935425,
+      "learning_rate": 1.271307404507982e-08,
+      "loss": 0.0648,
+      "step": 795
+    },
+    {
+      "epoch": 2.986915887850467,
+      "grad_norm": 0.8247569799423218,
+      "learning_rate": 3.1786052956578326e-09,
+      "loss": 0.0657,
+      "step": 800
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 804,
+  "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": 1.0372137333418885e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

119_128_e3_3e-5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e246f1c0fbb7c146f539dbda8869570dc93518ab81b9e049175f45fb16266be3
+size 8209

119_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)