Upload folder using huggingface_hub

107_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": [
+    "o_proj",
+    "q_proj",
+    "v_proj",
+    "k_proj",
+    "down_proj",
+    "gate_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

107_128_e3_3e-5/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3cccca95c6f888ffd63c655e004d0d326523ff7b56584a505616f224c74e74c
+size 791751704

107_128_e3_3e-5/latest

@@ -0,0 +1 @@
+global_step453

107_128_e3_3e-5/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aebbed2bd1d103dfc5310248ac4e600853dc76c8dba8f4f2daac68b59fcc4e55
+size 16389

107_128_e3_3e-5/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ecaf90bbe7bec723fc8b69951e3988690d136c52403ec79996ca755dfe06633
+size 16389

107_128_e3_3e-5/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67ad43cb4adc78dcc05ff56a33591eddd701a3ef884ed4c64493d7374a7c08a2
+size 16389

107_128_e3_3e-5/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37363fdad85663aa5db3792d1e55e57fd87e8c394c47495d605d84e9b9b479c8
+size 16389

107_128_e3_3e-5/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37a5ef4abebbaa0808b6f3a4907dad31bff2d8c0872b69bba0ab3f9a8ed50d91
+size 16389

107_128_e3_3e-5/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:876ed929a570ae9abeedeb10925a3c7a6aa7a9c144ffad1b9541f3cf36eb4057
+size 16389

107_128_e3_3e-5/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddd3af0795accf5485628c5441c94dbfd7d6cc45ce0670665a0aae4985f606ba
+size 16389

107_128_e3_3e-5/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dadb3876e2bc644952d44273ccc697de4d664ae66879fe1d5ba6a20300244c2
+size 16389

107_128_e3_3e-5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e46c915f319cce2bcacc1ddb7d1debe78c750fc91a243181a09e5d56a6247f7
+size 1401

107_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
+  }
+}

107_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
+}

107_128_e3_3e-5/trainer_state.json

@@ -0,0 +1,664 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 453,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.033112582781456956,
+      "grad_norm": 0.879943311214447,
+      "learning_rate": 5.2173913043478265e-06,
+      "loss": 1.245,
+      "step": 5
+    },
+    {
+      "epoch": 0.06622516556291391,
+      "grad_norm": 0.7352291345596313,
+      "learning_rate": 1.173913043478261e-05,
+      "loss": 1.2722,
+      "step": 10
+    },
+    {
+      "epoch": 0.09933774834437085,
+      "grad_norm": 0.5713614821434021,
+      "learning_rate": 1.8260869565217393e-05,
+      "loss": 1.1823,
+      "step": 15
+    },
+    {
+      "epoch": 0.13245033112582782,
+      "grad_norm": 0.6902788877487183,
+      "learning_rate": 2.4782608695652175e-05,
+      "loss": 1.1925,
+      "step": 20
+    },
+    {
+      "epoch": 0.16556291390728478,
+      "grad_norm": 0.6207479238510132,
+      "learning_rate": 2.999959966628585e-05,
+      "loss": 1.1543,
+      "step": 25
+    },
+    {
+      "epoch": 0.1986754966887417,
+      "grad_norm": 0.49078088998794556,
+      "learning_rate": 2.9985590229902075e-05,
+      "loss": 1.0957,
+      "step": 30
+    },
+    {
+      "epoch": 0.23178807947019867,
+      "grad_norm": 0.611196756362915,
+      "learning_rate": 2.9951585471628692e-05,
+      "loss": 1.0346,
+      "step": 35
+    },
+    {
+      "epoch": 0.26490066225165565,
+      "grad_norm": 0.5425114035606384,
+      "learning_rate": 2.9897630764125496e-05,
+      "loss": 1.0243,
+      "step": 40
+    },
+    {
+      "epoch": 0.2980132450331126,
+      "grad_norm": 0.5257452726364136,
+      "learning_rate": 2.982379809933459e-05,
+      "loss": 0.9972,
+      "step": 45
+    },
+    {
+      "epoch": 0.33112582781456956,
+      "grad_norm": 0.6245226860046387,
+      "learning_rate": 2.9730185992421254e-05,
+      "loss": 0.9768,
+      "step": 50
+    },
+    {
+      "epoch": 0.36423841059602646,
+      "grad_norm": 0.5445970296859741,
+      "learning_rate": 2.9616919350324878e-05,
+      "loss": 0.9047,
+      "step": 55
+    },
+    {
+      "epoch": 0.3973509933774834,
+      "grad_norm": 0.6791722178459167,
+      "learning_rate": 2.9484149305095205e-05,
+      "loss": 0.8798,
+      "step": 60
+    },
+    {
+      "epoch": 0.4304635761589404,
+      "grad_norm": 0.6363832950592041,
+      "learning_rate": 2.9332053012236348e-05,
+      "loss": 0.8165,
+      "step": 65
+    },
+    {
+      "epoch": 0.46357615894039733,
+      "grad_norm": 0.6689126491546631,
+      "learning_rate": 2.9160833414327637e-05,
+      "loss": 0.7956,
+      "step": 70
+    },
+    {
+      "epoch": 0.4966887417218543,
+      "grad_norm": 0.7589102387428284,
+      "learning_rate": 2.8970718970236647e-05,
+      "loss": 0.7531,
+      "step": 75
+    },
+    {
+      "epoch": 0.5298013245033113,
+      "grad_norm": 0.8603280186653137,
+      "learning_rate": 2.8761963350285812e-05,
+      "loss": 0.7106,
+      "step": 80
+    },
+    {
+      "epoch": 0.5629139072847682,
+      "grad_norm": 0.8996940851211548,
+      "learning_rate": 2.8534845097779284e-05,
+      "loss": 0.7398,
+      "step": 85
+    },
+    {
+      "epoch": 0.5960264900662252,
+      "grad_norm": 1.104547381401062,
+      "learning_rate": 2.828966725734167e-05,
+      "loss": 0.6413,
+      "step": 90
+    },
+    {
+      "epoch": 0.6291390728476821,
+      "grad_norm": 0.9328771829605103,
+      "learning_rate": 2.8026756970564636e-05,
+      "loss": 0.6549,
+      "step": 95
+    },
+    {
+      "epoch": 0.6622516556291391,
+      "grad_norm": 0.9276598691940308,
+      "learning_rate": 2.7746465039500786e-05,
+      "loss": 0.586,
+      "step": 100
+    },
+    {
+      "epoch": 0.695364238410596,
+      "grad_norm": 1.0182416439056396,
+      "learning_rate": 2.7449165458587357e-05,
+      "loss": 0.5804,
+      "step": 105
+    },
+    {
+      "epoch": 0.7284768211920529,
+      "grad_norm": 1.2231544256210327,
+      "learning_rate": 2.7135254915624213e-05,
+      "loss": 0.5718,
+      "step": 110
+    },
+    {
+      "epoch": 0.7615894039735099,
+      "grad_norm": 1.004658579826355,
+      "learning_rate": 2.680515226247203e-05,
+      "loss": 0.5028,
+      "step": 115
+    },
+    {
+      "epoch": 0.7947019867549668,
+      "grad_norm": 1.023861050605774,
+      "learning_rate": 2.6459297956176887e-05,
+      "loss": 0.4672,
+      "step": 120
+    },
+    {
+      "epoch": 0.8278145695364238,
+      "grad_norm": 1.2049216032028198,
+      "learning_rate": 2.6098153471266992e-05,
+      "loss": 0.5199,
+      "step": 125
+    },
+    {
+      "epoch": 0.8609271523178808,
+      "grad_norm": 0.9748395085334778,
+      "learning_rate": 2.5722200684005718e-05,
+      "loss": 0.4724,
+      "step": 130
+    },
+    {
+      "epoch": 0.8940397350993378,
+      "grad_norm": 0.9443665146827698,
+      "learning_rate": 2.5331941229422532e-05,
+      "loss": 0.4733,
+      "step": 135
+    },
+    {
+      "epoch": 0.9271523178807947,
+      "grad_norm": 1.1110583543777466,
+      "learning_rate": 2.492789583197975e-05,
+      "loss": 0.3931,
+      "step": 140
+    },
+    {
+      "epoch": 0.9602649006622517,
+      "grad_norm": 1.1699163913726807,
+      "learning_rate": 2.4510603610768167e-05,
+      "loss": 0.4133,
+      "step": 145
+    },
+    {
+      "epoch": 0.9933774834437086,
+      "grad_norm": 1.117264986038208,
+      "learning_rate": 2.408062136015872e-05,
+      "loss": 0.3616,
+      "step": 150
+    },
+    {
+      "epoch": 1.0264900662251655,
+      "grad_norm": 1.0720939636230469,
+      "learning_rate": 2.363852280686994e-05,
+      "loss": 0.3222,
+      "step": 155
+    },
+    {
+      "epoch": 1.0596026490066226,
+      "grad_norm": 1.1632972955703735,
+      "learning_rate": 2.3184897844442498e-05,
+      "loss": 0.3306,
+      "step": 160
+    },
+    {
+      "epoch": 1.0927152317880795,
+      "grad_norm": 1.1449373960494995,
+      "learning_rate": 2.2720351746142368e-05,
+      "loss": 0.2811,
+      "step": 165
+    },
+    {
+      "epoch": 1.1258278145695364,
+      "grad_norm": 1.1481462717056274,
+      "learning_rate": 2.224550435734272e-05,
+      "loss": 0.2858,
+      "step": 170
+    },
+    {
+      "epoch": 1.1589403973509933,
+      "grad_norm": 1.1522884368896484,
+      "learning_rate": 2.1760989268462184e-05,
+      "loss": 0.3392,
+      "step": 175
+    },
+    {
+      "epoch": 1.1920529801324504,
+      "grad_norm": 1.1070661544799805,
+      "learning_rate": 2.1267452969563156e-05,
+      "loss": 0.2888,
+      "step": 180
+    },
+    {
+      "epoch": 1.2251655629139073,
+      "grad_norm": 1.217092514038086,
+      "learning_rate": 2.0765553987737942e-05,
+      "loss": 0.2751,
+      "step": 185
+    },
+    {
+      "epoch": 1.2582781456953642,
+      "grad_norm": 1.221022129058838,
+      "learning_rate": 2.025596200843394e-05,
+      "loss": 0.2688,
+      "step": 190
+    },
+    {
+      "epoch": 1.2913907284768211,
+      "grad_norm": 1.154313087463379,
+      "learning_rate": 1.973935698189017e-05,
+      "loss": 0.247,
+      "step": 195
+    },
+    {
+      "epoch": 1.3245033112582782,
+      "grad_norm": 1.1461505889892578,
+      "learning_rate": 1.9216428215877427e-05,
+      "loss": 0.2456,
+      "step": 200
+    },
+    {
+      "epoch": 1.3576158940397351,
+      "grad_norm": 1.1467838287353516,
+      "learning_rate": 1.86878734559527e-05,
+      "loss": 0.2513,
+      "step": 205
+    },
+    {
+      "epoch": 1.390728476821192,
+      "grad_norm": 1.1764097213745117,
+      "learning_rate": 1.8154397954454996e-05,
+      "loss": 0.2284,
+      "step": 210
+    },
+    {
+      "epoch": 1.423841059602649,
+      "grad_norm": 1.1026008129119873,
+      "learning_rate": 1.7616713529484835e-05,
+      "loss": 0.2102,
+      "step": 215
+    },
+    {
+      "epoch": 1.4569536423841059,
+      "grad_norm": 1.086315631866455,
+      "learning_rate": 1.7075537615123045e-05,
+      "loss": 0.2171,
+      "step": 220
+    },
+    {
+      "epoch": 1.490066225165563,
+      "grad_norm": 1.1874076128005981,
+      "learning_rate": 1.653159230415614e-05,
+      "loss": 0.1902,
+      "step": 225
+    },
+    {
+      "epoch": 1.5231788079470199,
+      "grad_norm": 1.1898187398910522,
+      "learning_rate": 1.5985603384585545e-05,
+      "loss": 0.1759,
+      "step": 230
+    },
+    {
+      "epoch": 1.5562913907284768,
+      "grad_norm": 1.0575146675109863,
+      "learning_rate": 1.54382993712063e-05,
+      "loss": 0.1951,
+      "step": 235
+    },
+    {
+      "epoch": 1.589403973509934,
+      "grad_norm": 1.2608692646026611,
+      "learning_rate": 1.4890410533547405e-05,
+      "loss": 0.1806,
+      "step": 240
+    },
+    {
+      "epoch": 1.6225165562913908,
+      "grad_norm": 1.1632177829742432,
+      "learning_rate": 1.4342667921470813e-05,
+      "loss": 0.188,
+      "step": 245
+    },
+    {
+      "epoch": 1.6556291390728477,
+      "grad_norm": 0.9755818843841553,
+      "learning_rate": 1.3795802389729185e-05,
+      "loss": 0.1786,
+      "step": 250
+    },
+    {
+      "epoch": 1.6887417218543046,
+      "grad_norm": 1.2531871795654297,
+      "learning_rate": 1.3250543622784051e-05,
+      "loss": 0.1528,
+      "step": 255
+    },
+    {
+      "epoch": 1.7218543046357615,
+      "grad_norm": 0.994903028011322,
+      "learning_rate": 1.2707619161185425e-05,
+      "loss": 0.1558,
+      "step": 260
+    },
+    {
+      "epoch": 1.7549668874172184,
+      "grad_norm": 1.3092632293701172,
+      "learning_rate": 1.2167753430812096e-05,
+      "loss": 0.1515,
+      "step": 265
+    },
+    {
+      "epoch": 1.7880794701986755,
+      "grad_norm": 1.3809027671813965,
+      "learning_rate": 1.1631666776267804e-05,
+      "loss": 0.1594,
+      "step": 270
+    },
+    {
+      "epoch": 1.8211920529801324,
+      "grad_norm": 1.1267093420028687,
+      "learning_rate": 1.110007449972309e-05,
+      "loss": 0.1658,
+      "step": 275
+    },
+    {
+      "epoch": 1.8543046357615895,
+      "grad_norm": 1.0481507778167725,
+      "learning_rate": 1.0573685906485283e-05,
+      "loss": 0.1427,
+      "step": 280
+    },
+    {
+      "epoch": 1.8874172185430464,
+      "grad_norm": 1.0553224086761475,
+      "learning_rate": 1.0053203358570116e-05,
+      "loss": 0.1255,
+      "step": 285
+    },
+    {
+      "epoch": 1.9205298013245033,
+      "grad_norm": 0.893339216709137,
+      "learning_rate": 9.5393213375378e-06,
+      "loss": 0.1267,
+      "step": 290
+    },
+    {
+      "epoch": 1.9536423841059603,
+      "grad_norm": 1.1446415185928345,
+      "learning_rate": 9.032725517844005e-06,
+      "loss": 0.1274,
+      "step": 295
+    },
+    {
+      "epoch": 1.9867549668874172,
+      "grad_norm": 1.0354291200637817,
+      "learning_rate": 8.534091851942215e-06,
+      "loss": 0.1151,
+      "step": 300
+    },
+    {
+      "epoch": 2.019867549668874,
+      "grad_norm": 0.8198172450065613,
+      "learning_rate": 8.04408566835814e-06,
+      "loss": 0.1147,
+      "step": 305
+    },
+    {
+      "epoch": 2.052980132450331,
+      "grad_norm": 0.8697104454040527,
+      "learning_rate": 7.563360783939724e-06,
+      "loss": 0.0933,
+      "step": 310
+    },
+    {
+      "epoch": 2.0860927152317883,
+      "grad_norm": 0.9183566570281982,
+      "learning_rate": 7.092558631467178e-06,
+      "loss": 0.1061,
+      "step": 315
+    },
+    {
+      "epoch": 2.119205298013245,
+      "grad_norm": 0.950411319732666,
+      "learning_rate": 6.632307403787139e-06,
+      "loss": 0.0941,
+      "step": 320
+    },
+    {
+      "epoch": 2.152317880794702,
+      "grad_norm": 0.7919628024101257,
+      "learning_rate": 6.1832212156129045e-06,
+      "loss": 0.1004,
+      "step": 325
+    },
+    {
+      "epoch": 2.185430463576159,
+      "grad_norm": 0.9648961424827576,
+      "learning_rate": 5.7458992841091544e-06,
+      "loss": 0.103,
+      "step": 330
+    },
+    {
+      "epoch": 2.218543046357616,
+      "grad_norm": 0.8730937242507935,
+      "learning_rate": 5.320925129354552e-06,
+      "loss": 0.0969,
+      "step": 335
+    },
+    {
+      "epoch": 2.251655629139073,
+      "grad_norm": 0.789699912071228,
+      "learning_rate": 4.908865795749e-06,
+      "loss": 0.0818,
+      "step": 340
+    },
+    {
+      "epoch": 2.2847682119205297,
+      "grad_norm": 0.8547818064689636,
+      "learning_rate": 4.5102710954044495e-06,
+      "loss": 0.0822,
+      "step": 345
+    },
+    {
+      "epoch": 2.3178807947019866,
+      "grad_norm": 1.0699576139450073,
+      "learning_rate": 4.125672874528797e-06,
+      "loss": 0.0791,
+      "step": 350
+    },
+    {
+      "epoch": 2.3509933774834435,
+      "grad_norm": 0.7335050702095032,
+      "learning_rate": 3.755584303781763e-06,
+      "loss": 0.0816,
+      "step": 355
+    },
+    {
+      "epoch": 2.384105960264901,
+      "grad_norm": 0.7833991050720215,
+      "learning_rate": 3.4004991935496005e-06,
+      "loss": 0.0768,
+      "step": 360
+    },
+    {
+      "epoch": 2.4172185430463577,
+      "grad_norm": 0.8509645462036133,
+      "learning_rate": 3.060891335052303e-06,
+      "loss": 0.0798,
+      "step": 365
+    },
+    {
+      "epoch": 2.4503311258278146,
+      "grad_norm": 0.7322204113006592,
+      "learning_rate": 2.7372138681624246e-06,
+      "loss": 0.0759,
+      "step": 370
+    },
+    {
+      "epoch": 2.4834437086092715,
+      "grad_norm": 0.7538129091262817,
+      "learning_rate": 2.429898676779072e-06,
+      "loss": 0.0876,
+      "step": 375
+    },
+    {
+      "epoch": 2.5165562913907285,
+      "grad_norm": 0.710992157459259,
+      "learning_rate": 2.1393558125638067e-06,
+      "loss": 0.0826,
+      "step": 380
+    },
+    {
+      "epoch": 2.5496688741721854,
+      "grad_norm": 0.7199347615242004,
+      "learning_rate": 1.8659729478073768e-06,
+      "loss": 0.0739,
+      "step": 385
+    },
+    {
+      "epoch": 2.5827814569536423,
+      "grad_norm": 0.7115665674209595,
+      "learning_rate": 1.6101148581573276e-06,
+      "loss": 0.072,
+      "step": 390
+    },
+    {
+      "epoch": 2.6158940397350996,
+      "grad_norm": 0.6447179913520813,
+      "learning_rate": 1.3721229358966547e-06,
+      "loss": 0.0709,
+      "step": 395
+    },
+    {
+      "epoch": 2.6490066225165565,
+      "grad_norm": 0.7293813228607178,
+      "learning_rate": 1.1523147344229717e-06,
+      "loss": 0.0672,
+      "step": 400
+    },
+    {
+      "epoch": 2.6821192052980134,
+      "grad_norm": 0.7068458199501038,
+      "learning_rate": 9.509835445359738e-07,
+      "loss": 0.0749,
+      "step": 405
+    },
+    {
+      "epoch": 2.7152317880794703,
+      "grad_norm": 0.7104700803756714,
+      "learning_rate": 7.683980030985655e-07,
+      "loss": 0.0734,
+      "step": 410
+    },
+    {
+      "epoch": 2.748344370860927,
+      "grad_norm": 0.6446296572685242,
+      "learning_rate": 6.048017345938339e-07,
+      "loss": 0.0712,
+      "step": 415
+    },
+    {
+      "epoch": 2.781456953642384,
+      "grad_norm": 0.5779967308044434,
+      "learning_rate": 4.604130260560874e-07,
+      "loss": 0.0729,
+      "step": 420
+    },
+    {
+      "epoch": 2.814569536423841,
+      "grad_norm": 0.5231285691261292,
+      "learning_rate": 3.3542453580976795e-07,
+      "loss": 0.0667,
+      "step": 425
+    },
+    {
+      "epoch": 2.847682119205298,
+      "grad_norm": 0.7247755527496338,
+      "learning_rate": 2.300030364048139e-07,
+      "loss": 0.0749,
+      "step": 430
+    },
+    {
+      "epoch": 2.880794701986755,
+      "grad_norm": 0.6383816003799438,
+      "learning_rate": 1.4428919209150294e-07,
+      "loss": 0.0665,
+      "step": 435
+    },
+    {
+      "epoch": 2.9139072847682117,
+      "grad_norm": 0.5755408406257629,
+      "learning_rate": 7.839737113168932e-08,
+      "loss": 0.0696,
+      "step": 440
+    },
+    {
+      "epoch": 2.9470198675496686,
+      "grad_norm": 0.6622664332389832,
+      "learning_rate": 3.241549319685622e-08,
+      "loss": 0.0715,
+      "step": 445
+    },
+    {
+      "epoch": 2.980132450331126,
+      "grad_norm": 0.6709860563278198,
+      "learning_rate": 6.404912056610091e-09,
+      "loss": 0.0604,
+      "step": 450
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 453,
+  "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": 6.581688593536778e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

107_128_e3_3e-5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40913fc512111ea5bf6fda576f93f5d3fb26f4d28227e216498a139196035339
+size 8209

107_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)