Upload folder using huggingface_hub

36/README.md

@@ -0,0 +1,202 @@
+---
+base_model: ibm-granite/granite-3.3-8b-base
+library_name: peft
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.15.2

36/adapter_config.json

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

36/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62f3b9c8b7e0d1d70a352ffaa3e79a7ab33127559b366c24b6469c371642bb94
+size 791751704

36/latest

@@ -0,0 +1 @@
+global_step432

36/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39ecf7ddaeb26b9688a5a4050d35d5c4de5959d654a11a87c8375d4c81b76f77
+size 15920

36/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae80fce53798dcfe2fedf8b01f67dabfe21a644da454def949273a158f0e467a
+size 15920

36/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4cfbd75008afc5f9a3d4529df305c67b6008479048374cc65f46b09fe758177
+size 15920

36/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c9a591057653f5080b4ef2b0e1a3d77dcc85f4d5e57474a3494b9edd673dab6
+size 15920

36/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4d2e83286959764c54b104e377387ff7675c238eb7146d9da70188d4c3cfb5f
+size 15920

36/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f52ac2ac9caeb90657b3a2ba395ea613ea82487e276022468078dcfaf445883e
+size 15920

36/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0ed36e325b73c5e6cbfdd263ba152ad5518acdd4c771256a42f3eb3cad7e156
+size 15920

36/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a4846a856c91a8166610193e525207a1c9270a126317fca155caaba8c81f58d3
+size 15920

36/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0003d2fabd1dd4e6c5547cbfee3f8e14c7347988ce2e132457bf7c992e3d72b5
+size 1064

36/special_tokens_map.json

@@ -0,0 +1,51 @@
+{
+  "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": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

36/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": "<|endoftext|>",
+  "padding_side": "left",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}

36/trainer_state.json

@@ -0,0 +1,636 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 432,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02320185614849188,
+      "grad_norm": 1.2721507549285889,
+      "learning_rate": 2.222222222222222e-06,
+      "loss": 1.3929,
+      "step": 5
+    },
+    {
+      "epoch": 0.04640371229698376,
+      "grad_norm": 0.835289478302002,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 1.3396,
+      "step": 10
+    },
+    {
+      "epoch": 0.06960556844547564,
+      "grad_norm": 0.6244049668312073,
+      "learning_rate": 7.777777777777777e-06,
+      "loss": 1.3651,
+      "step": 15
+    },
+    {
+      "epoch": 0.09280742459396751,
+      "grad_norm": 0.5461729764938354,
+      "learning_rate": 1.0555555555555555e-05,
+      "loss": 1.3004,
+      "step": 20
+    },
+    {
+      "epoch": 0.11600928074245939,
+      "grad_norm": 0.6112794280052185,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 1.3393,
+      "step": 25
+    },
+    {
+      "epoch": 0.13921113689095127,
+      "grad_norm": 0.5857144594192505,
+      "learning_rate": 1.6111111111111115e-05,
+      "loss": 1.2507,
+      "step": 30
+    },
+    {
+      "epoch": 0.16241299303944315,
+      "grad_norm": 0.7531876564025879,
+      "learning_rate": 1.888888888888889e-05,
+      "loss": 1.1736,
+      "step": 35
+    },
+    {
+      "epoch": 0.18561484918793503,
+      "grad_norm": 0.520007848739624,
+      "learning_rate": 2.1666666666666667e-05,
+      "loss": 1.2801,
+      "step": 40
+    },
+    {
+      "epoch": 0.2088167053364269,
+      "grad_norm": 0.49689608812332153,
+      "learning_rate": 2.4444444444444445e-05,
+      "loss": 1.1952,
+      "step": 45
+    },
+    {
+      "epoch": 0.23201856148491878,
+      "grad_norm": 0.48155874013900757,
+      "learning_rate": 2.7222222222222223e-05,
+      "loss": 1.1944,
+      "step": 50
+    },
+    {
+      "epoch": 0.2552204176334107,
+      "grad_norm": 0.5951055288314819,
+      "learning_rate": 3e-05,
+      "loss": 1.1798,
+      "step": 55
+    },
+    {
+      "epoch": 0.27842227378190254,
+      "grad_norm": 0.5722975134849548,
+      "learning_rate": 2.999824208523885e-05,
+      "loss": 1.1342,
+      "step": 60
+    },
+    {
+      "epoch": 0.30162412993039445,
+      "grad_norm": 0.5526897311210632,
+      "learning_rate": 2.9992968752990648e-05,
+      "loss": 1.0696,
+      "step": 65
+    },
+    {
+      "epoch": 0.3248259860788863,
+      "grad_norm": 0.5819699168205261,
+      "learning_rate": 2.9984181239264532e-05,
+      "loss": 1.0669,
+      "step": 70
+    },
+    {
+      "epoch": 0.3480278422273782,
+      "grad_norm": 0.5398482084274292,
+      "learning_rate": 2.997188160375385e-05,
+      "loss": 1.0719,
+      "step": 75
+    },
+    {
+      "epoch": 0.37122969837587005,
+      "grad_norm": 0.5633794069290161,
+      "learning_rate": 2.9956072729353382e-05,
+      "loss": 1.0754,
+      "step": 80
+    },
+    {
+      "epoch": 0.39443155452436196,
+      "grad_norm": 0.5473197102546692,
+      "learning_rate": 2.9936758321483616e-05,
+      "loss": 1.0495,
+      "step": 85
+    },
+    {
+      "epoch": 0.4176334106728538,
+      "grad_norm": 0.6184537410736084,
+      "learning_rate": 2.9913942907222238e-05,
+      "loss": 1.0552,
+      "step": 90
+    },
+    {
+      "epoch": 0.4408352668213457,
+      "grad_norm": 0.6393222212791443,
+      "learning_rate": 2.9887631834243058e-05,
+      "loss": 0.9652,
+      "step": 95
+    },
+    {
+      "epoch": 0.46403712296983757,
+      "grad_norm": 0.7783088684082031,
+      "learning_rate": 2.985783126956255e-05,
+      "loss": 0.9692,
+      "step": 100
+    },
+    {
+      "epoch": 0.4872389791183295,
+      "grad_norm": 0.6314108371734619,
+      "learning_rate": 2.9824548198094386e-05,
+      "loss": 0.9768,
+      "step": 105
+    },
+    {
+      "epoch": 0.5104408352668214,
+      "grad_norm": 0.7619579434394836,
+      "learning_rate": 2.9787790421012247e-05,
+      "loss": 0.9689,
+      "step": 110
+    },
+    {
+      "epoch": 0.5336426914153132,
+      "grad_norm": 0.5916560888290405,
+      "learning_rate": 2.9747566553921326e-05,
+      "loss": 0.9671,
+      "step": 115
+    },
+    {
+      "epoch": 0.5568445475638051,
+      "grad_norm": 0.7228941321372986,
+      "learning_rate": 2.9703886024838916e-05,
+      "loss": 0.9565,
+      "step": 120
+    },
+    {
+      "epoch": 0.580046403712297,
+      "grad_norm": 0.732020378112793,
+      "learning_rate": 2.9656759071984602e-05,
+      "loss": 0.9067,
+      "step": 125
+    },
+    {
+      "epoch": 0.6032482598607889,
+      "grad_norm": 0.7976245880126953,
+      "learning_rate": 2.960619674138052e-05,
+      "loss": 0.9196,
+      "step": 130
+    },
+    {
+      "epoch": 0.6264501160092807,
+      "grad_norm": 0.720694899559021,
+      "learning_rate": 2.955221088426231e-05,
+      "loss": 0.8133,
+      "step": 135
+    },
+    {
+      "epoch": 0.6496519721577726,
+      "grad_norm": 1.1036807298660278,
+      "learning_rate": 2.9494814154301328e-05,
+      "loss": 0.8346,
+      "step": 140
+    },
+    {
+      "epoch": 0.6728538283062645,
+      "grad_norm": 0.8336039781570435,
+      "learning_rate": 2.9434020004638757e-05,
+      "loss": 0.8046,
+      "step": 145
+    },
+    {
+      "epoch": 0.6960556844547564,
+      "grad_norm": 0.7478963732719421,
+      "learning_rate": 2.9369842684732334e-05,
+      "loss": 0.8392,
+      "step": 150
+    },
+    {
+      "epoch": 0.7192575406032483,
+      "grad_norm": 0.9218920469284058,
+      "learning_rate": 2.9302297237016462e-05,
+      "loss": 0.8374,
+      "step": 155
+    },
+    {
+      "epoch": 0.7424593967517401,
+      "grad_norm": 0.8461490869522095,
+      "learning_rate": 2.9231399493376414e-05,
+      "loss": 0.8535,
+      "step": 160
+    },
+    {
+      "epoch": 0.765661252900232,
+      "grad_norm": 0.9020415544509888,
+      "learning_rate": 2.9157166071437542e-05,
+      "loss": 0.7843,
+      "step": 165
+    },
+    {
+      "epoch": 0.7888631090487239,
+      "grad_norm": 0.9536502361297607,
+      "learning_rate": 2.9079614370670268e-05,
+      "loss": 0.8528,
+      "step": 170
+    },
+    {
+      "epoch": 0.8120649651972158,
+      "grad_norm": 1.046362280845642,
+      "learning_rate": 2.8998762568311857e-05,
+      "loss": 0.7654,
+      "step": 175
+    },
+    {
+      "epoch": 0.8352668213457076,
+      "grad_norm": 1.2142820358276367,
+      "learning_rate": 2.89146296151059e-05,
+      "loss": 0.7513,
+      "step": 180
+    },
+    {
+      "epoch": 0.8584686774941995,
+      "grad_norm": 1.0387227535247803,
+      "learning_rate": 2.8827235230860425e-05,
+      "loss": 0.7403,
+      "step": 185
+    },
+    {
+      "epoch": 0.8816705336426914,
+      "grad_norm": 0.9896584153175354,
+      "learning_rate": 2.873659989982586e-05,
+      "loss": 0.6818,
+      "step": 190
+    },
+    {
+      "epoch": 0.9048723897911833,
+      "grad_norm": 0.9425647854804993,
+      "learning_rate": 2.864274486589371e-05,
+      "loss": 0.7047,
+      "step": 195
+    },
+    {
+      "epoch": 0.9280742459396751,
+      "grad_norm": 1.160969853401184,
+      "learning_rate": 2.8545692127617248e-05,
+      "loss": 0.754,
+      "step": 200
+    },
+    {
+      "epoch": 0.951276102088167,
+      "grad_norm": 0.9628556370735168,
+      "learning_rate": 2.8445464433055303e-05,
+      "loss": 0.7192,
+      "step": 205
+    },
+    {
+      "epoch": 0.974477958236659,
+      "grad_norm": 1.0352585315704346,
+      "learning_rate": 2.8342085274440372e-05,
+      "loss": 0.7091,
+      "step": 210
+    },
+    {
+      "epoch": 0.9976798143851509,
+      "grad_norm": 1.056439995765686,
+      "learning_rate": 2.823557888267232e-05,
+      "loss": 0.6401,
+      "step": 215
+    },
+    {
+      "epoch": 1.0185614849187936,
+      "grad_norm": 1.0733691453933716,
+      "learning_rate": 2.8125970221638906e-05,
+      "loss": 0.62,
+      "step": 220
+    },
+    {
+      "epoch": 1.0417633410672853,
+      "grad_norm": 1.0413997173309326,
+      "learning_rate": 2.8013284982364558e-05,
+      "loss": 0.545,
+      "step": 225
+    },
+    {
+      "epoch": 1.0649651972157772,
+      "grad_norm": 1.149565577507019,
+      "learning_rate": 2.7897549576988668e-05,
+      "loss": 0.544,
+      "step": 230
+    },
+    {
+      "epoch": 1.088167053364269,
+      "grad_norm": 1.1546003818511963,
+      "learning_rate": 2.7778791132574908e-05,
+      "loss": 0.5455,
+      "step": 235
+    },
+    {
+      "epoch": 1.111368909512761,
+      "grad_norm": 1.1283059120178223,
+      "learning_rate": 2.7657037484752935e-05,
+      "loss": 0.5429,
+      "step": 240
+    },
+    {
+      "epoch": 1.134570765661253,
+      "grad_norm": 1.011425256729126,
+      "learning_rate": 2.753231717119405e-05,
+      "loss": 0.5551,
+      "step": 245
+    },
+    {
+      "epoch": 1.1577726218097448,
+      "grad_norm": 1.31667160987854,
+      "learning_rate": 2.7404659424922274e-05,
+      "loss": 0.5576,
+      "step": 250
+    },
+    {
+      "epoch": 1.1809744779582367,
+      "grad_norm": 1.1013916730880737,
+      "learning_rate": 2.727409416746249e-05,
+      "loss": 0.5663,
+      "step": 255
+    },
+    {
+      "epoch": 1.2041763341067284,
+      "grad_norm": 1.1836365461349487,
+      "learning_rate": 2.714065200182714e-05,
+      "loss": 0.545,
+      "step": 260
+    },
+    {
+      "epoch": 1.2273781902552203,
+      "grad_norm": 1.192248821258545,
+      "learning_rate": 2.700436420534326e-05,
+      "loss": 0.5675,
+      "step": 265
+    },
+    {
+      "epoch": 1.2505800464037122,
+      "grad_norm": 1.334416151046753,
+      "learning_rate": 2.686526272232141e-05,
+      "loss": 0.5414,
+      "step": 270
+    },
+    {
+      "epoch": 1.2737819025522041,
+      "grad_norm": 1.1702522039413452,
+      "learning_rate": 2.67233801565683e-05,
+      "loss": 0.5769,
+      "step": 275
+    },
+    {
+      "epoch": 1.296983758700696,
+      "grad_norm": 1.1251720190048218,
+      "learning_rate": 2.6578749763744815e-05,
+      "loss": 0.4754,
+      "step": 280
+    },
+    {
+      "epoch": 1.320185614849188,
+      "grad_norm": 1.154903769493103,
+      "learning_rate": 2.6431405443571282e-05,
+      "loss": 0.4881,
+      "step": 285
+    },
+    {
+      "epoch": 1.3433874709976799,
+      "grad_norm": 1.1724737882614136,
+      "learning_rate": 2.6281381731881763e-05,
+      "loss": 0.5294,
+      "step": 290
+    },
+    {
+      "epoch": 1.3665893271461718,
+      "grad_norm": 1.5480215549468994,
+      "learning_rate": 2.6128713792529225e-05,
+      "loss": 0.4428,
+      "step": 295
+    },
+    {
+      "epoch": 1.3897911832946637,
+      "grad_norm": 1.135866641998291,
+      "learning_rate": 2.5973437409143557e-05,
+      "loss": 0.4511,
+      "step": 300
+    },
+    {
+      "epoch": 1.4129930394431556,
+      "grad_norm": 1.2271608114242554,
+      "learning_rate": 2.5815588976744274e-05,
+      "loss": 0.4658,
+      "step": 305
+    },
+    {
+      "epoch": 1.4361948955916473,
+      "grad_norm": 1.1785272359848022,
+      "learning_rate": 2.565520549320996e-05,
+      "loss": 0.4432,
+      "step": 310
+    },
+    {
+      "epoch": 1.4593967517401392,
+      "grad_norm": 1.1631721258163452,
+      "learning_rate": 2.549232455060637e-05,
+      "loss": 0.458,
+      "step": 315
+    },
+    {
+      "epoch": 1.482598607888631,
+      "grad_norm": 1.085505723953247,
+      "learning_rate": 2.532698432637528e-05,
+      "loss": 0.4551,
+      "step": 320
+    },
+    {
+      "epoch": 1.505800464037123,
+      "grad_norm": 1.1535762548446655,
+      "learning_rate": 2.5159223574386117e-05,
+      "loss": 0.4009,
+      "step": 325
+    },
+    {
+      "epoch": 1.5290023201856149,
+      "grad_norm": 1.1571379899978638,
+      "learning_rate": 2.498908161585253e-05,
+      "loss": 0.4474,
+      "step": 330
+    },
+    {
+      "epoch": 1.5522041763341066,
+      "grad_norm": 1.3281868696212769,
+      "learning_rate": 2.48165983301159e-05,
+      "loss": 0.4294,
+      "step": 335
+    },
+    {
+      "epoch": 1.5754060324825985,
+      "grad_norm": 1.22118079662323,
+      "learning_rate": 2.464181414529809e-05,
+      "loss": 0.4245,
+      "step": 340
+    },
+    {
+      "epoch": 1.5986078886310904,
+      "grad_norm": 1.2129508256912231,
+      "learning_rate": 2.4464770028825586e-05,
+      "loss": 0.4219,
+      "step": 345
+    },
+    {
+      "epoch": 1.6218097447795823,
+      "grad_norm": 1.099057912826538,
+      "learning_rate": 2.4285507477827138e-05,
+      "loss": 0.4172,
+      "step": 350
+    },
+    {
+      "epoch": 1.6450116009280742,
+      "grad_norm": 1.1964019536972046,
+      "learning_rate": 2.410406850940735e-05,
+      "loss": 0.456,
+      "step": 355
+    },
+    {
+      "epoch": 1.668213457076566,
+      "grad_norm": 1.3184633255004883,
+      "learning_rate": 2.3920495650798343e-05,
+      "loss": 0.4586,
+      "step": 360
+    },
+    {
+      "epoch": 1.691415313225058,
+      "grad_norm": 1.2149478197097778,
+      "learning_rate": 2.3734831929391824e-05,
+      "loss": 0.4056,
+      "step": 365
+    },
+    {
+      "epoch": 1.71461716937355,
+      "grad_norm": 1.3322263956069946,
+      "learning_rate": 2.3547120862653995e-05,
+      "loss": 0.3892,
+      "step": 370
+    },
+    {
+      "epoch": 1.7378190255220418,
+      "grad_norm": 1.0912957191467285,
+      "learning_rate": 2.3357406447925528e-05,
+      "loss": 0.4038,
+      "step": 375
+    },
+    {
+      "epoch": 1.7610208816705337,
+      "grad_norm": 1.262276291847229,
+      "learning_rate": 2.3165733152109097e-05,
+      "loss": 0.3873,
+      "step": 380
+    },
+    {
+      "epoch": 1.7842227378190256,
+      "grad_norm": 1.278756022453308,
+      "learning_rate": 2.297214590124684e-05,
+      "loss": 0.4077,
+      "step": 385
+    },
+    {
+      "epoch": 1.8074245939675175,
+      "grad_norm": 1.5225976705551147,
+      "learning_rate": 2.2776690069990208e-05,
+      "loss": 0.3684,
+      "step": 390
+    },
+    {
+      "epoch": 1.8306264501160094,
+      "grad_norm": 1.2496668100357056,
+      "learning_rate": 2.257941147096465e-05,
+      "loss": 0.3693,
+      "step": 395
+    },
+    {
+      "epoch": 1.8538283062645011,
+      "grad_norm": 1.1463340520858765,
+      "learning_rate": 2.2380356344031677e-05,
+      "loss": 0.4008,
+      "step": 400
+    },
+    {
+      "epoch": 1.877030162412993,
+      "grad_norm": 1.3657644987106323,
+      "learning_rate": 2.217957134545074e-05,
+      "loss": 0.4095,
+      "step": 405
+    },
+    {
+      "epoch": 1.900232018561485,
+      "grad_norm": 1.323555588722229,
+      "learning_rate": 2.197710353694355e-05,
+      "loss": 0.4053,
+      "step": 410
+    },
+    {
+      "epoch": 1.9234338747099768,
+      "grad_norm": 1.3700615167617798,
+      "learning_rate": 2.177300037466334e-05,
+      "loss": 0.3756,
+      "step": 415
+    },
+    {
+      "epoch": 1.9466357308584685,
+      "grad_norm": 1.1665352582931519,
+      "learning_rate": 2.156730969807168e-05,
+      "loss": 0.3428,
+      "step": 420
+    },
+    {
+      "epoch": 1.9698375870069604,
+      "grad_norm": 1.6480296850204468,
+      "learning_rate": 2.1360079718725452e-05,
+      "loss": 0.3577,
+      "step": 425
+    },
+    {
+      "epoch": 1.9930394431554523,
+      "grad_norm": 1.230934977531433,
+      "learning_rate": 2.1151359008976604e-05,
+      "loss": 0.3337,
+      "step": 430
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1080,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.057540665170985e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

36/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76926f46f1fc7ddf335bce9a0af52a4f52ec0d78cd0b578b34ba102b64303108
+size 7736

36/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)