Upload folder using huggingface_hub

121/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

121/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": [
+    "gate_proj",
+    "up_proj",
+    "q_proj",
+    "v_proj",
+    "k_proj",
+    "down_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

121/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a7bec5ebed8bc079c308100cfc08cb5a401b9df74490488fbc5c456db6735f4
+size 791751704

121/latest

@@ -0,0 +1 @@
+global_step451

121/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e39e5b0752c5cad6571a4c83bb09902bc1109ded34fac64f9826f25d7c272a3f
+size 15920

121/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a0838a7b030bc5fdb45362ac7a5cb48453705031edfe3ce7921e52c400dc0b
+size 15920

121/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73a3320358f15a66c4f1b33ef8f5f75c13e49be7bd092d522ee56186fd0effb1
+size 15920

121/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2155ebc3d204b52188cd386d48e8a706afab403cb03596d4fea1d740d3f4e0d
+size 15920

121/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cdb6db42e613849b9a049ef6684cdb4b95d096c1d2a3e4e0c525ca2731960fb
+size 15920

121/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41e9625a0ca041f30c2026f837366730348b4c9d30730f3281462eb655235877
+size 15920

121/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abd2eb7daf7d269f76c47c249b66ff4d83acc69912233d8094d37b3a8f8e77d9
+size 15920

121/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3717a5778a10ea9a10f63fd4df3dd634749ab6eafa48d9188c4a1b1000f232ec
+size 15920

121/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:793778190bd3bb2aabcba1c77158f20934d7967f8bb3c3624708c7cc738a0411
+size 1064

121/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
+  }
+}

121/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
+}

121/trainer_state.json

@@ -0,0 +1,664 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 451,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.011098779134295227,
+      "grad_norm": 1.020947813987732,
+      "learning_rate": 1.0619469026548673e-06,
+      "loss": 1.2866,
+      "step": 5
+    },
+    {
+      "epoch": 0.022197558268590455,
+      "grad_norm": 1.0182945728302002,
+      "learning_rate": 2.389380530973451e-06,
+      "loss": 1.242,
+      "step": 10
+    },
+    {
+      "epoch": 0.033296337402885685,
+      "grad_norm": 0.6542662382125854,
+      "learning_rate": 3.7168141592920353e-06,
+      "loss": 1.2928,
+      "step": 15
+    },
+    {
+      "epoch": 0.04439511653718091,
+      "grad_norm": 0.588236391544342,
+      "learning_rate": 5.04424778761062e-06,
+      "loss": 1.242,
+      "step": 20
+    },
+    {
+      "epoch": 0.05549389567147614,
+      "grad_norm": 0.579189121723175,
+      "learning_rate": 6.371681415929204e-06,
+      "loss": 1.222,
+      "step": 25
+    },
+    {
+      "epoch": 0.06659267480577137,
+      "grad_norm": 0.516274094581604,
+      "learning_rate": 7.699115044247788e-06,
+      "loss": 1.1735,
+      "step": 30
+    },
+    {
+      "epoch": 0.07769145394006659,
+      "grad_norm": 0.5116409659385681,
+      "learning_rate": 9.026548672566373e-06,
+      "loss": 1.2008,
+      "step": 35
+    },
+    {
+      "epoch": 0.08879023307436182,
+      "grad_norm": 0.566781222820282,
+      "learning_rate": 1.0353982300884955e-05,
+      "loss": 1.206,
+      "step": 40
+    },
+    {
+      "epoch": 0.09988901220865705,
+      "grad_norm": 0.48452886939048767,
+      "learning_rate": 1.168141592920354e-05,
+      "loss": 1.177,
+      "step": 45
+    },
+    {
+      "epoch": 0.11098779134295228,
+      "grad_norm": 0.4358895719051361,
+      "learning_rate": 1.3008849557522125e-05,
+      "loss": 1.1648,
+      "step": 50
+    },
+    {
+      "epoch": 0.1220865704772475,
+      "grad_norm": 0.5665825605392456,
+      "learning_rate": 1.433628318584071e-05,
+      "loss": 1.1316,
+      "step": 55
+    },
+    {
+      "epoch": 0.13318534961154274,
+      "grad_norm": 0.481657475233078,
+      "learning_rate": 1.5663716814159295e-05,
+      "loss": 1.1461,
+      "step": 60
+    },
+    {
+      "epoch": 0.14428412874583796,
+      "grad_norm": 0.5076817274093628,
+      "learning_rate": 1.6991150442477876e-05,
+      "loss": 1.1245,
+      "step": 65
+    },
+    {
+      "epoch": 0.15538290788013318,
+      "grad_norm": 0.4932589530944824,
+      "learning_rate": 1.8318584070796458e-05,
+      "loss": 1.1608,
+      "step": 70
+    },
+    {
+      "epoch": 0.16648168701442842,
+      "grad_norm": 0.488295316696167,
+      "learning_rate": 1.9646017699115046e-05,
+      "loss": 1.1016,
+      "step": 75
+    },
+    {
+      "epoch": 0.17758046614872364,
+      "grad_norm": 0.48246464133262634,
+      "learning_rate": 2.0973451327433628e-05,
+      "loss": 1.1259,
+      "step": 80
+    },
+    {
+      "epoch": 0.18867924528301888,
+      "grad_norm": 0.48559442162513733,
+      "learning_rate": 2.2300884955752213e-05,
+      "loss": 1.0816,
+      "step": 85
+    },
+    {
+      "epoch": 0.1997780244173141,
+      "grad_norm": 0.4469965398311615,
+      "learning_rate": 2.3628318584070798e-05,
+      "loss": 1.1392,
+      "step": 90
+    },
+    {
+      "epoch": 0.21087680355160932,
+      "grad_norm": 0.5929376482963562,
+      "learning_rate": 2.495575221238938e-05,
+      "loss": 1.1227,
+      "step": 95
+    },
+    {
+      "epoch": 0.22197558268590456,
+      "grad_norm": 0.42551663517951965,
+      "learning_rate": 2.6283185840707968e-05,
+      "loss": 1.14,
+      "step": 100
+    },
+    {
+      "epoch": 0.23307436182019978,
+      "grad_norm": 0.5455872416496277,
+      "learning_rate": 2.761061946902655e-05,
+      "loss": 1.0756,
+      "step": 105
+    },
+    {
+      "epoch": 0.244173140954495,
+      "grad_norm": 0.5717443823814392,
+      "learning_rate": 2.893805309734513e-05,
+      "loss": 1.0195,
+      "step": 110
+    },
+    {
+      "epoch": 0.25527192008879024,
+      "grad_norm": 0.5069417953491211,
+      "learning_rate": 2.999998386674501e-05,
+      "loss": 1.0218,
+      "step": 115
+    },
+    {
+      "epoch": 0.2663706992230855,
+      "grad_norm": 0.6419563293457031,
+      "learning_rate": 2.999941920646437e-05,
+      "loss": 1.0718,
+      "step": 120
+    },
+    {
+      "epoch": 0.27746947835738067,
+      "grad_norm": 0.5803301930427551,
+      "learning_rate": 2.9998047918138238e-05,
+      "loss": 1.0622,
+      "step": 125
+    },
+    {
+      "epoch": 0.2885682574916759,
+      "grad_norm": 0.5580577850341797,
+      "learning_rate": 2.9995870075510795e-05,
+      "loss": 1.0556,
+      "step": 130
+    },
+    {
+      "epoch": 0.29966703662597116,
+      "grad_norm": 0.518118679523468,
+      "learning_rate": 2.9992885795700492e-05,
+      "loss": 1.0543,
+      "step": 135
+    },
+    {
+      "epoch": 0.31076581576026635,
+      "grad_norm": 0.7111470699310303,
+      "learning_rate": 2.9989095239193793e-05,
+      "loss": 1.0242,
+      "step": 140
+    },
+    {
+      "epoch": 0.3218645948945616,
+      "grad_norm": 0.5507873296737671,
+      "learning_rate": 2.998449860983655e-05,
+      "loss": 0.9969,
+      "step": 145
+    },
+    {
+      "epoch": 0.33296337402885684,
+      "grad_norm": 0.5861643552780151,
+      "learning_rate": 2.9979096154823008e-05,
+      "loss": 0.9685,
+      "step": 150
+    },
+    {
+      "epoch": 0.34406215316315203,
+      "grad_norm": 0.5630763173103333,
+      "learning_rate": 2.997288816468252e-05,
+      "loss": 1.0296,
+      "step": 155
+    },
+    {
+      "epoch": 0.3551609322974473,
+      "grad_norm": 0.6153825521469116,
+      "learning_rate": 2.9965874973263953e-05,
+      "loss": 0.9819,
+      "step": 160
+    },
+    {
+      "epoch": 0.3662597114317425,
+      "grad_norm": 0.6237099170684814,
+      "learning_rate": 2.99580569577177e-05,
+      "loss": 1.0431,
+      "step": 165
+    },
+    {
+      "epoch": 0.37735849056603776,
+      "grad_norm": 0.6736854314804077,
+      "learning_rate": 2.9949434538475414e-05,
+      "loss": 0.9211,
+      "step": 170
+    },
+    {
+      "epoch": 0.38845726970033295,
+      "grad_norm": 0.7440330386161804,
+      "learning_rate": 2.9940008179227398e-05,
+      "loss": 0.9114,
+      "step": 175
+    },
+    {
+      "epoch": 0.3995560488346282,
+      "grad_norm": 0.7021297812461853,
+      "learning_rate": 2.9929778386897653e-05,
+      "loss": 0.959,
+      "step": 180
+    },
+    {
+      "epoch": 0.41065482796892344,
+      "grad_norm": 0.6943163871765137,
+      "learning_rate": 2.9918745711616647e-05,
+      "loss": 0.9401,
+      "step": 185
+    },
+    {
+      "epoch": 0.42175360710321863,
+      "grad_norm": 0.5902981758117676,
+      "learning_rate": 2.99069107466917e-05,
+      "loss": 0.9254,
+      "step": 190
+    },
+    {
+      "epoch": 0.4328523862375139,
+      "grad_norm": 0.7625190615653992,
+      "learning_rate": 2.9894274128575104e-05,
+      "loss": 1.0178,
+      "step": 195
+    },
+    {
+      "epoch": 0.4439511653718091,
+      "grad_norm": 0.9426499009132385,
+      "learning_rate": 2.9880836536829867e-05,
+      "loss": 0.9321,
+      "step": 200
+    },
+    {
+      "epoch": 0.4550499445061043,
+      "grad_norm": 0.6531440019607544,
+      "learning_rate": 2.9866598694093204e-05,
+      "loss": 0.9645,
+      "step": 205
+    },
+    {
+      "epoch": 0.46614872364039955,
+      "grad_norm": 0.7278143167495728,
+      "learning_rate": 2.985156136603764e-05,
+      "loss": 0.8816,
+      "step": 210
+    },
+    {
+      "epoch": 0.4772475027746948,
+      "grad_norm": 0.6720770001411438,
+      "learning_rate": 2.983572536132985e-05,
+      "loss": 0.9271,
+      "step": 215
+    },
+    {
+      "epoch": 0.48834628190899,
+      "grad_norm": 0.7253017425537109,
+      "learning_rate": 2.9819091531587184e-05,
+      "loss": 0.92,
+      "step": 220
+    },
+    {
+      "epoch": 0.49944506104328523,
+      "grad_norm": 0.7579580545425415,
+      "learning_rate": 2.9801660771331855e-05,
+      "loss": 0.9163,
+      "step": 225
+    },
+    {
+      "epoch": 0.5105438401775805,
+      "grad_norm": 0.7624631524085999,
+      "learning_rate": 2.9783434017942814e-05,
+      "loss": 0.8707,
+      "step": 230
+    },
+    {
+      "epoch": 0.5216426193118757,
+      "grad_norm": 0.8809816837310791,
+      "learning_rate": 2.9764412251605386e-05,
+      "loss": 0.8942,
+      "step": 235
+    },
+    {
+      "epoch": 0.532741398446171,
+      "grad_norm": 0.8497355580329895,
+      "learning_rate": 2.974459649525853e-05,
+      "loss": 0.9032,
+      "step": 240
+    },
+    {
+      "epoch": 0.5438401775804661,
+      "grad_norm": 0.771935224533081,
+      "learning_rate": 2.9723987814539818e-05,
+      "loss": 0.8595,
+      "step": 245
+    },
+    {
+      "epoch": 0.5549389567147613,
+      "grad_norm": 0.7510399222373962,
+      "learning_rate": 2.9702587317728157e-05,
+      "loss": 0.8882,
+      "step": 250
+    },
+    {
+      "epoch": 0.5660377358490566,
+      "grad_norm": 0.7542237043380737,
+      "learning_rate": 2.968039615568417e-05,
+      "loss": 0.8684,
+      "step": 255
+    },
+    {
+      "epoch": 0.5771365149833518,
+      "grad_norm": 0.8170939087867737,
+      "learning_rate": 2.965741552178831e-05,
+      "loss": 0.869,
+      "step": 260
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.7694569826126099,
+      "learning_rate": 2.9633646651876678e-05,
+      "loss": 0.8339,
+      "step": 265
+    },
+    {
+      "epoch": 0.5993340732519423,
+      "grad_norm": 0.8420746922492981,
+      "learning_rate": 2.960909082417457e-05,
+      "loss": 0.8849,
+      "step": 270
+    },
+    {
+      "epoch": 0.6104328523862376,
+      "grad_norm": 0.90952068567276,
+      "learning_rate": 2.9583749359227744e-05,
+      "loss": 0.8696,
+      "step": 275
+    },
+    {
+      "epoch": 0.6215316315205327,
+      "grad_norm": 0.9462668299674988,
+      "learning_rate": 2.9557623619831385e-05,
+      "loss": 0.8323,
+      "step": 280
+    },
+    {
+      "epoch": 0.632630410654828,
+      "grad_norm": 0.8263661861419678,
+      "learning_rate": 2.9530715010956828e-05,
+      "loss": 0.8225,
+      "step": 285
+    },
+    {
+      "epoch": 0.6437291897891232,
+      "grad_norm": 0.9636655449867249,
+      "learning_rate": 2.950302497967602e-05,
+      "loss": 0.8116,
+      "step": 290
+    },
+    {
+      "epoch": 0.6548279689234184,
+      "grad_norm": 0.792352020740509,
+      "learning_rate": 2.9474555015083665e-05,
+      "loss": 0.8045,
+      "step": 295
+    },
+    {
+      "epoch": 0.6659267480577137,
+      "grad_norm": 0.8013424873352051,
+      "learning_rate": 2.944530664821717e-05,
+      "loss": 0.7529,
+      "step": 300
+    },
+    {
+      "epoch": 0.6770255271920089,
+      "grad_norm": 0.8466354608535767,
+      "learning_rate": 2.9415281451974305e-05,
+      "loss": 0.7782,
+      "step": 305
+    },
+    {
+      "epoch": 0.6881243063263041,
+      "grad_norm": 0.8911815285682678,
+      "learning_rate": 2.938448104102862e-05,
+      "loss": 0.7746,
+      "step": 310
+    },
+    {
+      "epoch": 0.6992230854605993,
+      "grad_norm": 0.9244343638420105,
+      "learning_rate": 2.935290707174259e-05,
+      "loss": 0.7962,
+      "step": 315
+    },
+    {
+      "epoch": 0.7103218645948945,
+      "grad_norm": 0.9099295735359192,
+      "learning_rate": 2.9320561242078572e-05,
+      "loss": 0.7738,
+      "step": 320
+    },
+    {
+      "epoch": 0.7214206437291898,
+      "grad_norm": 0.8415182828903198,
+      "learning_rate": 2.928744529150748e-05,
+      "loss": 0.7366,
+      "step": 325
+    },
+    {
+      "epoch": 0.732519422863485,
+      "grad_norm": 0.9733411073684692,
+      "learning_rate": 2.9253561000915226e-05,
+      "loss": 0.7519,
+      "step": 330
+    },
+    {
+      "epoch": 0.7436182019977803,
+      "grad_norm": 0.8663511872291565,
+      "learning_rate": 2.9218910192506977e-05,
+      "loss": 0.7463,
+      "step": 335
+    },
+    {
+      "epoch": 0.7547169811320755,
+      "grad_norm": 0.8752774000167847,
+      "learning_rate": 2.918349472970914e-05,
+      "loss": 0.7216,
+      "step": 340
+    },
+    {
+      "epoch": 0.7658157602663707,
+      "grad_norm": 0.898760974407196,
+      "learning_rate": 2.9147316517069164e-05,
+      "loss": 0.805,
+      "step": 345
+    },
+    {
+      "epoch": 0.7769145394006659,
+      "grad_norm": 0.8114638328552246,
+      "learning_rate": 2.9110377500153103e-05,
+      "loss": 0.7402,
+      "step": 350
+    },
+    {
+      "epoch": 0.7880133185349611,
+      "grad_norm": 0.9255645871162415,
+      "learning_rate": 2.9072679665441016e-05,
+      "loss": 0.7046,
+      "step": 355
+    },
+    {
+      "epoch": 0.7991120976692564,
+      "grad_norm": 0.8243072628974915,
+      "learning_rate": 2.9034225040220105e-05,
+      "loss": 0.7653,
+      "step": 360
+    },
+    {
+      "epoch": 0.8102108768035516,
+      "grad_norm": 0.8807098269462585,
+      "learning_rate": 2.899501569247573e-05,
+      "loss": 0.7149,
+      "step": 365
+    },
+    {
+      "epoch": 0.8213096559378469,
+      "grad_norm": 0.8475015163421631,
+      "learning_rate": 2.895505373078016e-05,
+      "loss": 0.6763,
+      "step": 370
+    },
+    {
+      "epoch": 0.832408435072142,
+      "grad_norm": 0.8343256711959839,
+      "learning_rate": 2.891434130417922e-05,
+      "loss": 0.7348,
+      "step": 375
+    },
+    {
+      "epoch": 0.8435072142064373,
+      "grad_norm": 0.9187164306640625,
+      "learning_rate": 2.8872880602076675e-05,
+      "loss": 0.7555,
+      "step": 380
+    },
+    {
+      "epoch": 0.8546059933407325,
+      "grad_norm": 0.955761194229126,
+      "learning_rate": 2.8830673854116543e-05,
+      "loss": 0.6992,
+      "step": 385
+    },
+    {
+      "epoch": 0.8657047724750278,
+      "grad_norm": 0.9569207429885864,
+      "learning_rate": 2.878772333006314e-05,
+      "loss": 0.6657,
+      "step": 390
+    },
+    {
+      "epoch": 0.876803551609323,
+      "grad_norm": 0.9878470301628113,
+      "learning_rate": 2.874403133967906e-05,
+      "loss": 0.6982,
+      "step": 395
+    },
+    {
+      "epoch": 0.8879023307436182,
+      "grad_norm": 1.106117844581604,
+      "learning_rate": 2.869960023260092e-05,
+      "loss": 0.6987,
+      "step": 400
+    },
+    {
+      "epoch": 0.8990011098779135,
+      "grad_norm": 0.9092835187911987,
+      "learning_rate": 2.865443239821305e-05,
+      "loss": 0.6365,
+      "step": 405
+    },
+    {
+      "epoch": 0.9100998890122086,
+      "grad_norm": 1.0136357545852661,
+      "learning_rate": 2.860853026551896e-05,
+      "loss": 0.681,
+      "step": 410
+    },
+    {
+      "epoch": 0.9211986681465039,
+      "grad_norm": 0.9235714673995972,
+      "learning_rate": 2.8561896303010737e-05,
+      "loss": 0.6689,
+      "step": 415
+    },
+    {
+      "epoch": 0.9322974472807991,
+      "grad_norm": 0.8986316919326782,
+      "learning_rate": 2.8514533018536286e-05,
+      "loss": 0.6377,
+      "step": 420
+    },
+    {
+      "epoch": 0.9433962264150944,
+      "grad_norm": 1.089402675628662,
+      "learning_rate": 2.8466442959164467e-05,
+      "loss": 0.652,
+      "step": 425
+    },
+    {
+      "epoch": 0.9544950055493896,
+      "grad_norm": 0.9313281774520874,
+      "learning_rate": 2.841762871104812e-05,
+      "loss": 0.6775,
+      "step": 430
+    },
+    {
+      "epoch": 0.9655937846836848,
+      "grad_norm": 0.9396001100540161,
+      "learning_rate": 2.8368092899284992e-05,
+      "loss": 0.6707,
+      "step": 435
+    },
+    {
+      "epoch": 0.97669256381798,
+      "grad_norm": 0.8905186057090759,
+      "learning_rate": 2.8317838187776565e-05,
+      "loss": 0.6125,
+      "step": 440
+    },
+    {
+      "epoch": 0.9877913429522752,
+      "grad_norm": 0.9526168704032898,
+      "learning_rate": 2.826686727908479e-05,
+      "loss": 0.6681,
+      "step": 445
+    },
+    {
+      "epoch": 0.9988901220865705,
+      "grad_norm": 1.046419620513916,
+      "learning_rate": 2.8215182914286768e-05,
+      "loss": 0.6802,
+      "step": 450
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 2255,
+  "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": 6.499013219183493e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

121/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2046ae4ce5e7a2addbeaf11fa72be90c025d058631180fa9e5c08192e75f3ef7
+size 7736

121/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)