Upload folder using huggingface_hub

5/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

5/adapter_config.json

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

5/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9e029b8e65283e739bad24bf8d9178a90d2cb52f9ce246962d884d5f6fe2452
+size 791751704

5/latest

@@ -0,0 +1 @@
+global_step610

5/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f40b8f0341a72887a26e3a03818b04a7280996fe7cda836787db1dc89dda2af8
+size 15920

5/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e85f29352d7eeb23918df570fe377803eb3e305456f9985377e0a84aa2b21827
+size 15920

5/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa30fb6e977382599a7d5bb72702bb75bbdac542325d9e653bdbf8c4c973c210
+size 15920

5/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f957c4179974cdf4f1065432bae297c3552d662c0e4bb2d21efb149e4ef8cdbb
+size 15920

5/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a93277d9253735c25204efcfc508d73041cc19597d017b2d527b79038ca70f5
+size 15920

5/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9418a0736cd4f663a809bda9c7d2d71b5b255b4af8bf920d3312838e8b37488
+size 15920

5/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05458a0be19cfec75c7b30a67bc607500e9602e59e1af6b4bb7daf492840d9e2
+size 15920

5/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04dcfcebe1c1829d437aeb79535ca1b698cb9a548b97c21f0dd58ae7de829c3c
+size 15920

5/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce518e8efad5ba429fd3998ba368d4ae23d8d0e4ac5e214e0574eeb363be3c6f
+size 1064

5/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
+  }
+}

5/tokenizer_config.json

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

5/trainer_state.json

@@ -0,0 +1,888 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 610,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01639344262295082,
+      "grad_norm": 1.4445847272872925,
+      "learning_rate": 1.5584415584415587e-06,
+      "loss": 1.2372,
+      "step": 5
+    },
+    {
+      "epoch": 0.03278688524590164,
+      "grad_norm": 0.7953591346740723,
+      "learning_rate": 3.5064935064935066e-06,
+      "loss": 1.2195,
+      "step": 10
+    },
+    {
+      "epoch": 0.04918032786885246,
+      "grad_norm": 2.0549397468566895,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 1.2416,
+      "step": 15
+    },
+    {
+      "epoch": 0.06557377049180328,
+      "grad_norm": 0.5585774183273315,
+      "learning_rate": 7.402597402597403e-06,
+      "loss": 1.2289,
+      "step": 20
+    },
+    {
+      "epoch": 0.08196721311475409,
+      "grad_norm": 0.6144344210624695,
+      "learning_rate": 9.35064935064935e-06,
+      "loss": 1.2251,
+      "step": 25
+    },
+    {
+      "epoch": 0.09836065573770492,
+      "grad_norm": 0.5836143493652344,
+      "learning_rate": 1.12987012987013e-05,
+      "loss": 1.1654,
+      "step": 30
+    },
+    {
+      "epoch": 0.11475409836065574,
+      "grad_norm": 0.5931638479232788,
+      "learning_rate": 1.3246753246753247e-05,
+      "loss": 1.1853,
+      "step": 35
+    },
+    {
+      "epoch": 0.13114754098360656,
+      "grad_norm": 0.4809166193008423,
+      "learning_rate": 1.5194805194805194e-05,
+      "loss": 1.1453,
+      "step": 40
+    },
+    {
+      "epoch": 0.14754098360655737,
+      "grad_norm": 0.4646892845630646,
+      "learning_rate": 1.7142857142857142e-05,
+      "loss": 1.0871,
+      "step": 45
+    },
+    {
+      "epoch": 0.16393442622950818,
+      "grad_norm": 0.5099896192550659,
+      "learning_rate": 1.909090909090909e-05,
+      "loss": 1.1506,
+      "step": 50
+    },
+    {
+      "epoch": 0.18032786885245902,
+      "grad_norm": 0.501319169998169,
+      "learning_rate": 2.103896103896104e-05,
+      "loss": 1.1107,
+      "step": 55
+    },
+    {
+      "epoch": 0.19672131147540983,
+      "grad_norm": 0.43489858508110046,
+      "learning_rate": 2.298701298701299e-05,
+      "loss": 1.118,
+      "step": 60
+    },
+    {
+      "epoch": 0.21311475409836064,
+      "grad_norm": 0.5342327356338501,
+      "learning_rate": 2.493506493506494e-05,
+      "loss": 1.1286,
+      "step": 65
+    },
+    {
+      "epoch": 0.22950819672131148,
+      "grad_norm": 0.4543209969997406,
+      "learning_rate": 2.6883116883116883e-05,
+      "loss": 1.0512,
+      "step": 70
+    },
+    {
+      "epoch": 0.2459016393442623,
+      "grad_norm": 0.45664912462234497,
+      "learning_rate": 2.8831168831168832e-05,
+      "loss": 1.0923,
+      "step": 75
+    },
+    {
+      "epoch": 0.26229508196721313,
+      "grad_norm": 0.5187792181968689,
+      "learning_rate": 2.999985878423113e-05,
+      "loss": 1.0963,
+      "step": 80
+    },
+    {
+      "epoch": 0.2786885245901639,
+      "grad_norm": 0.6245763897895813,
+      "learning_rate": 2.999827013736722e-05,
+      "loss": 1.0923,
+      "step": 85
+    },
+    {
+      "epoch": 0.29508196721311475,
+      "grad_norm": 0.5452545285224915,
+      "learning_rate": 2.9994916511501615e-05,
+      "loss": 1.0095,
+      "step": 90
+    },
+    {
+      "epoch": 0.3114754098360656,
+      "grad_norm": 0.584423303604126,
+      "learning_rate": 2.9989798301285106e-05,
+      "loss": 1.0663,
+      "step": 95
+    },
+    {
+      "epoch": 0.32786885245901637,
+      "grad_norm": 0.5853920578956604,
+      "learning_rate": 2.9982916109022723e-05,
+      "loss": 0.9832,
+      "step": 100
+    },
+    {
+      "epoch": 0.3442622950819672,
+      "grad_norm": 0.5287838578224182,
+      "learning_rate": 2.9974270744602862e-05,
+      "loss": 1.0344,
+      "step": 105
+    },
+    {
+      "epoch": 0.36065573770491804,
+      "grad_norm": 0.7069889903068542,
+      "learning_rate": 2.9963863225401957e-05,
+      "loss": 0.9816,
+      "step": 110
+    },
+    {
+      "epoch": 0.3770491803278688,
+      "grad_norm": 0.6540791392326355,
+      "learning_rate": 2.9951694776164774e-05,
+      "loss": 0.9736,
+      "step": 115
+    },
+    {
+      "epoch": 0.39344262295081966,
+      "grad_norm": 0.5601514577865601,
+      "learning_rate": 2.9937766828860283e-05,
+      "loss": 0.9369,
+      "step": 120
+    },
+    {
+      "epoch": 0.4098360655737705,
+      "grad_norm": 0.6108849048614502,
+      "learning_rate": 2.9922081022513134e-05,
+      "loss": 0.9638,
+      "step": 125
+    },
+    {
+      "epoch": 0.4262295081967213,
+      "grad_norm": 0.5754315853118896,
+      "learning_rate": 2.9904639203010794e-05,
+      "loss": 0.9342,
+      "step": 130
+    },
+    {
+      "epoch": 0.4426229508196721,
+      "grad_norm": 0.6295889616012573,
+      "learning_rate": 2.988544342288631e-05,
+      "loss": 0.97,
+      "step": 135
+    },
+    {
+      "epoch": 0.45901639344262296,
+      "grad_norm": 0.6687360405921936,
+      "learning_rate": 2.9864495941076784e-05,
+      "loss": 0.9167,
+      "step": 140
+    },
+    {
+      "epoch": 0.47540983606557374,
+      "grad_norm": 0.6689046621322632,
+      "learning_rate": 2.984179922265754e-05,
+      "loss": 0.8656,
+      "step": 145
+    },
+    {
+      "epoch": 0.4918032786885246,
+      "grad_norm": 0.8281727433204651,
+      "learning_rate": 2.981735593855202e-05,
+      "loss": 0.8761,
+      "step": 150
+    },
+    {
+      "epoch": 0.5081967213114754,
+      "grad_norm": 0.8986142873764038,
+      "learning_rate": 2.9791168965217495e-05,
+      "loss": 0.8899,
+      "step": 155
+    },
+    {
+      "epoch": 0.5245901639344263,
+      "grad_norm": 0.7329938411712646,
+      "learning_rate": 2.976324138430656e-05,
+      "loss": 0.8305,
+      "step": 160
+    },
+    {
+      "epoch": 0.5409836065573771,
+      "grad_norm": 0.8260165452957153,
+      "learning_rate": 2.9733576482304487e-05,
+      "loss": 0.8892,
+      "step": 165
+    },
+    {
+      "epoch": 0.5573770491803278,
+      "grad_norm": 0.8703463077545166,
+      "learning_rate": 2.970217775014247e-05,
+      "loss": 0.8721,
+      "step": 170
+    },
+    {
+      "epoch": 0.5737704918032787,
+      "grad_norm": 0.9152560234069824,
+      "learning_rate": 2.9669048882786823e-05,
+      "loss": 0.8314,
+      "step": 175
+    },
+    {
+      "epoch": 0.5901639344262295,
+      "grad_norm": 0.8130025863647461,
+      "learning_rate": 2.9634193778804165e-05,
+      "loss": 0.847,
+      "step": 180
+    },
+    {
+      "epoch": 0.6065573770491803,
+      "grad_norm": 0.8214168548583984,
+      "learning_rate": 2.9597616539902622e-05,
+      "loss": 0.8146,
+      "step": 185
+    },
+    {
+      "epoch": 0.6229508196721312,
+      "grad_norm": 0.8313847780227661,
+      "learning_rate": 2.9559321470449178e-05,
+      "loss": 0.8047,
+      "step": 190
+    },
+    {
+      "epoch": 0.639344262295082,
+      "grad_norm": 0.8505757451057434,
+      "learning_rate": 2.951931307696309e-05,
+      "loss": 0.7699,
+      "step": 195
+    },
+    {
+      "epoch": 0.6557377049180327,
+      "grad_norm": 0.8449229598045349,
+      "learning_rate": 2.9477596067585633e-05,
+      "loss": 0.7856,
+      "step": 200
+    },
+    {
+      "epoch": 0.6721311475409836,
+      "grad_norm": 0.8714151978492737,
+      "learning_rate": 2.9434175351525983e-05,
+      "loss": 0.7814,
+      "step": 205
+    },
+    {
+      "epoch": 0.6885245901639344,
+      "grad_norm": 0.7892358303070068,
+      "learning_rate": 2.938905603848355e-05,
+      "loss": 0.7404,
+      "step": 210
+    },
+    {
+      "epoch": 0.7049180327868853,
+      "grad_norm": 0.8768689036369324,
+      "learning_rate": 2.9342243438046666e-05,
+      "loss": 0.7516,
+      "step": 215
+    },
+    {
+      "epoch": 0.7213114754098361,
+      "grad_norm": 0.9508657455444336,
+      "learning_rate": 2.9293743059067763e-05,
+      "loss": 0.7152,
+      "step": 220
+    },
+    {
+      "epoch": 0.7377049180327869,
+      "grad_norm": 0.9115400910377502,
+      "learning_rate": 2.924356060901507e-05,
+      "loss": 0.7078,
+      "step": 225
+    },
+    {
+      "epoch": 0.7540983606557377,
+      "grad_norm": 0.9279278516769409,
+      "learning_rate": 2.9191701993301e-05,
+      "loss": 0.8145,
+      "step": 230
+    },
+    {
+      "epoch": 0.7704918032786885,
+      "grad_norm": 0.8984402418136597,
+      "learning_rate": 2.913817331458719e-05,
+      "loss": 0.7177,
+      "step": 235
+    },
+    {
+      "epoch": 0.7868852459016393,
+      "grad_norm": 1.000724196434021,
+      "learning_rate": 2.9082980872066353e-05,
+      "loss": 0.7077,
+      "step": 240
+    },
+    {
+      "epoch": 0.8032786885245902,
+      "grad_norm": 0.968752920627594,
+      "learning_rate": 2.9026131160720993e-05,
+      "loss": 0.7409,
+      "step": 245
+    },
+    {
+      "epoch": 0.819672131147541,
+      "grad_norm": 1.0022119283676147,
+      "learning_rate": 2.8967630870559073e-05,
+      "loss": 0.6708,
+      "step": 250
+    },
+    {
+      "epoch": 0.8360655737704918,
+      "grad_norm": 0.9408950805664062,
+      "learning_rate": 2.890748688582677e-05,
+      "loss": 0.6916,
+      "step": 255
+    },
+    {
+      "epoch": 0.8524590163934426,
+      "grad_norm": 0.8202570080757141,
+      "learning_rate": 2.8845706284198314e-05,
+      "loss": 0.7148,
+      "step": 260
+    },
+    {
+      "epoch": 0.8688524590163934,
+      "grad_norm": 1.015702486038208,
+      "learning_rate": 2.8782296335943116e-05,
+      "loss": 0.6521,
+      "step": 265
+    },
+    {
+      "epoch": 0.8852459016393442,
+      "grad_norm": 1.0327638387680054,
+      "learning_rate": 2.871726450307021e-05,
+      "loss": 0.6667,
+      "step": 270
+    },
+    {
+      "epoch": 0.9016393442622951,
+      "grad_norm": 0.929453432559967,
+      "learning_rate": 2.865061843845011e-05,
+      "loss": 0.6014,
+      "step": 275
+    },
+    {
+      "epoch": 0.9180327868852459,
+      "grad_norm": 0.8924345970153809,
+      "learning_rate": 2.8582365984914265e-05,
+      "loss": 0.6254,
+      "step": 280
+    },
+    {
+      "epoch": 0.9344262295081968,
+      "grad_norm": 1.004958152770996,
+      "learning_rate": 2.85125151743321e-05,
+      "loss": 0.6633,
+      "step": 285
+    },
+    {
+      "epoch": 0.9508196721311475,
+      "grad_norm": 0.932119607925415,
+      "learning_rate": 2.8441074226665834e-05,
+      "loss": 0.619,
+      "step": 290
+    },
+    {
+      "epoch": 0.9672131147540983,
+      "grad_norm": 0.8984085917472839,
+      "learning_rate": 2.836805154900317e-05,
+      "loss": 0.6414,
+      "step": 295
+    },
+    {
+      "epoch": 0.9836065573770492,
+      "grad_norm": 1.0108938217163086,
+      "learning_rate": 2.8293455734567964e-05,
+      "loss": 0.6646,
+      "step": 300
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9058620929718018,
+      "learning_rate": 2.8217295561708986e-05,
+      "loss": 0.6386,
+      "step": 305
+    },
+    {
+      "epoch": 1.0163934426229508,
+      "grad_norm": 0.9871558547019958,
+      "learning_rate": 2.813957999286687e-05,
+      "loss": 0.5756,
+      "step": 310
+    },
+    {
+      "epoch": 1.0327868852459017,
+      "grad_norm": 1.0473088026046753,
+      "learning_rate": 2.8060318173519455e-05,
+      "loss": 0.5358,
+      "step": 315
+    },
+    {
+      "epoch": 1.0491803278688525,
+      "grad_norm": 1.2112137079238892,
+      "learning_rate": 2.7979519431105537e-05,
+      "loss": 0.5603,
+      "step": 320
+    },
+    {
+      "epoch": 1.0655737704918034,
+      "grad_norm": 1.0239282846450806,
+      "learning_rate": 2.7897193273927242e-05,
+      "loss": 0.5272,
+      "step": 325
+    },
+    {
+      "epoch": 1.0819672131147542,
+      "grad_norm": 1.1301106214523315,
+      "learning_rate": 2.7813349390031077e-05,
+      "loss": 0.5162,
+      "step": 330
+    },
+    {
+      "epoch": 1.098360655737705,
+      "grad_norm": 0.9581446051597595,
+      "learning_rate": 2.772799764606787e-05,
+      "loss": 0.5285,
+      "step": 335
+    },
+    {
+      "epoch": 1.1147540983606556,
+      "grad_norm": 1.120554804801941,
+      "learning_rate": 2.764114808613167e-05,
+      "loss": 0.5497,
+      "step": 340
+    },
+    {
+      "epoch": 1.1311475409836065,
+      "grad_norm": 1.1846710443496704,
+      "learning_rate": 2.755281093057778e-05,
+      "loss": 0.4792,
+      "step": 345
+    },
+    {
+      "epoch": 1.1475409836065573,
+      "grad_norm": 1.0247408151626587,
+      "learning_rate": 2.7462996574820014e-05,
+      "loss": 0.5178,
+      "step": 350
+    },
+    {
+      "epoch": 1.1639344262295082,
+      "grad_norm": 0.9996801614761353,
+      "learning_rate": 2.7371715588107388e-05,
+      "loss": 0.4987,
+      "step": 355
+    },
+    {
+      "epoch": 1.180327868852459,
+      "grad_norm": 1.0238093137741089,
+      "learning_rate": 2.7278978712280353e-05,
+      "loss": 0.456,
+      "step": 360
+    },
+    {
+      "epoch": 1.1967213114754098,
+      "grad_norm": 1.2357455492019653,
+      "learning_rate": 2.7184796860506697e-05,
+      "loss": 0.5292,
+      "step": 365
+    },
+    {
+      "epoch": 1.2131147540983607,
+      "grad_norm": 1.1693018674850464,
+      "learning_rate": 2.7089181115997292e-05,
+      "loss": 0.4745,
+      "step": 370
+    },
+    {
+      "epoch": 1.2295081967213115,
+      "grad_norm": 0.9463955760002136,
+      "learning_rate": 2.699214273070185e-05,
+      "loss": 0.4803,
+      "step": 375
+    },
+    {
+      "epoch": 1.2459016393442623,
+      "grad_norm": 1.2919942140579224,
+      "learning_rate": 2.689369312398479e-05,
+      "loss": 0.4721,
+      "step": 380
+    },
+    {
+      "epoch": 1.2622950819672132,
+      "grad_norm": 1.0499731302261353,
+      "learning_rate": 2.6793843881281435e-05,
+      "loss": 0.5081,
+      "step": 385
+    },
+    {
+      "epoch": 1.278688524590164,
+      "grad_norm": 1.1290088891983032,
+      "learning_rate": 2.6692606752734642e-05,
+      "loss": 0.4626,
+      "step": 390
+    },
+    {
+      "epoch": 1.2950819672131146,
+      "grad_norm": 1.0272178649902344,
+      "learning_rate": 2.6589993651812055e-05,
+      "loss": 0.4824,
+      "step": 395
+    },
+    {
+      "epoch": 1.3114754098360657,
+      "grad_norm": 1.1490875482559204,
+      "learning_rate": 2.648601665390416e-05,
+      "loss": 0.467,
+      "step": 400
+    },
+    {
+      "epoch": 1.3278688524590163,
+      "grad_norm": 1.0278301239013672,
+      "learning_rate": 2.638068799490324e-05,
+      "loss": 0.4236,
+      "step": 405
+    },
+    {
+      "epoch": 1.3442622950819672,
+      "grad_norm": 1.011580228805542,
+      "learning_rate": 2.6274020069763505e-05,
+      "loss": 0.4595,
+      "step": 410
+    },
+    {
+      "epoch": 1.360655737704918,
+      "grad_norm": 1.1207678318023682,
+      "learning_rate": 2.6166025431042427e-05,
+      "loss": 0.4877,
+      "step": 415
+    },
+    {
+      "epoch": 1.3770491803278688,
+      "grad_norm": 1.0334429740905762,
+      "learning_rate": 2.605671678742361e-05,
+      "loss": 0.416,
+      "step": 420
+    },
+    {
+      "epoch": 1.3934426229508197,
+      "grad_norm": 1.1720657348632812,
+      "learning_rate": 2.5946107002221206e-05,
+      "loss": 0.4295,
+      "step": 425
+    },
+    {
+      "epoch": 1.4098360655737705,
+      "grad_norm": 1.1176438331604004,
+      "learning_rate": 2.583420909186622e-05,
+      "loss": 0.4133,
+      "step": 430
+    },
+    {
+      "epoch": 1.4262295081967213,
+      "grad_norm": 1.1450058221817017,
+      "learning_rate": 2.572103622437471e-05,
+      "loss": 0.4412,
+      "step": 435
+    },
+    {
+      "epoch": 1.4426229508196722,
+      "grad_norm": 1.0903273820877075,
+      "learning_rate": 2.5606601717798212e-05,
+      "loss": 0.3953,
+      "step": 440
+    },
+    {
+      "epoch": 1.459016393442623,
+      "grad_norm": 1.2657668590545654,
+      "learning_rate": 2.5490919038656495e-05,
+      "loss": 0.4089,
+      "step": 445
+    },
+    {
+      "epoch": 1.4754098360655736,
+      "grad_norm": 1.1453659534454346,
+      "learning_rate": 2.537400180035281e-05,
+      "loss": 0.4526,
+      "step": 450
+    },
+    {
+      "epoch": 1.4918032786885247,
+      "grad_norm": 1.1551433801651,
+      "learning_rate": 2.525586376157191e-05,
+      "loss": 0.4247,
+      "step": 455
+    },
+    {
+      "epoch": 1.5081967213114753,
+      "grad_norm": 1.0655601024627686,
+      "learning_rate": 2.513651882466091e-05,
+      "loss": 0.4207,
+      "step": 460
+    },
+    {
+      "epoch": 1.5245901639344264,
+      "grad_norm": 1.0673174858093262,
+      "learning_rate": 2.5015981033993313e-05,
+      "loss": 0.3831,
+      "step": 465
+    },
+    {
+      "epoch": 1.540983606557377,
+      "grad_norm": 1.0724608898162842,
+      "learning_rate": 2.4894264574316258e-05,
+      "loss": 0.3997,
+      "step": 470
+    },
+    {
+      "epoch": 1.5573770491803278,
+      "grad_norm": 1.037182092666626,
+      "learning_rate": 2.4771383769081286e-05,
+      "loss": 0.3979,
+      "step": 475
+    },
+    {
+      "epoch": 1.5737704918032787,
+      "grad_norm": 1.1553537845611572,
+      "learning_rate": 2.4647353078758772e-05,
+      "loss": 0.3831,
+      "step": 480
+    },
+    {
+      "epoch": 1.5901639344262295,
+      "grad_norm": 1.154228687286377,
+      "learning_rate": 2.4522187099136223e-05,
+      "loss": 0.3843,
+      "step": 485
+    },
+    {
+      "epoch": 1.6065573770491803,
+      "grad_norm": 1.0977567434310913,
+      "learning_rate": 2.4395900559600676e-05,
+      "loss": 0.3558,
+      "step": 490
+    },
+    {
+      "epoch": 1.6229508196721312,
+      "grad_norm": 1.1280032396316528,
+      "learning_rate": 2.4268508321405363e-05,
+      "loss": 0.394,
+      "step": 495
+    },
+    {
+      "epoch": 1.639344262295082,
+      "grad_norm": 1.16530442237854,
+      "learning_rate": 2.414002537592084e-05,
+      "loss": 0.361,
+      "step": 500
+    },
+    {
+      "epoch": 1.6557377049180326,
+      "grad_norm": 1.2421867847442627,
+      "learning_rate": 2.401046684287084e-05,
+      "loss": 0.3906,
+      "step": 505
+    },
+    {
+      "epoch": 1.6721311475409837,
+      "grad_norm": 1.2887860536575317,
+      "learning_rate": 2.3879847968552983e-05,
+      "loss": 0.4039,
+      "step": 510
+    },
+    {
+      "epoch": 1.6885245901639343,
+      "grad_norm": 1.034765362739563,
+      "learning_rate": 2.3748184124044614e-05,
+      "loss": 0.3515,
+      "step": 515
+    },
+    {
+      "epoch": 1.7049180327868854,
+      "grad_norm": 1.0902974605560303,
+      "learning_rate": 2.3615490803393963e-05,
+      "loss": 0.393,
+      "step": 520
+    },
+    {
+      "epoch": 1.721311475409836,
+      "grad_norm": 1.1227500438690186,
+      "learning_rate": 2.3481783621796817e-05,
+      "loss": 0.3554,
+      "step": 525
+    },
+    {
+      "epoch": 1.737704918032787,
+      "grad_norm": 1.0987911224365234,
+      "learning_rate": 2.3347078313758932e-05,
+      "loss": 0.3646,
+      "step": 530
+    },
+    {
+      "epoch": 1.7540983606557377,
+      "grad_norm": 1.0380116701126099,
+      "learning_rate": 2.3211390731244437e-05,
+      "loss": 0.3497,
+      "step": 535
+    },
+    {
+      "epoch": 1.7704918032786885,
+      "grad_norm": 1.0511633157730103,
+      "learning_rate": 2.3074736841810365e-05,
+      "loss": 0.385,
+      "step": 540
+    },
+    {
+      "epoch": 1.7868852459016393,
+      "grad_norm": 1.018215537071228,
+      "learning_rate": 2.2937132726727633e-05,
+      "loss": 0.348,
+      "step": 545
+    },
+    {
+      "epoch": 1.8032786885245902,
+      "grad_norm": 1.085090160369873,
+      "learning_rate": 2.279859457908862e-05,
+      "loss": 0.3675,
+      "step": 550
+    },
+    {
+      "epoch": 1.819672131147541,
+      "grad_norm": 1.0416409969329834,
+      "learning_rate": 2.265913870190154e-05,
+      "loss": 0.3742,
+      "step": 555
+    },
+    {
+      "epoch": 1.8360655737704918,
+      "grad_norm": 1.280327558517456,
+      "learning_rate": 2.2518781506171975e-05,
+      "loss": 0.3629,
+      "step": 560
+    },
+    {
+      "epoch": 1.8524590163934427,
+      "grad_norm": 1.1138688325881958,
+      "learning_rate": 2.237753950897164e-05,
+      "loss": 0.3433,
+      "step": 565
+    },
+    {
+      "epoch": 1.8688524590163933,
+      "grad_norm": 1.2459837198257446,
+      "learning_rate": 2.2235429331494618e-05,
+      "loss": 0.3369,
+      "step": 570
+    },
+    {
+      "epoch": 1.8852459016393444,
+      "grad_norm": 1.0113780498504639,
+      "learning_rate": 2.209246769710148e-05,
+      "loss": 0.3501,
+      "step": 575
+    },
+    {
+      "epoch": 1.901639344262295,
+      "grad_norm": 1.1586419343948364,
+      "learning_rate": 2.1948671429351232e-05,
+      "loss": 0.3459,
+      "step": 580
+    },
+    {
+      "epoch": 1.918032786885246,
+      "grad_norm": 1.0399360656738281,
+      "learning_rate": 2.1804057450021564e-05,
+      "loss": 0.3283,
+      "step": 585
+    },
+    {
+      "epoch": 1.9344262295081966,
+      "grad_norm": 1.1013191938400269,
+      "learning_rate": 2.1658642777117534e-05,
+      "loss": 0.3214,
+      "step": 590
+    },
+    {
+      "epoch": 1.9508196721311475,
+      "grad_norm": 1.1633572578430176,
+      "learning_rate": 2.151244452286887e-05,
+      "loss": 0.3399,
+      "step": 595
+    },
+    {
+      "epoch": 1.9672131147540983,
+      "grad_norm": 0.9520934224128723,
+      "learning_rate": 2.136547989171625e-05,
+      "loss": 0.3468,
+      "step": 600
+    },
+    {
+      "epoch": 1.9836065573770492,
+      "grad_norm": 1.1227986812591553,
+      "learning_rate": 2.1217766178286696e-05,
+      "loss": 0.2954,
+      "step": 605
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.099863886833191,
+      "learning_rate": 2.1069320765358354e-05,
+      "loss": 0.2882,
+      "step": 610
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1525,
+  "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": 9.493903381060649e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

5/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50c4beae811068ed7d96ca51f489ce08a8bbeb03dfecd79e5bb7753d0ab991b8
+size 7736

5/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)