Upload folder using huggingface_hub

31/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

31/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": [
+    "q_proj",
+    "o_proj",
+    "v_proj",
+    "gate_proj",
+    "up_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

31/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f61cde273bf5ab0491b5e53c189bfde3103e6ef23b1ce4467f1f5963b324ed5a
+size 791751704

31/latest

@@ -0,0 +1 @@
+global_step618

31/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fd14d2c29491e41dd2c08ce5ab9a24a167d5d86465eb7aaf3e0371511422a65
+size 15920

31/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:703aedab46e6ff410b6f430983afc65ee803af5a987db99865658c25e4ae20da
+size 15920

31/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0bb270b086a06a4c866d1e63924621b7c9d6ee293835126367075426f3005e5a
+size 15920

31/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:587d9b9bc63bd53c70767de473202fc82ffa774ba1e5e62e35d30ece8dc4da15
+size 15920

31/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:204ad7ea692a1c6ac635ec65db5a6f79acb2de70373de50ecba9063b875375ad
+size 15920

31/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f6a684fafe100a09324afeaa49748ece77447f21bc2b114d5195998d96239e1
+size 15920

31/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60d8f1860024a39e7eb59274618cf03604ad7dd303d3520788bb19b4d5bad093
+size 15920

31/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fed6f4233ac722cca58dd8e75b98be2c5f7881fa61a2ac70c04a16c962e30c2
+size 15920

31/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29254aefbb16799e3a35c79b3d21369504f3c19a83825f04725d9450a5c7038f
+size 1064

31/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
+  }
+}

31/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
+}

31/trainer_state.json

@@ -0,0 +1,895 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 618,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01620745542949757,
+      "grad_norm": 1.2666090726852417,
+      "learning_rate": 1.5384615384615385e-06,
+      "loss": 1.3575,
+      "step": 5
+    },
+    {
+      "epoch": 0.03241491085899514,
+      "grad_norm": 0.9747076630592346,
+      "learning_rate": 3.4615384615384617e-06,
+      "loss": 1.3319,
+      "step": 10
+    },
+    {
+      "epoch": 0.04862236628849271,
+      "grad_norm": 0.9286755919456482,
+      "learning_rate": 5.384615384615385e-06,
+      "loss": 1.2949,
+      "step": 15
+    },
+    {
+      "epoch": 0.06482982171799027,
+      "grad_norm": 0.8217413425445557,
+      "learning_rate": 7.307692307692308e-06,
+      "loss": 1.3087,
+      "step": 20
+    },
+    {
+      "epoch": 0.08103727714748785,
+      "grad_norm": 0.5836331844329834,
+      "learning_rate": 9.230769230769232e-06,
+      "loss": 1.3194,
+      "step": 25
+    },
+    {
+      "epoch": 0.09724473257698542,
+      "grad_norm": 0.6233932375907898,
+      "learning_rate": 1.1153846153846154e-05,
+      "loss": 1.2975,
+      "step": 30
+    },
+    {
+      "epoch": 0.11345218800648298,
+      "grad_norm": 0.5077695250511169,
+      "learning_rate": 1.3076923076923078e-05,
+      "loss": 1.2869,
+      "step": 35
+    },
+    {
+      "epoch": 0.12965964343598055,
+      "grad_norm": 0.4717237055301666,
+      "learning_rate": 1.5e-05,
+      "loss": 1.2447,
+      "step": 40
+    },
+    {
+      "epoch": 0.1458670988654781,
+      "grad_norm": 0.5701112151145935,
+      "learning_rate": 1.6923076923076924e-05,
+      "loss": 1.2602,
+      "step": 45
+    },
+    {
+      "epoch": 0.1620745542949757,
+      "grad_norm": 0.573088526725769,
+      "learning_rate": 1.8846153846153846e-05,
+      "loss": 1.2537,
+      "step": 50
+    },
+    {
+      "epoch": 0.17828200972447325,
+      "grad_norm": 0.6250770092010498,
+      "learning_rate": 2.076923076923077e-05,
+      "loss": 1.2161,
+      "step": 55
+    },
+    {
+      "epoch": 0.19448946515397084,
+      "grad_norm": 0.4536466896533966,
+      "learning_rate": 2.269230769230769e-05,
+      "loss": 1.1942,
+      "step": 60
+    },
+    {
+      "epoch": 0.2106969205834684,
+      "grad_norm": 0.5518207550048828,
+      "learning_rate": 2.4615384615384616e-05,
+      "loss": 1.1764,
+      "step": 65
+    },
+    {
+      "epoch": 0.22690437601296595,
+      "grad_norm": 0.49097076058387756,
+      "learning_rate": 2.6538461538461538e-05,
+      "loss": 1.1654,
+      "step": 70
+    },
+    {
+      "epoch": 0.24311183144246354,
+      "grad_norm": 0.7466371059417725,
+      "learning_rate": 2.846153846153846e-05,
+      "loss": 1.2078,
+      "step": 75
+    },
+    {
+      "epoch": 0.2593192868719611,
+      "grad_norm": 0.5528755187988281,
+      "learning_rate": 2.999996560458015e-05,
+      "loss": 1.0975,
+      "step": 80
+    },
+    {
+      "epoch": 0.2755267423014587,
+      "grad_norm": 0.573396623134613,
+      "learning_rate": 2.999876178144779e-05,
+      "loss": 1.1408,
+      "step": 85
+    },
+    {
+      "epoch": 0.2917341977309562,
+      "grad_norm": 0.5396841764450073,
+      "learning_rate": 2.9995838345058782e-05,
+      "loss": 1.1354,
+      "step": 90
+    },
+    {
+      "epoch": 0.3079416531604538,
+      "grad_norm": 0.6083248257637024,
+      "learning_rate": 2.999119563058612e-05,
+      "loss": 1.1492,
+      "step": 95
+    },
+    {
+      "epoch": 0.3241491085899514,
+      "grad_norm": 0.6934502720832825,
+      "learning_rate": 2.9984834170318635e-05,
+      "loss": 1.1186,
+      "step": 100
+    },
+    {
+      "epoch": 0.34035656401944897,
+      "grad_norm": 0.6632962822914124,
+      "learning_rate": 2.9976754693599964e-05,
+      "loss": 1.0607,
+      "step": 105
+    },
+    {
+      "epoch": 0.3565640194489465,
+      "grad_norm": 0.7568578720092773,
+      "learning_rate": 2.9966958126744923e-05,
+      "loss": 1.1135,
+      "step": 110
+    },
+    {
+      "epoch": 0.3727714748784441,
+      "grad_norm": 0.8758991956710815,
+      "learning_rate": 2.9955445592933296e-05,
+      "loss": 1.0556,
+      "step": 115
+    },
+    {
+      "epoch": 0.3889789303079417,
+      "grad_norm": 0.7006877064704895,
+      "learning_rate": 2.994221841208111e-05,
+      "loss": 0.9963,
+      "step": 120
+    },
+    {
+      "epoch": 0.4051863857374392,
+      "grad_norm": 0.7291510105133057,
+      "learning_rate": 2.9927278100689243e-05,
+      "loss": 1.0383,
+      "step": 125
+    },
+    {
+      "epoch": 0.4213938411669368,
+      "grad_norm": 0.6858572959899902,
+      "learning_rate": 2.9910626371669593e-05,
+      "loss": 1.0833,
+      "step": 130
+    },
+    {
+      "epoch": 0.4376012965964344,
+      "grad_norm": 0.788310170173645,
+      "learning_rate": 2.9892265134148686e-05,
+      "loss": 0.9907,
+      "step": 135
+    },
+    {
+      "epoch": 0.4538087520259319,
+      "grad_norm": 0.6967231631278992,
+      "learning_rate": 2.9872196493248794e-05,
+      "loss": 1.0198,
+      "step": 140
+    },
+    {
+      "epoch": 0.4700162074554295,
+      "grad_norm": 0.7951614856719971,
+      "learning_rate": 2.9850422749846577e-05,
+      "loss": 0.9352,
+      "step": 145
+    },
+    {
+      "epoch": 0.4862236628849271,
+      "grad_norm": 0.7685369849205017,
+      "learning_rate": 2.9826946400309295e-05,
+      "loss": 0.9808,
+      "step": 150
+    },
+    {
+      "epoch": 0.5024311183144247,
+      "grad_norm": 0.8352431654930115,
+      "learning_rate": 2.980177013620858e-05,
+      "loss": 0.9216,
+      "step": 155
+    },
+    {
+      "epoch": 0.5186385737439222,
+      "grad_norm": 0.827575147151947,
+      "learning_rate": 2.9774896844011887e-05,
+      "loss": 0.9008,
+      "step": 160
+    },
+    {
+      "epoch": 0.5348460291734197,
+      "grad_norm": 0.8291617035865784,
+      "learning_rate": 2.97463296047515e-05,
+      "loss": 0.8954,
+      "step": 165
+    },
+    {
+      "epoch": 0.5510534846029174,
+      "grad_norm": 0.7617440223693848,
+      "learning_rate": 2.9716071693671353e-05,
+      "loss": 0.9752,
+      "step": 170
+    },
+    {
+      "epoch": 0.5672609400324149,
+      "grad_norm": 0.7175273299217224,
+      "learning_rate": 2.9684126579851468e-05,
+      "loss": 0.9052,
+      "step": 175
+    },
+    {
+      "epoch": 0.5834683954619124,
+      "grad_norm": 0.8096367120742798,
+      "learning_rate": 2.9650497925810266e-05,
+      "loss": 0.9176,
+      "step": 180
+    },
+    {
+      "epoch": 0.5996758508914101,
+      "grad_norm": 0.91055828332901,
+      "learning_rate": 2.9615189587084628e-05,
+      "loss": 0.925,
+      "step": 185
+    },
+    {
+      "epoch": 0.6158833063209076,
+      "grad_norm": 0.9182955026626587,
+      "learning_rate": 2.9578205611787877e-05,
+      "loss": 0.9157,
+      "step": 190
+    },
+    {
+      "epoch": 0.6320907617504052,
+      "grad_norm": 0.9966630935668945,
+      "learning_rate": 2.953955024014565e-05,
+      "loss": 0.8842,
+      "step": 195
+    },
+    {
+      "epoch": 0.6482982171799028,
+      "grad_norm": 0.7878755927085876,
+      "learning_rate": 2.9499227904009748e-05,
+      "loss": 0.8493,
+      "step": 200
+    },
+    {
+      "epoch": 0.6645056726094003,
+      "grad_norm": 0.841929018497467,
+      "learning_rate": 2.945724322635004e-05,
+      "loss": 0.7851,
+      "step": 205
+    },
+    {
+      "epoch": 0.6807131280388979,
+      "grad_norm": 0.8280230164527893,
+      "learning_rate": 2.9413601020724435e-05,
+      "loss": 0.8042,
+      "step": 210
+    },
+    {
+      "epoch": 0.6969205834683955,
+      "grad_norm": 1.0004469156265259,
+      "learning_rate": 2.9368306290726984e-05,
+      "loss": 0.8014,
+      "step": 215
+    },
+    {
+      "epoch": 0.713128038897893,
+      "grad_norm": 1.0178419351577759,
+      "learning_rate": 2.932136422941424e-05,
+      "loss": 0.8118,
+      "step": 220
+    },
+    {
+      "epoch": 0.7293354943273906,
+      "grad_norm": 0.9624495506286621,
+      "learning_rate": 2.927278021870987e-05,
+      "loss": 0.8164,
+      "step": 225
+    },
+    {
+      "epoch": 0.7455429497568882,
+      "grad_norm": 0.9296135306358337,
+      "learning_rate": 2.92225598287876e-05,
+      "loss": 0.8124,
+      "step": 230
+    },
+    {
+      "epoch": 0.7617504051863857,
+      "grad_norm": 0.922747015953064,
+      "learning_rate": 2.9170708817432612e-05,
+      "loss": 0.8082,
+      "step": 235
+    },
+    {
+      "epoch": 0.7779578606158833,
+      "grad_norm": 1.0309699773788452,
+      "learning_rate": 2.9117233129381393e-05,
+      "loss": 0.7887,
+      "step": 240
+    },
+    {
+      "epoch": 0.7941653160453809,
+      "grad_norm": 0.9075000882148743,
+      "learning_rate": 2.9062138895640185e-05,
+      "loss": 0.7794,
+      "step": 245
+    },
+    {
+      "epoch": 0.8103727714748784,
+      "grad_norm": 1.001808524131775,
+      "learning_rate": 2.900543243278206e-05,
+      "loss": 0.7868,
+      "step": 250
+    },
+    {
+      "epoch": 0.826580226904376,
+      "grad_norm": 1.077176809310913,
+      "learning_rate": 2.8947120242222706e-05,
+      "loss": 0.7489,
+      "step": 255
+    },
+    {
+      "epoch": 0.8427876823338736,
+      "grad_norm": 1.210211992263794,
+      "learning_rate": 2.8887209009475064e-05,
+      "loss": 0.7702,
+      "step": 260
+    },
+    {
+      "epoch": 0.8589951377633711,
+      "grad_norm": 0.9452102780342102,
+      "learning_rate": 2.882570560338281e-05,
+      "loss": 0.7382,
+      "step": 265
+    },
+    {
+      "epoch": 0.8752025931928687,
+      "grad_norm": 1.0349738597869873,
+      "learning_rate": 2.8762617075332855e-05,
+      "loss": 0.74,
+      "step": 270
+    },
+    {
+      "epoch": 0.8914100486223663,
+      "grad_norm": 0.987579882144928,
+      "learning_rate": 2.8697950658446884e-05,
+      "loss": 0.7508,
+      "step": 275
+    },
+    {
+      "epoch": 0.9076175040518638,
+      "grad_norm": 0.9751459360122681,
+      "learning_rate": 2.8631713766752097e-05,
+      "loss": 0.7146,
+      "step": 280
+    },
+    {
+      "epoch": 0.9238249594813615,
+      "grad_norm": 1.062954306602478,
+      "learning_rate": 2.8563913994331172e-05,
+      "loss": 0.7068,
+      "step": 285
+    },
+    {
+      "epoch": 0.940032414910859,
+      "grad_norm": 1.1477030515670776,
+      "learning_rate": 2.8494559114451605e-05,
+      "loss": 0.6545,
+      "step": 290
+    },
+    {
+      "epoch": 0.9562398703403565,
+      "grad_norm": 1.0274162292480469,
+      "learning_rate": 2.84236570786745e-05,
+      "loss": 0.6998,
+      "step": 295
+    },
+    {
+      "epoch": 0.9724473257698542,
+      "grad_norm": 1.110557198524475,
+      "learning_rate": 2.8351216015942933e-05,
+      "loss": 0.6747,
+      "step": 300
+    },
+    {
+      "epoch": 0.9886547811993517,
+      "grad_norm": 1.1557697057724,
+      "learning_rate": 2.827724423164995e-05,
+      "loss": 0.6515,
+      "step": 305
+    },
+    {
+      "epoch": 1.0032414910858996,
+      "grad_norm": 1.0226082801818848,
+      "learning_rate": 2.820175020668635e-05,
+      "loss": 0.6719,
+      "step": 310
+    },
+    {
+      "epoch": 1.019448946515397,
+      "grad_norm": 1.129378318786621,
+      "learning_rate": 2.812474259646837e-05,
+      "loss": 0.6247,
+      "step": 315
+    },
+    {
+      "epoch": 1.0356564019448946,
+      "grad_norm": 1.1172395944595337,
+      "learning_rate": 2.804623022994531e-05,
+      "loss": 0.6146,
+      "step": 320
+    },
+    {
+      "epoch": 1.0518638573743921,
+      "grad_norm": 1.1870160102844238,
+      "learning_rate": 2.7966222108587307e-05,
+      "loss": 0.5694,
+      "step": 325
+    },
+    {
+      "epoch": 1.0680713128038897,
+      "grad_norm": 1.1029253005981445,
+      "learning_rate": 2.788472740535331e-05,
+      "loss": 0.5751,
+      "step": 330
+    },
+    {
+      "epoch": 1.0842787682333874,
+      "grad_norm": 1.449513554573059,
+      "learning_rate": 2.780175546363941e-05,
+      "loss": 0.5374,
+      "step": 335
+    },
+    {
+      "epoch": 1.100486223662885,
+      "grad_norm": 1.0343148708343506,
+      "learning_rate": 2.7717315796207576e-05,
+      "loss": 0.5688,
+      "step": 340
+    },
+    {
+      "epoch": 1.1166936790923825,
+      "grad_norm": 1.0800855159759521,
+      "learning_rate": 2.7631418084095064e-05,
+      "loss": 0.582,
+      "step": 345
+    },
+    {
+      "epoch": 1.13290113452188,
+      "grad_norm": 1.2004995346069336,
+      "learning_rate": 2.7544072175504457e-05,
+      "loss": 0.5486,
+      "step": 350
+    },
+    {
+      "epoch": 1.1491085899513775,
+      "grad_norm": 1.1442773342132568,
+      "learning_rate": 2.7455288084674565e-05,
+      "loss": 0.5438,
+      "step": 355
+    },
+    {
+      "epoch": 1.1653160453808753,
+      "grad_norm": 1.1458081007003784,
+      "learning_rate": 2.7365075990732285e-05,
+      "loss": 0.5537,
+      "step": 360
+    },
+    {
+      "epoch": 1.1815235008103728,
+      "grad_norm": 1.1584341526031494,
+      "learning_rate": 2.727344623652558e-05,
+      "loss": 0.5659,
+      "step": 365
+    },
+    {
+      "epoch": 1.1977309562398704,
+      "grad_norm": 1.389176845550537,
+      "learning_rate": 2.7180409327437648e-05,
+      "loss": 0.5922,
+      "step": 370
+    },
+    {
+      "epoch": 1.213938411669368,
+      "grad_norm": 1.3196877241134644,
+      "learning_rate": 2.708597593018248e-05,
+      "loss": 0.496,
+      "step": 375
+    },
+    {
+      "epoch": 1.2301458670988654,
+      "grad_norm": 1.1960793733596802,
+      "learning_rate": 2.6990156871581938e-05,
+      "loss": 0.5452,
+      "step": 380
+    },
+    {
+      "epoch": 1.246353322528363,
+      "grad_norm": 1.1414562463760376,
+      "learning_rate": 2.689296313732442e-05,
+      "loss": 0.5696,
+      "step": 385
+    },
+    {
+      "epoch": 1.2625607779578605,
+      "grad_norm": 1.173238754272461,
+      "learning_rate": 2.679440587070538e-05,
+      "loss": 0.4971,
+      "step": 390
+    },
+    {
+      "epoch": 1.2787682333873582,
+      "grad_norm": 1.245996117591858,
+      "learning_rate": 2.6694496371349723e-05,
+      "loss": 0.5369,
+      "step": 395
+    },
+    {
+      "epoch": 1.2949756888168558,
+      "grad_norm": 1.3267337083816528,
+      "learning_rate": 2.6593246093916307e-05,
+      "loss": 0.4883,
+      "step": 400
+    },
+    {
+      "epoch": 1.3111831442463533,
+      "grad_norm": 1.1124933958053589,
+      "learning_rate": 2.649066664678467e-05,
+      "loss": 0.4963,
+      "step": 405
+    },
+    {
+      "epoch": 1.3273905996758508,
+      "grad_norm": 1.1307417154312134,
+      "learning_rate": 2.638676979072412e-05,
+      "loss": 0.486,
+      "step": 410
+    },
+    {
+      "epoch": 1.3435980551053484,
+      "grad_norm": 1.1167956590652466,
+      "learning_rate": 2.6281567437545347e-05,
+      "loss": 0.5253,
+      "step": 415
+    },
+    {
+      "epoch": 1.359805510534846,
+      "grad_norm": 1.1781648397445679,
+      "learning_rate": 2.6175071648734752e-05,
+      "loss": 0.5155,
+      "step": 420
+    },
+    {
+      "epoch": 1.3760129659643436,
+      "grad_norm": 1.1858192682266235,
+      "learning_rate": 2.606729463407156e-05,
+      "loss": 0.5228,
+      "step": 425
+    },
+    {
+      "epoch": 1.3922204213938412,
+      "grad_norm": 1.1593049764633179,
+      "learning_rate": 2.5958248750228018e-05,
+      "loss": 0.5411,
+      "step": 430
+    },
+    {
+      "epoch": 1.4084278768233387,
+      "grad_norm": 1.190211296081543,
+      "learning_rate": 2.5847946499352637e-05,
+      "loss": 0.4975,
+      "step": 435
+    },
+    {
+      "epoch": 1.4246353322528362,
+      "grad_norm": 1.0861825942993164,
+      "learning_rate": 2.573640052763686e-05,
+      "loss": 0.517,
+      "step": 440
+    },
+    {
+      "epoch": 1.440842787682334,
+      "grad_norm": 1.1397205591201782,
+      "learning_rate": 2.5623623623865152e-05,
+      "loss": 0.4718,
+      "step": 445
+    },
+    {
+      "epoch": 1.4570502431118315,
+      "grad_norm": 1.305219054222107,
+      "learning_rate": 2.550962871794877e-05,
+      "loss": 0.4437,
+      "step": 450
+    },
+    {
+      "epoch": 1.473257698541329,
+      "grad_norm": 1.154456615447998,
+      "learning_rate": 2.5394428879443333e-05,
+      "loss": 0.4669,
+      "step": 455
+    },
+    {
+      "epoch": 1.4894651539708266,
+      "grad_norm": 1.1383739709854126,
+      "learning_rate": 2.5278037316050417e-05,
+      "loss": 0.4817,
+      "step": 460
+    },
+    {
+      "epoch": 1.505672609400324,
+      "grad_norm": 1.2615864276885986,
+      "learning_rate": 2.516046737210325e-05,
+      "loss": 0.4006,
+      "step": 465
+    },
+    {
+      "epoch": 1.5218800648298219,
+      "grad_norm": 1.257574200630188,
+      "learning_rate": 2.5041732527036817e-05,
+      "loss": 0.429,
+      "step": 470
+    },
+    {
+      "epoch": 1.5380875202593192,
+      "grad_norm": 1.2761658430099487,
+      "learning_rate": 2.4921846393842414e-05,
+      "loss": 0.4609,
+      "step": 475
+    },
+    {
+      "epoch": 1.554294975688817,
+      "grad_norm": 1.2191444635391235,
+      "learning_rate": 2.480082271750692e-05,
+      "loss": 0.4211,
+      "step": 480
+    },
+    {
+      "epoch": 1.5705024311183144,
+      "grad_norm": 1.2483482360839844,
+      "learning_rate": 2.4678675373436938e-05,
+      "loss": 0.4495,
+      "step": 485
+    },
+    {
+      "epoch": 1.586709886547812,
+      "grad_norm": 1.2474972009658813,
+      "learning_rate": 2.4555418365867965e-05,
+      "loss": 0.471,
+      "step": 490
+    },
+    {
+      "epoch": 1.6029173419773097,
+      "grad_norm": 1.5780339241027832,
+      "learning_rate": 2.443106582625879e-05,
+      "loss": 0.4464,
+      "step": 495
+    },
+    {
+      "epoch": 1.619124797406807,
+      "grad_norm": 1.1278367042541504,
+      "learning_rate": 2.430563201167136e-05,
+      "loss": 0.3857,
+      "step": 500
+    },
+    {
+      "epoch": 1.6353322528363048,
+      "grad_norm": 1.1629952192306519,
+      "learning_rate": 2.4179131303136146e-05,
+      "loss": 0.4811,
+      "step": 505
+    },
+    {
+      "epoch": 1.6515397082658023,
+      "grad_norm": 1.193259596824646,
+      "learning_rate": 2.4051578204003405e-05,
+      "loss": 0.3986,
+      "step": 510
+    },
+    {
+      "epoch": 1.6677471636952999,
+      "grad_norm": 1.2777222394943237,
+      "learning_rate": 2.3922987338280326e-05,
+      "loss": 0.3735,
+      "step": 515
+    },
+    {
+      "epoch": 1.6839546191247974,
+      "grad_norm": 1.2153775691986084,
+      "learning_rate": 2.3793373448954406e-05,
+      "loss": 0.4203,
+      "step": 520
+    },
+    {
+      "epoch": 1.700162074554295,
+      "grad_norm": 1.3885631561279297,
+      "learning_rate": 2.366275139630315e-05,
+      "loss": 0.397,
+      "step": 525
+    },
+    {
+      "epoch": 1.7163695299837927,
+      "grad_norm": 1.1484588384628296,
+      "learning_rate": 2.3531136156190335e-05,
+      "loss": 0.387,
+      "step": 530
+    },
+    {
+      "epoch": 1.73257698541329,
+      "grad_norm": 1.1026580333709717,
+      "learning_rate": 2.3398542818349042e-05,
+      "loss": 0.3726,
+      "step": 535
+    },
+    {
+      "epoch": 1.7487844408427877,
+      "grad_norm": 1.323022484779358,
+      "learning_rate": 2.326498658465158e-05,
+      "loss": 0.424,
+      "step": 540
+    },
+    {
+      "epoch": 1.7649918962722853,
+      "grad_norm": 1.2276582717895508,
+      "learning_rate": 2.3130482767366614e-05,
+      "loss": 0.4325,
+      "step": 545
+    },
+    {
+      "epoch": 1.7811993517017828,
+      "grad_norm": 1.324051022529602,
+      "learning_rate": 2.299504678740359e-05,
+      "loss": 0.3644,
+      "step": 550
+    },
+    {
+      "epoch": 1.7974068071312805,
+      "grad_norm": 1.2084301710128784,
+      "learning_rate": 2.2858694172544733e-05,
+      "loss": 0.3649,
+      "step": 555
+    },
+    {
+      "epoch": 1.8136142625607778,
+      "grad_norm": 1.2286015748977661,
+      "learning_rate": 2.2721440555664776e-05,
+      "loss": 0.3838,
+      "step": 560
+    },
+    {
+      "epoch": 1.8298217179902756,
+      "grad_norm": 1.2307486534118652,
+      "learning_rate": 2.2583301672938648e-05,
+      "loss": 0.3661,
+      "step": 565
+    },
+    {
+      "epoch": 1.8460291734197731,
+      "grad_norm": 1.1322344541549683,
+      "learning_rate": 2.2444293362037317e-05,
+      "loss": 0.3824,
+      "step": 570
+    },
+    {
+      "epoch": 1.8622366288492707,
+      "grad_norm": 1.1923706531524658,
+      "learning_rate": 2.2304431560311984e-05,
+      "loss": 0.3626,
+      "step": 575
+    },
+    {
+      "epoch": 1.8784440842787682,
+      "grad_norm": 1.1578867435455322,
+      "learning_rate": 2.216373230296689e-05,
+      "loss": 0.4028,
+      "step": 580
+    },
+    {
+      "epoch": 1.8946515397082657,
+      "grad_norm": 1.2379870414733887,
+      "learning_rate": 2.2022211721220834e-05,
+      "loss": 0.3252,
+      "step": 585
+    },
+    {
+      "epoch": 1.9108589951377635,
+      "grad_norm": 1.1819502115249634,
+      "learning_rate": 2.1879886040457755e-05,
+      "loss": 0.3459,
+      "step": 590
+    },
+    {
+      "epoch": 1.9270664505672608,
+      "grad_norm": 1.0207310914993286,
+      "learning_rate": 2.1736771578366472e-05,
+      "loss": 0.3855,
+      "step": 595
+    },
+    {
+      "epoch": 1.9432739059967585,
+      "grad_norm": 1.2070367336273193,
+      "learning_rate": 2.1592884743069855e-05,
+      "loss": 0.3465,
+      "step": 600
+    },
+    {
+      "epoch": 1.959481361426256,
+      "grad_norm": 1.172386646270752,
+      "learning_rate": 2.1448242031243624e-05,
+      "loss": 0.3583,
+      "step": 605
+    },
+    {
+      "epoch": 1.9756888168557536,
+      "grad_norm": 1.183351993560791,
+      "learning_rate": 2.1302860026225027e-05,
+      "loss": 0.3374,
+      "step": 610
+    },
+    {
+      "epoch": 1.9918962722852513,
+      "grad_norm": 1.4326958656311035,
+      "learning_rate": 2.1156755396111516e-05,
+      "loss": 0.3243,
+      "step": 615
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1545,
+  "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": 8.55238528743768e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

31/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea99dff72a8a1a1f6141bf70b09da0e3e2182ca1eb8c97e605c91acb58eb5aba
+size 7736

31/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)