Upload folder using huggingface_hub

12/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

12/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",
+    "q_proj",
+    "down_proj",
+    "up_proj",
+    "o_proj",
+    "gate_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

12/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:447e8b29d354cb5466f5c199d01804acf0a7b8a88ecbe7c483c1588cce2e31b0
+size 791751704

12/latest

@@ -0,0 +1 @@
+global_step740

12/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50f8a49f038344adda1890aa97c6492e5ff0f19c942d003ecc62de12a64d0bc8
+size 15920

12/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e60d5d94cc6f112707fce881530098e40391287ac1ef9830d94ed88b5f9dabc
+size 15920

12/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e56a9d7c087bf9928542d9760542548eba4f5f1b420cda496edd94a0d2e4dc1
+size 15920

12/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ead2b70311e002862e747e91e5875497642ee18e6c3edab438a88e4836b04cc
+size 15920

12/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e7a87c09f612d76d325366ce98ca010e73dcb4bba8f77630692688ab6a7b361
+size 15920

12/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f59ddac0af56d1f542778db63c9631bc0ac49c862cdf74d68febfd843e3b9cb5
+size 15920

12/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a849849ac9093b945bfa92eeb6dda3b4345d1908c80d24d87b6ab3de7babee78
+size 15920

12/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9401116121be32940670aa3196bd5f6fe1d696466f391d45f1ff5ff9a5fc38ee
+size 15920

12/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f574b2b20bd7d97e9890b11b949cbb96ff7330d34d8e046c92961cf239c2b50
+size 1064

12/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
+  }
+}

12/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
+}

12/trainer_state.json

@@ -0,0 +1,1070 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 740,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013531799729364006,
+      "grad_norm": 1.4412583112716675,
+      "learning_rate": 1.2903225806451614e-06,
+      "loss": 1.3064,
+      "step": 5
+    },
+    {
+      "epoch": 0.02706359945872801,
+      "grad_norm": 0.9369994401931763,
+      "learning_rate": 2.9032258064516128e-06,
+      "loss": 1.3176,
+      "step": 10
+    },
+    {
+      "epoch": 0.04059539918809202,
+      "grad_norm": 0.6337246298789978,
+      "learning_rate": 4.516129032258065e-06,
+      "loss": 1.3047,
+      "step": 15
+    },
+    {
+      "epoch": 0.05412719891745602,
+      "grad_norm": 0.7102669477462769,
+      "learning_rate": 6.129032258064517e-06,
+      "loss": 1.2226,
+      "step": 20
+    },
+    {
+      "epoch": 0.06765899864682003,
+      "grad_norm": 0.5792216062545776,
+      "learning_rate": 7.741935483870968e-06,
+      "loss": 1.2671,
+      "step": 25
+    },
+    {
+      "epoch": 0.08119079837618404,
+      "grad_norm": 0.5476118326187134,
+      "learning_rate": 9.35483870967742e-06,
+      "loss": 1.2585,
+      "step": 30
+    },
+    {
+      "epoch": 0.09472259810554803,
+      "grad_norm": 0.6799878478050232,
+      "learning_rate": 1.0967741935483872e-05,
+      "loss": 1.2151,
+      "step": 35
+    },
+    {
+      "epoch": 0.10825439783491204,
+      "grad_norm": 0.5542110800743103,
+      "learning_rate": 1.2580645161290324e-05,
+      "loss": 1.185,
+      "step": 40
+    },
+    {
+      "epoch": 0.12178619756427606,
+      "grad_norm": 0.4575681686401367,
+      "learning_rate": 1.4193548387096776e-05,
+      "loss": 1.1688,
+      "step": 45
+    },
+    {
+      "epoch": 0.13531799729364005,
+      "grad_norm": 0.5137224197387695,
+      "learning_rate": 1.5806451612903226e-05,
+      "loss": 1.1865,
+      "step": 50
+    },
+    {
+      "epoch": 0.14884979702300405,
+      "grad_norm": 0.4554082453250885,
+      "learning_rate": 1.741935483870968e-05,
+      "loss": 1.163,
+      "step": 55
+    },
+    {
+      "epoch": 0.16238159675236807,
+      "grad_norm": 0.5306389331817627,
+      "learning_rate": 1.903225806451613e-05,
+      "loss": 1.1513,
+      "step": 60
+    },
+    {
+      "epoch": 0.17591339648173207,
+      "grad_norm": 0.5879004597663879,
+      "learning_rate": 2.0645161290322582e-05,
+      "loss": 1.1948,
+      "step": 65
+    },
+    {
+      "epoch": 0.18944519621109607,
+      "grad_norm": 0.4350433647632599,
+      "learning_rate": 2.2258064516129034e-05,
+      "loss": 1.0976,
+      "step": 70
+    },
+    {
+      "epoch": 0.2029769959404601,
+      "grad_norm": 0.49727171659469604,
+      "learning_rate": 2.3870967741935483e-05,
+      "loss": 1.1021,
+      "step": 75
+    },
+    {
+      "epoch": 0.2165087956698241,
+      "grad_norm": 0.551313042640686,
+      "learning_rate": 2.548387096774194e-05,
+      "loss": 1.1962,
+      "step": 80
+    },
+    {
+      "epoch": 0.23004059539918809,
+      "grad_norm": 0.4979248046875,
+      "learning_rate": 2.7096774193548387e-05,
+      "loss": 1.1318,
+      "step": 85
+    },
+    {
+      "epoch": 0.2435723951285521,
+      "grad_norm": 0.5184155702590942,
+      "learning_rate": 2.870967741935484e-05,
+      "loss": 1.1268,
+      "step": 90
+    },
+    {
+      "epoch": 0.2571041948579161,
+      "grad_norm": 0.6341415643692017,
+      "learning_rate": 2.9999976021756284e-05,
+      "loss": 1.0454,
+      "step": 95
+    },
+    {
+      "epoch": 0.2706359945872801,
+      "grad_norm": 0.4915357530117035,
+      "learning_rate": 2.9999136791275564e-05,
+      "loss": 1.078,
+      "step": 100
+    },
+    {
+      "epoch": 0.28416779431664413,
+      "grad_norm": 0.541904091835022,
+      "learning_rate": 2.999709872526874e-05,
+      "loss": 1.0623,
+      "step": 105
+    },
+    {
+      "epoch": 0.2976995940460081,
+      "grad_norm": 0.5630597472190857,
+      "learning_rate": 2.999386198663225e-05,
+      "loss": 1.0517,
+      "step": 110
+    },
+    {
+      "epoch": 0.3112313937753721,
+      "grad_norm": 0.7628896832466125,
+      "learning_rate": 2.9989426834068792e-05,
+      "loss": 1.084,
+      "step": 115
+    },
+    {
+      "epoch": 0.32476319350473615,
+      "grad_norm": 0.6299301981925964,
+      "learning_rate": 2.9983793622066668e-05,
+      "loss": 1.0186,
+      "step": 120
+    },
+    {
+      "epoch": 0.3382949932341001,
+      "grad_norm": 0.6008714437484741,
+      "learning_rate": 2.9976962800871434e-05,
+      "loss": 1.0228,
+      "step": 125
+    },
+    {
+      "epoch": 0.35182679296346414,
+      "grad_norm": 0.6700873374938965,
+      "learning_rate": 2.9968934916449923e-05,
+      "loss": 0.9923,
+      "step": 130
+    },
+    {
+      "epoch": 0.36535859269282817,
+      "grad_norm": 0.5688751339912415,
+      "learning_rate": 2.9959710610446577e-05,
+      "loss": 0.998,
+      "step": 135
+    },
+    {
+      "epoch": 0.37889039242219213,
+      "grad_norm": 0.7533664703369141,
+      "learning_rate": 2.9949290620132225e-05,
+      "loss": 1.0353,
+      "step": 140
+    },
+    {
+      "epoch": 0.39242219215155616,
+      "grad_norm": 0.7014450430870056,
+      "learning_rate": 2.99376757783451e-05,
+      "loss": 0.9277,
+      "step": 145
+    },
+    {
+      "epoch": 0.4059539918809202,
+      "grad_norm": 0.6594902873039246,
+      "learning_rate": 2.992486701342427e-05,
+      "loss": 0.9636,
+      "step": 150
+    },
+    {
+      "epoch": 0.41948579161028415,
+      "grad_norm": 0.7603819966316223,
+      "learning_rate": 2.9910865349135498e-05,
+      "loss": 0.9665,
+      "step": 155
+    },
+    {
+      "epoch": 0.4330175913396482,
+      "grad_norm": 0.642242968082428,
+      "learning_rate": 2.989567190458935e-05,
+      "loss": 0.9616,
+      "step": 160
+    },
+    {
+      "epoch": 0.4465493910690122,
+      "grad_norm": 0.6365484595298767,
+      "learning_rate": 2.9879287894151786e-05,
+      "loss": 0.979,
+      "step": 165
+    },
+    {
+      "epoch": 0.46008119079837617,
+      "grad_norm": 0.6748781800270081,
+      "learning_rate": 2.9861714627347076e-05,
+      "loss": 0.9437,
+      "step": 170
+    },
+    {
+      "epoch": 0.4736129905277402,
+      "grad_norm": 0.6489437222480774,
+      "learning_rate": 2.984295350875316e-05,
+      "loss": 0.9036,
+      "step": 175
+    },
+    {
+      "epoch": 0.4871447902571042,
+      "grad_norm": 0.6802551746368408,
+      "learning_rate": 2.9823006037889358e-05,
+      "loss": 0.8769,
+      "step": 180
+    },
+    {
+      "epoch": 0.5006765899864682,
+      "grad_norm": 0.7979145050048828,
+      "learning_rate": 2.9801873809096543e-05,
+      "loss": 0.9136,
+      "step": 185
+    },
+    {
+      "epoch": 0.5142083897158322,
+      "grad_norm": 0.774574875831604,
+      "learning_rate": 2.9779558511409678e-05,
+      "loss": 0.8767,
+      "step": 190
+    },
+    {
+      "epoch": 0.5277401894451962,
+      "grad_norm": 0.724077045917511,
+      "learning_rate": 2.9756061928422857e-05,
+      "loss": 0.913,
+      "step": 195
+    },
+    {
+      "epoch": 0.5412719891745602,
+      "grad_norm": 0.7980031967163086,
+      "learning_rate": 2.973138593814671e-05,
+      "loss": 0.9224,
+      "step": 200
+    },
+    {
+      "epoch": 0.5548037889039242,
+      "grad_norm": 0.900132417678833,
+      "learning_rate": 2.9705532512858324e-05,
+      "loss": 0.8389,
+      "step": 205
+    },
+    {
+      "epoch": 0.5683355886332883,
+      "grad_norm": 0.8545295596122742,
+      "learning_rate": 2.9678503718943594e-05,
+      "loss": 0.8391,
+      "step": 210
+    },
+    {
+      "epoch": 0.5818673883626523,
+      "grad_norm": 0.78533935546875,
+      "learning_rate": 2.965030171673207e-05,
+      "loss": 0.8716,
+      "step": 215
+    },
+    {
+      "epoch": 0.5953991880920162,
+      "grad_norm": 0.9080139994621277,
+      "learning_rate": 2.962092876032427e-05,
+      "loss": 0.8158,
+      "step": 220
+    },
+    {
+      "epoch": 0.6089309878213802,
+      "grad_norm": 0.9167734384536743,
+      "learning_rate": 2.9590387197411547e-05,
+      "loss": 0.8416,
+      "step": 225
+    },
+    {
+      "epoch": 0.6224627875507442,
+      "grad_norm": 1.0270551443099976,
+      "learning_rate": 2.9558679469088423e-05,
+      "loss": 0.8628,
+      "step": 230
+    },
+    {
+      "epoch": 0.6359945872801083,
+      "grad_norm": 0.8939360976219177,
+      "learning_rate": 2.9525808109657485e-05,
+      "loss": 0.8487,
+      "step": 235
+    },
+    {
+      "epoch": 0.6495263870094723,
+      "grad_norm": 0.8669422268867493,
+      "learning_rate": 2.949177574642682e-05,
+      "loss": 0.8317,
+      "step": 240
+    },
+    {
+      "epoch": 0.6630581867388363,
+      "grad_norm": 0.7396143674850464,
+      "learning_rate": 2.9456585099500036e-05,
+      "loss": 0.784,
+      "step": 245
+    },
+    {
+      "epoch": 0.6765899864682002,
+      "grad_norm": 0.9191597700119019,
+      "learning_rate": 2.942023898155885e-05,
+      "loss": 0.8148,
+      "step": 250
+    },
+    {
+      "epoch": 0.6901217861975643,
+      "grad_norm": 1.05917227268219,
+      "learning_rate": 2.938274029763826e-05,
+      "loss": 0.7824,
+      "step": 255
+    },
+    {
+      "epoch": 0.7036535859269283,
+      "grad_norm": 0.8561118245124817,
+      "learning_rate": 2.934409204489438e-05,
+      "loss": 0.8054,
+      "step": 260
+    },
+    {
+      "epoch": 0.7171853856562923,
+      "grad_norm": 0.7686528563499451,
+      "learning_rate": 2.9304297312364865e-05,
+      "loss": 0.7765,
+      "step": 265
+    },
+    {
+      "epoch": 0.7307171853856563,
+      "grad_norm": 0.9166774749755859,
+      "learning_rate": 2.926335928072203e-05,
+      "loss": 0.7964,
+      "step": 270
+    },
+    {
+      "epoch": 0.7442489851150202,
+      "grad_norm": 0.9427902102470398,
+      "learning_rate": 2.922128122201862e-05,
+      "loss": 0.7521,
+      "step": 275
+    },
+    {
+      "epoch": 0.7577807848443843,
+      "grad_norm": 0.8347809314727783,
+      "learning_rate": 2.9178066499426284e-05,
+      "loss": 0.7492,
+      "step": 280
+    },
+    {
+      "epoch": 0.7713125845737483,
+      "grad_norm": 0.9658071994781494,
+      "learning_rate": 2.9133718566966773e-05,
+      "loss": 0.6946,
+      "step": 285
+    },
+    {
+      "epoch": 0.7848443843031123,
+      "grad_norm": 0.8596900105476379,
+      "learning_rate": 2.9088240969235864e-05,
+      "loss": 0.7395,
+      "step": 290
+    },
+    {
+      "epoch": 0.7983761840324763,
+      "grad_norm": 0.9865032434463501,
+      "learning_rate": 2.9041637341120054e-05,
+      "loss": 0.7497,
+      "step": 295
+    },
+    {
+      "epoch": 0.8119079837618404,
+      "grad_norm": 0.8877797722816467,
+      "learning_rate": 2.8993911407506037e-05,
+      "loss": 0.7162,
+      "step": 300
+    },
+    {
+      "epoch": 0.8254397834912043,
+      "grad_norm": 0.9064735174179077,
+      "learning_rate": 2.8945066982982984e-05,
+      "loss": 0.6864,
+      "step": 305
+    },
+    {
+      "epoch": 0.8389715832205683,
+      "grad_norm": 0.9386357665061951,
+      "learning_rate": 2.889510797153764e-05,
+      "loss": 0.6957,
+      "step": 310
+    },
+    {
+      "epoch": 0.8525033829499323,
+      "grad_norm": 1.0564672946929932,
+      "learning_rate": 2.8844038366242326e-05,
+      "loss": 0.698,
+      "step": 315
+    },
+    {
+      "epoch": 0.8660351826792964,
+      "grad_norm": 0.9778911471366882,
+      "learning_rate": 2.879186224893574e-05,
+      "loss": 0.6916,
+      "step": 320
+    },
+    {
+      "epoch": 0.8795669824086604,
+      "grad_norm": 0.9041974544525146,
+      "learning_rate": 2.8738583789896743e-05,
+      "loss": 0.6482,
+      "step": 325
+    },
+    {
+      "epoch": 0.8930987821380244,
+      "grad_norm": 1.026167392730713,
+      "learning_rate": 2.8684207247511025e-05,
+      "loss": 0.7138,
+      "step": 330
+    },
+    {
+      "epoch": 0.9066305818673883,
+      "grad_norm": 0.8488349318504333,
+      "learning_rate": 2.8628736967930747e-05,
+      "loss": 0.6957,
+      "step": 335
+    },
+    {
+      "epoch": 0.9201623815967523,
+      "grad_norm": 0.9476014375686646,
+      "learning_rate": 2.8572177384727167e-05,
+      "loss": 0.6485,
+      "step": 340
+    },
+    {
+      "epoch": 0.9336941813261164,
+      "grad_norm": 0.9190114140510559,
+      "learning_rate": 2.8514533018536286e-05,
+      "loss": 0.6747,
+      "step": 345
+    },
+    {
+      "epoch": 0.9472259810554804,
+      "grad_norm": 0.9507735371589661,
+      "learning_rate": 2.8455808476697513e-05,
+      "loss": 0.6732,
+      "step": 350
+    },
+    {
+      "epoch": 0.9607577807848444,
+      "grad_norm": 1.0128976106643677,
+      "learning_rate": 2.8396008452885426e-05,
+      "loss": 0.6633,
+      "step": 355
+    },
+    {
+      "epoch": 0.9742895805142084,
+      "grad_norm": 1.2091327905654907,
+      "learning_rate": 2.8335137726734608e-05,
+      "loss": 0.6888,
+      "step": 360
+    },
+    {
+      "epoch": 0.9878213802435724,
+      "grad_norm": 0.998440682888031,
+      "learning_rate": 2.827320116345764e-05,
+      "loss": 0.6605,
+      "step": 365
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.5365850925445557,
+      "learning_rate": 2.821020371345624e-05,
+      "loss": 0.6687,
+      "step": 370
+    },
+    {
+      "epoch": 1.013531799729364,
+      "grad_norm": 1.3685508966445923,
+      "learning_rate": 2.8146150411925568e-05,
+      "loss": 0.541,
+      "step": 375
+    },
+    {
+      "epoch": 1.027063599458728,
+      "grad_norm": 1.0007332563400269,
+      "learning_rate": 2.8081046378451807e-05,
+      "loss": 0.5494,
+      "step": 380
+    },
+    {
+      "epoch": 1.040595399188092,
+      "grad_norm": 1.3241759538650513,
+      "learning_rate": 2.801489681660296e-05,
+      "loss": 0.5614,
+      "step": 385
+    },
+    {
+      "epoch": 1.054127198917456,
+      "grad_norm": 1.0600066184997559,
+      "learning_rate": 2.7947707013512936e-05,
+      "loss": 0.5606,
+      "step": 390
+    },
+    {
+      "epoch": 1.0676589986468201,
+      "grad_norm": 0.9737664461135864,
+      "learning_rate": 2.7879482339458974e-05,
+      "loss": 0.5386,
+      "step": 395
+    },
+    {
+      "epoch": 1.0811907983761841,
+      "grad_norm": 0.9769577980041504,
+      "learning_rate": 2.7810228247432415e-05,
+      "loss": 0.5397,
+      "step": 400
+    },
+    {
+      "epoch": 1.094722598105548,
+      "grad_norm": 0.8770543336868286,
+      "learning_rate": 2.7739950272702856e-05,
+      "loss": 0.5319,
+      "step": 405
+    },
+    {
+      "epoch": 1.108254397834912,
+      "grad_norm": 0.9559663534164429,
+      "learning_rate": 2.7668654032375733e-05,
+      "loss": 0.5952,
+      "step": 410
+    },
+    {
+      "epoch": 1.121786197564276,
+      "grad_norm": 0.9816983938217163,
+      "learning_rate": 2.7596345224943357e-05,
+      "loss": 0.5308,
+      "step": 415
+    },
+    {
+      "epoch": 1.13531799729364,
+      "grad_norm": 1.0468382835388184,
+      "learning_rate": 2.7523029629829478e-05,
+      "loss": 0.5199,
+      "step": 420
+    },
+    {
+      "epoch": 1.148849797023004,
+      "grad_norm": 1.1081461906433105,
+      "learning_rate": 2.744871310692731e-05,
+      "loss": 0.5094,
+      "step": 425
+    },
+    {
+      "epoch": 1.162381596752368,
+      "grad_norm": 1.0275734663009644,
+      "learning_rate": 2.73734015961312e-05,
+      "loss": 0.5744,
+      "step": 430
+    },
+    {
+      "epoch": 1.175913396481732,
+      "grad_norm": 0.9587875604629517,
+      "learning_rate": 2.7297101116861862e-05,
+      "loss": 0.5095,
+      "step": 435
+    },
+    {
+      "epoch": 1.1894451962110961,
+      "grad_norm": 1.1627024412155151,
+      "learning_rate": 2.721981776758526e-05,
+      "loss": 0.4957,
+      "step": 440
+    },
+    {
+      "epoch": 1.2029769959404601,
+      "grad_norm": 0.9614389538764954,
+      "learning_rate": 2.714155772532518e-05,
+      "loss": 0.506,
+      "step": 445
+    },
+    {
+      "epoch": 1.2165087956698242,
+      "grad_norm": 1.0214952230453491,
+      "learning_rate": 2.7062327245169506e-05,
+      "loss": 0.5174,
+      "step": 450
+    },
+    {
+      "epoch": 1.230040595399188,
+      "grad_norm": 1.1152983903884888,
+      "learning_rate": 2.6982132659770298e-05,
+      "loss": 0.5119,
+      "step": 455
+    },
+    {
+      "epoch": 1.243572395128552,
+      "grad_norm": 0.9631413817405701,
+      "learning_rate": 2.6900980378837614e-05,
+      "loss": 0.5094,
+      "step": 460
+    },
+    {
+      "epoch": 1.257104194857916,
+      "grad_norm": 1.333390712738037,
+      "learning_rate": 2.6818876888627204e-05,
+      "loss": 0.5279,
+      "step": 465
+    },
+    {
+      "epoch": 1.27063599458728,
+      "grad_norm": 1.1421831846237183,
+      "learning_rate": 2.6735828751422117e-05,
+      "loss": 0.5038,
+      "step": 470
+    },
+    {
+      "epoch": 1.284167794316644,
+      "grad_norm": 1.093878149986267,
+      "learning_rate": 2.6651842605008142e-05,
+      "loss": 0.4847,
+      "step": 475
+    },
+    {
+      "epoch": 1.297699594046008,
+      "grad_norm": 1.0584564208984375,
+      "learning_rate": 2.6566925162143322e-05,
+      "loss": 0.463,
+      "step": 480
+    },
+    {
+      "epoch": 1.3112313937753721,
+      "grad_norm": 1.1284931898117065,
+      "learning_rate": 2.6481083210021396e-05,
+      "loss": 0.5409,
+      "step": 485
+    },
+    {
+      "epoch": 1.3247631935047361,
+      "grad_norm": 1.6687848567962646,
+      "learning_rate": 2.6394323609729317e-05,
+      "loss": 0.5144,
+      "step": 490
+    },
+    {
+      "epoch": 1.3382949932341002,
+      "grad_norm": 1.1010960340499878,
+      "learning_rate": 2.6306653295698885e-05,
+      "loss": 0.526,
+      "step": 495
+    },
+    {
+      "epoch": 1.3518267929634642,
+      "grad_norm": 1.0190247297286987,
+      "learning_rate": 2.6218079275152485e-05,
+      "loss": 0.4194,
+      "step": 500
+    },
+    {
+      "epoch": 1.3653585926928282,
+      "grad_norm": 1.2479708194732666,
+      "learning_rate": 2.6128608627543012e-05,
+      "loss": 0.4678,
+      "step": 505
+    },
+    {
+      "epoch": 1.3788903924221922,
+      "grad_norm": 1.0126714706420898,
+      "learning_rate": 2.6038248503988058e-05,
+      "loss": 0.4241,
+      "step": 510
+    },
+    {
+      "epoch": 1.3924221921515563,
+      "grad_norm": 1.1457223892211914,
+      "learning_rate": 2.5947006126698325e-05,
+      "loss": 0.4553,
+      "step": 515
+    },
+    {
+      "epoch": 1.4059539918809203,
+      "grad_norm": 0.9630009531974792,
+      "learning_rate": 2.5854888788400384e-05,
+      "loss": 0.4664,
+      "step": 520
+    },
+    {
+      "epoch": 1.419485791610284,
+      "grad_norm": 1.067337989807129,
+      "learning_rate": 2.5761903851753783e-05,
+      "loss": 0.4379,
+      "step": 525
+    },
+    {
+      "epoch": 1.4330175913396481,
+      "grad_norm": 1.0295122861862183,
+      "learning_rate": 2.5668058748762574e-05,
+      "loss": 0.4428,
+      "step": 530
+    },
+    {
+      "epoch": 1.4465493910690121,
+      "grad_norm": 0.9696286916732788,
+      "learning_rate": 2.5573360980181297e-05,
+      "loss": 0.4245,
+      "step": 535
+    },
+    {
+      "epoch": 1.4600811907983762,
+      "grad_norm": 1.2453484535217285,
+      "learning_rate": 2.5477818114915477e-05,
+      "loss": 0.462,
+      "step": 540
+    },
+    {
+      "epoch": 1.4736129905277402,
+      "grad_norm": 1.038318395614624,
+      "learning_rate": 2.5381437789416643e-05,
+      "loss": 0.4367,
+      "step": 545
+    },
+    {
+      "epoch": 1.4871447902571042,
+      "grad_norm": 1.1128343343734741,
+      "learning_rate": 2.5284227707071986e-05,
+      "loss": 0.4184,
+      "step": 550
+    },
+    {
+      "epoch": 1.5006765899864682,
+      "grad_norm": 1.2020708322525024,
+      "learning_rate": 2.518619563758864e-05,
+      "loss": 0.4334,
+      "step": 555
+    },
+    {
+      "epoch": 1.514208389715832,
+      "grad_norm": 1.0298750400543213,
+      "learning_rate": 2.5087349416372696e-05,
+      "loss": 0.4125,
+      "step": 560
+    },
+    {
+      "epoch": 1.527740189445196,
+      "grad_norm": 1.0778883695602417,
+      "learning_rate": 2.49876969439029e-05,
+      "loss": 0.4133,
+      "step": 565
+    },
+    {
+      "epoch": 1.54127198917456,
+      "grad_norm": 1.1158068180084229,
+      "learning_rate": 2.4887246185099237e-05,
+      "loss": 0.4366,
+      "step": 570
+    },
+    {
+      "epoch": 1.5548037889039241,
+      "grad_norm": 1.1676713228225708,
+      "learning_rate": 2.4786005168686286e-05,
+      "loss": 0.436,
+      "step": 575
+    },
+    {
+      "epoch": 1.5683355886332881,
+      "grad_norm": 1.2565547227859497,
+      "learning_rate": 2.4683981986551526e-05,
+      "loss": 0.4557,
+      "step": 580
+    },
+    {
+      "epoch": 1.5818673883626522,
+      "grad_norm": 1.133944034576416,
+      "learning_rate": 2.458118479309857e-05,
+      "loss": 0.4026,
+      "step": 585
+    },
+    {
+      "epoch": 1.5953991880920162,
+      "grad_norm": 1.0183407068252563,
+      "learning_rate": 2.4477621804595402e-05,
+      "loss": 0.4533,
+      "step": 590
+    },
+    {
+      "epoch": 1.6089309878213802,
+      "grad_norm": 1.26309335231781,
+      "learning_rate": 2.4373301298517696e-05,
+      "loss": 0.4314,
+      "step": 595
+    },
+    {
+      "epoch": 1.6224627875507442,
+      "grad_norm": 1.1744800806045532,
+      "learning_rate": 2.42682316128872e-05,
+      "loss": 0.3831,
+      "step": 600
+    },
+    {
+      "epoch": 1.6359945872801083,
+      "grad_norm": 1.1486276388168335,
+      "learning_rate": 2.4162421145605308e-05,
+      "loss": 0.4609,
+      "step": 605
+    },
+    {
+      "epoch": 1.6495263870094723,
+      "grad_norm": 1.1308343410491943,
+      "learning_rate": 2.4055878353781858e-05,
+      "loss": 0.3715,
+      "step": 610
+    },
+    {
+      "epoch": 1.6630581867388363,
+      "grad_norm": 1.048828125,
+      "learning_rate": 2.3948611753059155e-05,
+      "loss": 0.3978,
+      "step": 615
+    },
+    {
+      "epoch": 1.6765899864682003,
+      "grad_norm": 1.2727230787277222,
+      "learning_rate": 2.3840629916931362e-05,
+      "loss": 0.3986,
+      "step": 620
+    },
+    {
+      "epoch": 1.6901217861975644,
+      "grad_norm": 1.1679140329360962,
+      "learning_rate": 2.3731941476059243e-05,
+      "loss": 0.3896,
+      "step": 625
+    },
+    {
+      "epoch": 1.7036535859269284,
+      "grad_norm": 1.1558784246444702,
+      "learning_rate": 2.362255511758033e-05,
+      "loss": 0.3888,
+      "step": 630
+    },
+    {
+      "epoch": 1.7171853856562924,
+      "grad_norm": 1.337999939918518,
+      "learning_rate": 2.351247958441459e-05,
+      "loss": 0.3811,
+      "step": 635
+    },
+    {
+      "epoch": 1.7307171853856564,
+      "grad_norm": 1.1977120637893677,
+      "learning_rate": 2.340172367456564e-05,
+      "loss": 0.3987,
+      "step": 640
+    },
+    {
+      "epoch": 1.7442489851150202,
+      "grad_norm": 1.2168259620666504,
+      "learning_rate": 2.3290296240417544e-05,
+      "loss": 0.3497,
+      "step": 645
+    },
+    {
+      "epoch": 1.7577807848443843,
+      "grad_norm": 1.1579980850219727,
+      "learning_rate": 2.3178206188027265e-05,
+      "loss": 0.3342,
+      "step": 650
+    },
+    {
+      "epoch": 1.7713125845737483,
+      "grad_norm": 0.9696447253227234,
+      "learning_rate": 2.3065462476412825e-05,
+      "loss": 0.3763,
+      "step": 655
+    },
+    {
+      "epoch": 1.7848443843031123,
+      "grad_norm": 1.072059154510498,
+      "learning_rate": 2.295207411683725e-05,
+      "loss": 0.3668,
+      "step": 660
+    },
+    {
+      "epoch": 1.7983761840324763,
+      "grad_norm": 1.0907052755355835,
+      "learning_rate": 2.283805017208834e-05,
+      "loss": 0.369,
+      "step": 665
+    },
+    {
+      "epoch": 1.8119079837618404,
+      "grad_norm": 1.1626112461090088,
+      "learning_rate": 2.2723399755754262e-05,
+      "loss": 0.349,
+      "step": 670
+    },
+    {
+      "epoch": 1.8254397834912042,
+      "grad_norm": 1.0268486738204956,
+      "learning_rate": 2.2608132031495184e-05,
+      "loss": 0.3214,
+      "step": 675
+    },
+    {
+      "epoch": 1.8389715832205682,
+      "grad_norm": 1.1764811277389526,
+      "learning_rate": 2.2492256212310805e-05,
+      "loss": 0.3133,
+      "step": 680
+    },
+    {
+      "epoch": 1.8525033829499322,
+      "grad_norm": 1.0286613702774048,
+      "learning_rate": 2.2375781559804012e-05,
+      "loss": 0.3362,
+      "step": 685
+    },
+    {
+      "epoch": 1.8660351826792962,
+      "grad_norm": 1.2203805446624756,
+      "learning_rate": 2.2258717383440632e-05,
+      "loss": 0.3344,
+      "step": 690
+    },
+    {
+      "epoch": 1.8795669824086603,
+      "grad_norm": 1.0343270301818848,
+      "learning_rate": 2.2141073039805344e-05,
+      "loss": 0.3352,
+      "step": 695
+    },
+    {
+      "epoch": 1.8930987821380243,
+      "grad_norm": 1.082728385925293,
+      "learning_rate": 2.202285793185383e-05,
+      "loss": 0.3419,
+      "step": 700
+    },
+    {
+      "epoch": 1.9066305818673883,
+      "grad_norm": 1.1293714046478271,
+      "learning_rate": 2.1904081508161236e-05,
+      "loss": 0.3589,
+      "step": 705
+    },
+    {
+      "epoch": 1.9201623815967523,
+      "grad_norm": 1.090536117553711,
+      "learning_rate": 2.1784753262166984e-05,
+      "loss": 0.3487,
+      "step": 710
+    },
+    {
+      "epoch": 1.9336941813261164,
+      "grad_norm": 1.2215139865875244,
+      "learning_rate": 2.166488273141597e-05,
+      "loss": 0.3917,
+      "step": 715
+    },
+    {
+      "epoch": 1.9472259810554804,
+      "grad_norm": 1.2397185564041138,
+      "learning_rate": 2.1544479496796258e-05,
+      "loss": 0.3715,
+      "step": 720
+    },
+    {
+      "epoch": 1.9607577807848444,
+      "grad_norm": 1.0956600904464722,
+      "learning_rate": 2.1423553181773336e-05,
+      "loss": 0.3448,
+      "step": 725
+    },
+    {
+      "epoch": 1.9742895805142084,
+      "grad_norm": 1.1384000778198242,
+      "learning_rate": 2.130211345162091e-05,
+      "loss": 0.3248,
+      "step": 730
+    },
+    {
+      "epoch": 1.9878213802435725,
+      "grad_norm": 1.2017817497253418,
+      "learning_rate": 2.1180170012648406e-05,
+      "loss": 0.3016,
+      "step": 735
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.444764256477356,
+      "learning_rate": 2.105773261142516e-05,
+      "loss": 0.3309,
+      "step": 740
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1850,
+  "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": 1.0801623899381432e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

12/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6ed3482f3cec4167aae1918c4bbe78f0aacfff0b2e970894f0dfc6b17bfa72c
+size 7736

12/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)