Upload folder using huggingface_hub

21/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

21/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": [
+    "o_proj",
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

21/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0f3891a03264a560e94721be4ebf1e97cd8e87da12bad407df66ad94c7cbf4b
+size 791751704

21/latest

@@ -0,0 +1 @@
+global_step730

21/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d34c2377e3008d9aa4bdd95fd52480be28487832db858031545440f725486801
+size 15920

21/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cb0bed2f91f25f1c593b586a000e1dda7188ba90b95824943e96b15476a96f7
+size 15920

21/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c93be5b60daaa4b5240b24e0db1efb7890971a5cf476c32898886a450f27e8e
+size 15920

21/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d04d432fa7809d535a87dac0407f4e166736f21f12a89b5bab4936628d6d509
+size 15920

21/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff3ec71318bdce48138f4cb4184175c8694f6fcfa0bca69a55b69786db97b2ba
+size 15920

21/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5f740367f752432e1350fa84de7253f790786939035fa06a2e18617a295ade5
+size 15920

21/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc1d8157be76f37972d096c0735a87bb9922d1879b9984124685ba47336c2fdf
+size 15920

21/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9385eb60c80b254fa66b1425bfcf32a0341358cde5b9ba0e443122afa494eff2
+size 15920

21/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb313940fafe3ac8f9a2ea6ed3e874ba63f38499d21569bd8f98d9cec5669379
+size 1064

21/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
+  }
+}

21/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
+}

21/trainer_state.json

@@ -0,0 +1,1056 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 730,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013717421124828532,
+      "grad_norm": 1.2567085027694702,
+      "learning_rate": 1.3043478260869566e-06,
+      "loss": 1.2424,
+      "step": 5
+    },
+    {
+      "epoch": 0.027434842249657063,
+      "grad_norm": 1.181019902229309,
+      "learning_rate": 2.9347826086956523e-06,
+      "loss": 1.2733,
+      "step": 10
+    },
+    {
+      "epoch": 0.0411522633744856,
+      "grad_norm": 0.6036953926086426,
+      "learning_rate": 4.565217391304348e-06,
+      "loss": 1.2413,
+      "step": 15
+    },
+    {
+      "epoch": 0.05486968449931413,
+      "grad_norm": 0.8439048528671265,
+      "learning_rate": 6.195652173913044e-06,
+      "loss": 1.2214,
+      "step": 20
+    },
+    {
+      "epoch": 0.06858710562414266,
+      "grad_norm": 0.5841037631034851,
+      "learning_rate": 7.826086956521738e-06,
+      "loss": 1.2187,
+      "step": 25
+    },
+    {
+      "epoch": 0.0823045267489712,
+      "grad_norm": 0.6085624694824219,
+      "learning_rate": 9.456521739130436e-06,
+      "loss": 1.1609,
+      "step": 30
+    },
+    {
+      "epoch": 0.09602194787379972,
+      "grad_norm": 0.6230579018592834,
+      "learning_rate": 1.108695652173913e-05,
+      "loss": 1.1653,
+      "step": 35
+    },
+    {
+      "epoch": 0.10973936899862825,
+      "grad_norm": 0.5125408172607422,
+      "learning_rate": 1.2717391304347827e-05,
+      "loss": 1.1442,
+      "step": 40
+    },
+    {
+      "epoch": 0.12345679012345678,
+      "grad_norm": 0.4418923258781433,
+      "learning_rate": 1.4347826086956522e-05,
+      "loss": 1.135,
+      "step": 45
+    },
+    {
+      "epoch": 0.13717421124828533,
+      "grad_norm": 0.39918941259384155,
+      "learning_rate": 1.597826086956522e-05,
+      "loss": 1.15,
+      "step": 50
+    },
+    {
+      "epoch": 0.15089163237311384,
+      "grad_norm": 0.4432643949985504,
+      "learning_rate": 1.7608695652173915e-05,
+      "loss": 1.1159,
+      "step": 55
+    },
+    {
+      "epoch": 0.1646090534979424,
+      "grad_norm": 0.7146240472793579,
+      "learning_rate": 1.9239130434782607e-05,
+      "loss": 1.1433,
+      "step": 60
+    },
+    {
+      "epoch": 0.17832647462277093,
+      "grad_norm": 0.6695220470428467,
+      "learning_rate": 2.0869565217391306e-05,
+      "loss": 1.1032,
+      "step": 65
+    },
+    {
+      "epoch": 0.19204389574759945,
+      "grad_norm": 0.489704966545105,
+      "learning_rate": 2.25e-05,
+      "loss": 1.0569,
+      "step": 70
+    },
+    {
+      "epoch": 0.205761316872428,
+      "grad_norm": 0.4658520817756653,
+      "learning_rate": 2.4130434782608697e-05,
+      "loss": 1.0715,
+      "step": 75
+    },
+    {
+      "epoch": 0.2194787379972565,
+      "grad_norm": 0.5942860245704651,
+      "learning_rate": 2.5760869565217392e-05,
+      "loss": 1.0534,
+      "step": 80
+    },
+    {
+      "epoch": 0.23319615912208505,
+      "grad_norm": 0.48524120450019836,
+      "learning_rate": 2.7391304347826085e-05,
+      "loss": 1.1362,
+      "step": 85
+    },
+    {
+      "epoch": 0.24691358024691357,
+      "grad_norm": 0.41874566674232483,
+      "learning_rate": 2.9021739130434783e-05,
+      "loss": 1.0455,
+      "step": 90
+    },
+    {
+      "epoch": 0.2606310013717421,
+      "grad_norm": 0.49967625737190247,
+      "learning_rate": 2.999990141214925e-05,
+      "loss": 1.0053,
+      "step": 95
+    },
+    {
+      "epoch": 0.27434842249657065,
+      "grad_norm": 0.529108464717865,
+      "learning_rate": 2.999879231371134e-05,
+      "loss": 1.0228,
+      "step": 100
+    },
+    {
+      "epoch": 0.2880658436213992,
+      "grad_norm": 0.5769373774528503,
+      "learning_rate": 2.9996450973444988e-05,
+      "loss": 1.0495,
+      "step": 105
+    },
+    {
+      "epoch": 0.3017832647462277,
+      "grad_norm": 0.6768115162849426,
+      "learning_rate": 2.999287758370551e-05,
+      "loss": 0.9451,
+      "step": 110
+    },
+    {
+      "epoch": 0.31550068587105623,
+      "grad_norm": 0.6151679754257202,
+      "learning_rate": 2.998807243806856e-05,
+      "loss": 1.0238,
+      "step": 115
+    },
+    {
+      "epoch": 0.3292181069958848,
+      "grad_norm": 0.6618012189865112,
+      "learning_rate": 2.998203593130602e-05,
+      "loss": 1.0144,
+      "step": 120
+    },
+    {
+      "epoch": 0.3429355281207133,
+      "grad_norm": 0.5754362344741821,
+      "learning_rate": 2.9974768559353564e-05,
+      "loss": 0.9812,
+      "step": 125
+    },
+    {
+      "epoch": 0.35665294924554186,
+      "grad_norm": 0.614632785320282,
+      "learning_rate": 2.99662709192699e-05,
+      "loss": 0.9152,
+      "step": 130
+    },
+    {
+      "epoch": 0.37037037037037035,
+      "grad_norm": 0.5445948839187622,
+      "learning_rate": 2.995654370918775e-05,
+      "loss": 1.0159,
+      "step": 135
+    },
+    {
+      "epoch": 0.3840877914951989,
+      "grad_norm": 0.592310905456543,
+      "learning_rate": 2.9945587728256456e-05,
+      "loss": 0.9158,
+      "step": 140
+    },
+    {
+      "epoch": 0.39780521262002744,
+      "grad_norm": 0.6413975954055786,
+      "learning_rate": 2.9933403876576364e-05,
+      "loss": 0.938,
+      "step": 145
+    },
+    {
+      "epoch": 0.411522633744856,
+      "grad_norm": 0.5976924300193787,
+      "learning_rate": 2.9919993155124834e-05,
+      "loss": 0.8984,
+      "step": 150
+    },
+    {
+      "epoch": 0.4252400548696845,
+      "grad_norm": 0.6372507810592651,
+      "learning_rate": 2.990535666567403e-05,
+      "loss": 0.8719,
+      "step": 155
+    },
+    {
+      "epoch": 0.438957475994513,
+      "grad_norm": 0.5930208563804626,
+      "learning_rate": 2.9889495610700416e-05,
+      "loss": 0.9289,
+      "step": 160
+    },
+    {
+      "epoch": 0.45267489711934156,
+      "grad_norm": 0.6256594061851501,
+      "learning_rate": 2.9872411293285916e-05,
+      "loss": 0.852,
+      "step": 165
+    },
+    {
+      "epoch": 0.4663923182441701,
+      "grad_norm": 0.6492271423339844,
+      "learning_rate": 2.985410511701092e-05,
+      "loss": 0.8581,
+      "step": 170
+    },
+    {
+      "epoch": 0.48010973936899864,
+      "grad_norm": 0.7370251417160034,
+      "learning_rate": 2.9834578585838907e-05,
+      "loss": 0.9052,
+      "step": 175
+    },
+    {
+      "epoch": 0.49382716049382713,
+      "grad_norm": 0.7324991226196289,
+      "learning_rate": 2.9813833303992948e-05,
+      "loss": 0.8242,
+      "step": 180
+    },
+    {
+      "epoch": 0.5075445816186557,
+      "grad_norm": 0.7410762906074524,
+      "learning_rate": 2.979187097582386e-05,
+      "loss": 0.8182,
+      "step": 185
+    },
+    {
+      "epoch": 0.5212620027434842,
+      "grad_norm": 0.6918640732765198,
+      "learning_rate": 2.976869340567021e-05,
+      "loss": 0.838,
+      "step": 190
+    },
+    {
+      "epoch": 0.5349794238683128,
+      "grad_norm": 2.690075159072876,
+      "learning_rate": 2.9744302497710076e-05,
+      "loss": 0.8393,
+      "step": 195
+    },
+    {
+      "epoch": 0.5486968449931413,
+      "grad_norm": 0.7638638019561768,
+      "learning_rate": 2.9718700255804588e-05,
+      "loss": 0.8171,
+      "step": 200
+    },
+    {
+      "epoch": 0.5624142661179699,
+      "grad_norm": 0.8737165331840515,
+      "learning_rate": 2.969188878333332e-05,
+      "loss": 0.8264,
+      "step": 205
+    },
+    {
+      "epoch": 0.5761316872427984,
+      "grad_norm": 0.7183387875556946,
+      "learning_rate": 2.9663870283021477e-05,
+      "loss": 0.8421,
+      "step": 210
+    },
+    {
+      "epoch": 0.5898491083676269,
+      "grad_norm": 0.8239722847938538,
+      "learning_rate": 2.9634647056758927e-05,
+      "loss": 0.7993,
+      "step": 215
+    },
+    {
+      "epoch": 0.6035665294924554,
+      "grad_norm": 0.8498063683509827,
+      "learning_rate": 2.960422150541109e-05,
+      "loss": 0.8239,
+      "step": 220
+    },
+    {
+      "epoch": 0.6172839506172839,
+      "grad_norm": 1.0519356727600098,
+      "learning_rate": 2.9572596128621683e-05,
+      "loss": 0.7706,
+      "step": 225
+    },
+    {
+      "epoch": 0.6310013717421125,
+      "grad_norm": 0.844680666923523,
+      "learning_rate": 2.9539773524607373e-05,
+      "loss": 0.7471,
+      "step": 230
+    },
+    {
+      "epoch": 0.644718792866941,
+      "grad_norm": 0.7600374221801758,
+      "learning_rate": 2.95057563899443e-05,
+      "loss": 0.7894,
+      "step": 235
+    },
+    {
+      "epoch": 0.6584362139917695,
+      "grad_norm": 0.8541322946548462,
+      "learning_rate": 2.947054751934656e-05,
+      "loss": 0.7903,
+      "step": 240
+    },
+    {
+      "epoch": 0.6721536351165981,
+      "grad_norm": 0.7854955196380615,
+      "learning_rate": 2.9434149805436586e-05,
+      "loss": 0.7754,
+      "step": 245
+    },
+    {
+      "epoch": 0.6858710562414266,
+      "grad_norm": 0.7242818474769592,
+      "learning_rate": 2.9396566238507496e-05,
+      "loss": 0.7455,
+      "step": 250
+    },
+    {
+      "epoch": 0.6995884773662552,
+      "grad_norm": 0.9281851053237915,
+      "learning_rate": 2.935779990627744e-05,
+      "loss": 0.7562,
+      "step": 255
+    },
+    {
+      "epoch": 0.7133058984910837,
+      "grad_norm": 0.8336049318313599,
+      "learning_rate": 2.931785399363592e-05,
+      "loss": 0.7557,
+      "step": 260
+    },
+    {
+      "epoch": 0.7270233196159122,
+      "grad_norm": 0.967589795589447,
+      "learning_rate": 2.9276731782382123e-05,
+      "loss": 0.7062,
+      "step": 265
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 0.8542262315750122,
+      "learning_rate": 2.9234436650955297e-05,
+      "loss": 0.7184,
+      "step": 270
+    },
+    {
+      "epoch": 0.7544581618655692,
+      "grad_norm": 0.9164705276489258,
+      "learning_rate": 2.9190972074157232e-05,
+      "loss": 0.6814,
+      "step": 275
+    },
+    {
+      "epoch": 0.7681755829903978,
+      "grad_norm": 0.8561894297599792,
+      "learning_rate": 2.9146341622866716e-05,
+      "loss": 0.6944,
+      "step": 280
+    },
+    {
+      "epoch": 0.7818930041152263,
+      "grad_norm": 0.9264963865280151,
+      "learning_rate": 2.910054896374623e-05,
+      "loss": 0.6571,
+      "step": 285
+    },
+    {
+      "epoch": 0.7956104252400549,
+      "grad_norm": 0.8652317523956299,
+      "learning_rate": 2.9053597858940666e-05,
+      "loss": 0.7355,
+      "step": 290
+    },
+    {
+      "epoch": 0.8093278463648834,
+      "grad_norm": 0.9738150835037231,
+      "learning_rate": 2.9005492165768278e-05,
+      "loss": 0.6453,
+      "step": 295
+    },
+    {
+      "epoch": 0.823045267489712,
+      "grad_norm": 1.012303352355957,
+      "learning_rate": 2.895623583640375e-05,
+      "loss": 0.6413,
+      "step": 300
+    },
+    {
+      "epoch": 0.8367626886145405,
+      "grad_norm": 0.8940839767456055,
+      "learning_rate": 2.890583291755351e-05,
+      "loss": 0.7145,
+      "step": 305
+    },
+    {
+      "epoch": 0.850480109739369,
+      "grad_norm": 0.8137922883033752,
+      "learning_rate": 2.8854287550123278e-05,
+      "loss": 0.6835,
+      "step": 310
+    },
+    {
+      "epoch": 0.8641975308641975,
+      "grad_norm": 0.8530572056770325,
+      "learning_rate": 2.880160396887787e-05,
+      "loss": 0.6702,
+      "step": 315
+    },
+    {
+      "epoch": 0.877914951989026,
+      "grad_norm": 0.9513605237007141,
+      "learning_rate": 2.8747786502093258e-05,
+      "loss": 0.7024,
+      "step": 320
+    },
+    {
+      "epoch": 0.8916323731138546,
+      "grad_norm": 0.899515688419342,
+      "learning_rate": 2.8692839571201e-05,
+      "loss": 0.6845,
+      "step": 325
+    },
+    {
+      "epoch": 0.9053497942386831,
+      "grad_norm": 0.9002428650856018,
+      "learning_rate": 2.863676769042498e-05,
+      "loss": 0.6447,
+      "step": 330
+    },
+    {
+      "epoch": 0.9190672153635117,
+      "grad_norm": 0.8433024883270264,
+      "learning_rate": 2.8579575466410566e-05,
+      "loss": 0.6325,
+      "step": 335
+    },
+    {
+      "epoch": 0.9327846364883402,
+      "grad_norm": 0.9360433220863342,
+      "learning_rate": 2.8521267597846094e-05,
+      "loss": 0.5826,
+      "step": 340
+    },
+    {
+      "epoch": 0.9465020576131687,
+      "grad_norm": 0.9888073205947876,
+      "learning_rate": 2.8461848875076884e-05,
+      "loss": 0.6349,
+      "step": 345
+    },
+    {
+      "epoch": 0.9602194787379973,
+      "grad_norm": 0.914045512676239,
+      "learning_rate": 2.8401324179711678e-05,
+      "loss": 0.5755,
+      "step": 350
+    },
+    {
+      "epoch": 0.9739368998628258,
+      "grad_norm": 0.9859684705734253,
+      "learning_rate": 2.8339698484221574e-05,
+      "loss": 0.6325,
+      "step": 355
+    },
+    {
+      "epoch": 0.9876543209876543,
+      "grad_norm": 0.9974327683448792,
+      "learning_rate": 2.827697685153151e-05,
+      "loss": 0.5818,
+      "step": 360
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.362316370010376,
+      "learning_rate": 2.8213164434604316e-05,
+      "loss": 0.5783,
+      "step": 365
+    },
+    {
+      "epoch": 1.0137174211248285,
+      "grad_norm": 1.4332367181777954,
+      "learning_rate": 2.814826647601738e-05,
+      "loss": 0.5402,
+      "step": 370
+    },
+    {
+      "epoch": 1.027434842249657,
+      "grad_norm": 1.2112072706222534,
+      "learning_rate": 2.8082288307531914e-05,
+      "loss": 0.5368,
+      "step": 375
+    },
+    {
+      "epoch": 1.0411522633744856,
+      "grad_norm": 1.2344884872436523,
+      "learning_rate": 2.8015235349654938e-05,
+      "loss": 0.5097,
+      "step": 380
+    },
+    {
+      "epoch": 1.0548696844993142,
+      "grad_norm": 0.9244922995567322,
+      "learning_rate": 2.7947113111193936e-05,
+      "loss": 0.5583,
+      "step": 385
+    },
+    {
+      "epoch": 1.0685871056241427,
+      "grad_norm": 0.980315089225769,
+      "learning_rate": 2.7877927188804288e-05,
+      "loss": 0.51,
+      "step": 390
+    },
+    {
+      "epoch": 1.0823045267489713,
+      "grad_norm": 0.9603582620620728,
+      "learning_rate": 2.7807683266529466e-05,
+      "loss": 0.5517,
+      "step": 395
+    },
+    {
+      "epoch": 1.0960219478737998,
+      "grad_norm": 1.0928096771240234,
+      "learning_rate": 2.773638711533405e-05,
+      "loss": 0.5018,
+      "step": 400
+    },
+    {
+      "epoch": 1.1097393689986284,
+      "grad_norm": 1.0439014434814453,
+      "learning_rate": 2.7664044592629615e-05,
+      "loss": 0.5139,
+      "step": 405
+    },
+    {
+      "epoch": 1.123456790123457,
+      "grad_norm": 0.9458956718444824,
+      "learning_rate": 2.7590661641793513e-05,
+      "loss": 0.5583,
+      "step": 410
+    },
+    {
+      "epoch": 1.1371742112482854,
+      "grad_norm": 0.9447291493415833,
+      "learning_rate": 2.7516244291680565e-05,
+      "loss": 0.4616,
+      "step": 415
+    },
+    {
+      "epoch": 1.1508916323731138,
+      "grad_norm": 1.076687216758728,
+      "learning_rate": 2.7440798656127792e-05,
+      "loss": 0.4695,
+      "step": 420
+    },
+    {
+      "epoch": 1.1646090534979423,
+      "grad_norm": 1.0236420631408691,
+      "learning_rate": 2.7364330933452094e-05,
+      "loss": 0.5455,
+      "step": 425
+    },
+    {
+      "epoch": 1.1783264746227708,
+      "grad_norm": 1.0733507871627808,
+      "learning_rate": 2.7286847405941024e-05,
+      "loss": 0.4956,
+      "step": 430
+    },
+    {
+      "epoch": 1.1920438957475994,
+      "grad_norm": 0.9793803095817566,
+      "learning_rate": 2.720835443933669e-05,
+      "loss": 0.5202,
+      "step": 435
+    },
+    {
+      "epoch": 1.205761316872428,
+      "grad_norm": 1.0080183744430542,
+      "learning_rate": 2.712885848231273e-05,
+      "loss": 0.4865,
+      "step": 440
+    },
+    {
+      "epoch": 1.2194787379972565,
+      "grad_norm": 0.9846312999725342,
+      "learning_rate": 2.7048366065944538e-05,
+      "loss": 0.4843,
+      "step": 445
+    },
+    {
+      "epoch": 1.233196159122085,
+      "grad_norm": 1.0748393535614014,
+      "learning_rate": 2.6966883803172698e-05,
+      "loss": 0.4714,
+      "step": 450
+    },
+    {
+      "epoch": 1.2469135802469136,
+      "grad_norm": 1.0638792514801025,
+      "learning_rate": 2.6884418388259675e-05,
+      "loss": 0.5295,
+      "step": 455
+    },
+    {
+      "epoch": 1.260631001371742,
+      "grad_norm": 1.0708755254745483,
+      "learning_rate": 2.6800976596239855e-05,
+      "loss": 0.4713,
+      "step": 460
+    },
+    {
+      "epoch": 1.2743484224965707,
+      "grad_norm": 0.9797709584236145,
+      "learning_rate": 2.6716565282362928e-05,
+      "loss": 0.4489,
+      "step": 465
+    },
+    {
+      "epoch": 1.2880658436213992,
+      "grad_norm": 1.017024278640747,
+      "learning_rate": 2.663119138153069e-05,
+      "loss": 0.4699,
+      "step": 470
+    },
+    {
+      "epoch": 1.3017832647462277,
+      "grad_norm": 1.1177973747253418,
+      "learning_rate": 2.654486190772729e-05,
+      "loss": 0.4354,
+      "step": 475
+    },
+    {
+      "epoch": 1.3155006858710563,
+      "grad_norm": 0.9993549585342407,
+      "learning_rate": 2.6457583953443022e-05,
+      "loss": 0.4882,
+      "step": 480
+    },
+    {
+      "epoch": 1.3292181069958848,
+      "grad_norm": 1.1262377500534058,
+      "learning_rate": 2.636936468909158e-05,
+      "loss": 0.41,
+      "step": 485
+    },
+    {
+      "epoch": 1.3429355281207134,
+      "grad_norm": 0.964245080947876,
+      "learning_rate": 2.628021136242101e-05,
+      "loss": 0.4364,
+      "step": 490
+    },
+    {
+      "epoch": 1.356652949245542,
+      "grad_norm": 1.0607110261917114,
+      "learning_rate": 2.619013129791823e-05,
+      "loss": 0.4891,
+      "step": 495
+    },
+    {
+      "epoch": 1.3703703703703702,
+      "grad_norm": 1.0203349590301514,
+      "learning_rate": 2.6099131896207327e-05,
+      "loss": 0.4341,
+      "step": 500
+    },
+    {
+      "epoch": 1.3840877914951988,
+      "grad_norm": 0.9529250860214233,
+      "learning_rate": 2.6007220633441486e-05,
+      "loss": 0.4266,
+      "step": 505
+    },
+    {
+      "epoch": 1.3978052126200273,
+      "grad_norm": 0.9978516697883606,
+      "learning_rate": 2.591440506068883e-05,
+      "loss": 0.434,
+      "step": 510
+    },
+    {
+      "epoch": 1.4115226337448559,
+      "grad_norm": 1.03233003616333,
+      "learning_rate": 2.582069280331204e-05,
+      "loss": 0.3978,
+      "step": 515
+    },
+    {
+      "epoch": 1.4252400548696844,
+      "grad_norm": 1.2834852933883667,
+      "learning_rate": 2.5726091560341873e-05,
+      "loss": 0.4496,
+      "step": 520
+    },
+    {
+      "epoch": 1.438957475994513,
+      "grad_norm": 1.0178697109222412,
+      "learning_rate": 2.5630609103844646e-05,
+      "loss": 0.433,
+      "step": 525
+    },
+    {
+      "epoch": 1.4526748971193415,
+      "grad_norm": 1.0839297771453857,
+      "learning_rate": 2.5534253278283725e-05,
+      "loss": 0.4494,
+      "step": 530
+    },
+    {
+      "epoch": 1.46639231824417,
+      "grad_norm": 0.9672828316688538,
+      "learning_rate": 2.5437031999875047e-05,
+      "loss": 0.4438,
+      "step": 535
+    },
+    {
+      "epoch": 1.4801097393689986,
+      "grad_norm": 0.9954227805137634,
+      "learning_rate": 2.533895325593674e-05,
+      "loss": 0.4253,
+      "step": 540
+    },
+    {
+      "epoch": 1.4938271604938271,
+      "grad_norm": 0.9949136972427368,
+      "learning_rate": 2.5240025104232938e-05,
+      "loss": 0.4565,
+      "step": 545
+    },
+    {
+      "epoch": 1.5075445816186557,
+      "grad_norm": 1.2475636005401611,
+      "learning_rate": 2.514025567231178e-05,
+      "loss": 0.397,
+      "step": 550
+    },
+    {
+      "epoch": 1.5212620027434842,
+      "grad_norm": 1.0037999153137207,
+      "learning_rate": 2.5039653156837686e-05,
+      "loss": 0.3955,
+      "step": 555
+    },
+    {
+      "epoch": 1.5349794238683128,
+      "grad_norm": 1.2002379894256592,
+      "learning_rate": 2.4938225822917932e-05,
+      "loss": 0.3541,
+      "step": 560
+    },
+    {
+      "epoch": 1.5486968449931413,
+      "grad_norm": 1.0180314779281616,
+      "learning_rate": 2.4835982003423654e-05,
+      "loss": 0.403,
+      "step": 565
+    },
+    {
+      "epoch": 1.5624142661179699,
+      "grad_norm": 1.017683982849121,
+      "learning_rate": 2.473293009830522e-05,
+      "loss": 0.4082,
+      "step": 570
+    },
+    {
+      "epoch": 1.5761316872427984,
+      "grad_norm": 1.055148720741272,
+      "learning_rate": 2.4629078573902136e-05,
+      "loss": 0.4118,
+      "step": 575
+    },
+    {
+      "epoch": 1.589849108367627,
+      "grad_norm": 1.082885980606079,
+      "learning_rate": 2.45244359622475e-05,
+      "loss": 0.3646,
+      "step": 580
+    },
+    {
+      "epoch": 1.6035665294924555,
+      "grad_norm": 0.9864196181297302,
+      "learning_rate": 2.4419010860367013e-05,
+      "loss": 0.3726,
+      "step": 585
+    },
+    {
+      "epoch": 1.617283950617284,
+      "grad_norm": 1.0728785991668701,
+      "learning_rate": 2.431281192957271e-05,
+      "loss": 0.3665,
+      "step": 590
+    },
+    {
+      "epoch": 1.6310013717421126,
+      "grad_norm": 1.0340756177902222,
+      "learning_rate": 2.4205847894751358e-05,
+      "loss": 0.402,
+      "step": 595
+    },
+    {
+      "epoch": 1.6447187928669411,
+      "grad_norm": 0.9732591509819031,
+      "learning_rate": 2.409812754364768e-05,
+      "loss": 0.3913,
+      "step": 600
+    },
+    {
+      "epoch": 1.6584362139917697,
+      "grad_norm": 0.9975690245628357,
+      "learning_rate": 2.398965972614235e-05,
+      "loss": 0.356,
+      "step": 605
+    },
+    {
+      "epoch": 1.6721536351165982,
+      "grad_norm": 1.0766946077346802,
+      "learning_rate": 2.3880453353524963e-05,
+      "loss": 0.3921,
+      "step": 610
+    },
+    {
+      "epoch": 1.6858710562414267,
+      "grad_norm": 1.0128309726715088,
+      "learning_rate": 2.377051739776189e-05,
+      "loss": 0.4264,
+      "step": 615
+    },
+    {
+      "epoch": 1.6995884773662553,
+      "grad_norm": 1.0055245161056519,
+      "learning_rate": 2.3659860890759184e-05,
+      "loss": 0.3296,
+      "step": 620
+    },
+    {
+      "epoch": 1.7133058984910838,
+      "grad_norm": 1.058605432510376,
+      "learning_rate": 2.3548492923620567e-05,
+      "loss": 0.3617,
+      "step": 625
+    },
+    {
+      "epoch": 1.7270233196159122,
+      "grad_norm": 1.0829980373382568,
+      "learning_rate": 2.343642264590051e-05,
+      "loss": 0.3801,
+      "step": 630
+    },
+    {
+      "epoch": 1.7407407407407407,
+      "grad_norm": 1.2426400184631348,
+      "learning_rate": 2.3323659264852586e-05,
+      "loss": 0.3491,
+      "step": 635
+    },
+    {
+      "epoch": 1.7544581618655692,
+      "grad_norm": 1.0566037893295288,
+      "learning_rate": 2.3210212044672995e-05,
+      "loss": 0.3547,
+      "step": 640
+    },
+    {
+      "epoch": 1.7681755829903978,
+      "grad_norm": 0.9220610857009888,
+      "learning_rate": 2.3096090305739476e-05,
+      "loss": 0.3682,
+      "step": 645
+    },
+    {
+      "epoch": 1.7818930041152263,
+      "grad_norm": 1.124056100845337,
+      "learning_rate": 2.298130342384559e-05,
+      "loss": 0.3236,
+      "step": 650
+    },
+    {
+      "epoch": 1.7956104252400549,
+      "grad_norm": 1.10678231716156,
+      "learning_rate": 2.2865860829430405e-05,
+      "loss": 0.359,
+      "step": 655
+    },
+    {
+      "epoch": 1.8093278463648834,
+      "grad_norm": 1.1028835773468018,
+      "learning_rate": 2.2749772006803782e-05,
+      "loss": 0.3276,
+      "step": 660
+    },
+    {
+      "epoch": 1.823045267489712,
+      "grad_norm": 0.9391087889671326,
+      "learning_rate": 2.2633046493367128e-05,
+      "loss": 0.3589,
+      "step": 665
+    },
+    {
+      "epoch": 1.8367626886145405,
+      "grad_norm": 1.010254979133606,
+      "learning_rate": 2.2515693878829872e-05,
+      "loss": 0.3517,
+      "step": 670
+    },
+    {
+      "epoch": 1.850480109739369,
+      "grad_norm": 1.0423787832260132,
+      "learning_rate": 2.2397723804421613e-05,
+      "loss": 0.32,
+      "step": 675
+    },
+    {
+      "epoch": 1.8641975308641974,
+      "grad_norm": 1.1941534280776978,
+      "learning_rate": 2.227914596210002e-05,
+      "loss": 0.3128,
+      "step": 680
+    },
+    {
+      "epoch": 1.877914951989026,
+      "grad_norm": 0.9325462579727173,
+      "learning_rate": 2.2159970093754583e-05,
+      "loss": 0.3595,
+      "step": 685
+    },
+    {
+      "epoch": 1.8916323731138545,
+      "grad_norm": 1.2271301746368408,
+      "learning_rate": 2.2040205990406257e-05,
+      "loss": 0.3118,
+      "step": 690
+    },
+    {
+      "epoch": 1.905349794238683,
+      "grad_norm": 0.9768636226654053,
+      "learning_rate": 2.1919863491403083e-05,
+      "loss": 0.3858,
+      "step": 695
+    },
+    {
+      "epoch": 1.9190672153635115,
+      "grad_norm": 1.3226031064987183,
+      "learning_rate": 2.1798952483611812e-05,
+      "loss": 0.3268,
+      "step": 700
+    },
+    {
+      "epoch": 1.93278463648834,
+      "grad_norm": 0.8924455046653748,
+      "learning_rate": 2.167748290060564e-05,
+      "loss": 0.2887,
+      "step": 705
+    },
+    {
+      "epoch": 1.9465020576131686,
+      "grad_norm": 1.220363974571228,
+      "learning_rate": 2.1555464721848107e-05,
+      "loss": 0.3174,
+      "step": 710
+    },
+    {
+      "epoch": 1.9602194787379972,
+      "grad_norm": 1.1989498138427734,
+      "learning_rate": 2.1432907971873225e-05,
+      "loss": 0.3026,
+      "step": 715
+    },
+    {
+      "epoch": 1.9739368998628257,
+      "grad_norm": 1.065783977508545,
+      "learning_rate": 2.1309822719461905e-05,
+      "loss": 0.3121,
+      "step": 720
+    },
+    {
+      "epoch": 1.9876543209876543,
+      "grad_norm": 1.019674301147461,
+      "learning_rate": 2.118621907681474e-05,
+      "loss": 0.2984,
+      "step": 725
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.3716942071914673,
+      "learning_rate": 2.106210719872121e-05,
+      "loss": 0.2699,
+      "step": 730
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1825,
+  "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": 1.1214531076208722e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

21/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79b8d795e752f31f335de08fb94265816c3f36b065968d4f22e3d63f40c1eec6
+size 7736

21/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)