Upload folder using huggingface_hub

10/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

10/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": [
+    "v_proj",
+    "up_proj",
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "down_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

10/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d13170a560a778cc1454d0db008bcd41bfd0a56007b09d55f12ed6dd04ff8d10
+size 791751704

10/latest

@@ -0,0 +1 @@
+global_step534

10/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eed80c4d62dc3354df1cb5acb53515c03fe07f1aafc5a0b8cdf02b28eecf03bd
+size 15920

10/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4956292562d757416029337f29acf99e4eb4b8bb8b4fbc8abd48b2486e80e811
+size 15920

10/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1eb93c42df34c41c3628841ee8e8b6fa93833697757694e7548c3ea8ad036283
+size 15920

10/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e78cd1dedbf828b3274240e6ebab32f95914484d3b5a5cab968f01c8d519719
+size 15920

10/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:528d7406f74ab17656815a0203eb414102533857767fa6163c425285ef974399
+size 15920

10/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f87f6a40d48f4ed30133ef41b53f23d86aeda753d2c9a61a5796622932ab1263
+size 15920

10/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:859dbf19ff5d05d639c5a9005e0c33f031c76fdeb5dd0b2f4fd859f12dc8305e
+size 15920

10/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5c4914a4be988e86fae24ef8da0b6a8bf49cf722c6d0f5caa12a8b568c4a949
+size 15920

10/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b3f3ed0594d2d1ba26a4fec3a8ca282953aee31ba3ecbec688509917a1fcfca
+size 1064

10/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
+  }
+}

10/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
+}

10/trainer_state.json

@@ -0,0 +1,776 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 534,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01876172607879925,
+      "grad_norm": 1.1220635175704956,
+      "learning_rate": 1.791044776119403e-06,
+      "loss": 1.3297,
+      "step": 5
+    },
+    {
+      "epoch": 0.0375234521575985,
+      "grad_norm": 0.7967448234558105,
+      "learning_rate": 4.029850746268657e-06,
+      "loss": 1.3423,
+      "step": 10
+    },
+    {
+      "epoch": 0.05628517823639775,
+      "grad_norm": 0.6561905145645142,
+      "learning_rate": 6.268656716417911e-06,
+      "loss": 1.3378,
+      "step": 15
+    },
+    {
+      "epoch": 0.075046904315197,
+      "grad_norm": 0.5100926160812378,
+      "learning_rate": 8.507462686567164e-06,
+      "loss": 1.2788,
+      "step": 20
+    },
+    {
+      "epoch": 0.09380863039399624,
+      "grad_norm": 0.5022882223129272,
+      "learning_rate": 1.0746268656716418e-05,
+      "loss": 1.2822,
+      "step": 25
+    },
+    {
+      "epoch": 0.1125703564727955,
+      "grad_norm": 0.5697208642959595,
+      "learning_rate": 1.2985074626865672e-05,
+      "loss": 1.2556,
+      "step": 30
+    },
+    {
+      "epoch": 0.13133208255159476,
+      "grad_norm": 0.5847183465957642,
+      "learning_rate": 1.5223880597014927e-05,
+      "loss": 1.2403,
+      "step": 35
+    },
+    {
+      "epoch": 0.150093808630394,
+      "grad_norm": 0.47744235396385193,
+      "learning_rate": 1.746268656716418e-05,
+      "loss": 1.2096,
+      "step": 40
+    },
+    {
+      "epoch": 0.16885553470919323,
+      "grad_norm": 0.47178810834884644,
+      "learning_rate": 1.9701492537313435e-05,
+      "loss": 1.2533,
+      "step": 45
+    },
+    {
+      "epoch": 0.18761726078799248,
+      "grad_norm": 0.44852447509765625,
+      "learning_rate": 2.194029850746269e-05,
+      "loss": 1.2555,
+      "step": 50
+    },
+    {
+      "epoch": 0.20637898686679174,
+      "grad_norm": 0.44277751445770264,
+      "learning_rate": 2.417910447761194e-05,
+      "loss": 1.1888,
+      "step": 55
+    },
+    {
+      "epoch": 0.225140712945591,
+      "grad_norm": 0.4661575257778168,
+      "learning_rate": 2.6417910447761193e-05,
+      "loss": 1.2168,
+      "step": 60
+    },
+    {
+      "epoch": 0.24390243902439024,
+      "grad_norm": 0.5210767984390259,
+      "learning_rate": 2.8656716417910447e-05,
+      "loss": 1.1508,
+      "step": 65
+    },
+    {
+      "epoch": 0.2626641651031895,
+      "grad_norm": 0.5324119925498962,
+      "learning_rate": 2.9999815845810124e-05,
+      "loss": 1.1554,
+      "step": 70
+    },
+    {
+      "epoch": 0.28142589118198874,
+      "grad_norm": 0.5865416526794434,
+      "learning_rate": 2.9997744163102546e-05,
+      "loss": 1.1346,
+      "step": 75
+    },
+    {
+      "epoch": 0.300187617260788,
+      "grad_norm": 0.5545316338539124,
+      "learning_rate": 2.9993370923930618e-05,
+      "loss": 1.1047,
+      "step": 80
+    },
+    {
+      "epoch": 0.31894934333958724,
+      "grad_norm": 0.6227930784225464,
+      "learning_rate": 2.9986696799412394e-05,
+      "loss": 1.1855,
+      "step": 85
+    },
+    {
+      "epoch": 0.33771106941838647,
+      "grad_norm": 0.5252574682235718,
+      "learning_rate": 2.9977722813760195e-05,
+      "loss": 1.1261,
+      "step": 90
+    },
+    {
+      "epoch": 0.35647279549718575,
+      "grad_norm": 0.6158416867256165,
+      "learning_rate": 2.9966450344123444e-05,
+      "loss": 1.1142,
+      "step": 95
+    },
+    {
+      "epoch": 0.37523452157598497,
+      "grad_norm": 0.7427587509155273,
+      "learning_rate": 2.9952881120377314e-05,
+      "loss": 1.0015,
+      "step": 100
+    },
+    {
+      "epoch": 0.39399624765478425,
+      "grad_norm": 0.6637995839118958,
+      "learning_rate": 2.993701722485728e-05,
+      "loss": 1.0785,
+      "step": 105
+    },
+    {
+      "epoch": 0.41275797373358347,
+      "grad_norm": 0.7029932737350464,
+      "learning_rate": 2.9918861092039548e-05,
+      "loss": 1.057,
+      "step": 110
+    },
+    {
+      "epoch": 0.43151969981238275,
+      "grad_norm": 0.6611917018890381,
+      "learning_rate": 2.9898415508167465e-05,
+      "loss": 1.0607,
+      "step": 115
+    },
+    {
+      "epoch": 0.450281425891182,
+      "grad_norm": 0.539310872554779,
+      "learning_rate": 2.987568361082392e-05,
+      "loss": 1.0257,
+      "step": 120
+    },
+    {
+      "epoch": 0.46904315196998125,
+      "grad_norm": 0.6124554872512817,
+      "learning_rate": 2.9850668888449912e-05,
+      "loss": 0.9757,
+      "step": 125
+    },
+    {
+      "epoch": 0.4878048780487805,
+      "grad_norm": 0.6248557567596436,
+      "learning_rate": 2.9823375179809135e-05,
+      "loss": 1.0127,
+      "step": 130
+    },
+    {
+      "epoch": 0.5065666041275797,
+      "grad_norm": 0.993503987789154,
+      "learning_rate": 2.9793806673398934e-05,
+      "loss": 1.0222,
+      "step": 135
+    },
+    {
+      "epoch": 0.525328330206379,
+      "grad_norm": 0.7925176620483398,
+      "learning_rate": 2.9761967906807514e-05,
+      "loss": 0.9856,
+      "step": 140
+    },
+    {
+      "epoch": 0.5440900562851783,
+      "grad_norm": 0.7589050531387329,
+      "learning_rate": 2.9727863766017616e-05,
+      "loss": 0.9949,
+      "step": 145
+    },
+    {
+      "epoch": 0.5628517823639775,
+      "grad_norm": 0.7123038172721863,
+      "learning_rate": 2.9691499484656697e-05,
+      "loss": 0.939,
+      "step": 150
+    },
+    {
+      "epoch": 0.5816135084427767,
+      "grad_norm": 0.8624748587608337,
+      "learning_rate": 2.9652880643193788e-05,
+      "loss": 0.9972,
+      "step": 155
+    },
+    {
+      "epoch": 0.600375234521576,
+      "grad_norm": 0.732760488986969,
+      "learning_rate": 2.9612013168083113e-05,
+      "loss": 1.005,
+      "step": 160
+    },
+    {
+      "epoch": 0.6191369606003753,
+      "grad_norm": 0.820073127746582,
+      "learning_rate": 2.9568903330854612e-05,
+      "loss": 0.9217,
+      "step": 165
+    },
+    {
+      "epoch": 0.6378986866791745,
+      "grad_norm": 0.8361268043518066,
+      "learning_rate": 2.9523557747151505e-05,
+      "loss": 0.8491,
+      "step": 170
+    },
+    {
+      "epoch": 0.6566604127579737,
+      "grad_norm": 0.8558568954467773,
+      "learning_rate": 2.9475983375715068e-05,
+      "loss": 0.8825,
+      "step": 175
+    },
+    {
+      "epoch": 0.6754221388367729,
+      "grad_norm": 0.8939546942710876,
+      "learning_rate": 2.9426187517316723e-05,
+      "loss": 0.8438,
+      "step": 180
+    },
+    {
+      "epoch": 0.6941838649155723,
+      "grad_norm": 0.8571194410324097,
+      "learning_rate": 2.93741778136377e-05,
+      "loss": 0.8708,
+      "step": 185
+    },
+    {
+      "epoch": 0.7129455909943715,
+      "grad_norm": 0.906139075756073,
+      "learning_rate": 2.931996224609628e-05,
+      "loss": 0.8978,
+      "step": 190
+    },
+    {
+      "epoch": 0.7317073170731707,
+      "grad_norm": 1.2007263898849487,
+      "learning_rate": 2.926354913462303e-05,
+      "loss": 0.8055,
+      "step": 195
+    },
+    {
+      "epoch": 0.7504690431519699,
+      "grad_norm": 1.034984827041626,
+      "learning_rate": 2.9204947136383984e-05,
+      "loss": 0.8532,
+      "step": 200
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.9235329031944275,
+      "learning_rate": 2.9144165244452125e-05,
+      "loss": 0.7931,
+      "step": 205
+    },
+    {
+      "epoch": 0.7879924953095685,
+      "grad_norm": 0.8924856781959534,
+      "learning_rate": 2.9081212786427314e-05,
+      "loss": 0.8213,
+      "step": 210
+    },
+    {
+      "epoch": 0.8067542213883677,
+      "grad_norm": 0.9607811570167542,
+      "learning_rate": 2.9016099423004854e-05,
+      "loss": 0.801,
+      "step": 215
+    },
+    {
+      "epoch": 0.8255159474671669,
+      "grad_norm": 1.1123484373092651,
+      "learning_rate": 2.8948835146492985e-05,
+      "loss": 0.8307,
+      "step": 220
+    },
+    {
+      "epoch": 0.8442776735459663,
+      "grad_norm": 0.9834749698638916,
+      "learning_rate": 2.8879430279279443e-05,
+      "loss": 0.7948,
+      "step": 225
+    },
+    {
+      "epoch": 0.8630393996247655,
+      "grad_norm": 1.0126004219055176,
+      "learning_rate": 2.8807895472247393e-05,
+      "loss": 0.8065,
+      "step": 230
+    },
+    {
+      "epoch": 0.8818011257035647,
+      "grad_norm": 1.1847103834152222,
+      "learning_rate": 2.873424170314093e-05,
+      "loss": 0.7728,
+      "step": 235
+    },
+    {
+      "epoch": 0.900562851782364,
+      "grad_norm": 0.9110663533210754,
+      "learning_rate": 2.8658480274880462e-05,
+      "loss": 0.7374,
+      "step": 240
+    },
+    {
+      "epoch": 0.9193245778611632,
+      "grad_norm": 1.0707428455352783,
+      "learning_rate": 2.8580622813828123e-05,
+      "loss": 0.7819,
+      "step": 245
+    },
+    {
+      "epoch": 0.9380863039399625,
+      "grad_norm": 1.02041494846344,
+      "learning_rate": 2.8500681268003624e-05,
+      "loss": 0.8076,
+      "step": 250
+    },
+    {
+      "epoch": 0.9568480300187617,
+      "grad_norm": 1.0658625364303589,
+      "learning_rate": 2.8418667905250683e-05,
+      "loss": 0.7103,
+      "step": 255
+    },
+    {
+      "epoch": 0.975609756097561,
+      "grad_norm": 0.9454028606414795,
+      "learning_rate": 2.833459531135441e-05,
+      "loss": 0.7534,
+      "step": 260
+    },
+    {
+      "epoch": 0.9943714821763602,
+      "grad_norm": 1.212715744972229,
+      "learning_rate": 2.8248476388109897e-05,
+      "loss": 0.7299,
+      "step": 265
+    },
+    {
+      "epoch": 1.0112570356472796,
+      "grad_norm": 1.012231469154358,
+      "learning_rate": 2.8160324351342288e-05,
+      "loss": 0.6632,
+      "step": 270
+    },
+    {
+      "epoch": 1.0300187617260788,
+      "grad_norm": 1.054259181022644,
+      "learning_rate": 2.80701527288787e-05,
+      "loss": 0.6126,
+      "step": 275
+    },
+    {
+      "epoch": 1.048780487804878,
+      "grad_norm": 1.0729856491088867,
+      "learning_rate": 2.797797535847221e-05,
+      "loss": 0.6336,
+      "step": 280
+    },
+    {
+      "epoch": 1.0675422138836772,
+      "grad_norm": 1.1964389085769653,
+      "learning_rate": 2.788380638567835e-05,
+      "loss": 0.663,
+      "step": 285
+    },
+    {
+      "epoch": 1.0863039399624765,
+      "grad_norm": 1.1786532402038574,
+      "learning_rate": 2.778766026168428e-05,
+      "loss": 0.5962,
+      "step": 290
+    },
+    {
+      "epoch": 1.1050656660412759,
+      "grad_norm": 1.171257495880127,
+      "learning_rate": 2.768955174109115e-05,
+      "loss": 0.5722,
+      "step": 295
+    },
+    {
+      "epoch": 1.123827392120075,
+      "grad_norm": 1.0211284160614014,
+      "learning_rate": 2.758949587964983e-05,
+      "loss": 0.5923,
+      "step": 300
+    },
+    {
+      "epoch": 1.1425891181988743,
+      "grad_norm": 1.18953275680542,
+      "learning_rate": 2.748750803195047e-05,
+      "loss": 0.597,
+      "step": 305
+    },
+    {
+      "epoch": 1.1613508442776737,
+      "grad_norm": 1.2630354166030884,
+      "learning_rate": 2.7383603849066167e-05,
+      "loss": 0.5956,
+      "step": 310
+    },
+    {
+      "epoch": 1.1801125703564728,
+      "grad_norm": 1.1608085632324219,
+      "learning_rate": 2.7277799276151137e-05,
+      "loss": 0.6085,
+      "step": 315
+    },
+    {
+      "epoch": 1.198874296435272,
+      "grad_norm": 1.1372995376586914,
+      "learning_rate": 2.7170110549993817e-05,
+      "loss": 0.5848,
+      "step": 320
+    },
+    {
+      "epoch": 1.2176360225140712,
+      "grad_norm": 1.081286072731018,
+      "learning_rate": 2.7060554196525105e-05,
+      "loss": 0.5666,
+      "step": 325
+    },
+    {
+      "epoch": 1.2363977485928705,
+      "grad_norm": 1.418300747871399,
+      "learning_rate": 2.694914702828233e-05,
+      "loss": 0.5474,
+      "step": 330
+    },
+    {
+      "epoch": 1.2551594746716699,
+      "grad_norm": 1.1424428224563599,
+      "learning_rate": 2.6835906141829173e-05,
+      "loss": 0.5782,
+      "step": 335
+    },
+    {
+      "epoch": 1.273921200750469,
+      "grad_norm": 1.208847165107727,
+      "learning_rate": 2.6720848915132053e-05,
+      "loss": 0.5132,
+      "step": 340
+    },
+    {
+      "epoch": 1.2926829268292683,
+      "grad_norm": 1.0693706274032593,
+      "learning_rate": 2.660399300489326e-05,
+      "loss": 0.5607,
+      "step": 345
+    },
+    {
+      "epoch": 1.3114446529080674,
+      "grad_norm": 1.2664192914962769,
+      "learning_rate": 2.6485356343841395e-05,
+      "loss": 0.5611,
+      "step": 350
+    },
+    {
+      "epoch": 1.3302063789868668,
+      "grad_norm": 1.3361153602600098,
+      "learning_rate": 2.636495713797938e-05,
+      "loss": 0.5923,
+      "step": 355
+    },
+    {
+      "epoch": 1.3489681050656661,
+      "grad_norm": 1.1549376249313354,
+      "learning_rate": 2.624281386379059e-05,
+      "loss": 0.5911,
+      "step": 360
+    },
+    {
+      "epoch": 1.3677298311444652,
+      "grad_norm": 1.3522552251815796,
+      "learning_rate": 2.6118945265403413e-05,
+      "loss": 0.5859,
+      "step": 365
+    },
+    {
+      "epoch": 1.3864915572232646,
+      "grad_norm": 1.36016047000885,
+      "learning_rate": 2.599337035171481e-05,
+      "loss": 0.5155,
+      "step": 370
+    },
+    {
+      "epoch": 1.4052532833020637,
+      "grad_norm": 1.268315076828003,
+      "learning_rate": 2.5866108393473182e-05,
+      "loss": 0.4985,
+      "step": 375
+    },
+    {
+      "epoch": 1.424015009380863,
+      "grad_norm": 1.075866937637329,
+      "learning_rate": 2.573717892032109e-05,
+      "loss": 0.4856,
+      "step": 380
+    },
+    {
+      "epoch": 1.4427767354596623,
+      "grad_norm": 1.2592428922653198,
+      "learning_rate": 2.5606601717798212e-05,
+      "loss": 0.5281,
+      "step": 385
+    },
+    {
+      "epoch": 1.4615384615384617,
+      "grad_norm": 1.2148767709732056,
+      "learning_rate": 2.547439682430509e-05,
+      "loss": 0.5133,
+      "step": 390
+    },
+    {
+      "epoch": 1.4803001876172608,
+      "grad_norm": 1.2412337064743042,
+      "learning_rate": 2.534058452802798e-05,
+      "loss": 0.5065,
+      "step": 395
+    },
+    {
+      "epoch": 1.49906191369606,
+      "grad_norm": 1.2149518728256226,
+      "learning_rate": 2.5205185363825482e-05,
+      "loss": 0.4689,
+      "step": 400
+    },
+    {
+      "epoch": 1.5178236397748592,
+      "grad_norm": 1.579555869102478,
+      "learning_rate": 2.506822011007721e-05,
+      "loss": 0.5183,
+      "step": 405
+    },
+    {
+      "epoch": 1.5365853658536586,
+      "grad_norm": 1.134460687637329,
+      "learning_rate": 2.492970978549515e-05,
+      "loss": 0.5026,
+      "step": 410
+    },
+    {
+      "epoch": 1.555347091932458,
+      "grad_norm": 1.2218743562698364,
+      "learning_rate": 2.478967564589816e-05,
+      "loss": 0.5281,
+      "step": 415
+    },
+    {
+      "epoch": 1.574108818011257,
+      "grad_norm": 1.143458366394043,
+      "learning_rate": 2.464813918094999e-05,
+      "loss": 0.4916,
+      "step": 420
+    },
+    {
+      "epoch": 1.5928705440900561,
+      "grad_norm": 1.1755683422088623,
+      "learning_rate": 2.450512211086151e-05,
+      "loss": 0.5052,
+      "step": 425
+    },
+    {
+      "epoch": 1.6116322701688555,
+      "grad_norm": 1.4553024768829346,
+      "learning_rate": 2.4360646383057522e-05,
+      "loss": 0.4152,
+      "step": 430
+    },
+    {
+      "epoch": 1.6303939962476548,
+      "grad_norm": 1.1782385110855103,
+      "learning_rate": 2.4214734168808693e-05,
+      "loss": 0.4832,
+      "step": 435
+    },
+    {
+      "epoch": 1.6491557223264541,
+      "grad_norm": 1.4304338693618774,
+      "learning_rate": 2.4067407859829157e-05,
+      "loss": 0.459,
+      "step": 440
+    },
+    {
+      "epoch": 1.6679174484052532,
+      "grad_norm": 1.1110352277755737,
+      "learning_rate": 2.3918690064840282e-05,
+      "loss": 0.464,
+      "step": 445
+    },
+    {
+      "epoch": 1.6866791744840526,
+      "grad_norm": 1.200399398803711,
+      "learning_rate": 2.37686036061011e-05,
+      "loss": 0.4133,
+      "step": 450
+    },
+    {
+      "epoch": 1.7054409005628517,
+      "grad_norm": 1.1901819705963135,
+      "learning_rate": 2.361717151590605e-05,
+      "loss": 0.4412,
+      "step": 455
+    },
+    {
+      "epoch": 1.724202626641651,
+      "grad_norm": 1.2039285898208618,
+      "learning_rate": 2.3464417033050394e-05,
+      "loss": 0.4763,
+      "step": 460
+    },
+    {
+      "epoch": 1.7429643527204504,
+      "grad_norm": 1.3061219453811646,
+      "learning_rate": 2.331036359926402e-05,
+      "loss": 0.4514,
+      "step": 465
+    },
+    {
+      "epoch": 1.7617260787992497,
+      "grad_norm": 1.2160491943359375,
+      "learning_rate": 2.315503485561404e-05,
+      "loss": 0.4169,
+      "step": 470
+    },
+    {
+      "epoch": 1.7804878048780488,
+      "grad_norm": 1.1981985569000244,
+      "learning_rate": 2.299845463887685e-05,
+      "loss": 0.4207,
+      "step": 475
+    },
+    {
+      "epoch": 1.799249530956848,
+      "grad_norm": 1.164445400238037,
+      "learning_rate": 2.2840646977880132e-05,
+      "loss": 0.4327,
+      "step": 480
+    },
+    {
+      "epoch": 1.8180112570356473,
+      "grad_norm": 1.268277883529663,
+      "learning_rate": 2.268163608981536e-05,
+      "loss": 0.434,
+      "step": 485
+    },
+    {
+      "epoch": 1.8367729831144466,
+      "grad_norm": 1.2116129398345947,
+      "learning_rate": 2.2521446376521476e-05,
+      "loss": 0.396,
+      "step": 490
+    },
+    {
+      "epoch": 1.855534709193246,
+      "grad_norm": 1.208999752998352,
+      "learning_rate": 2.236010242074013e-05,
+      "loss": 0.472,
+      "step": 495
+    },
+    {
+      "epoch": 1.874296435272045,
+      "grad_norm": 1.2668207883834839,
+      "learning_rate": 2.2197628982343224e-05,
+      "loss": 0.3733,
+      "step": 500
+    },
+    {
+      "epoch": 1.8930581613508441,
+      "grad_norm": 1.347996711730957,
+      "learning_rate": 2.2034050994533275e-05,
+      "loss": 0.407,
+      "step": 505
+    },
+    {
+      "epoch": 1.9118198874296435,
+      "grad_norm": 1.42654287815094,
+      "learning_rate": 2.1869393560017143e-05,
+      "loss": 0.3818,
+      "step": 510
+    },
+    {
+      "epoch": 1.9305816135084428,
+      "grad_norm": 1.1722033023834229,
+      "learning_rate": 2.1703681947153768e-05,
+      "loss": 0.4143,
+      "step": 515
+    },
+    {
+      "epoch": 1.9493433395872422,
+      "grad_norm": 1.4157124757766724,
+      "learning_rate": 2.153694158607648e-05,
+      "loss": 0.4325,
+      "step": 520
+    },
+    {
+      "epoch": 1.9681050656660413,
+      "grad_norm": 1.2454453706741333,
+      "learning_rate": 2.136919806479051e-05,
+      "loss": 0.3982,
+      "step": 525
+    },
+    {
+      "epoch": 1.9868667917448404,
+      "grad_norm": 1.444549798965454,
+      "learning_rate": 2.1200477125246203e-05,
+      "loss": 0.3563,
+      "step": 530
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1335,
+  "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": 7.513490532954276e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

10/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30ad5c503d731fc47c9cc32fc1877a74454731b1176cc47739f8566aabfeb3f9
+size 7736

10/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)