Upload folder using huggingface_hub

50/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

50/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",
+    "down_proj",
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

50/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34ef5f7c4fb7c2d80cfd0199de9e9e9d31b4f8944ea747792521c028a9d9f43c
+size 791751704

50/latest

@@ -0,0 +1 @@
+global_step670

50/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1d41b97e5aa5dadc498363c55e8f2bf32713108d05fcb9519e3642d51b7e57c
+size 15920

50/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:542cf6c2b2ea10a6ae2f96b47f655507e762d09d059bc6ca045184ab777a1184
+size 15920

50/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ee829fbfb83b0ee14260045fe02875ee827499a4f48ef01dc40db38528181f1
+size 15920

50/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0673487b8ecb4b448d37a43b09b553373973eac454790df2a2e5d65695dd903c
+size 15920

50/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6914aaabf4db514d18d0966fb3682b2fea0cfc712136fc1e3786fd7cf15e98a8
+size 15920

50/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9492ebc9965adc49289a4fcf8ef1c44d80b3376ca6108db171e04f5b2dd359f3
+size 15920

50/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:201b5d0ad9c512d402bcc085b13a8510b656c56e60766ecb645dabceb913aa7b
+size 15920

50/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb7901b8cd491974bcec52a516c95d7824434757bc8c1ce7c0df28d69cc8698b
+size 15920

50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47b2a66e8f763048744b242bac181f3e4d1645b3daf782a7196a91922fbb0e84
+size 1064

50/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
+  }
+}

50/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
+}

50/trainer_state.json

@@ -0,0 +1,972 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 670,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.014925373134328358,
+      "grad_norm": 1.263885259628296,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 1.3606,
+      "step": 5
+    },
+    {
+      "epoch": 0.029850746268656716,
+      "grad_norm": 1.0558431148529053,
+      "learning_rate": 3.2142857142857143e-06,
+      "loss": 1.3011,
+      "step": 10
+    },
+    {
+      "epoch": 0.04477611940298507,
+      "grad_norm": 0.8035950064659119,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 1.2785,
+      "step": 15
+    },
+    {
+      "epoch": 0.05970149253731343,
+      "grad_norm": 0.7125612497329712,
+      "learning_rate": 6.785714285714286e-06,
+      "loss": 1.3308,
+      "step": 20
+    },
+    {
+      "epoch": 0.07462686567164178,
+      "grad_norm": 0.6430915594100952,
+      "learning_rate": 8.571428571428571e-06,
+      "loss": 1.2538,
+      "step": 25
+    },
+    {
+      "epoch": 0.08955223880597014,
+      "grad_norm": 0.7023003697395325,
+      "learning_rate": 1.0357142857142857e-05,
+      "loss": 1.2544,
+      "step": 30
+    },
+    {
+      "epoch": 0.1044776119402985,
+      "grad_norm": 0.7934185862541199,
+      "learning_rate": 1.2142857142857144e-05,
+      "loss": 1.2752,
+      "step": 35
+    },
+    {
+      "epoch": 0.11940298507462686,
+      "grad_norm": 0.5246712565422058,
+      "learning_rate": 1.3928571428571429e-05,
+      "loss": 1.2569,
+      "step": 40
+    },
+    {
+      "epoch": 0.13432835820895522,
+      "grad_norm": 0.5122241973876953,
+      "learning_rate": 1.5714285714285715e-05,
+      "loss": 1.1751,
+      "step": 45
+    },
+    {
+      "epoch": 0.14925373134328357,
+      "grad_norm": 0.7071592807769775,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.2481,
+      "step": 50
+    },
+    {
+      "epoch": 0.16417910447761194,
+      "grad_norm": 0.5430818200111389,
+      "learning_rate": 1.928571428571429e-05,
+      "loss": 1.2299,
+      "step": 55
+    },
+    {
+      "epoch": 0.1791044776119403,
+      "grad_norm": 0.49823200702667236,
+      "learning_rate": 2.107142857142857e-05,
+      "loss": 1.1872,
+      "step": 60
+    },
+    {
+      "epoch": 0.19402985074626866,
+      "grad_norm": 0.5978754758834839,
+      "learning_rate": 2.2857142857142858e-05,
+      "loss": 1.1698,
+      "step": 65
+    },
+    {
+      "epoch": 0.208955223880597,
+      "grad_norm": 0.572367787361145,
+      "learning_rate": 2.464285714285714e-05,
+      "loss": 1.1965,
+      "step": 70
+    },
+    {
+      "epoch": 0.22388059701492538,
+      "grad_norm": 0.583101212978363,
+      "learning_rate": 2.6428571428571428e-05,
+      "loss": 1.1288,
+      "step": 75
+    },
+    {
+      "epoch": 0.23880597014925373,
+      "grad_norm": 0.640312671661377,
+      "learning_rate": 2.8214285714285714e-05,
+      "loss": 1.1795,
+      "step": 80
+    },
+    {
+      "epoch": 0.2537313432835821,
+      "grad_norm": 0.4734359085559845,
+      "learning_rate": 3e-05,
+      "loss": 1.1263,
+      "step": 85
+    },
+    {
+      "epoch": 0.26865671641791045,
+      "grad_norm": 0.5374538898468018,
+      "learning_rate": 2.999926893308441e-05,
+      "loss": 1.1391,
+      "step": 90
+    },
+    {
+      "epoch": 0.2835820895522388,
+      "grad_norm": 0.5650818943977356,
+      "learning_rate": 2.9997075803598805e-05,
+      "loss": 1.1529,
+      "step": 95
+    },
+    {
+      "epoch": 0.29850746268656714,
+      "grad_norm": 0.558725893497467,
+      "learning_rate": 2.9993420825319784e-05,
+      "loss": 1.1537,
+      "step": 100
+    },
+    {
+      "epoch": 0.31343283582089554,
+      "grad_norm": 0.617064356803894,
+      "learning_rate": 2.9988304354518498e-05,
+      "loss": 1.0812,
+      "step": 105
+    },
+    {
+      "epoch": 0.3283582089552239,
+      "grad_norm": 0.5848479270935059,
+      "learning_rate": 2.9981726889925956e-05,
+      "loss": 1.0247,
+      "step": 110
+    },
+    {
+      "epoch": 0.34328358208955223,
+      "grad_norm": 0.753408670425415,
+      "learning_rate": 2.997368907268439e-05,
+      "loss": 1.0811,
+      "step": 115
+    },
+    {
+      "epoch": 0.3582089552238806,
+      "grad_norm": 0.6939697861671448,
+      "learning_rate": 2.9964191686284768e-05,
+      "loss": 0.9471,
+      "step": 120
+    },
+    {
+      "epoch": 0.373134328358209,
+      "grad_norm": 0.6099917888641357,
+      "learning_rate": 2.995323565649042e-05,
+      "loss": 1.0416,
+      "step": 125
+    },
+    {
+      "epoch": 0.3880597014925373,
+      "grad_norm": 0.7474483251571655,
+      "learning_rate": 2.99408220512468e-05,
+      "loss": 0.9866,
+      "step": 130
+    },
+    {
+      "epoch": 0.40298507462686567,
+      "grad_norm": 0.6931478381156921,
+      "learning_rate": 2.9926952080577392e-05,
+      "loss": 1.0037,
+      "step": 135
+    },
+    {
+      "epoch": 0.417910447761194,
+      "grad_norm": 0.7909236550331116,
+      "learning_rate": 2.991162709646575e-05,
+      "loss": 0.9312,
+      "step": 140
+    },
+    {
+      "epoch": 0.43283582089552236,
+      "grad_norm": 0.666280210018158,
+      "learning_rate": 2.9894848592723724e-05,
+      "loss": 0.98,
+      "step": 145
+    },
+    {
+      "epoch": 0.44776119402985076,
+      "grad_norm": 0.6665749549865723,
+      "learning_rate": 2.9876618204845835e-05,
+      "loss": 1.0119,
+      "step": 150
+    },
+    {
+      "epoch": 0.4626865671641791,
+      "grad_norm": 0.7398273348808289,
+      "learning_rate": 2.985693770984989e-05,
+      "loss": 0.9827,
+      "step": 155
+    },
+    {
+      "epoch": 0.47761194029850745,
+      "grad_norm": 0.8117846846580505,
+      "learning_rate": 2.9835809026103717e-05,
+      "loss": 0.9971,
+      "step": 160
+    },
+    {
+      "epoch": 0.4925373134328358,
+      "grad_norm": 0.760555624961853,
+      "learning_rate": 2.98132342131382e-05,
+      "loss": 0.8978,
+      "step": 165
+    },
+    {
+      "epoch": 0.5074626865671642,
+      "grad_norm": 0.7810600399971008,
+      "learning_rate": 2.9789215471446528e-05,
+      "loss": 0.875,
+      "step": 170
+    },
+    {
+      "epoch": 0.5223880597014925,
+      "grad_norm": 0.8079382181167603,
+      "learning_rate": 2.9763755142269695e-05,
+      "loss": 0.9043,
+      "step": 175
+    },
+    {
+      "epoch": 0.5373134328358209,
+      "grad_norm": 0.7788965106010437,
+      "learning_rate": 2.9736855707368266e-05,
+      "loss": 0.8991,
+      "step": 180
+    },
+    {
+      "epoch": 0.5522388059701493,
+      "grad_norm": 0.78010094165802,
+      "learning_rate": 2.97085197887805e-05,
+      "loss": 0.9647,
+      "step": 185
+    },
+    {
+      "epoch": 0.5671641791044776,
+      "grad_norm": 1.0148367881774902,
+      "learning_rate": 2.9678750148566748e-05,
+      "loss": 0.9089,
+      "step": 190
+    },
+    {
+      "epoch": 0.582089552238806,
+      "grad_norm": 0.8195189833641052,
+      "learning_rate": 2.9647549688540206e-05,
+      "loss": 0.8643,
+      "step": 195
+    },
+    {
+      "epoch": 0.5970149253731343,
+      "grad_norm": 0.848161518573761,
+      "learning_rate": 2.9614921449984092e-05,
+      "loss": 0.8712,
+      "step": 200
+    },
+    {
+      "epoch": 0.6119402985074627,
+      "grad_norm": 0.8245360255241394,
+      "learning_rate": 2.958086861335517e-05,
+      "loss": 0.8555,
+      "step": 205
+    },
+    {
+      "epoch": 0.6268656716417911,
+      "grad_norm": 0.9479895830154419,
+      "learning_rate": 2.9545394497973736e-05,
+      "loss": 0.8398,
+      "step": 210
+    },
+    {
+      "epoch": 0.6417910447761194,
+      "grad_norm": 0.9396081566810608,
+      "learning_rate": 2.950850256170007e-05,
+      "loss": 0.8513,
+      "step": 215
+    },
+    {
+      "epoch": 0.6567164179104478,
+      "grad_norm": 0.9190545082092285,
+      "learning_rate": 2.9470196400597385e-05,
+      "loss": 0.8284,
+      "step": 220
+    },
+    {
+      "epoch": 0.6716417910447762,
+      "grad_norm": 0.8513995409011841,
+      "learning_rate": 2.9430479748581286e-05,
+      "loss": 0.7965,
+      "step": 225
+    },
+    {
+      "epoch": 0.6865671641791045,
+      "grad_norm": 0.916506826877594,
+      "learning_rate": 2.9389356477055807e-05,
+      "loss": 0.7795,
+      "step": 230
+    },
+    {
+      "epoch": 0.7014925373134329,
+      "grad_norm": 0.9645065665245056,
+      "learning_rate": 2.934683059453605e-05,
+      "loss": 0.7978,
+      "step": 235
+    },
+    {
+      "epoch": 0.7164179104477612,
+      "grad_norm": 1.0454235076904297,
+      "learning_rate": 2.9302906246257462e-05,
+      "loss": 0.7961,
+      "step": 240
+    },
+    {
+      "epoch": 0.7313432835820896,
+      "grad_norm": 0.8453083634376526,
+      "learning_rate": 2.925758771377174e-05,
+      "loss": 0.7732,
+      "step": 245
+    },
+    {
+      "epoch": 0.746268656716418,
+      "grad_norm": 0.8657544255256653,
+      "learning_rate": 2.9210879414529526e-05,
+      "loss": 0.7631,
+      "step": 250
+    },
+    {
+      "epoch": 0.7611940298507462,
+      "grad_norm": 0.9750953316688538,
+      "learning_rate": 2.9162785901449783e-05,
+      "loss": 0.7337,
+      "step": 255
+    },
+    {
+      "epoch": 0.7761194029850746,
+      "grad_norm": 0.923392117023468,
+      "learning_rate": 2.9113311862476023e-05,
+      "loss": 0.7294,
+      "step": 260
+    },
+    {
+      "epoch": 0.7910447761194029,
+      "grad_norm": 0.8583614230155945,
+      "learning_rate": 2.906246212011931e-05,
+      "loss": 0.7727,
+      "step": 265
+    },
+    {
+      "epoch": 0.8059701492537313,
+      "grad_norm": 1.0172728300094604,
+      "learning_rate": 2.901024163098822e-05,
+      "loss": 0.7335,
+      "step": 270
+    },
+    {
+      "epoch": 0.8208955223880597,
+      "grad_norm": 1.0450495481491089,
+      "learning_rate": 2.8956655485305684e-05,
+      "loss": 0.7365,
+      "step": 275
+    },
+    {
+      "epoch": 0.835820895522388,
+      "grad_norm": 0.915125846862793,
+      "learning_rate": 2.8901708906412793e-05,
+      "loss": 0.7908,
+      "step": 280
+    },
+    {
+      "epoch": 0.8507462686567164,
+      "grad_norm": 0.8688738942146301,
+      "learning_rate": 2.8845407250259678e-05,
+      "loss": 0.7055,
+      "step": 285
+    },
+    {
+      "epoch": 0.8656716417910447,
+      "grad_norm": 1.1249737739562988,
+      "learning_rate": 2.8787756004883418e-05,
+      "loss": 0.6999,
+      "step": 290
+    },
+    {
+      "epoch": 0.8805970149253731,
+      "grad_norm": 1.0666416883468628,
+      "learning_rate": 2.8728760789873096e-05,
+      "loss": 0.6972,
+      "step": 295
+    },
+    {
+      "epoch": 0.8955223880597015,
+      "grad_norm": 0.9456201195716858,
+      "learning_rate": 2.8668427355822036e-05,
+      "loss": 0.7061,
+      "step": 300
+    },
+    {
+      "epoch": 0.9104477611940298,
+      "grad_norm": 1.0024287700653076,
+      "learning_rate": 2.860676158376724e-05,
+      "loss": 0.6921,
+      "step": 305
+    },
+    {
+      "epoch": 0.9253731343283582,
+      "grad_norm": 1.1162575483322144,
+      "learning_rate": 2.8543769484616136e-05,
+      "loss": 0.6779,
+      "step": 310
+    },
+    {
+      "epoch": 0.9402985074626866,
+      "grad_norm": 0.901252269744873,
+      "learning_rate": 2.8479457198560684e-05,
+      "loss": 0.6988,
+      "step": 315
+    },
+    {
+      "epoch": 0.9552238805970149,
+      "grad_norm": 1.011564016342163,
+      "learning_rate": 2.841383099447883e-05,
+      "loss": 0.6379,
+      "step": 320
+    },
+    {
+      "epoch": 0.9701492537313433,
+      "grad_norm": 1.080054521560669,
+      "learning_rate": 2.8346897269323456e-05,
+      "loss": 0.6503,
+      "step": 325
+    },
+    {
+      "epoch": 0.9850746268656716,
+      "grad_norm": 1.0466442108154297,
+      "learning_rate": 2.827866254749882e-05,
+      "loss": 0.6311,
+      "step": 330
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.9857845306396484,
+      "learning_rate": 2.8209133480224612e-05,
+      "loss": 0.6904,
+      "step": 335
+    },
+    {
+      "epoch": 1.0149253731343284,
+      "grad_norm": 1.1252492666244507,
+      "learning_rate": 2.81383168448876e-05,
+      "loss": 0.583,
+      "step": 340
+    },
+    {
+      "epoch": 1.0298507462686568,
+      "grad_norm": 1.1118468046188354,
+      "learning_rate": 2.8066219544381002e-05,
+      "loss": 0.5952,
+      "step": 345
+    },
+    {
+      "epoch": 1.044776119402985,
+      "grad_norm": 1.0845764875411987,
+      "learning_rate": 2.7992848606431635e-05,
+      "loss": 0.5363,
+      "step": 350
+    },
+    {
+      "epoch": 1.0597014925373134,
+      "grad_norm": 1.0071251392364502,
+      "learning_rate": 2.7918211182914864e-05,
+      "loss": 0.4835,
+      "step": 355
+    },
+    {
+      "epoch": 1.0746268656716418,
+      "grad_norm": 1.0085487365722656,
+      "learning_rate": 2.7842314549157505e-05,
+      "loss": 0.5614,
+      "step": 360
+    },
+    {
+      "epoch": 1.0895522388059702,
+      "grad_norm": 1.3661624193191528,
+      "learning_rate": 2.7765166103228607e-05,
+      "loss": 0.5464,
+      "step": 365
+    },
+    {
+      "epoch": 1.1044776119402986,
+      "grad_norm": 1.0594074726104736,
+      "learning_rate": 2.7686773365218354e-05,
+      "loss": 0.5516,
+      "step": 370
+    },
+    {
+      "epoch": 1.1194029850746268,
+      "grad_norm": 1.1637259721755981,
+      "learning_rate": 2.7607143976505047e-05,
+      "loss": 0.5379,
+      "step": 375
+    },
+    {
+      "epoch": 1.1343283582089552,
+      "grad_norm": 1.1152710914611816,
+      "learning_rate": 2.7526285699010223e-05,
+      "loss": 0.5577,
+      "step": 380
+    },
+    {
+      "epoch": 1.1492537313432836,
+      "grad_norm": 1.1225759983062744,
+      "learning_rate": 2.7444206414442095e-05,
+      "loss": 0.5277,
+      "step": 385
+    },
+    {
+      "epoch": 1.164179104477612,
+      "grad_norm": 1.0522161722183228,
+      "learning_rate": 2.7360914123527242e-05,
+      "loss": 0.5051,
+      "step": 390
+    },
+    {
+      "epoch": 1.1791044776119404,
+      "grad_norm": 1.2671774625778198,
+      "learning_rate": 2.7276416945230763e-05,
+      "loss": 0.54,
+      "step": 395
+    },
+    {
+      "epoch": 1.1940298507462686,
+      "grad_norm": 1.2012797594070435,
+      "learning_rate": 2.719072311596486e-05,
+      "loss": 0.5534,
+      "step": 400
+    },
+    {
+      "epoch": 1.208955223880597,
+      "grad_norm": 1.1887849569320679,
+      "learning_rate": 2.710384098878599e-05,
+      "loss": 0.5133,
+      "step": 405
+    },
+    {
+      "epoch": 1.2238805970149254,
+      "grad_norm": 1.2068318128585815,
+      "learning_rate": 2.7015779032580654e-05,
+      "loss": 0.4882,
+      "step": 410
+    },
+    {
+      "epoch": 1.2388059701492538,
+      "grad_norm": 1.216110110282898,
+      "learning_rate": 2.692654583123988e-05,
+      "loss": 0.4902,
+      "step": 415
+    },
+    {
+      "epoch": 1.2537313432835822,
+      "grad_norm": 1.2037813663482666,
+      "learning_rate": 2.6836150082822498e-05,
+      "loss": 0.572,
+      "step": 420
+    },
+    {
+      "epoch": 1.2686567164179103,
+      "grad_norm": 1.0787304639816284,
+      "learning_rate": 2.674460059870731e-05,
+      "loss": 0.4691,
+      "step": 425
+    },
+    {
+      "epoch": 1.2835820895522387,
+      "grad_norm": 1.005724549293518,
+      "learning_rate": 2.665190630273418e-05,
+      "loss": 0.4779,
+      "step": 430
+    },
+    {
+      "epoch": 1.2985074626865671,
+      "grad_norm": 1.1548197269439697,
+      "learning_rate": 2.6558076230334178e-05,
+      "loss": 0.488,
+      "step": 435
+    },
+    {
+      "epoch": 1.3134328358208955,
+      "grad_norm": 1.105469822883606,
+      "learning_rate": 2.6463119527648864e-05,
+      "loss": 0.472,
+      "step": 440
+    },
+    {
+      "epoch": 1.328358208955224,
+      "grad_norm": 1.1275166273117065,
+      "learning_rate": 2.636704545063872e-05,
+      "loss": 0.4679,
+      "step": 445
+    },
+    {
+      "epoch": 1.3432835820895521,
+      "grad_norm": 1.081590175628662,
+      "learning_rate": 2.6269863364180978e-05,
+      "loss": 0.4763,
+      "step": 450
+    },
+    {
+      "epoch": 1.3582089552238805,
+      "grad_norm": 1.4201221466064453,
+      "learning_rate": 2.617158274115673e-05,
+      "loss": 0.4488,
+      "step": 455
+    },
+    {
+      "epoch": 1.373134328358209,
+      "grad_norm": 1.0977258682250977,
+      "learning_rate": 2.6072213161527563e-05,
+      "loss": 0.4801,
+      "step": 460
+    },
+    {
+      "epoch": 1.3880597014925373,
+      "grad_norm": 1.101251482963562,
+      "learning_rate": 2.5971764311401755e-05,
+      "loss": 0.4574,
+      "step": 465
+    },
+    {
+      "epoch": 1.4029850746268657,
+      "grad_norm": 1.1963441371917725,
+      "learning_rate": 2.587024598209012e-05,
+      "loss": 0.4831,
+      "step": 470
+    },
+    {
+      "epoch": 1.417910447761194,
+      "grad_norm": 1.2703346014022827,
+      "learning_rate": 2.5767668069151566e-05,
+      "loss": 0.4436,
+      "step": 475
+    },
+    {
+      "epoch": 1.4328358208955223,
+      "grad_norm": 1.0840184688568115,
+      "learning_rate": 2.5664040571428548e-05,
+      "loss": 0.4529,
+      "step": 480
+    },
+    {
+      "epoch": 1.4477611940298507,
+      "grad_norm": 1.1298258304595947,
+      "learning_rate": 2.555937359007243e-05,
+      "loss": 0.5147,
+      "step": 485
+    },
+    {
+      "epoch": 1.462686567164179,
+      "grad_norm": 1.090550422668457,
+      "learning_rate": 2.5453677327558825e-05,
+      "loss": 0.4533,
+      "step": 490
+    },
+    {
+      "epoch": 1.4776119402985075,
+      "grad_norm": 1.0793935060501099,
+      "learning_rate": 2.534696208669316e-05,
+      "loss": 0.4676,
+      "step": 495
+    },
+    {
+      "epoch": 1.4925373134328357,
+      "grad_norm": 1.0123590230941772,
+      "learning_rate": 2.5239238269606367e-05,
+      "loss": 0.4632,
+      "step": 500
+    },
+    {
+      "epoch": 1.5074626865671643,
+      "grad_norm": 1.2274974584579468,
+      "learning_rate": 2.5130516376740936e-05,
+      "loss": 0.4766,
+      "step": 505
+    },
+    {
+      "epoch": 1.5223880597014925,
+      "grad_norm": 1.2372727394104004,
+      "learning_rate": 2.502080700582738e-05,
+      "loss": 0.4131,
+      "step": 510
+    },
+    {
+      "epoch": 1.537313432835821,
+      "grad_norm": 1.1837670803070068,
+      "learning_rate": 2.491012085085122e-05,
+      "loss": 0.4241,
+      "step": 515
+    },
+    {
+      "epoch": 1.5522388059701493,
+      "grad_norm": 1.090791940689087,
+      "learning_rate": 2.4798468701010592e-05,
+      "loss": 0.4379,
+      "step": 520
+    },
+    {
+      "epoch": 1.5671641791044775,
+      "grad_norm": 1.2342928647994995,
+      "learning_rate": 2.4685861439664513e-05,
+      "loss": 0.4063,
+      "step": 525
+    },
+    {
+      "epoch": 1.582089552238806,
+      "grad_norm": 1.076446533203125,
+      "learning_rate": 2.45723100432721e-05,
+      "loss": 0.4322,
+      "step": 530
+    },
+    {
+      "epoch": 1.5970149253731343,
+      "grad_norm": 1.175313115119934,
+      "learning_rate": 2.4457825580322564e-05,
+      "loss": 0.4192,
+      "step": 535
+    },
+    {
+      "epoch": 1.6119402985074627,
+      "grad_norm": 0.9952613115310669,
+      "learning_rate": 2.4342419210256327e-05,
+      "loss": 0.417,
+      "step": 540
+    },
+    {
+      "epoch": 1.626865671641791,
+      "grad_norm": 1.2897940874099731,
+      "learning_rate": 2.4226102182377263e-05,
+      "loss": 0.4099,
+      "step": 545
+    },
+    {
+      "epoch": 1.6417910447761193,
+      "grad_norm": 1.1653212308883667,
+      "learning_rate": 2.4108885834756144e-05,
+      "loss": 0.3811,
+      "step": 550
+    },
+    {
+      "epoch": 1.6567164179104479,
+      "grad_norm": 1.1938152313232422,
+      "learning_rate": 2.399078159312546e-05,
+      "loss": 0.4095,
+      "step": 555
+    },
+    {
+      "epoch": 1.671641791044776,
+      "grad_norm": 1.1645913124084473,
+      "learning_rate": 2.38718009697657e-05,
+      "loss": 0.4013,
+      "step": 560
+    },
+    {
+      "epoch": 1.6865671641791045,
+      "grad_norm": 1.1362707614898682,
+      "learning_rate": 2.3751955562383184e-05,
+      "loss": 0.3792,
+      "step": 565
+    },
+    {
+      "epoch": 1.7014925373134329,
+      "grad_norm": 1.2568191289901733,
+      "learning_rate": 2.3631257052979543e-05,
+      "loss": 0.3956,
+      "step": 570
+    },
+    {
+      "epoch": 1.716417910447761,
+      "grad_norm": 1.2282624244689941,
+      "learning_rate": 2.3509717206713047e-05,
+      "loss": 0.3899,
+      "step": 575
+    },
+    {
+      "epoch": 1.7313432835820897,
+      "grad_norm": 1.0997339487075806,
+      "learning_rate": 2.3387347870751768e-05,
+      "loss": 0.3831,
+      "step": 580
+    },
+    {
+      "epoch": 1.7462686567164178,
+      "grad_norm": 1.3019473552703857,
+      "learning_rate": 2.3264160973118772e-05,
+      "loss": 0.4055,
+      "step": 585
+    },
+    {
+      "epoch": 1.7611940298507462,
+      "grad_norm": 1.0819166898727417,
+      "learning_rate": 2.314016852152943e-05,
+      "loss": 0.3918,
+      "step": 590
+    },
+    {
+      "epoch": 1.7761194029850746,
+      "grad_norm": 1.1915706396102905,
+      "learning_rate": 2.301538260222096e-05,
+      "loss": 0.3752,
+      "step": 595
+    },
+    {
+      "epoch": 1.7910447761194028,
+      "grad_norm": 1.0722540616989136,
+      "learning_rate": 2.2889815378774316e-05,
+      "loss": 0.4141,
+      "step": 600
+    },
+    {
+      "epoch": 1.8059701492537314,
+      "grad_norm": 1.354112982749939,
+      "learning_rate": 2.2763479090928535e-05,
+      "loss": 0.3551,
+      "step": 605
+    },
+    {
+      "epoch": 1.8208955223880596,
+      "grad_norm": 0.9958921074867249,
+      "learning_rate": 2.263638605338765e-05,
+      "loss": 0.3487,
+      "step": 610
+    },
+    {
+      "epoch": 1.835820895522388,
+      "grad_norm": 1.121872901916504,
+      "learning_rate": 2.2508548654620317e-05,
+      "loss": 0.3864,
+      "step": 615
+    },
+    {
+      "epoch": 1.8507462686567164,
+      "grad_norm": 1.0944926738739014,
+      "learning_rate": 2.237997935565225e-05,
+      "loss": 0.3777,
+      "step": 620
+    },
+    {
+      "epoch": 1.8656716417910446,
+      "grad_norm": 1.1265228986740112,
+      "learning_rate": 2.2250690688851556e-05,
+      "loss": 0.3509,
+      "step": 625
+    },
+    {
+      "epoch": 1.8805970149253732,
+      "grad_norm": 1.1528276205062866,
+      "learning_rate": 2.2120695256707158e-05,
+      "loss": 0.3487,
+      "step": 630
+    },
+    {
+      "epoch": 1.8955223880597014,
+      "grad_norm": 1.1243373155593872,
+      "learning_rate": 2.1990005730600332e-05,
+      "loss": 0.3644,
+      "step": 635
+    },
+    {
+      "epoch": 1.9104477611940298,
+      "grad_norm": 1.0203492641448975,
+      "learning_rate": 2.1858634849569578e-05,
+      "loss": 0.3431,
+      "step": 640
+    },
+    {
+      "epoch": 1.9253731343283582,
+      "grad_norm": 1.1616039276123047,
+      "learning_rate": 2.1726595419068872e-05,
+      "loss": 0.3708,
+      "step": 645
+    },
+    {
+      "epoch": 1.9402985074626866,
+      "grad_norm": 1.0977258682250977,
+      "learning_rate": 2.1593900309719437e-05,
+      "loss": 0.3326,
+      "step": 650
+    },
+    {
+      "epoch": 1.955223880597015,
+      "grad_norm": 1.1156553030014038,
+      "learning_rate": 2.146056245605518e-05,
+      "loss": 0.3305,
+      "step": 655
+    },
+    {
+      "epoch": 1.9701492537313432,
+      "grad_norm": 1.0591223239898682,
+      "learning_rate": 2.132659485526189e-05,
+      "loss": 0.3748,
+      "step": 660
+    },
+    {
+      "epoch": 1.9850746268656716,
+      "grad_norm": 1.0911786556243896,
+      "learning_rate": 2.1192010565910326e-05,
+      "loss": 0.3238,
+      "step": 665
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.167986273765564,
+      "learning_rate": 2.1056822706683333e-05,
+      "loss": 0.3146,
+      "step": 670
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1675,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9.471314918616596e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1211c9ad2e05e0d12300f7320e7ad2d526a2165a4daddf925fb958daabacfa5
+size 7736

50/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)