Upload folder using huggingface_hub

27/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

27/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",
+    "down_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

27/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:494218783a251c922467f3f88df4ada609109f7fb6ac0c535a5a4cb1f51eece1
+size 791751704

27/latest

@@ -0,0 +1 @@
+global_step720

27/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7b5747b62c6e06edb449c7ad1bbcc513d74f6c0d6c0970f8bcd56f0c9735faf
+size 15920

27/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0d3bd348a2ad9049af206ab3f0c4705fc3ddc13155e4b0aaa73758e541ad517
+size 15920

27/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65196f1accd5d2dc84fb17f655f1478b48fd3514e347d436b428d66c4c0dd331
+size 15920

27/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b48524bce8399fd71f5b7f7fe93926fce8296dcba54fc374874144fc3d014208
+size 15920

27/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cb286ce25dbc87ae8cf431d35aeb9a2204e3258c47819df1602b71808a41e2a
+size 15920

27/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9322a431ac952098632b7802e8f8e77271d0a6d6e73520fdb39ed0251868e36
+size 15920

27/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fc34be4c2db6238badd249195d3e1ddce31b6c42dc0bf5fff39ae4a81730097
+size 15920

27/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:360467b86ae062225abbc6cb1dec5b41a7fc205de58595545eedd129d551bb1e
+size 15920

27/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9ea344970f130e604584d4fa45cdf1de07469dec9f9862908d041c3cbfade24
+size 1064

27/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
+  }
+}

27/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
+}

27/trainer_state.json

@@ -0,0 +1,1042 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 720,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013888888888888888,
+      "grad_norm": 1.3303340673446655,
+      "learning_rate": 1.3333333333333334e-06,
+      "loss": 1.2758,
+      "step": 5
+    },
+    {
+      "epoch": 0.027777777777777776,
+      "grad_norm": 0.8942481279373169,
+      "learning_rate": 3e-06,
+      "loss": 1.261,
+      "step": 10
+    },
+    {
+      "epoch": 0.041666666666666664,
+      "grad_norm": 0.69441157579422,
+      "learning_rate": 4.666666666666667e-06,
+      "loss": 1.2394,
+      "step": 15
+    },
+    {
+      "epoch": 0.05555555555555555,
+      "grad_norm": 0.5783836841583252,
+      "learning_rate": 6.333333333333333e-06,
+      "loss": 1.28,
+      "step": 20
+    },
+    {
+      "epoch": 0.06944444444444445,
+      "grad_norm": 0.6185364127159119,
+      "learning_rate": 8e-06,
+      "loss": 1.2652,
+      "step": 25
+    },
+    {
+      "epoch": 0.08333333333333333,
+      "grad_norm": 0.5661951899528503,
+      "learning_rate": 9.666666666666667e-06,
+      "loss": 1.2474,
+      "step": 30
+    },
+    {
+      "epoch": 0.09722222222222222,
+      "grad_norm": 0.5511808395385742,
+      "learning_rate": 1.1333333333333334e-05,
+      "loss": 1.204,
+      "step": 35
+    },
+    {
+      "epoch": 0.1111111111111111,
+      "grad_norm": 0.46661698818206787,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 1.2092,
+      "step": 40
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 0.4330252707004547,
+      "learning_rate": 1.4666666666666666e-05,
+      "loss": 1.1796,
+      "step": 45
+    },
+    {
+      "epoch": 0.1388888888888889,
+      "grad_norm": 0.4823797941207886,
+      "learning_rate": 1.633333333333333e-05,
+      "loss": 1.2358,
+      "step": 50
+    },
+    {
+      "epoch": 0.1527777777777778,
+      "grad_norm": 0.45977669954299927,
+      "learning_rate": 1.8e-05,
+      "loss": 1.179,
+      "step": 55
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 0.5074872374534607,
+      "learning_rate": 1.9666666666666666e-05,
+      "loss": 1.1703,
+      "step": 60
+    },
+    {
+      "epoch": 0.18055555555555555,
+      "grad_norm": 0.47606608271598816,
+      "learning_rate": 2.1333333333333335e-05,
+      "loss": 1.1341,
+      "step": 65
+    },
+    {
+      "epoch": 0.19444444444444445,
+      "grad_norm": 0.5941349864006042,
+      "learning_rate": 2.3000000000000003e-05,
+      "loss": 1.1718,
+      "step": 70
+    },
+    {
+      "epoch": 0.20833333333333334,
+      "grad_norm": 0.5346580743789673,
+      "learning_rate": 2.4666666666666665e-05,
+      "loss": 1.151,
+      "step": 75
+    },
+    {
+      "epoch": 0.2222222222222222,
+      "grad_norm": 0.49426817893981934,
+      "learning_rate": 2.6333333333333334e-05,
+      "loss": 1.0661,
+      "step": 80
+    },
+    {
+      "epoch": 0.2361111111111111,
+      "grad_norm": 0.425722599029541,
+      "learning_rate": 2.8e-05,
+      "loss": 1.044,
+      "step": 85
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.5695877075195312,
+      "learning_rate": 2.966666666666667e-05,
+      "loss": 1.1016,
+      "step": 90
+    },
+    {
+      "epoch": 0.2638888888888889,
+      "grad_norm": 0.5956630706787109,
+      "learning_rate": 2.999959497035049e-05,
+      "loss": 1.0176,
+      "step": 95
+    },
+    {
+      "epoch": 0.2777777777777778,
+      "grad_norm": 0.5605266690254211,
+      "learning_rate": 2.9997949574887035e-05,
+      "loss": 1.0613,
+      "step": 100
+    },
+    {
+      "epoch": 0.2916666666666667,
+      "grad_norm": 0.6029469966888428,
+      "learning_rate": 2.99950386379898e-05,
+      "loss": 1.0369,
+      "step": 105
+    },
+    {
+      "epoch": 0.3055555555555556,
+      "grad_norm": 0.5131080150604248,
+      "learning_rate": 2.9990862405286438e-05,
+      "loss": 1.0467,
+      "step": 110
+    },
+    {
+      "epoch": 0.3194444444444444,
+      "grad_norm": 0.6137083172798157,
+      "learning_rate": 2.998542122917149e-05,
+      "loss": 1.0145,
+      "step": 115
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5534991025924683,
+      "learning_rate": 2.997871556877664e-05,
+      "loss": 1.0217,
+      "step": 120
+    },
+    {
+      "epoch": 0.3472222222222222,
+      "grad_norm": 0.7161787152290344,
+      "learning_rate": 2.997074598993197e-05,
+      "loss": 1.0402,
+      "step": 125
+    },
+    {
+      "epoch": 0.3611111111111111,
+      "grad_norm": 0.594761312007904,
+      "learning_rate": 2.9961513165118224e-05,
+      "loss": 1.0464,
+      "step": 130
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.6094787120819092,
+      "learning_rate": 2.9951017873410053e-05,
+      "loss": 0.9971,
+      "step": 135
+    },
+    {
+      "epoch": 0.3888888888888889,
+      "grad_norm": 0.5699455142021179,
+      "learning_rate": 2.993926100041028e-05,
+      "loss": 0.9726,
+      "step": 140
+    },
+    {
+      "epoch": 0.4027777777777778,
+      "grad_norm": 0.726179301738739,
+      "learning_rate": 2.9926243538175172e-05,
+      "loss": 0.9568,
+      "step": 145
+    },
+    {
+      "epoch": 0.4166666666666667,
+      "grad_norm": 0.7146620154380798,
+      "learning_rate": 2.9911966585130737e-05,
+      "loss": 0.9367,
+      "step": 150
+    },
+    {
+      "epoch": 0.4305555555555556,
+      "grad_norm": 0.7823082804679871,
+      "learning_rate": 2.9896431345980032e-05,
+      "loss": 0.9897,
+      "step": 155
+    },
+    {
+      "epoch": 0.4444444444444444,
+      "grad_norm": 0.6647103428840637,
+      "learning_rate": 2.9879639131601504e-05,
+      "loss": 0.9398,
+      "step": 160
+    },
+    {
+      "epoch": 0.4583333333333333,
+      "grad_norm": 0.6053484678268433,
+      "learning_rate": 2.9861591358938368e-05,
+      "loss": 0.958,
+      "step": 165
+    },
+    {
+      "epoch": 0.4722222222222222,
+      "grad_norm": 0.629128634929657,
+      "learning_rate": 2.9842289550879072e-05,
+      "loss": 0.9099,
+      "step": 170
+    },
+    {
+      "epoch": 0.4861111111111111,
+      "grad_norm": 0.7081131935119629,
+      "learning_rate": 2.9821735336128774e-05,
+      "loss": 0.8751,
+      "step": 175
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.7313970923423767,
+      "learning_rate": 2.9799930449071926e-05,
+      "loss": 0.8703,
+      "step": 180
+    },
+    {
+      "epoch": 0.5138888888888888,
+      "grad_norm": 0.8824923038482666,
+      "learning_rate": 2.9776876729625894e-05,
+      "loss": 0.9141,
+      "step": 185
+    },
+    {
+      "epoch": 0.5277777777777778,
+      "grad_norm": 0.6370046734809875,
+      "learning_rate": 2.9752576123085737e-05,
+      "loss": 0.8878,
+      "step": 190
+    },
+    {
+      "epoch": 0.5416666666666666,
+      "grad_norm": 0.7419236898422241,
+      "learning_rate": 2.9727030679960043e-05,
+      "loss": 0.8464,
+      "step": 195
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.8200927376747131,
+      "learning_rate": 2.970024255579792e-05,
+      "loss": 0.8303,
+      "step": 200
+    },
+    {
+      "epoch": 0.5694444444444444,
+      "grad_norm": 0.7647119760513306,
+      "learning_rate": 2.9672214011007087e-05,
+      "loss": 0.9101,
+      "step": 205
+    },
+    {
+      "epoch": 0.5833333333333334,
+      "grad_norm": 0.8843978643417358,
+      "learning_rate": 2.9642947410663154e-05,
+      "loss": 0.8749,
+      "step": 210
+    },
+    {
+      "epoch": 0.5972222222222222,
+      "grad_norm": 0.8577505350112915,
+      "learning_rate": 2.9612445224310053e-05,
+      "loss": 0.8437,
+      "step": 215
+    },
+    {
+      "epoch": 0.6111111111111112,
+      "grad_norm": 0.8507093191146851,
+      "learning_rate": 2.958071002575166e-05,
+      "loss": 0.8313,
+      "step": 220
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 0.861611545085907,
+      "learning_rate": 2.9547744492834593e-05,
+      "loss": 0.8744,
+      "step": 225
+    },
+    {
+      "epoch": 0.6388888888888888,
+      "grad_norm": 0.7713214159011841,
+      "learning_rate": 2.951355140722228e-05,
+      "loss": 0.8769,
+      "step": 230
+    },
+    {
+      "epoch": 0.6527777777777778,
+      "grad_norm": 0.9293366074562073,
+      "learning_rate": 2.947813365416023e-05,
+      "loss": 0.8354,
+      "step": 235
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.8537072539329529,
+      "learning_rate": 2.9441494222232568e-05,
+      "loss": 0.7734,
+      "step": 240
+    },
+    {
+      "epoch": 0.6805555555555556,
+      "grad_norm": 0.8877134323120117,
+      "learning_rate": 2.940363620310985e-05,
+      "loss": 0.8141,
+      "step": 245
+    },
+    {
+      "epoch": 0.6944444444444444,
+      "grad_norm": 0.9355044960975647,
+      "learning_rate": 2.9364562791288207e-05,
+      "loss": 0.7648,
+      "step": 250
+    },
+    {
+      "epoch": 0.7083333333333334,
+      "grad_norm": 0.8050118088722229,
+      "learning_rate": 2.9324277283819768e-05,
+      "loss": 0.8067,
+      "step": 255
+    },
+    {
+      "epoch": 0.7222222222222222,
+      "grad_norm": 0.7927704453468323,
+      "learning_rate": 2.9282783080034466e-05,
+      "loss": 0.7921,
+      "step": 260
+    },
+    {
+      "epoch": 0.7361111111111112,
+      "grad_norm": 0.7963441610336304,
+      "learning_rate": 2.9240083681253192e-05,
+      "loss": 0.8044,
+      "step": 265
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.9507607817649841,
+      "learning_rate": 2.9196182690492342e-05,
+      "loss": 0.758,
+      "step": 270
+    },
+    {
+      "epoch": 0.7638888888888888,
+      "grad_norm": 0.8225581049919128,
+      "learning_rate": 2.915108381215982e-05,
+      "loss": 0.7465,
+      "step": 275
+    },
+    {
+      "epoch": 0.7777777777777778,
+      "grad_norm": 0.9361241459846497,
+      "learning_rate": 2.910479085174242e-05,
+      "loss": 0.8049,
+      "step": 280
+    },
+    {
+      "epoch": 0.7916666666666666,
+      "grad_norm": 0.9086710214614868,
+      "learning_rate": 2.9057307715484746e-05,
+      "loss": 0.7478,
+      "step": 285
+    },
+    {
+      "epoch": 0.8055555555555556,
+      "grad_norm": 0.9152787327766418,
+      "learning_rate": 2.900863841005957e-05,
+      "loss": 0.7627,
+      "step": 290
+    },
+    {
+      "epoch": 0.8194444444444444,
+      "grad_norm": 0.9706355333328247,
+      "learning_rate": 2.895878704222978e-05,
+      "loss": 0.7092,
+      "step": 295
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.8699019551277161,
+      "learning_rate": 2.890775781850181e-05,
+      "loss": 0.7574,
+      "step": 300
+    },
+    {
+      "epoch": 0.8472222222222222,
+      "grad_norm": 0.9554646015167236,
+      "learning_rate": 2.8855555044770728e-05,
+      "loss": 0.7224,
+      "step": 305
+    },
+    {
+      "epoch": 0.8611111111111112,
+      "grad_norm": 0.9026999473571777,
+      "learning_rate": 2.880218312595687e-05,
+      "loss": 0.6757,
+      "step": 310
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 0.8771184086799622,
+      "learning_rate": 2.8747646565634157e-05,
+      "loss": 0.7325,
+      "step": 315
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 1.1735718250274658,
+      "learning_rate": 2.8691949965650092e-05,
+      "loss": 0.6886,
+      "step": 320
+    },
+    {
+      "epoch": 0.9027777777777778,
+      "grad_norm": 0.8841818571090698,
+      "learning_rate": 2.863509802573744e-05,
+      "loss": 0.6573,
+      "step": 325
+    },
+    {
+      "epoch": 0.9166666666666666,
+      "grad_norm": 0.9687197804450989,
+      "learning_rate": 2.8577095543117655e-05,
+      "loss": 0.6765,
+      "step": 330
+    },
+    {
+      "epoch": 0.9305555555555556,
+      "grad_norm": 0.9504878520965576,
+      "learning_rate": 2.8517947412096107e-05,
+      "loss": 0.6933,
+      "step": 335
+    },
+    {
+      "epoch": 0.9444444444444444,
+      "grad_norm": 1.0276504755020142,
+      "learning_rate": 2.8457658623649063e-05,
+      "loss": 0.6655,
+      "step": 340
+    },
+    {
+      "epoch": 0.9583333333333334,
+      "grad_norm": 0.9590686559677124,
+      "learning_rate": 2.8396234265002583e-05,
+      "loss": 0.6441,
+      "step": 345
+    },
+    {
+      "epoch": 0.9722222222222222,
+      "grad_norm": 0.9926571249961853,
+      "learning_rate": 2.8333679519203223e-05,
+      "loss": 0.6189,
+      "step": 350
+    },
+    {
+      "epoch": 0.9861111111111112,
+      "grad_norm": 1.0292668342590332,
+      "learning_rate": 2.826999966468069e-05,
+      "loss": 0.6611,
+      "step": 355
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0787709951400757,
+      "learning_rate": 2.820520007480246e-05,
+      "loss": 0.7084,
+      "step": 360
+    },
+    {
+      "epoch": 1.0138888888888888,
+      "grad_norm": 0.9844677448272705,
+      "learning_rate": 2.8139286217420356e-05,
+      "loss": 0.5665,
+      "step": 365
+    },
+    {
+      "epoch": 1.0277777777777777,
+      "grad_norm": 0.9413930773735046,
+      "learning_rate": 2.807226365440916e-05,
+      "loss": 0.5811,
+      "step": 370
+    },
+    {
+      "epoch": 1.0416666666666667,
+      "grad_norm": 1.0975089073181152,
+      "learning_rate": 2.8004138041197307e-05,
+      "loss": 0.5766,
+      "step": 375
+    },
+    {
+      "epoch": 1.0555555555555556,
+      "grad_norm": 1.1106640100479126,
+      "learning_rate": 2.7934915126289675e-05,
+      "loss": 0.5595,
+      "step": 380
+    },
+    {
+      "epoch": 1.0694444444444444,
+      "grad_norm": 0.9964764714241028,
+      "learning_rate": 2.7864600750782507e-05,
+      "loss": 0.5701,
+      "step": 385
+    },
+    {
+      "epoch": 1.0833333333333333,
+      "grad_norm": 1.0286003351211548,
+      "learning_rate": 2.7793200847870544e-05,
+      "loss": 0.5003,
+      "step": 390
+    },
+    {
+      "epoch": 1.0972222222222223,
+      "grad_norm": 1.0146455764770508,
+      "learning_rate": 2.772072144234639e-05,
+      "loss": 0.5854,
+      "step": 395
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 1.1295380592346191,
+      "learning_rate": 2.76471686500921e-05,
+      "loss": 0.6124,
+      "step": 400
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 1.013516902923584,
+      "learning_rate": 2.757254867756314e-05,
+      "loss": 0.5504,
+      "step": 405
+    },
+    {
+      "epoch": 1.1388888888888888,
+      "grad_norm": 1.05109441280365,
+      "learning_rate": 2.7496867821264688e-05,
+      "loss": 0.5176,
+      "step": 410
+    },
+    {
+      "epoch": 1.1527777777777777,
+      "grad_norm": 0.8996031880378723,
+      "learning_rate": 2.74201324672203e-05,
+      "loss": 0.5452,
+      "step": 415
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 0.9900648593902588,
+      "learning_rate": 2.734234909043308e-05,
+      "loss": 0.5308,
+      "step": 420
+    },
+    {
+      "epoch": 1.1805555555555556,
+      "grad_norm": 1.0687711238861084,
+      "learning_rate": 2.7263524254339294e-05,
+      "loss": 0.5086,
+      "step": 425
+    },
+    {
+      "epoch": 1.1944444444444444,
+      "grad_norm": 1.0682910680770874,
+      "learning_rate": 2.718366461025455e-05,
+      "loss": 0.4955,
+      "step": 430
+    },
+    {
+      "epoch": 1.2083333333333333,
+      "grad_norm": 1.0465431213378906,
+      "learning_rate": 2.7102776896812554e-05,
+      "loss": 0.5502,
+      "step": 435
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 1.0831612348556519,
+      "learning_rate": 2.7020867939396483e-05,
+      "loss": 0.5047,
+      "step": 440
+    },
+    {
+      "epoch": 1.2361111111111112,
+      "grad_norm": 0.9753193855285645,
+      "learning_rate": 2.6937944649563078e-05,
+      "loss": 0.5268,
+      "step": 445
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 1.2069863080978394,
+      "learning_rate": 2.685401402445942e-05,
+      "loss": 0.4902,
+      "step": 450
+    },
+    {
+      "epoch": 1.2638888888888888,
+      "grad_norm": 1.0545098781585693,
+      "learning_rate": 2.6769083146232514e-05,
+      "loss": 0.4788,
+      "step": 455
+    },
+    {
+      "epoch": 1.2777777777777777,
+      "grad_norm": 1.0795210599899292,
+      "learning_rate": 2.668315918143169e-05,
+      "loss": 0.5468,
+      "step": 460
+    },
+    {
+      "epoch": 1.2916666666666667,
+      "grad_norm": 1.2133684158325195,
+      "learning_rate": 2.6596249380403864e-05,
+      "loss": 0.48,
+      "step": 465
+    },
+    {
+      "epoch": 1.3055555555555556,
+      "grad_norm": 1.0929421186447144,
+      "learning_rate": 2.6508361076681776e-05,
+      "loss": 0.4918,
+      "step": 470
+    },
+    {
+      "epoch": 1.3194444444444444,
+      "grad_norm": 0.9951267838478088,
+      "learning_rate": 2.641950168636517e-05,
+      "loss": 0.4974,
+      "step": 475
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 1.131229043006897,
+      "learning_rate": 2.6329678707495002e-05,
+      "loss": 0.4925,
+      "step": 480
+    },
+    {
+      "epoch": 1.3472222222222223,
+      "grad_norm": 1.0469568967819214,
+      "learning_rate": 2.623889971942076e-05,
+      "loss": 0.4916,
+      "step": 485
+    },
+    {
+      "epoch": 1.3611111111111112,
+      "grad_norm": 1.0473977327346802,
+      "learning_rate": 2.6147172382160913e-05,
+      "loss": 0.4861,
+      "step": 490
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 1.0654410123825073,
+      "learning_rate": 2.605450443575655e-05,
+      "loss": 0.5142,
+      "step": 495
+    },
+    {
+      "epoch": 1.3888888888888888,
+      "grad_norm": 1.1624093055725098,
+      "learning_rate": 2.5960903699618263e-05,
+      "loss": 0.477,
+      "step": 500
+    },
+    {
+      "epoch": 1.4027777777777777,
+      "grad_norm": 1.1483854055404663,
+      "learning_rate": 2.5866378071866338e-05,
+      "loss": 0.5209,
+      "step": 505
+    },
+    {
+      "epoch": 1.4166666666666667,
+      "grad_norm": 0.9577386975288391,
+      "learning_rate": 2.5770935528664304e-05,
+      "loss": 0.4941,
+      "step": 510
+    },
+    {
+      "epoch": 1.4305555555555556,
+      "grad_norm": 1.0291787385940552,
+      "learning_rate": 2.5674584123545908e-05,
+      "loss": 0.4468,
+      "step": 515
+    },
+    {
+      "epoch": 1.4444444444444444,
+      "grad_norm": 1.153069019317627,
+      "learning_rate": 2.5577331986735536e-05,
+      "loss": 0.4491,
+      "step": 520
+    },
+    {
+      "epoch": 1.4583333333333333,
+      "grad_norm": 1.2133917808532715,
+      "learning_rate": 2.5479187324462184e-05,
+      "loss": 0.4373,
+      "step": 525
+    },
+    {
+      "epoch": 1.4722222222222223,
+      "grad_norm": 1.123104453086853,
+      "learning_rate": 2.5380158418267e-05,
+      "loss": 0.4666,
+      "step": 530
+    },
+    {
+      "epoch": 1.4861111111111112,
+      "grad_norm": 0.8907086849212646,
+      "learning_rate": 2.52802536243045e-05,
+      "loss": 0.4511,
+      "step": 535
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.0025827884674072,
+      "learning_rate": 2.517948137263743e-05,
+      "loss": 0.4249,
+      "step": 540
+    },
+    {
+      "epoch": 1.5138888888888888,
+      "grad_norm": 1.0645135641098022,
+      "learning_rate": 2.5077850166525484e-05,
+      "loss": 0.4483,
+      "step": 545
+    },
+    {
+      "epoch": 1.5277777777777777,
+      "grad_norm": 1.1788437366485596,
+      "learning_rate": 2.4975368581707724e-05,
+      "loss": 0.4472,
+      "step": 550
+    },
+    {
+      "epoch": 1.5416666666666665,
+      "grad_norm": 1.0517628192901611,
+      "learning_rate": 2.4872045265679024e-05,
+      "loss": 0.4178,
+      "step": 555
+    },
+    {
+      "epoch": 1.5555555555555556,
+      "grad_norm": 1.0361683368682861,
+      "learning_rate": 2.4767888936960312e-05,
+      "loss": 0.4179,
+      "step": 560
+    },
+    {
+      "epoch": 1.5694444444444444,
+      "grad_norm": 1.1737172603607178,
+      "learning_rate": 2.4662908384362964e-05,
+      "loss": 0.4989,
+      "step": 565
+    },
+    {
+      "epoch": 1.5833333333333335,
+      "grad_norm": 1.106797695159912,
+      "learning_rate": 2.455711246624713e-05,
+      "loss": 0.3939,
+      "step": 570
+    },
+    {
+      "epoch": 1.5972222222222223,
+      "grad_norm": 1.0679595470428467,
+      "learning_rate": 2.4450510109774314e-05,
+      "loss": 0.4174,
+      "step": 575
+    },
+    {
+      "epoch": 1.6111111111111112,
+      "grad_norm": 1.3460017442703247,
+      "learning_rate": 2.434311031015404e-05,
+      "loss": 0.4243,
+      "step": 580
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 1.1296610832214355,
+      "learning_rate": 2.4234922129884873e-05,
+      "loss": 0.4456,
+      "step": 585
+    },
+    {
+      "epoch": 1.6388888888888888,
+      "grad_norm": 1.2325292825698853,
+      "learning_rate": 2.4125954697989694e-05,
+      "loss": 0.4114,
+      "step": 590
+    },
+    {
+      "epoch": 1.6527777777777777,
+      "grad_norm": 1.159067988395691,
+      "learning_rate": 2.4016217209245377e-05,
+      "loss": 0.4596,
+      "step": 595
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.2470641136169434,
+      "learning_rate": 2.3905718923406947e-05,
+      "loss": 0.3935,
+      "step": 600
+    },
+    {
+      "epoch": 1.6805555555555556,
+      "grad_norm": 1.0686368942260742,
+      "learning_rate": 2.37944691644262e-05,
+      "loss": 0.418,
+      "step": 605
+    },
+    {
+      "epoch": 1.6944444444444444,
+      "grad_norm": 1.0233144760131836,
+      "learning_rate": 2.368247731966499e-05,
+      "loss": 0.3572,
+      "step": 610
+    },
+    {
+      "epoch": 1.7083333333333335,
+      "grad_norm": 1.0826553106307983,
+      "learning_rate": 2.356975283910306e-05,
+      "loss": 0.4114,
+      "step": 615
+    },
+    {
+      "epoch": 1.7222222222222223,
+      "grad_norm": 1.1858890056610107,
+      "learning_rate": 2.3456305234540677e-05,
+      "loss": 0.3592,
+      "step": 620
+    },
+    {
+      "epoch": 1.7361111111111112,
+      "grad_norm": 1.117847204208374,
+      "learning_rate": 2.3342144078796007e-05,
+      "loss": 0.4112,
+      "step": 625
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 1.130576491355896,
+      "learning_rate": 2.3227279004897358e-05,
+      "loss": 0.4087,
+      "step": 630
+    },
+    {
+      "epoch": 1.7638888888888888,
+      "grad_norm": 1.0268139839172363,
+      "learning_rate": 2.3111719705270314e-05,
+      "loss": 0.4078,
+      "step": 635
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 1.1426600217819214,
+      "learning_rate": 2.2995475930919907e-05,
+      "loss": 0.403,
+      "step": 640
+    },
+    {
+      "epoch": 1.7916666666666665,
+      "grad_norm": 1.0689858198165894,
+      "learning_rate": 2.2878557490607802e-05,
+      "loss": 0.3588,
+      "step": 645
+    },
+    {
+      "epoch": 1.8055555555555556,
+      "grad_norm": 1.174478530883789,
+      "learning_rate": 2.276097425002464e-05,
+      "loss": 0.3848,
+      "step": 650
+    },
+    {
+      "epoch": 1.8194444444444444,
+      "grad_norm": 1.0673271417617798,
+      "learning_rate": 2.2642736130957522e-05,
+      "loss": 0.4206,
+      "step": 655
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 1.1517634391784668,
+      "learning_rate": 2.2523853110452853e-05,
+      "loss": 0.3915,
+      "step": 660
+    },
+    {
+      "epoch": 1.8472222222222223,
+      "grad_norm": 1.1096245050430298,
+      "learning_rate": 2.240433521997443e-05,
+      "loss": 0.3316,
+      "step": 665
+    },
+    {
+      "epoch": 1.8611111111111112,
+      "grad_norm": 1.089471459388733,
+      "learning_rate": 2.2284192544556988e-05,
+      "loss": 0.3748,
+      "step": 670
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 1.1131129264831543,
+      "learning_rate": 2.216343522195523e-05,
+      "loss": 0.3758,
+      "step": 675
+    },
+    {
+      "epoch": 1.8888888888888888,
+      "grad_norm": 1.0454002618789673,
+      "learning_rate": 2.2042073441788363e-05,
+      "loss": 0.354,
+      "step": 680
+    },
+    {
+      "epoch": 1.9027777777777777,
+      "grad_norm": 1.0819425582885742,
+      "learning_rate": 2.1920117444680317e-05,
+      "loss": 0.3419,
+      "step": 685
+    },
+    {
+      "epoch": 1.9166666666666665,
+      "grad_norm": 1.09736967086792,
+      "learning_rate": 2.179757752139562e-05,
+      "loss": 0.3783,
+      "step": 690
+    },
+    {
+      "epoch": 1.9305555555555556,
+      "grad_norm": 1.1822260618209839,
+      "learning_rate": 2.1674464011971048e-05,
+      "loss": 0.3788,
+      "step": 695
+    },
+    {
+      "epoch": 1.9444444444444444,
+      "grad_norm": 1.1565293073654175,
+      "learning_rate": 2.155078730484313e-05,
+      "loss": 0.3549,
+      "step": 700
+    },
+    {
+      "epoch": 1.9583333333333335,
+      "grad_norm": 1.2186641693115234,
+      "learning_rate": 2.1426557835971566e-05,
+      "loss": 0.3476,
+      "step": 705
+    },
+    {
+      "epoch": 1.9722222222222223,
+      "grad_norm": 1.2447794675827026,
+      "learning_rate": 2.1301786087958617e-05,
+      "loss": 0.3758,
+      "step": 710
+    },
+    {
+      "epoch": 1.9861111111111112,
+      "grad_norm": 1.1461338996887207,
+      "learning_rate": 2.1176482589164575e-05,
+      "loss": 0.3405,
+      "step": 715
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.123814582824707,
+      "learning_rate": 2.105065791281939e-05,
+      "loss": 0.345,
+      "step": 720
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1800,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 2000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.065684080953131e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

27/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:baa806d9569558dc03da587c03a5993bc97491c487ed60278c44398b0e71b6a0
+size 7736

27/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)