Upload folder using huggingface_hub

34/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

34/adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "ibm-granite/granite-3.3-8b-base",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 256,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "down_proj",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

34/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d181d73243496a8a98adfa431a9c30164d0f95905aa6c5bb9b4644ab1ee3027e
+size 791751704

34/latest

@@ -0,0 +1 @@
+global_step666

34/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1943a27163e7cce2967ad50cdb874c75516727f47c45e9bad9e5b9e2df8fb00
+size 15920

34/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:228464833836dad78e0f687e9da0ec0a0aa4925360486a31c851608e8693d544
+size 15920

34/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36918f26b95835ef283543583fb75bf9f542608a0eaa6170f0f1dbb11ecc4d42
+size 15920

34/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0390bf3f8166bb9554880c2094ad5354092a9bb4d50704c5b3d871286a1400db
+size 15920

34/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1bbd9256a548d8f349e665083bcc3820e575305b3023df2d049f09a9b007cfc
+size 15920

34/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0479620dba94e3057b0e1adef63bb31b89fa5a7a419ff415bae5b1bef3753e63
+size 15920

34/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4b70b8e0248275bc8edc305d199b9dd9dda691275748239ac58cb5cb4a92cdf
+size 15920

34/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84e6caae28ffb7f15c5f70be342942c8be22a1b0ea4ab167689410c864c64ccc
+size 15920

34/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fdab190bc94fbc18bf170f3c1be64387168c72222dac52e4214a1d152fa3cd2c
+size 1064

34/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
+  }
+}

34/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
+}

34/trainer_state.json

@@ -0,0 +1,965 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 666,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.015015015015015015,
+      "grad_norm": 1.168221116065979,
+      "learning_rate": 1.4285714285714286e-06,
+      "loss": 1.4071,
+      "step": 5
+    },
+    {
+      "epoch": 0.03003003003003003,
+      "grad_norm": 0.9412075281143188,
+      "learning_rate": 3.2142857142857143e-06,
+      "loss": 1.4071,
+      "step": 10
+    },
+    {
+      "epoch": 0.04504504504504504,
+      "grad_norm": 0.5820316672325134,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 1.4308,
+      "step": 15
+    },
+    {
+      "epoch": 0.06006006006006006,
+      "grad_norm": 0.7234917283058167,
+      "learning_rate": 6.785714285714286e-06,
+      "loss": 1.403,
+      "step": 20
+    },
+    {
+      "epoch": 0.07507507507507508,
+      "grad_norm": 0.7134122252464294,
+      "learning_rate": 8.571428571428571e-06,
+      "loss": 1.3703,
+      "step": 25
+    },
+    {
+      "epoch": 0.09009009009009009,
+      "grad_norm": 0.5349926352500916,
+      "learning_rate": 1.0357142857142857e-05,
+      "loss": 1.3204,
+      "step": 30
+    },
+    {
+      "epoch": 0.10510510510510511,
+      "grad_norm": 0.5786008238792419,
+      "learning_rate": 1.2142857142857144e-05,
+      "loss": 1.2867,
+      "step": 35
+    },
+    {
+      "epoch": 0.12012012012012012,
+      "grad_norm": 0.506987452507019,
+      "learning_rate": 1.3928571428571429e-05,
+      "loss": 1.3269,
+      "step": 40
+    },
+    {
+      "epoch": 0.13513513513513514,
+      "grad_norm": 0.5051132440567017,
+      "learning_rate": 1.5714285714285715e-05,
+      "loss": 1.3177,
+      "step": 45
+    },
+    {
+      "epoch": 0.15015015015015015,
+      "grad_norm": 0.4060744643211365,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 1.1999,
+      "step": 50
+    },
+    {
+      "epoch": 0.16516516516516516,
+      "grad_norm": 0.5537149906158447,
+      "learning_rate": 1.928571428571429e-05,
+      "loss": 1.304,
+      "step": 55
+    },
+    {
+      "epoch": 0.18018018018018017,
+      "grad_norm": 0.5138696432113647,
+      "learning_rate": 2.107142857142857e-05,
+      "loss": 1.2876,
+      "step": 60
+    },
+    {
+      "epoch": 0.19519519519519518,
+      "grad_norm": 0.501071035861969,
+      "learning_rate": 2.2857142857142858e-05,
+      "loss": 1.2592,
+      "step": 65
+    },
+    {
+      "epoch": 0.21021021021021022,
+      "grad_norm": 0.505533754825592,
+      "learning_rate": 2.464285714285714e-05,
+      "loss": 1.2247,
+      "step": 70
+    },
+    {
+      "epoch": 0.22522522522522523,
+      "grad_norm": 0.5126124620437622,
+      "learning_rate": 2.6428571428571428e-05,
+      "loss": 1.2843,
+      "step": 75
+    },
+    {
+      "epoch": 0.24024024024024024,
+      "grad_norm": 0.5336039066314697,
+      "learning_rate": 2.8214285714285714e-05,
+      "loss": 1.2725,
+      "step": 80
+    },
+    {
+      "epoch": 0.2552552552552553,
+      "grad_norm": 0.5905837416648865,
+      "learning_rate": 3e-05,
+      "loss": 1.2371,
+      "step": 85
+    },
+    {
+      "epoch": 0.2702702702702703,
+      "grad_norm": 1.3594383001327515,
+      "learning_rate": 2.9999259655754585e-05,
+      "loss": 1.1873,
+      "step": 90
+    },
+    {
+      "epoch": 0.2852852852852853,
+      "grad_norm": 0.5035964846611023,
+      "learning_rate": 2.9997038696099626e-05,
+      "loss": 1.1599,
+      "step": 95
+    },
+    {
+      "epoch": 0.3003003003003003,
+      "grad_norm": 0.5989760756492615,
+      "learning_rate": 2.9993337340271743e-05,
+      "loss": 1.2299,
+      "step": 100
+    },
+    {
+      "epoch": 0.3153153153153153,
+      "grad_norm": 0.6236782073974609,
+      "learning_rate": 2.9988155953641272e-05,
+      "loss": 1.125,
+      "step": 105
+    },
+    {
+      "epoch": 0.3303303303303303,
+      "grad_norm": 0.635779082775116,
+      "learning_rate": 2.998149504767618e-05,
+      "loss": 1.1218,
+      "step": 110
+    },
+    {
+      "epoch": 0.34534534534534533,
+      "grad_norm": 0.6790189146995544,
+      "learning_rate": 2.9973355279891595e-05,
+      "loss": 1.1742,
+      "step": 115
+    },
+    {
+      "epoch": 0.36036036036036034,
+      "grad_norm": 0.5752054452896118,
+      "learning_rate": 2.996373745378487e-05,
+      "loss": 1.1375,
+      "step": 120
+    },
+    {
+      "epoch": 0.37537537537537535,
+      "grad_norm": 0.6132190227508545,
+      "learning_rate": 2.995264251875631e-05,
+      "loss": 1.1516,
+      "step": 125
+    },
+    {
+      "epoch": 0.39039039039039036,
+      "grad_norm": 0.6133520603179932,
+      "learning_rate": 2.9940071570015415e-05,
+      "loss": 1.0994,
+      "step": 130
+    },
+    {
+      "epoch": 0.40540540540540543,
+      "grad_norm": 0.634513795375824,
+      "learning_rate": 2.9926025848472798e-05,
+      "loss": 1.101,
+      "step": 135
+    },
+    {
+      "epoch": 0.42042042042042044,
+      "grad_norm": 0.6447832584381104,
+      "learning_rate": 2.991050674061767e-05,
+      "loss": 1.059,
+      "step": 140
+    },
+    {
+      "epoch": 0.43543543543543545,
+      "grad_norm": 0.671330451965332,
+      "learning_rate": 2.9893515778380997e-05,
+      "loss": 1.1333,
+      "step": 145
+    },
+    {
+      "epoch": 0.45045045045045046,
+      "grad_norm": 0.7105020880699158,
+      "learning_rate": 2.9875054638984253e-05,
+      "loss": 1.0852,
+      "step": 150
+    },
+    {
+      "epoch": 0.46546546546546547,
+      "grad_norm": 0.6580174565315247,
+      "learning_rate": 2.9855125144773885e-05,
+      "loss": 1.0793,
+      "step": 155
+    },
+    {
+      "epoch": 0.4804804804804805,
+      "grad_norm": 0.6566060185432434,
+      "learning_rate": 2.9833729263041407e-05,
+      "loss": 1.0913,
+      "step": 160
+    },
+    {
+      "epoch": 0.4954954954954955,
+      "grad_norm": 0.7263543605804443,
+      "learning_rate": 2.9810869105829202e-05,
+      "loss": 1.0181,
+      "step": 165
+    },
+    {
+      "epoch": 0.5105105105105106,
+      "grad_norm": 0.7267442345619202,
+      "learning_rate": 2.9786546929722055e-05,
+      "loss": 1.032,
+      "step": 170
+    },
+    {
+      "epoch": 0.5255255255255256,
+      "grad_norm": 0.8429233431816101,
+      "learning_rate": 2.9760765135624387e-05,
+      "loss": 0.9719,
+      "step": 175
+    },
+    {
+      "epoch": 0.5405405405405406,
+      "grad_norm": 0.7601214051246643,
+      "learning_rate": 2.9733526268523238e-05,
+      "loss": 1.0216,
+      "step": 180
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.7685151100158691,
+      "learning_rate": 2.9704833017237077e-05,
+      "loss": 0.9199,
+      "step": 185
+    },
+    {
+      "epoch": 0.5705705705705706,
+      "grad_norm": 0.9268845915794373,
+      "learning_rate": 2.967468821415038e-05,
+      "loss": 1.0086,
+      "step": 190
+    },
+    {
+      "epoch": 0.5855855855855856,
+      "grad_norm": 0.9145250916481018,
+      "learning_rate": 2.9643094834933997e-05,
+      "loss": 0.946,
+      "step": 195
+    },
+    {
+      "epoch": 0.6006006006006006,
+      "grad_norm": 0.9080163240432739,
+      "learning_rate": 2.9610055998251473e-05,
+      "loss": 0.9132,
+      "step": 200
+    },
+    {
+      "epoch": 0.6156156156156156,
+      "grad_norm": 0.9134447574615479,
+      "learning_rate": 2.9575574965451156e-05,
+      "loss": 0.8778,
+      "step": 205
+    },
+    {
+      "epoch": 0.6306306306306306,
+      "grad_norm": 0.8887650370597839,
+      "learning_rate": 2.9539655140244263e-05,
+      "loss": 0.9295,
+      "step": 210
+    },
+    {
+      "epoch": 0.6456456456456456,
+      "grad_norm": 1.0790126323699951,
+      "learning_rate": 2.9502300068368922e-05,
+      "loss": 0.8752,
+      "step": 215
+    },
+    {
+      "epoch": 0.6606606606606606,
+      "grad_norm": 0.9577323794364929,
+      "learning_rate": 2.946351343724013e-05,
+      "loss": 0.8857,
+      "step": 220
+    },
+    {
+      "epoch": 0.6756756756756757,
+      "grad_norm": 0.9615657329559326,
+      "learning_rate": 2.9423299075585775e-05,
+      "loss": 0.9026,
+      "step": 225
+    },
+    {
+      "epoch": 0.6906906906906907,
+      "grad_norm": 0.8258851766586304,
+      "learning_rate": 2.9381660953068686e-05,
+      "loss": 0.9374,
+      "step": 230
+    },
+    {
+      "epoch": 0.7057057057057057,
+      "grad_norm": 0.8674429655075073,
+      "learning_rate": 2.9338603179894784e-05,
+      "loss": 0.9496,
+      "step": 235
+    },
+    {
+      "epoch": 0.7207207207207207,
+      "grad_norm": 0.9960718750953674,
+      "learning_rate": 2.929413000640735e-05,
+      "loss": 0.8618,
+      "step": 240
+    },
+    {
+      "epoch": 0.7357357357357357,
+      "grad_norm": 0.9150928258895874,
+      "learning_rate": 2.9248245822667457e-05,
+      "loss": 0.9003,
+      "step": 245
+    },
+    {
+      "epoch": 0.7507507507507507,
+      "grad_norm": 0.9323129057884216,
+      "learning_rate": 2.920095515802062e-05,
+      "loss": 0.8615,
+      "step": 250
+    },
+    {
+      "epoch": 0.7657657657657657,
+      "grad_norm": 1.002323031425476,
+      "learning_rate": 2.9152262680649704e-05,
+      "loss": 0.876,
+      "step": 255
+    },
+    {
+      "epoch": 0.7807807807807807,
+      "grad_norm": 1.0666371583938599,
+      "learning_rate": 2.9102173197114094e-05,
+      "loss": 0.8508,
+      "step": 260
+    },
+    {
+      "epoch": 0.7957957957957958,
+      "grad_norm": 0.9196386933326721,
+      "learning_rate": 2.9050691651875243e-05,
+      "loss": 0.8039,
+      "step": 265
+    },
+    {
+      "epoch": 0.8108108108108109,
+      "grad_norm": 0.9974318742752075,
+      "learning_rate": 2.8997823126808583e-05,
+      "loss": 0.832,
+      "step": 270
+    },
+    {
+      "epoch": 0.8258258258258259,
+      "grad_norm": 0.93741375207901,
+      "learning_rate": 2.894357284070189e-05,
+      "loss": 0.8235,
+      "step": 275
+    },
+    {
+      "epoch": 0.8408408408408409,
+      "grad_norm": 1.0261346101760864,
+      "learning_rate": 2.888794614874011e-05,
+      "loss": 0.803,
+      "step": 280
+    },
+    {
+      "epoch": 0.8558558558558559,
+      "grad_norm": 0.9743596911430359,
+      "learning_rate": 2.883094854197676e-05,
+      "loss": 0.7929,
+      "step": 285
+    },
+    {
+      "epoch": 0.8708708708708709,
+      "grad_norm": 1.0855077505111694,
+      "learning_rate": 2.877258564679185e-05,
+      "loss": 0.7591,
+      "step": 290
+    },
+    {
+      "epoch": 0.8858858858858859,
+      "grad_norm": 0.9973369836807251,
+      "learning_rate": 2.8712863224336533e-05,
+      "loss": 0.7941,
+      "step": 295
+    },
+    {
+      "epoch": 0.9009009009009009,
+      "grad_norm": 1.0347039699554443,
+      "learning_rate": 2.8651787169964374e-05,
+      "loss": 0.7256,
+      "step": 300
+    },
+    {
+      "epoch": 0.9159159159159159,
+      "grad_norm": 1.189362645149231,
+      "learning_rate": 2.8589363512649432e-05,
+      "loss": 0.7328,
+      "step": 305
+    },
+    {
+      "epoch": 0.9309309309309309,
+      "grad_norm": 1.0265212059020996,
+      "learning_rate": 2.8525598414391104e-05,
+      "loss": 0.7552,
+      "step": 310
+    },
+    {
+      "epoch": 0.9459459459459459,
+      "grad_norm": 1.0244401693344116,
+      "learning_rate": 2.846049816960585e-05,
+      "loss": 0.7563,
+      "step": 315
+    },
+    {
+      "epoch": 0.960960960960961,
+      "grad_norm": 1.2380430698394775,
+      "learning_rate": 2.83940692045059e-05,
+      "loss": 0.779,
+      "step": 320
+    },
+    {
+      "epoch": 0.975975975975976,
+      "grad_norm": 1.1021960973739624,
+      "learning_rate": 2.8326318076464852e-05,
+      "loss": 0.7572,
+      "step": 325
+    },
+    {
+      "epoch": 0.990990990990991,
+      "grad_norm": 1.007034182548523,
+      "learning_rate": 2.8257251473370408e-05,
+      "loss": 0.7074,
+      "step": 330
+    },
+    {
+      "epoch": 1.006006006006006,
+      "grad_norm": 1.062544345855713,
+      "learning_rate": 2.8186876212964185e-05,
+      "loss": 0.6919,
+      "step": 335
+    },
+    {
+      "epoch": 1.021021021021021,
+      "grad_norm": 1.149069905281067,
+      "learning_rate": 2.811519924216873e-05,
+      "loss": 0.6393,
+      "step": 340
+    },
+    {
+      "epoch": 1.0360360360360361,
+      "grad_norm": 1.0859824419021606,
+      "learning_rate": 2.8042227636401757e-05,
+      "loss": 0.6682,
+      "step": 345
+    },
+    {
+      "epoch": 1.0510510510510511,
+      "grad_norm": 1.0685064792633057,
+      "learning_rate": 2.796796859887772e-05,
+      "loss": 0.6079,
+      "step": 350
+    },
+    {
+      "epoch": 1.0660660660660661,
+      "grad_norm": 1.0562701225280762,
+      "learning_rate": 2.7892429459896766e-05,
+      "loss": 0.6154,
+      "step": 355
+    },
+    {
+      "epoch": 1.0810810810810811,
+      "grad_norm": 1.1275193691253662,
+      "learning_rate": 2.7815617676121138e-05,
+      "loss": 0.6308,
+      "step": 360
+    },
+    {
+      "epoch": 1.0960960960960962,
+      "grad_norm": 1.0415915250778198,
+      "learning_rate": 2.773754082983912e-05,
+      "loss": 0.6441,
+      "step": 365
+    },
+    {
+      "epoch": 1.1111111111111112,
+      "grad_norm": 1.1084564924240112,
+      "learning_rate": 2.7658206628216556e-05,
+      "loss": 0.5814,
+      "step": 370
+    },
+    {
+      "epoch": 1.1261261261261262,
+      "grad_norm": 1.0700006484985352,
+      "learning_rate": 2.7577622902536064e-05,
+      "loss": 0.5895,
+      "step": 375
+    },
+    {
+      "epoch": 1.1411411411411412,
+      "grad_norm": 1.1039676666259766,
+      "learning_rate": 2.7495797607423986e-05,
+      "loss": 0.6004,
+      "step": 380
+    },
+    {
+      "epoch": 1.1561561561561562,
+      "grad_norm": 1.235087513923645,
+      "learning_rate": 2.7412738820065173e-05,
+      "loss": 0.6174,
+      "step": 385
+    },
+    {
+      "epoch": 1.1711711711711712,
+      "grad_norm": 1.1676405668258667,
+      "learning_rate": 2.732845473940566e-05,
+      "loss": 0.5685,
+      "step": 390
+    },
+    {
+      "epoch": 1.1861861861861862,
+      "grad_norm": 1.2204903364181519,
+      "learning_rate": 2.7242953685343327e-05,
+      "loss": 0.6005,
+      "step": 395
+    },
+    {
+      "epoch": 1.2012012012012012,
+      "grad_norm": 1.1344414949417114,
+      "learning_rate": 2.7156244097906614e-05,
+      "loss": 0.6182,
+      "step": 400
+    },
+    {
+      "epoch": 1.2162162162162162,
+      "grad_norm": 1.2518144845962524,
+      "learning_rate": 2.7068334536421408e-05,
+      "loss": 0.5759,
+      "step": 405
+    },
+    {
+      "epoch": 1.2312312312312312,
+      "grad_norm": 1.248169183731079,
+      "learning_rate": 2.6979233678666102e-05,
+      "loss": 0.5642,
+      "step": 410
+    },
+    {
+      "epoch": 1.2462462462462462,
+      "grad_norm": 1.1738685369491577,
+      "learning_rate": 2.6888950320014993e-05,
+      "loss": 0.5953,
+      "step": 415
+    },
+    {
+      "epoch": 1.2612612612612613,
+      "grad_norm": 1.3666422367095947,
+      "learning_rate": 2.6797493372570098e-05,
+      "loss": 0.5547,
+      "step": 420
+    },
+    {
+      "epoch": 1.2762762762762763,
+      "grad_norm": 1.2075399160385132,
+      "learning_rate": 2.6704871864281377e-05,
+      "loss": 0.5817,
+      "step": 425
+    },
+    {
+      "epoch": 1.2912912912912913,
+      "grad_norm": 1.1854655742645264,
+      "learning_rate": 2.6611094938055586e-05,
+      "loss": 0.5291,
+      "step": 430
+    },
+    {
+      "epoch": 1.3063063063063063,
+      "grad_norm": 1.2468748092651367,
+      "learning_rate": 2.651617185085375e-05,
+      "loss": 0.5862,
+      "step": 435
+    },
+    {
+      "epoch": 1.3213213213213213,
+      "grad_norm": 1.2312390804290771,
+      "learning_rate": 2.642011197277738e-05,
+      "loss": 0.5579,
+      "step": 440
+    },
+    {
+      "epoch": 1.3363363363363363,
+      "grad_norm": 1.2079895734786987,
+      "learning_rate": 2.6322924786143544e-05,
+      "loss": 0.5381,
+      "step": 445
+    },
+    {
+      "epoch": 1.3513513513513513,
+      "grad_norm": 1.2430604696273804,
+      "learning_rate": 2.6224619884548814e-05,
+      "loss": 0.5199,
+      "step": 450
+    },
+    {
+      "epoch": 1.3663663663663663,
+      "grad_norm": 1.3518890142440796,
+      "learning_rate": 2.612520697192229e-05,
+      "loss": 0.5885,
+      "step": 455
+    },
+    {
+      "epoch": 1.3813813813813813,
+      "grad_norm": 1.2342760562896729,
+      "learning_rate": 2.6024695861567675e-05,
+      "loss": 0.5597,
+      "step": 460
+    },
+    {
+      "epoch": 1.3963963963963963,
+      "grad_norm": 1.205000638961792,
+      "learning_rate": 2.592309647519458e-05,
+      "loss": 0.5427,
+      "step": 465
+    },
+    {
+      "epoch": 1.4114114114114114,
+      "grad_norm": 1.304811716079712,
+      "learning_rate": 2.5820418841939152e-05,
+      "loss": 0.5439,
+      "step": 470
+    },
+    {
+      "epoch": 1.4264264264264264,
+      "grad_norm": 1.3263685703277588,
+      "learning_rate": 2.5716673097374047e-05,
+      "loss": 0.5013,
+      "step": 475
+    },
+    {
+      "epoch": 1.4414414414414414,
+      "grad_norm": 1.227352261543274,
+      "learning_rate": 2.5611869482507924e-05,
+      "loss": 0.5695,
+      "step": 480
+    },
+    {
+      "epoch": 1.4564564564564564,
+      "grad_norm": 1.2734203338623047,
+      "learning_rate": 2.550601834277454e-05,
+      "loss": 0.5228,
+      "step": 485
+    },
+    {
+      "epoch": 1.4714714714714714,
+      "grad_norm": 1.147456169128418,
+      "learning_rate": 2.539913012701152e-05,
+      "loss": 0.5174,
+      "step": 490
+    },
+    {
+      "epoch": 1.4864864864864864,
+      "grad_norm": 1.2142225503921509,
+      "learning_rate": 2.529121538642892e-05,
+      "loss": 0.4837,
+      "step": 495
+    },
+    {
+      "epoch": 1.5015015015015014,
+      "grad_norm": 1.177823543548584,
+      "learning_rate": 2.51822847735677e-05,
+      "loss": 0.5203,
+      "step": 500
+    },
+    {
+      "epoch": 1.5165165165165164,
+      "grad_norm": 1.3306628465652466,
+      "learning_rate": 2.5072349041248175e-05,
+      "loss": 0.5125,
+      "step": 505
+    },
+    {
+      "epoch": 1.5315315315315314,
+      "grad_norm": 1.3492076396942139,
+      "learning_rate": 2.496141904150859e-05,
+      "loss": 0.5235,
+      "step": 510
+    },
+    {
+      "epoch": 1.5465465465465464,
+      "grad_norm": 1.0597307682037354,
+      "learning_rate": 2.484950572453386e-05,
+      "loss": 0.5086,
+      "step": 515
+    },
+    {
+      "epoch": 1.5615615615615615,
+      "grad_norm": 1.1566152572631836,
+      "learning_rate": 2.4736620137574686e-05,
+      "loss": 0.4862,
+      "step": 520
+    },
+    {
+      "epoch": 1.5765765765765765,
+      "grad_norm": 1.1374027729034424,
+      "learning_rate": 2.4622773423857032e-05,
+      "loss": 0.5468,
+      "step": 525
+    },
+    {
+      "epoch": 1.5915915915915915,
+      "grad_norm": 1.0723894834518433,
+      "learning_rate": 2.4507976821482138e-05,
+      "loss": 0.523,
+      "step": 530
+    },
+    {
+      "epoch": 1.6066066066066065,
+      "grad_norm": 1.4630274772644043,
+      "learning_rate": 2.4392241662317205e-05,
+      "loss": 0.4643,
+      "step": 535
+    },
+    {
+      "epoch": 1.6216216216216215,
+      "grad_norm": 1.3217425346374512,
+      "learning_rate": 2.4275579370876772e-05,
+      "loss": 0.489,
+      "step": 540
+    },
+    {
+      "epoch": 1.6366366366366365,
+      "grad_norm": 1.301561951637268,
+      "learning_rate": 2.4158001463194998e-05,
+      "loss": 0.4548,
+      "step": 545
+    },
+    {
+      "epoch": 1.6516516516516515,
+      "grad_norm": 1.132656455039978,
+      "learning_rate": 2.4039519545688848e-05,
+      "loss": 0.4805,
+      "step": 550
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 1.1595160961151123,
+      "learning_rate": 2.392014531401244e-05,
+      "loss": 0.5021,
+      "step": 555
+    },
+    {
+      "epoch": 1.6816816816816815,
+      "grad_norm": 1.075697660446167,
+      "learning_rate": 2.37998905519025e-05,
+      "loss": 0.4383,
+      "step": 560
+    },
+    {
+      "epoch": 1.6966966966966965,
+      "grad_norm": 1.2260234355926514,
+      "learning_rate": 2.3678767130015174e-05,
+      "loss": 0.4359,
+      "step": 565
+    },
+    {
+      "epoch": 1.7117117117117115,
+      "grad_norm": 1.1868109703063965,
+      "learning_rate": 2.3556787004754253e-05,
+      "loss": 0.4411,
+      "step": 570
+    },
+    {
+      "epoch": 1.7267267267267268,
+      "grad_norm": 1.1722155809402466,
+      "learning_rate": 2.3433962217090904e-05,
+      "loss": 0.4586,
+      "step": 575
+    },
+    {
+      "epoch": 1.7417417417417418,
+      "grad_norm": 1.1019175052642822,
+      "learning_rate": 2.3310304891375092e-05,
+      "loss": 0.4072,
+      "step": 580
+    },
+    {
+      "epoch": 1.7567567567567568,
+      "grad_norm": 1.2503912448883057,
+      "learning_rate": 2.3185827234138756e-05,
+      "loss": 0.4602,
+      "step": 585
+    },
+    {
+      "epoch": 1.7717717717717718,
+      "grad_norm": 1.1760993003845215,
+      "learning_rate": 2.306054153289085e-05,
+      "loss": 0.4525,
+      "step": 590
+    },
+    {
+      "epoch": 1.7867867867867868,
+      "grad_norm": 1.2905851602554321,
+      "learning_rate": 2.2934460154904436e-05,
+      "loss": 0.4767,
+      "step": 595
+    },
+    {
+      "epoch": 1.8018018018018018,
+      "grad_norm": 1.2083145380020142,
+      "learning_rate": 2.280759554599587e-05,
+      "loss": 0.4617,
+      "step": 600
+    },
+    {
+      "epoch": 1.8168168168168168,
+      "grad_norm": 1.2210261821746826,
+      "learning_rate": 2.2679960229296244e-05,
+      "loss": 0.4468,
+      "step": 605
+    },
+    {
+      "epoch": 1.8318318318318318,
+      "grad_norm": 1.238731026649475,
+      "learning_rate": 2.255156680401518e-05,
+      "loss": 0.4182,
+      "step": 610
+    },
+    {
+      "epoch": 1.8468468468468469,
+      "grad_norm": 1.33894944190979,
+      "learning_rate": 2.242242794419715e-05,
+      "loss": 0.4642,
+      "step": 615
+    },
+    {
+      "epoch": 1.8618618618618619,
+      "grad_norm": 1.2281477451324463,
+      "learning_rate": 2.2292556397470394e-05,
+      "loss": 0.445,
+      "step": 620
+    },
+    {
+      "epoch": 1.8768768768768769,
+      "grad_norm": 1.1752314567565918,
+      "learning_rate": 2.2161964983788535e-05,
+      "loss": 0.3754,
+      "step": 625
+    },
+    {
+      "epoch": 1.8918918918918919,
+      "grad_norm": 1.0684999227523804,
+      "learning_rate": 2.2030666594165135e-05,
+      "loss": 0.4132,
+      "step": 630
+    },
+    {
+      "epoch": 1.906906906906907,
+      "grad_norm": 1.2152754068374634,
+      "learning_rate": 2.1898674189401148e-05,
+      "loss": 0.3813,
+      "step": 635
+    },
+    {
+      "epoch": 1.921921921921922,
+      "grad_norm": 1.2044199705123901,
+      "learning_rate": 2.1766000798805542e-05,
+      "loss": 0.4362,
+      "step": 640
+    },
+    {
+      "epoch": 1.936936936936937,
+      "grad_norm": 1.1659032106399536,
+      "learning_rate": 2.1632659518909156e-05,
+      "loss": 0.3985,
+      "step": 645
+    },
+    {
+      "epoch": 1.951951951951952,
+      "grad_norm": 1.2562652826309204,
+      "learning_rate": 2.1498663512171885e-05,
+      "loss": 0.3964,
+      "step": 650
+    },
+    {
+      "epoch": 1.966966966966967,
+      "grad_norm": 1.1590440273284912,
+      "learning_rate": 2.13640260056834e-05,
+      "loss": 0.3917,
+      "step": 655
+    },
+    {
+      "epoch": 1.981981981981982,
+      "grad_norm": 1.2917598485946655,
+      "learning_rate": 2.1228760289857456e-05,
+      "loss": 0.3642,
+      "step": 660
+    },
+    {
+      "epoch": 1.996996996996997,
+      "grad_norm": 1.40131413936615,
+      "learning_rate": 2.1092879717119955e-05,
+      "loss": 0.4154,
+      "step": 665
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1665,
+  "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": 9.581609998216069e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

34/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5144faed7dcc41e0a564175cadcf4daf6c7415c0abc81efb880456894e37bb15
+size 7736

34/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)