Upload folder using huggingface_hub

44/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

44/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": [
+    "q_proj",
+    "down_proj",
+    "gate_proj",
+    "v_proj",
+    "o_proj",
+    "k_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

44/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:641401ebf9e263a8ed64c085d602984d479e349fbc4fa674a7ac4e3e97dbf834
+size 791751704

44/latest

@@ -0,0 +1 @@
+global_step686

44/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0865abd43dd4ecf7d26c3c689c035c59bfd6923e7ddd762a427531d6be9f1c84
+size 15920

44/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cda53ec14f1abeee9901804e7bfce464d8600000f4c398ceca92232aea338c6e
+size 15920

44/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2357e0f46eef858bc1db5dd8c5a44b66dcfbf964111168732e56f35efc99bbb
+size 15920

44/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84a6197734941b1ca03401329967d4189c48170a093ed8806af086c2b445d9c0
+size 15920

44/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7642f8befca9df55cd446ccf4ad77c9aea888ecc38ac4156cbc62b0016487bea
+size 15920

44/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cccaa8d384d0519b51a9e00a5d4e8b0ce49e918a3e4310cea8ea2bf8f6955445
+size 15920

44/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:474c94748c07ececfa0cacb1d3d4c2d2859a0d68377c80d24bbc4fc28ba1b8f8
+size 15920

44/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5db7f7860c8a0bc3aa2c91283bb72d5c8c93f4fa2161042b5b17b090b0e748e9
+size 15920

44/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:721d61d155bbf5fb5e431ce2efca154ea9f7e17ee3a979cd89ddbf384125cc6a
+size 1064

44/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
+  }
+}

44/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
+}

44/trainer_state.json

@@ -0,0 +1,993 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 686,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.014577259475218658,
+      "grad_norm": 1.1190299987792969,
+      "learning_rate": 1.3953488372093023e-06,
+      "loss": 1.2363,
+      "step": 5
+    },
+    {
+      "epoch": 0.029154518950437316,
+      "grad_norm": 1.1321505308151245,
+      "learning_rate": 3.1395348837209307e-06,
+      "loss": 1.2861,
+      "step": 10
+    },
+    {
+      "epoch": 0.043731778425655975,
+      "grad_norm": 0.7213772535324097,
+      "learning_rate": 4.883720930232559e-06,
+      "loss": 1.2781,
+      "step": 15
+    },
+    {
+      "epoch": 0.05830903790087463,
+      "grad_norm": 0.5360537767410278,
+      "learning_rate": 6.627906976744187e-06,
+      "loss": 1.2532,
+      "step": 20
+    },
+    {
+      "epoch": 0.0728862973760933,
+      "grad_norm": 0.6889077425003052,
+      "learning_rate": 8.372093023255815e-06,
+      "loss": 1.242,
+      "step": 25
+    },
+    {
+      "epoch": 0.08746355685131195,
+      "grad_norm": 0.5271703600883484,
+      "learning_rate": 1.0116279069767442e-05,
+      "loss": 1.2187,
+      "step": 30
+    },
+    {
+      "epoch": 0.10204081632653061,
+      "grad_norm": 0.5964151620864868,
+      "learning_rate": 1.1860465116279069e-05,
+      "loss": 1.1691,
+      "step": 35
+    },
+    {
+      "epoch": 0.11661807580174927,
+      "grad_norm": 0.507500410079956,
+      "learning_rate": 1.3604651162790698e-05,
+      "loss": 1.2055,
+      "step": 40
+    },
+    {
+      "epoch": 0.13119533527696792,
+      "grad_norm": 0.48948171734809875,
+      "learning_rate": 1.5348837209302328e-05,
+      "loss": 1.1319,
+      "step": 45
+    },
+    {
+      "epoch": 0.1457725947521866,
+      "grad_norm": 0.44432899355888367,
+      "learning_rate": 1.7093023255813955e-05,
+      "loss": 1.1077,
+      "step": 50
+    },
+    {
+      "epoch": 0.16034985422740525,
+      "grad_norm": 0.4004872143268585,
+      "learning_rate": 1.8837209302325582e-05,
+      "loss": 1.1243,
+      "step": 55
+    },
+    {
+      "epoch": 0.1749271137026239,
+      "grad_norm": 0.5246862173080444,
+      "learning_rate": 2.058139534883721e-05,
+      "loss": 1.2295,
+      "step": 60
+    },
+    {
+      "epoch": 0.18950437317784258,
+      "grad_norm": 0.44234415888786316,
+      "learning_rate": 2.2325581395348837e-05,
+      "loss": 1.1708,
+      "step": 65
+    },
+    {
+      "epoch": 0.20408163265306123,
+      "grad_norm": 0.47460269927978516,
+      "learning_rate": 2.4069767441860464e-05,
+      "loss": 1.1034,
+      "step": 70
+    },
+    {
+      "epoch": 0.21865889212827988,
+      "grad_norm": 0.5430591702461243,
+      "learning_rate": 2.5813953488372094e-05,
+      "loss": 1.1032,
+      "step": 75
+    },
+    {
+      "epoch": 0.23323615160349853,
+      "grad_norm": 0.6747642159461975,
+      "learning_rate": 2.755813953488372e-05,
+      "loss": 1.0462,
+      "step": 80
+    },
+    {
+      "epoch": 0.2478134110787172,
+      "grad_norm": 0.5223561525344849,
+      "learning_rate": 2.930232558139535e-05,
+      "loss": 1.0891,
+      "step": 85
+    },
+    {
+      "epoch": 0.26239067055393583,
+      "grad_norm": 0.5522071719169617,
+      "learning_rate": 2.999974895005061e-05,
+      "loss": 1.0963,
+      "step": 90
+    },
+    {
+      "epoch": 0.27696793002915454,
+      "grad_norm": 0.5590180158615112,
+      "learning_rate": 2.9998214786347694e-05,
+      "loss": 1.0226,
+      "step": 95
+    },
+    {
+      "epoch": 0.2915451895043732,
+      "grad_norm": 0.6090812087059021,
+      "learning_rate": 2.9995286073611126e-05,
+      "loss": 1.0539,
+      "step": 100
+    },
+    {
+      "epoch": 0.30612244897959184,
+      "grad_norm": 0.4404645562171936,
+      "learning_rate": 2.9990963084155556e-05,
+      "loss": 1.017,
+      "step": 105
+    },
+    {
+      "epoch": 0.3206997084548105,
+      "grad_norm": 0.5025774240493774,
+      "learning_rate": 2.9985246219936873e-05,
+      "loss": 1.0265,
+      "step": 110
+    },
+    {
+      "epoch": 0.33527696793002915,
+      "grad_norm": 0.5865061283111572,
+      "learning_rate": 2.9978136012514836e-05,
+      "loss": 1.0595,
+      "step": 115
+    },
+    {
+      "epoch": 0.3498542274052478,
+      "grad_norm": 0.5579649209976196,
+      "learning_rate": 2.9969633123003653e-05,
+      "loss": 1.0304,
+      "step": 120
+    },
+    {
+      "epoch": 0.36443148688046645,
+      "grad_norm": 0.5952175259590149,
+      "learning_rate": 2.9959738342010516e-05,
+      "loss": 1.0135,
+      "step": 125
+    },
+    {
+      "epoch": 0.37900874635568516,
+      "grad_norm": 0.71574467420578,
+      "learning_rate": 2.9948452589562062e-05,
+      "loss": 1.0132,
+      "step": 130
+    },
+    {
+      "epoch": 0.3935860058309038,
+      "grad_norm": 0.601050853729248,
+      "learning_rate": 2.9935776915018857e-05,
+      "loss": 1.0407,
+      "step": 135
+    },
+    {
+      "epoch": 0.40816326530612246,
+      "grad_norm": 0.5946221351623535,
+      "learning_rate": 2.9921712496977826e-05,
+      "loss": 0.9446,
+      "step": 140
+    },
+    {
+      "epoch": 0.4227405247813411,
+      "grad_norm": 0.7047356963157654,
+      "learning_rate": 2.9906260643162636e-05,
+      "loss": 0.9796,
+      "step": 145
+    },
+    {
+      "epoch": 0.43731778425655976,
+      "grad_norm": 0.6650790572166443,
+      "learning_rate": 2.9889422790302142e-05,
+      "loss": 0.918,
+      "step": 150
+    },
+    {
+      "epoch": 0.4518950437317784,
+      "grad_norm": 0.6386116743087769,
+      "learning_rate": 2.987120050399676e-05,
+      "loss": 0.9514,
+      "step": 155
+    },
+    {
+      "epoch": 0.46647230320699706,
+      "grad_norm": 0.6894969344139099,
+      "learning_rate": 2.9851595478572928e-05,
+      "loss": 0.9443,
+      "step": 160
+    },
+    {
+      "epoch": 0.48104956268221577,
+      "grad_norm": 0.7509779334068298,
+      "learning_rate": 2.9830609536925545e-05,
+      "loss": 0.9415,
+      "step": 165
+    },
+    {
+      "epoch": 0.4956268221574344,
+      "grad_norm": 0.7854268550872803,
+      "learning_rate": 2.9808244630348486e-05,
+      "loss": 0.8831,
+      "step": 170
+    },
+    {
+      "epoch": 0.5102040816326531,
+      "grad_norm": 0.6071781516075134,
+      "learning_rate": 2.9784502838353168e-05,
+      "loss": 0.9035,
+      "step": 175
+    },
+    {
+      "epoch": 0.5247813411078717,
+      "grad_norm": 0.7060590982437134,
+      "learning_rate": 2.975938636847519e-05,
+      "loss": 0.8842,
+      "step": 180
+    },
+    {
+      "epoch": 0.5393586005830904,
+      "grad_norm": 0.7780089378356934,
+      "learning_rate": 2.9732897556069073e-05,
+      "loss": 0.9195,
+      "step": 185
+    },
+    {
+      "epoch": 0.5539358600583091,
+      "grad_norm": 0.7674872875213623,
+      "learning_rate": 2.970503886409114e-05,
+      "loss": 0.9115,
+      "step": 190
+    },
+    {
+      "epoch": 0.5685131195335277,
+      "grad_norm": 0.7817531228065491,
+      "learning_rate": 2.9675812882870458e-05,
+      "loss": 0.854,
+      "step": 195
+    },
+    {
+      "epoch": 0.5830903790087464,
+      "grad_norm": 0.7957063913345337,
+      "learning_rate": 2.9645222329868052e-05,
+      "loss": 0.9491,
+      "step": 200
+    },
+    {
+      "epoch": 0.597667638483965,
+      "grad_norm": 0.743044376373291,
+      "learning_rate": 2.9613270049424174e-05,
+      "loss": 0.8417,
+      "step": 205
+    },
+    {
+      "epoch": 0.6122448979591837,
+      "grad_norm": 0.9406312108039856,
+      "learning_rate": 2.9579959012493857e-05,
+      "loss": 0.8611,
+      "step": 210
+    },
+    {
+      "epoch": 0.6268221574344023,
+      "grad_norm": 0.768855094909668,
+      "learning_rate": 2.9545292316370693e-05,
+      "loss": 0.8575,
+      "step": 215
+    },
+    {
+      "epoch": 0.641399416909621,
+      "grad_norm": 0.7359038591384888,
+      "learning_rate": 2.9509273184398803e-05,
+      "loss": 0.806,
+      "step": 220
+    },
+    {
+      "epoch": 0.6559766763848397,
+      "grad_norm": 0.8091476559638977,
+      "learning_rate": 2.947190496567316e-05,
+      "loss": 0.8204,
+      "step": 225
+    },
+    {
+      "epoch": 0.6705539358600583,
+      "grad_norm": 0.8504658341407776,
+      "learning_rate": 2.943319113472817e-05,
+      "loss": 0.7735,
+      "step": 230
+    },
+    {
+      "epoch": 0.685131195335277,
+      "grad_norm": 0.8491882085800171,
+      "learning_rate": 2.939313529121461e-05,
+      "loss": 0.7801,
+      "step": 235
+    },
+    {
+      "epoch": 0.6997084548104956,
+      "grad_norm": 0.8185837268829346,
+      "learning_rate": 2.9351741159564928e-05,
+      "loss": 0.8103,
+      "step": 240
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.8639430403709412,
+      "learning_rate": 2.9309012588646956e-05,
+      "loss": 0.816,
+      "step": 245
+    },
+    {
+      "epoch": 0.7288629737609329,
+      "grad_norm": 0.8384712934494019,
+      "learning_rate": 2.926495355140601e-05,
+      "loss": 0.7593,
+      "step": 250
+    },
+    {
+      "epoch": 0.7434402332361516,
+      "grad_norm": 0.9397363662719727,
+      "learning_rate": 2.9219568144495503e-05,
+      "loss": 0.7771,
+      "step": 255
+    },
+    {
+      "epoch": 0.7580174927113703,
+      "grad_norm": 0.9188200831413269,
+      "learning_rate": 2.917286058789603e-05,
+      "loss": 0.8083,
+      "step": 260
+    },
+    {
+      "epoch": 0.7725947521865889,
+      "grad_norm": 0.9108596444129944,
+      "learning_rate": 2.9124835224522995e-05,
+      "loss": 0.7288,
+      "step": 265
+    },
+    {
+      "epoch": 0.7871720116618076,
+      "grad_norm": 1.0203593969345093,
+      "learning_rate": 2.907549651982278e-05,
+      "loss": 0.749,
+      "step": 270
+    },
+    {
+      "epoch": 0.8017492711370262,
+      "grad_norm": 0.7860879898071289,
+      "learning_rate": 2.9024849061357574e-05,
+      "loss": 0.696,
+      "step": 275
+    },
+    {
+      "epoch": 0.8163265306122449,
+      "grad_norm": 0.9177045226097107,
+      "learning_rate": 2.897289755837879e-05,
+      "loss": 0.729,
+      "step": 280
+    },
+    {
+      "epoch": 0.8309037900874635,
+      "grad_norm": 0.8526281118392944,
+      "learning_rate": 2.891964684138921e-05,
+      "loss": 0.7788,
+      "step": 285
+    },
+    {
+      "epoch": 0.8454810495626822,
+      "grad_norm": 0.8337252736091614,
+      "learning_rate": 2.8865101861693846e-05,
+      "loss": 0.7203,
+      "step": 290
+    },
+    {
+      "epoch": 0.8600583090379009,
+      "grad_norm": 0.9173249006271362,
+      "learning_rate": 2.880926769093954e-05,
+      "loss": 0.678,
+      "step": 295
+    },
+    {
+      "epoch": 0.8746355685131195,
+      "grad_norm": 0.9064881205558777,
+      "learning_rate": 2.8752149520643413e-05,
+      "loss": 0.7479,
+      "step": 300
+    },
+    {
+      "epoch": 0.8892128279883382,
+      "grad_norm": 0.8854787349700928,
+      "learning_rate": 2.8693752661710156e-05,
+      "loss": 0.7213,
+      "step": 305
+    },
+    {
+      "epoch": 0.9037900874635568,
+      "grad_norm": 0.8375863432884216,
+      "learning_rate": 2.86340825439382e-05,
+      "loss": 0.6664,
+      "step": 310
+    },
+    {
+      "epoch": 0.9183673469387755,
+      "grad_norm": 0.9675496220588684,
+      "learning_rate": 2.8573144715514866e-05,
+      "loss": 0.6982,
+      "step": 315
+    },
+    {
+      "epoch": 0.9329446064139941,
+      "grad_norm": 0.8814638257026672,
+      "learning_rate": 2.8510944842500476e-05,
+      "loss": 0.6316,
+      "step": 320
+    },
+    {
+      "epoch": 0.9475218658892128,
+      "grad_norm": 0.9256383776664734,
+      "learning_rate": 2.844748870830153e-05,
+      "loss": 0.7049,
+      "step": 325
+    },
+    {
+      "epoch": 0.9620991253644315,
+      "grad_norm": 1.1202212572097778,
+      "learning_rate": 2.8382782213132937e-05,
+      "loss": 0.6481,
+      "step": 330
+    },
+    {
+      "epoch": 0.9766763848396501,
+      "grad_norm": 0.8997995257377625,
+      "learning_rate": 2.831683137346944e-05,
+      "loss": 0.6606,
+      "step": 335
+    },
+    {
+      "epoch": 0.9912536443148688,
+      "grad_norm": 1.1372482776641846,
+      "learning_rate": 2.8249642321486167e-05,
+      "loss": 0.6664,
+      "step": 340
+    },
+    {
+      "epoch": 1.0058309037900874,
+      "grad_norm": 0.9555745720863342,
+      "learning_rate": 2.8181221304488468e-05,
+      "loss": 0.6206,
+      "step": 345
+    },
+    {
+      "epoch": 1.0204081632653061,
+      "grad_norm": 1.1468002796173096,
+      "learning_rate": 2.8111574684331043e-05,
+      "loss": 0.5763,
+      "step": 350
+    },
+    {
+      "epoch": 1.0349854227405249,
+      "grad_norm": 1.0075445175170898,
+      "learning_rate": 2.8040708936826387e-05,
+      "loss": 0.5748,
+      "step": 355
+    },
+    {
+      "epoch": 1.0495626822157433,
+      "grad_norm": 1.0628175735473633,
+      "learning_rate": 2.7968630651142687e-05,
+      "loss": 0.5727,
+      "step": 360
+    },
+    {
+      "epoch": 1.064139941690962,
+      "grad_norm": 1.005645751953125,
+      "learning_rate": 2.789534652919114e-05,
+      "loss": 0.5928,
+      "step": 365
+    },
+    {
+      "epoch": 1.0787172011661808,
+      "grad_norm": 0.9667788147926331,
+      "learning_rate": 2.782086338500282e-05,
+      "loss": 0.5614,
+      "step": 370
+    },
+    {
+      "epoch": 1.0932944606413995,
+      "grad_norm": 0.9596889019012451,
+      "learning_rate": 2.774518814409507e-05,
+      "loss": 0.5471,
+      "step": 375
+    },
+    {
+      "epoch": 1.1078717201166182,
+      "grad_norm": 0.946275532245636,
+      "learning_rate": 2.766832784282759e-05,
+      "loss": 0.576,
+      "step": 380
+    },
+    {
+      "epoch": 1.1224489795918366,
+      "grad_norm": 1.1329126358032227,
+      "learning_rate": 2.7590289627748178e-05,
+      "loss": 0.5563,
+      "step": 385
+    },
+    {
+      "epoch": 1.1370262390670554,
+      "grad_norm": 1.0845911502838135,
+      "learning_rate": 2.7511080754928245e-05,
+      "loss": 0.5721,
+      "step": 390
+    },
+    {
+      "epoch": 1.151603498542274,
+      "grad_norm": 0.975816011428833,
+      "learning_rate": 2.7430708589288116e-05,
+      "loss": 0.5517,
+      "step": 395
+    },
+    {
+      "epoch": 1.1661807580174928,
+      "grad_norm": 0.9972317814826965,
+      "learning_rate": 2.734918060391226e-05,
+      "loss": 0.53,
+      "step": 400
+    },
+    {
+      "epoch": 1.1807580174927113,
+      "grad_norm": 1.0469579696655273,
+      "learning_rate": 2.7266504379354413e-05,
+      "loss": 0.5437,
+      "step": 405
+    },
+    {
+      "epoch": 1.19533527696793,
+      "grad_norm": 1.0206043720245361,
+      "learning_rate": 2.7182687602932757e-05,
+      "loss": 0.5698,
+      "step": 410
+    },
+    {
+      "epoch": 1.2099125364431487,
+      "grad_norm": 1.0890380144119263,
+      "learning_rate": 2.7097738068015113e-05,
+      "loss": 0.493,
+      "step": 415
+    },
+    {
+      "epoch": 1.2244897959183674,
+      "grad_norm": 1.0657082796096802,
+      "learning_rate": 2.7011663673294323e-05,
+      "loss": 0.5195,
+      "step": 420
+    },
+    {
+      "epoch": 1.239067055393586,
+      "grad_norm": 1.072803020477295,
+      "learning_rate": 2.6924472422053827e-05,
+      "loss": 0.5031,
+      "step": 425
+    },
+    {
+      "epoch": 1.2536443148688048,
+      "grad_norm": 0.9221733212471008,
+      "learning_rate": 2.6836172421423498e-05,
+      "loss": 0.5224,
+      "step": 430
+    },
+    {
+      "epoch": 1.2682215743440233,
+      "grad_norm": 0.9522933959960938,
+      "learning_rate": 2.674677188162584e-05,
+      "loss": 0.5029,
+      "step": 435
+    },
+    {
+      "epoch": 1.282798833819242,
+      "grad_norm": 1.0837171077728271,
+      "learning_rate": 2.6656279115212598e-05,
+      "loss": 0.452,
+      "step": 440
+    },
+    {
+      "epoch": 1.2973760932944607,
+      "grad_norm": 1.0236507654190063,
+      "learning_rate": 2.656470253629182e-05,
+      "loss": 0.4919,
+      "step": 445
+    },
+    {
+      "epoch": 1.3119533527696792,
+      "grad_norm": 1.0839462280273438,
+      "learning_rate": 2.647205065974555e-05,
+      "loss": 0.5256,
+      "step": 450
+    },
+    {
+      "epoch": 1.3265306122448979,
+      "grad_norm": 0.9910500049591064,
+      "learning_rate": 2.6378332100438076e-05,
+      "loss": 0.4818,
+      "step": 455
+    },
+    {
+      "epoch": 1.3411078717201166,
+      "grad_norm": 1.0026817321777344,
+      "learning_rate": 2.6283555572414905e-05,
+      "loss": 0.4987,
+      "step": 460
+    },
+    {
+      "epoch": 1.3556851311953353,
+      "grad_norm": 1.0966472625732422,
+      "learning_rate": 2.6187729888092554e-05,
+      "loss": 0.4388,
+      "step": 465
+    },
+    {
+      "epoch": 1.370262390670554,
+      "grad_norm": 1.0502007007598877,
+      "learning_rate": 2.609086395743912e-05,
+      "loss": 0.5116,
+      "step": 470
+    },
+    {
+      "epoch": 1.3848396501457727,
+      "grad_norm": 1.0499721765518188,
+      "learning_rate": 2.599296678714588e-05,
+      "loss": 0.461,
+      "step": 475
+    },
+    {
+      "epoch": 1.3994169096209912,
+      "grad_norm": 1.0597882270812988,
+      "learning_rate": 2.589404747978977e-05,
+      "loss": 0.478,
+      "step": 480
+    },
+    {
+      "epoch": 1.41399416909621,
+      "grad_norm": 1.0780905485153198,
+      "learning_rate": 2.5794115232987077e-05,
+      "loss": 0.4958,
+      "step": 485
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 1.0490083694458008,
+      "learning_rate": 2.56931793385382e-05,
+      "loss": 0.4706,
+      "step": 490
+    },
+    {
+      "epoch": 1.4431486880466473,
+      "grad_norm": 1.036818504333496,
+      "learning_rate": 2.559124918156371e-05,
+      "loss": 0.4354,
+      "step": 495
+    },
+    {
+      "epoch": 1.4577259475218658,
+      "grad_norm": 0.9886294603347778,
+      "learning_rate": 2.54883342396317e-05,
+      "loss": 0.464,
+      "step": 500
+    },
+    {
+      "epoch": 1.4723032069970845,
+      "grad_norm": 1.025623083114624,
+      "learning_rate": 2.5384444081876553e-05,
+      "loss": 0.4564,
+      "step": 505
+    },
+    {
+      "epoch": 1.4868804664723032,
+      "grad_norm": 0.9342569708824158,
+      "learning_rate": 2.52795883681092e-05,
+      "loss": 0.4644,
+      "step": 510
+    },
+    {
+      "epoch": 1.501457725947522,
+      "grad_norm": 0.9936116933822632,
+      "learning_rate": 2.517377684791893e-05,
+      "loss": 0.4466,
+      "step": 515
+    },
+    {
+      "epoch": 1.5160349854227406,
+      "grad_norm": 1.079041600227356,
+      "learning_rate": 2.506701935976688e-05,
+      "loss": 0.4661,
+      "step": 520
+    },
+    {
+      "epoch": 1.5306122448979593,
+      "grad_norm": 1.1008844375610352,
+      "learning_rate": 2.4959325830071226e-05,
+      "loss": 0.4741,
+      "step": 525
+    },
+    {
+      "epoch": 1.5451895043731778,
+      "grad_norm": 1.1267269849777222,
+      "learning_rate": 2.4850706272284232e-05,
+      "loss": 0.4356,
+      "step": 530
+    },
+    {
+      "epoch": 1.5597667638483965,
+      "grad_norm": 1.0247701406478882,
+      "learning_rate": 2.474117078596118e-05,
+      "loss": 0.444,
+      "step": 535
+    },
+    {
+      "epoch": 1.574344023323615,
+      "grad_norm": 0.9392454624176025,
+      "learning_rate": 2.4630729555821303e-05,
+      "loss": 0.4584,
+      "step": 540
+    },
+    {
+      "epoch": 1.5889212827988337,
+      "grad_norm": 1.2563761472702026,
+      "learning_rate": 2.4519392850800803e-05,
+      "loss": 0.4415,
+      "step": 545
+    },
+    {
+      "epoch": 1.6034985422740524,
+      "grad_norm": 1.0794179439544678,
+      "learning_rate": 2.440717102309803e-05,
+      "loss": 0.433,
+      "step": 550
+    },
+    {
+      "epoch": 1.6180758017492711,
+      "grad_norm": 1.0182685852050781,
+      "learning_rate": 2.429407450721093e-05,
+      "loss": 0.4477,
+      "step": 555
+    },
+    {
+      "epoch": 1.6326530612244898,
+      "grad_norm": 1.053568720817566,
+      "learning_rate": 2.418011381896683e-05,
+      "loss": 0.4113,
+      "step": 560
+    },
+    {
+      "epoch": 1.6472303206997085,
+      "grad_norm": 1.0692411661148071,
+      "learning_rate": 2.406529955454467e-05,
+      "loss": 0.3787,
+      "step": 565
+    },
+    {
+      "epoch": 1.6618075801749272,
+      "grad_norm": 1.0086205005645752,
+      "learning_rate": 2.3949642389489768e-05,
+      "loss": 0.41,
+      "step": 570
+    },
+    {
+      "epoch": 1.6763848396501457,
+      "grad_norm": 0.9602847695350647,
+      "learning_rate": 2.3833153077721167e-05,
+      "loss": 0.3696,
+      "step": 575
+    },
+    {
+      "epoch": 1.6909620991253644,
+      "grad_norm": 1.245957851409912,
+      "learning_rate": 2.3715842450531762e-05,
+      "loss": 0.4136,
+      "step": 580
+    },
+    {
+      "epoch": 1.7055393586005831,
+      "grad_norm": 1.0041766166687012,
+      "learning_rate": 2.3597721415581168e-05,
+      "loss": 0.4137,
+      "step": 585
+    },
+    {
+      "epoch": 1.7201166180758016,
+      "grad_norm": 0.9883555769920349,
+      "learning_rate": 2.347880095588153e-05,
+      "loss": 0.3773,
+      "step": 590
+    },
+    {
+      "epoch": 1.7346938775510203,
+      "grad_norm": 1.1427297592163086,
+      "learning_rate": 2.3359092128776306e-05,
+      "loss": 0.4156,
+      "step": 595
+    },
+    {
+      "epoch": 1.749271137026239,
+      "grad_norm": 1.0740915536880493,
+      "learning_rate": 2.323860606491215e-05,
+      "loss": 0.3906,
+      "step": 600
+    },
+    {
+      "epoch": 1.7638483965014577,
+      "grad_norm": 1.234735369682312,
+      "learning_rate": 2.311735396720396e-05,
+      "loss": 0.3973,
+      "step": 605
+    },
+    {
+      "epoch": 1.7784256559766765,
+      "grad_norm": 1.2047481536865234,
+      "learning_rate": 2.2995347109793232e-05,
+      "loss": 0.3841,
+      "step": 610
+    },
+    {
+      "epoch": 1.7930029154518952,
+      "grad_norm": 1.037848711013794,
+      "learning_rate": 2.2872596836999775e-05,
+      "loss": 0.3893,
+      "step": 615
+    },
+    {
+      "epoch": 1.8075801749271136,
+      "grad_norm": 1.202088475227356,
+      "learning_rate": 2.2749114562266905e-05,
+      "loss": 0.367,
+      "step": 620
+    },
+    {
+      "epoch": 1.8221574344023324,
+      "grad_norm": 1.0675629377365112,
+      "learning_rate": 2.2624911767100214e-05,
+      "loss": 0.3805,
+      "step": 625
+    },
+    {
+      "epoch": 1.836734693877551,
+      "grad_norm": 1.180262804031372,
+      "learning_rate": 2.25e-05,
+      "loss": 0.369,
+      "step": 630
+    },
+    {
+      "epoch": 1.8513119533527695,
+      "grad_norm": 1.1085103750228882,
+      "learning_rate": 2.2374390875387487e-05,
+      "loss": 0.3716,
+      "step": 635
+    },
+    {
+      "epoch": 1.8658892128279883,
+      "grad_norm": 1.0706039667129517,
+      "learning_rate": 2.224809607252489e-05,
+      "loss": 0.3897,
+      "step": 640
+    },
+    {
+      "epoch": 1.880466472303207,
+      "grad_norm": 1.0980050563812256,
+      "learning_rate": 2.2121127334429493e-05,
+      "loss": 0.3699,
+      "step": 645
+    },
+    {
+      "epoch": 1.8950437317784257,
+      "grad_norm": 1.1098896265029907,
+      "learning_rate": 2.1993496466781732e-05,
+      "loss": 0.3688,
+      "step": 650
+    },
+    {
+      "epoch": 1.9096209912536444,
+      "grad_norm": 1.005852460861206,
+      "learning_rate": 2.1865215336827523e-05,
+      "loss": 0.3974,
+      "step": 655
+    },
+    {
+      "epoch": 1.924198250728863,
+      "grad_norm": 1.1395288705825806,
+      "learning_rate": 2.1736295872274817e-05,
+      "loss": 0.3663,
+      "step": 660
+    },
+    {
+      "epoch": 1.9387755102040818,
+      "grad_norm": 1.0027847290039062,
+      "learning_rate": 2.1606750060184572e-05,
+      "loss": 0.3596,
+      "step": 665
+    },
+    {
+      "epoch": 1.9533527696793003,
+      "grad_norm": 1.208714246749878,
+      "learning_rate": 2.147658994585614e-05,
+      "loss": 0.316,
+      "step": 670
+    },
+    {
+      "epoch": 1.967930029154519,
+      "grad_norm": 1.1107137203216553,
+      "learning_rate": 2.1345827631707328e-05,
+      "loss": 0.3528,
+      "step": 675
+    },
+    {
+      "epoch": 1.9825072886297375,
+      "grad_norm": 1.093211054801941,
+      "learning_rate": 2.1214475276149085e-05,
+      "loss": 0.3402,
+      "step": 680
+    },
+    {
+      "epoch": 1.9970845481049562,
+      "grad_norm": 1.1834017038345337,
+      "learning_rate": 2.1082545092455005e-05,
+      "loss": 0.3469,
+      "step": 685
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1715,
+  "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": 1.0426632848193618e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

44/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c254b3d46ec8067c6cdab367c531c1aaa33b78525c315ee480895cfe43964843
+size 7736

44/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)