Upload folder using huggingface_hub

22/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

22/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",
+    "gate_proj",
+    "o_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

22/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99a0d01d9babd652bd770639963a677a5ee7ae33ec48411ae3efaf6c6e0771c3
+size 791751704

22/latest

@@ -0,0 +1 @@
+global_step802

22/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68265894e968d48c9a2775aca9988b08271c8ef088a22d164427874a36715776
+size 15920

22/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e7fb1cfaf5f2aef6f53f905eb298ccb9b8f8668cd2c1837157ebf82ab9197d3
+size 15920

22/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:361fb85a90206f28823d5e4a459fa3051931a37ac6fd4562faf2e73b2ce6c165
+size 15920

22/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a8e03ba136013eb239948ecd0ae99ec883a4b7210b43726063b1cb0a1fb062b
+size 15920

22/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a63a0dd8e75a0a7fa3f7c84dfc91a13f4375eef72a842bcaebcc09c3eb3e52a
+size 15920

22/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9e91c78462377794d8bc5835b1652418b025373c7a73829ace32c4861d952d9
+size 15920

22/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091108e4fc0bc607a672a0547b8934d0ebd4fd259294946ba788920bf5e53131
+size 15920

22/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e0976dfe506f1901f3faae7a56f03ca93d21a85d0e7526c40e6fb44eb7af51c
+size 15920

22/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7db24bb4d06a39c5caf0f62c4087a0d195f389194ff960bab1fa1f914c64b5f4
+size 1064

22/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
+  }
+}

22/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
+}

22/trainer_state.json

@@ -0,0 +1,1154 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 802,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.012484394506866416,
+      "grad_norm": 1.250030279159546,
+      "learning_rate": 1.1881188118811881e-06,
+      "loss": 1.3024,
+      "step": 5
+    },
+    {
+      "epoch": 0.024968789013732832,
+      "grad_norm": 1.0841675996780396,
+      "learning_rate": 2.673267326732673e-06,
+      "loss": 1.3147,
+      "step": 10
+    },
+    {
+      "epoch": 0.03745318352059925,
+      "grad_norm": 0.6254027485847473,
+      "learning_rate": 4.158415841584159e-06,
+      "loss": 1.2397,
+      "step": 15
+    },
+    {
+      "epoch": 0.049937578027465665,
+      "grad_norm": 0.5402944684028625,
+      "learning_rate": 5.643564356435644e-06,
+      "loss": 1.2073,
+      "step": 20
+    },
+    {
+      "epoch": 0.062421972534332085,
+      "grad_norm": 0.47610044479370117,
+      "learning_rate": 7.128712871287129e-06,
+      "loss": 1.2178,
+      "step": 25
+    },
+    {
+      "epoch": 0.0749063670411985,
+      "grad_norm": 0.5858327746391296,
+      "learning_rate": 8.613861386138613e-06,
+      "loss": 1.238,
+      "step": 30
+    },
+    {
+      "epoch": 0.08739076154806492,
+      "grad_norm": 0.5472700595855713,
+      "learning_rate": 1.00990099009901e-05,
+      "loss": 1.2016,
+      "step": 35
+    },
+    {
+      "epoch": 0.09987515605493133,
+      "grad_norm": 0.44305941462516785,
+      "learning_rate": 1.1584158415841584e-05,
+      "loss": 1.184,
+      "step": 40
+    },
+    {
+      "epoch": 0.11235955056179775,
+      "grad_norm": 0.4975726902484894,
+      "learning_rate": 1.306930693069307e-05,
+      "loss": 1.2251,
+      "step": 45
+    },
+    {
+      "epoch": 0.12484394506866417,
+      "grad_norm": 0.43187275528907776,
+      "learning_rate": 1.4554455445544554e-05,
+      "loss": 1.1542,
+      "step": 50
+    },
+    {
+      "epoch": 0.1373283395755306,
+      "grad_norm": 0.4613569974899292,
+      "learning_rate": 1.6039603960396042e-05,
+      "loss": 1.1681,
+      "step": 55
+    },
+    {
+      "epoch": 0.149812734082397,
+      "grad_norm": 0.44121506810188293,
+      "learning_rate": 1.7524752475247524e-05,
+      "loss": 1.1475,
+      "step": 60
+    },
+    {
+      "epoch": 0.16229712858926343,
+      "grad_norm": 0.4488328993320465,
+      "learning_rate": 1.900990099009901e-05,
+      "loss": 1.1619,
+      "step": 65
+    },
+    {
+      "epoch": 0.17478152309612985,
+      "grad_norm": 0.44015923142433167,
+      "learning_rate": 2.0495049504950496e-05,
+      "loss": 1.0965,
+      "step": 70
+    },
+    {
+      "epoch": 0.18726591760299627,
+      "grad_norm": 0.44825753569602966,
+      "learning_rate": 2.198019801980198e-05,
+      "loss": 1.1127,
+      "step": 75
+    },
+    {
+      "epoch": 0.19975031210986266,
+      "grad_norm": 0.4191741347312927,
+      "learning_rate": 2.3465346534653467e-05,
+      "loss": 1.1326,
+      "step": 80
+    },
+    {
+      "epoch": 0.21223470661672908,
+      "grad_norm": 0.47760850191116333,
+      "learning_rate": 2.495049504950495e-05,
+      "loss": 1.1003,
+      "step": 85
+    },
+    {
+      "epoch": 0.2247191011235955,
+      "grad_norm": 0.5718854069709778,
+      "learning_rate": 2.6435643564356436e-05,
+      "loss": 1.1227,
+      "step": 90
+    },
+    {
+      "epoch": 0.23720349563046192,
+      "grad_norm": 0.6666268110275269,
+      "learning_rate": 2.792079207920792e-05,
+      "loss": 1.1288,
+      "step": 95
+    },
+    {
+      "epoch": 0.24968789013732834,
+      "grad_norm": 0.44064071774482727,
+      "learning_rate": 2.9405940594059407e-05,
+      "loss": 1.065,
+      "step": 100
+    },
+    {
+      "epoch": 0.26217228464419473,
+      "grad_norm": 0.5092717409133911,
+      "learning_rate": 2.999981623248469e-05,
+      "loss": 1.0766,
+      "step": 105
+    },
+    {
+      "epoch": 0.2746566791510612,
+      "grad_norm": 0.5116708278656006,
+      "learning_rate": 2.9998693225086163e-05,
+      "loss": 1.0636,
+      "step": 110
+    },
+    {
+      "epoch": 0.28714107365792757,
+      "grad_norm": 0.5135843753814697,
+      "learning_rate": 2.9996549379694418e-05,
+      "loss": 1.0543,
+      "step": 115
+    },
+    {
+      "epoch": 0.299625468164794,
+      "grad_norm": 0.5474386811256409,
+      "learning_rate": 2.9993384842223426e-05,
+      "loss": 1.0784,
+      "step": 120
+    },
+    {
+      "epoch": 0.3121098626716604,
+      "grad_norm": 0.5349007844924927,
+      "learning_rate": 2.998919982805729e-05,
+      "loss": 1.0306,
+      "step": 125
+    },
+    {
+      "epoch": 0.32459425717852686,
+      "grad_norm": 0.5236003994941711,
+      "learning_rate": 2.998399462203559e-05,
+      "loss": 1.0462,
+      "step": 130
+    },
+    {
+      "epoch": 0.33707865168539325,
+      "grad_norm": 0.5239087343215942,
+      "learning_rate": 2.9977769578433987e-05,
+      "loss": 1.0233,
+      "step": 135
+    },
+    {
+      "epoch": 0.3495630461922597,
+      "grad_norm": 0.6136004328727722,
+      "learning_rate": 2.9970525120940132e-05,
+      "loss": 0.9779,
+      "step": 140
+    },
+    {
+      "epoch": 0.3620474406991261,
+      "grad_norm": 0.6154010891914368,
+      "learning_rate": 2.9962261742624813e-05,
+      "loss": 0.9944,
+      "step": 145
+    },
+    {
+      "epoch": 0.37453183520599254,
+      "grad_norm": 0.5972864627838135,
+      "learning_rate": 2.995298000590839e-05,
+      "loss": 1.0074,
+      "step": 150
+    },
+    {
+      "epoch": 0.38701622971285893,
+      "grad_norm": 0.5776309370994568,
+      "learning_rate": 2.9942680542522535e-05,
+      "loss": 0.9963,
+      "step": 155
+    },
+    {
+      "epoch": 0.3995006242197253,
+      "grad_norm": 0.6151773929595947,
+      "learning_rate": 2.993136405346722e-05,
+      "loss": 0.9857,
+      "step": 160
+    },
+    {
+      "epoch": 0.41198501872659177,
+      "grad_norm": 0.6129782795906067,
+      "learning_rate": 2.991903130896301e-05,
+      "loss": 1.0118,
+      "step": 165
+    },
+    {
+      "epoch": 0.42446941323345816,
+      "grad_norm": 0.6217514872550964,
+      "learning_rate": 2.9905683148398642e-05,
+      "loss": 0.9757,
+      "step": 170
+    },
+    {
+      "epoch": 0.4369538077403246,
+      "grad_norm": 0.6364917755126953,
+      "learning_rate": 2.9891320480273886e-05,
+      "loss": 0.9542,
+      "step": 175
+    },
+    {
+      "epoch": 0.449438202247191,
+      "grad_norm": 0.6444506049156189,
+      "learning_rate": 2.9875944282137734e-05,
+      "loss": 0.968,
+      "step": 180
+    },
+    {
+      "epoch": 0.46192259675405745,
+      "grad_norm": 0.6949779987335205,
+      "learning_rate": 2.9859555600521838e-05,
+      "loss": 0.9255,
+      "step": 185
+    },
+    {
+      "epoch": 0.47440699126092384,
+      "grad_norm": 0.719308614730835,
+      "learning_rate": 2.9842155550869294e-05,
+      "loss": 0.9439,
+      "step": 190
+    },
+    {
+      "epoch": 0.4868913857677903,
+      "grad_norm": 0.6572843194007874,
+      "learning_rate": 2.982374531745873e-05,
+      "loss": 0.9059,
+      "step": 195
+    },
+    {
+      "epoch": 0.4993757802746567,
+      "grad_norm": 0.7524131536483765,
+      "learning_rate": 2.9804326153323683e-05,
+      "loss": 0.9322,
+      "step": 200
+    },
+    {
+      "epoch": 0.5118601747815231,
+      "grad_norm": 0.7650299072265625,
+      "learning_rate": 2.9783899380167345e-05,
+      "loss": 0.9085,
+      "step": 205
+    },
+    {
+      "epoch": 0.5243445692883895,
+      "grad_norm": 0.7098513841629028,
+      "learning_rate": 2.976246638827257e-05,
+      "loss": 0.9223,
+      "step": 210
+    },
+    {
+      "epoch": 0.5368289637952559,
+      "grad_norm": 0.8576691150665283,
+      "learning_rate": 2.974002863640727e-05,
+      "loss": 0.9254,
+      "step": 215
+    },
+    {
+      "epoch": 0.5493133583021224,
+      "grad_norm": 0.7496089339256287,
+      "learning_rate": 2.9716587651725134e-05,
+      "loss": 0.9314,
+      "step": 220
+    },
+    {
+      "epoch": 0.5617977528089888,
+      "grad_norm": 0.6800470948219299,
+      "learning_rate": 2.9692145029661664e-05,
+      "loss": 0.8435,
+      "step": 225
+    },
+    {
+      "epoch": 0.5742821473158551,
+      "grad_norm": 0.8700007796287537,
+      "learning_rate": 2.9666702433825614e-05,
+      "loss": 0.9101,
+      "step": 230
+    },
+    {
+      "epoch": 0.5867665418227216,
+      "grad_norm": 0.6640142798423767,
+      "learning_rate": 2.9640261595885735e-05,
+      "loss": 0.875,
+      "step": 235
+    },
+    {
+      "epoch": 0.599250936329588,
+      "grad_norm": 0.8278601169586182,
+      "learning_rate": 2.9612824315452943e-05,
+      "loss": 0.8479,
+      "step": 240
+    },
+    {
+      "epoch": 0.6117353308364545,
+      "grad_norm": 0.9126909375190735,
+      "learning_rate": 2.958439245995781e-05,
+      "loss": 0.8526,
+      "step": 245
+    },
+    {
+      "epoch": 0.6242197253433208,
+      "grad_norm": 0.8736037015914917,
+      "learning_rate": 2.955496796452347e-05,
+      "loss": 0.8113,
+      "step": 250
+    },
+    {
+      "epoch": 0.6367041198501873,
+      "grad_norm": 0.8721530437469482,
+      "learning_rate": 2.9524552831833924e-05,
+      "loss": 0.8612,
+      "step": 255
+    },
+    {
+      "epoch": 0.6491885143570537,
+      "grad_norm": 0.8261131048202515,
+      "learning_rate": 2.9493149131997713e-05,
+      "loss": 0.7762,
+      "step": 260
+    },
+    {
+      "epoch": 0.66167290886392,
+      "grad_norm": 0.8957151174545288,
+      "learning_rate": 2.946075900240704e-05,
+      "loss": 0.832,
+      "step": 265
+    },
+    {
+      "epoch": 0.6741573033707865,
+      "grad_norm": 0.916682779788971,
+      "learning_rate": 2.942738464759229e-05,
+      "loss": 0.8213,
+      "step": 270
+    },
+    {
+      "epoch": 0.686641697877653,
+      "grad_norm": 0.7609239220619202,
+      "learning_rate": 2.939302833907197e-05,
+      "loss": 0.8061,
+      "step": 275
+    },
+    {
+      "epoch": 0.6991260923845194,
+      "grad_norm": 0.8781178593635559,
+      "learning_rate": 2.9357692415198136e-05,
+      "loss": 0.7798,
+      "step": 280
+    },
+    {
+      "epoch": 0.7116104868913857,
+      "grad_norm": 0.8101415038108826,
+      "learning_rate": 2.932137928099722e-05,
+      "loss": 0.8564,
+      "step": 285
+    },
+    {
+      "epoch": 0.7240948813982522,
+      "grad_norm": 0.8110167980194092,
+      "learning_rate": 2.928409140800634e-05,
+      "loss": 0.8175,
+      "step": 290
+    },
+    {
+      "epoch": 0.7365792759051186,
+      "grad_norm": 0.7818710803985596,
+      "learning_rate": 2.9245831334105084e-05,
+      "loss": 0.7912,
+      "step": 295
+    },
+    {
+      "epoch": 0.7490636704119851,
+      "grad_norm": 0.8244704008102417,
+      "learning_rate": 2.9206601663342786e-05,
+      "loss": 0.7646,
+      "step": 300
+    },
+    {
+      "epoch": 0.7615480649188514,
+      "grad_norm": 0.8379747867584229,
+      "learning_rate": 2.916640506576128e-05,
+      "loss": 0.7671,
+      "step": 305
+    },
+    {
+      "epoch": 0.7740324594257179,
+      "grad_norm": 0.8790676593780518,
+      "learning_rate": 2.9125244277213176e-05,
+      "loss": 0.7688,
+      "step": 310
+    },
+    {
+      "epoch": 0.7865168539325843,
+      "grad_norm": 0.9378462433815002,
+      "learning_rate": 2.908312209917565e-05,
+      "loss": 0.7718,
+      "step": 315
+    },
+    {
+      "epoch": 0.7990012484394506,
+      "grad_norm": 0.8543297052383423,
+      "learning_rate": 2.904004139855978e-05,
+      "loss": 0.7663,
+      "step": 320
+    },
+    {
+      "epoch": 0.8114856429463171,
+      "grad_norm": 0.8526532053947449,
+      "learning_rate": 2.8996005107515405e-05,
+      "loss": 0.7911,
+      "step": 325
+    },
+    {
+      "epoch": 0.8239700374531835,
+      "grad_norm": 0.8875757455825806,
+      "learning_rate": 2.8951016223231567e-05,
+      "loss": 0.7511,
+      "step": 330
+    },
+    {
+      "epoch": 0.83645443196005,
+      "grad_norm": 0.8844619393348694,
+      "learning_rate": 2.890507780773252e-05,
+      "loss": 0.7234,
+      "step": 335
+    },
+    {
+      "epoch": 0.8489388264669163,
+      "grad_norm": 0.896385908126831,
+      "learning_rate": 2.8858192987669303e-05,
+      "loss": 0.699,
+      "step": 340
+    },
+    {
+      "epoch": 0.8614232209737828,
+      "grad_norm": 1.157642126083374,
+      "learning_rate": 2.881036495410696e-05,
+      "loss": 0.7825,
+      "step": 345
+    },
+    {
+      "epoch": 0.8739076154806492,
+      "grad_norm": 0.9201584458351135,
+      "learning_rate": 2.8761596962307343e-05,
+      "loss": 0.6774,
+      "step": 350
+    },
+    {
+      "epoch": 0.8863920099875156,
+      "grad_norm": 0.924287736415863,
+      "learning_rate": 2.871189233150754e-05,
+      "loss": 0.7057,
+      "step": 355
+    },
+    {
+      "epoch": 0.898876404494382,
+      "grad_norm": 0.9572568535804749,
+      "learning_rate": 2.8661254444693975e-05,
+      "loss": 0.7201,
+      "step": 360
+    },
+    {
+      "epoch": 0.9113607990012484,
+      "grad_norm": 1.010705828666687,
+      "learning_rate": 2.8609686748372155e-05,
+      "loss": 0.776,
+      "step": 365
+    },
+    {
+      "epoch": 0.9238451935081149,
+      "grad_norm": 0.8497414588928223,
+      "learning_rate": 2.8557192752332082e-05,
+      "loss": 0.7348,
+      "step": 370
+    },
+    {
+      "epoch": 0.9363295880149812,
+      "grad_norm": 0.9973917603492737,
+      "learning_rate": 2.8503776029409398e-05,
+      "loss": 0.6799,
+      "step": 375
+    },
+    {
+      "epoch": 0.9488139825218477,
+      "grad_norm": 1.0082824230194092,
+      "learning_rate": 2.844944021524217e-05,
+      "loss": 0.6975,
+      "step": 380
+    },
+    {
+      "epoch": 0.9612983770287141,
+      "grad_norm": 1.0351635217666626,
+      "learning_rate": 2.8394189008023482e-05,
+      "loss": 0.715,
+      "step": 385
+    },
+    {
+      "epoch": 0.9737827715355806,
+      "grad_norm": 0.9392898082733154,
+      "learning_rate": 2.833802616824972e-05,
+      "loss": 0.7083,
+      "step": 390
+    },
+    {
+      "epoch": 0.9862671660424469,
+      "grad_norm": 0.9130701422691345,
+      "learning_rate": 2.8280955518464587e-05,
+      "loss": 0.671,
+      "step": 395
+    },
+    {
+      "epoch": 0.9987515605493134,
+      "grad_norm": 0.8589998483657837,
+      "learning_rate": 2.8222980942999005e-05,
+      "loss": 0.6443,
+      "step": 400
+    },
+    {
+      "epoch": 1.0099875156054932,
+      "grad_norm": 1.179226279258728,
+      "learning_rate": 2.8164106387706675e-05,
+      "loss": 0.5978,
+      "step": 405
+    },
+    {
+      "epoch": 1.0224719101123596,
+      "grad_norm": 1.1383295059204102,
+      "learning_rate": 2.8104335859695545e-05,
+      "loss": 0.6234,
+      "step": 410
+    },
+    {
+      "epoch": 1.034956304619226,
+      "grad_norm": 1.0137205123901367,
+      "learning_rate": 2.8043673427055076e-05,
+      "loss": 0.5565,
+      "step": 415
+    },
+    {
+      "epoch": 1.0474406991260923,
+      "grad_norm": 0.9573396444320679,
+      "learning_rate": 2.7982123218579373e-05,
+      "loss": 0.5922,
+      "step": 420
+    },
+    {
+      "epoch": 1.0599250936329587,
+      "grad_norm": 0.9142056107521057,
+      "learning_rate": 2.7919689423486136e-05,
+      "loss": 0.5344,
+      "step": 425
+    },
+    {
+      "epoch": 1.0724094881398252,
+      "grad_norm": 1.0959941148757935,
+      "learning_rate": 2.785637629113157e-05,
+      "loss": 0.5991,
+      "step": 430
+    },
+    {
+      "epoch": 1.0848938826466916,
+      "grad_norm": 1.0628936290740967,
+      "learning_rate": 2.779218813072116e-05,
+      "loss": 0.6041,
+      "step": 435
+    },
+    {
+      "epoch": 1.097378277153558,
+      "grad_norm": 1.1268197298049927,
+      "learning_rate": 2.7727129311016346e-05,
+      "loss": 0.5696,
+      "step": 440
+    },
+    {
+      "epoch": 1.1098626716604245,
+      "grad_norm": 1.011999249458313,
+      "learning_rate": 2.7661204260037234e-05,
+      "loss": 0.5735,
+      "step": 445
+    },
+    {
+      "epoch": 1.122347066167291,
+      "grad_norm": 0.9339777827262878,
+      "learning_rate": 2.7594417464761165e-05,
+      "loss": 0.5435,
+      "step": 450
+    },
+    {
+      "epoch": 1.1348314606741572,
+      "grad_norm": 0.9832940697669983,
+      "learning_rate": 2.7526773470817353e-05,
+      "loss": 0.5349,
+      "step": 455
+    },
+    {
+      "epoch": 1.1473158551810236,
+      "grad_norm": 0.9407443404197693,
+      "learning_rate": 2.745827688217749e-05,
+      "loss": 0.5582,
+      "step": 460
+    },
+    {
+      "epoch": 1.15980024968789,
+      "grad_norm": 1.006485939025879,
+      "learning_rate": 2.7388932360842394e-05,
+      "loss": 0.5492,
+      "step": 465
+    },
+    {
+      "epoch": 1.1722846441947565,
+      "grad_norm": 0.997183084487915,
+      "learning_rate": 2.7318744626524704e-05,
+      "loss": 0.5265,
+      "step": 470
+    },
+    {
+      "epoch": 1.184769038701623,
+      "grad_norm": 0.9414803385734558,
+      "learning_rate": 2.7247718456327656e-05,
+      "loss": 0.5636,
+      "step": 475
+    },
+    {
+      "epoch": 1.1972534332084894,
+      "grad_norm": 0.9664161801338196,
+      "learning_rate": 2.717585868441992e-05,
+      "loss": 0.5622,
+      "step": 480
+    },
+    {
+      "epoch": 1.2097378277153559,
+      "grad_norm": 1.2472707033157349,
+      "learning_rate": 2.7103170201706615e-05,
+      "loss": 0.517,
+      "step": 485
+    },
+    {
+      "epoch": 1.2222222222222223,
+      "grad_norm": 1.0070255994796753,
+      "learning_rate": 2.7029657955496392e-05,
+      "loss": 0.538,
+      "step": 490
+    },
+    {
+      "epoch": 1.2347066167290885,
+      "grad_norm": 1.0936633348464966,
+      "learning_rate": 2.6955326949164738e-05,
+      "loss": 0.5338,
+      "step": 495
+    },
+    {
+      "epoch": 1.247191011235955,
+      "grad_norm": 1.0601006746292114,
+      "learning_rate": 2.688018224181341e-05,
+      "loss": 0.5381,
+      "step": 500
+    },
+    {
+      "epoch": 1.2596754057428214,
+      "grad_norm": 1.0741894245147705,
+      "learning_rate": 2.6804228947926125e-05,
+      "loss": 0.4972,
+      "step": 505
+    },
+    {
+      "epoch": 1.2721598002496879,
+      "grad_norm": 0.8921645283699036,
+      "learning_rate": 2.672747223702045e-05,
+      "loss": 0.5096,
+      "step": 510
+    },
+    {
+      "epoch": 1.2846441947565543,
+      "grad_norm": 0.9685996174812317,
+      "learning_rate": 2.6649917333295956e-05,
+      "loss": 0.5174,
+      "step": 515
+    },
+    {
+      "epoch": 1.2971285892634208,
+      "grad_norm": 0.9764025807380676,
+      "learning_rate": 2.657156951527865e-05,
+      "loss": 0.5165,
+      "step": 520
+    },
+    {
+      "epoch": 1.309612983770287,
+      "grad_norm": 1.0654433965682983,
+      "learning_rate": 2.649243411546171e-05,
+      "loss": 0.5041,
+      "step": 525
+    },
+    {
+      "epoch": 1.3220973782771535,
+      "grad_norm": 1.050934910774231,
+      "learning_rate": 2.6412516519942554e-05,
+      "loss": 0.5104,
+      "step": 530
+    },
+    {
+      "epoch": 1.33458177278402,
+      "grad_norm": 1.0428624153137207,
+      "learning_rate": 2.6331822168056234e-05,
+      "loss": 0.4698,
+      "step": 535
+    },
+    {
+      "epoch": 1.3470661672908864,
+      "grad_norm": 1.0803771018981934,
+      "learning_rate": 2.6250356552005234e-05,
+      "loss": 0.4493,
+      "step": 540
+    },
+    {
+      "epoch": 1.3595505617977528,
+      "grad_norm": 1.0904477834701538,
+      "learning_rate": 2.6168125216485666e-05,
+      "loss": 0.4573,
+      "step": 545
+    },
+    {
+      "epoch": 1.3720349563046192,
+      "grad_norm": 1.0958645343780518,
+      "learning_rate": 2.6085133758309887e-05,
+      "loss": 0.5288,
+      "step": 550
+    },
+    {
+      "epoch": 1.3845193508114857,
+      "grad_norm": 1.12440824508667,
+      "learning_rate": 2.6001387826025564e-05,
+      "loss": 0.485,
+      "step": 555
+    },
+    {
+      "epoch": 1.3970037453183521,
+      "grad_norm": 1.0093923807144165,
+      "learning_rate": 2.591689311953122e-05,
+      "loss": 0.4735,
+      "step": 560
+    },
+    {
+      "epoch": 1.4094881398252186,
+      "grad_norm": 0.8993324041366577,
+      "learning_rate": 2.5831655389688308e-05,
+      "loss": 0.4851,
+      "step": 565
+    },
+    {
+      "epoch": 1.4219725343320848,
+      "grad_norm": 1.028627634048462,
+      "learning_rate": 2.5745680437929774e-05,
+      "loss": 0.4648,
+      "step": 570
+    },
+    {
+      "epoch": 1.4344569288389513,
+      "grad_norm": 1.0453159809112549,
+      "learning_rate": 2.565897411586522e-05,
+      "loss": 0.4848,
+      "step": 575
+    },
+    {
+      "epoch": 1.4469413233458177,
+      "grad_norm": 1.0226070880889893,
+      "learning_rate": 2.557154232488261e-05,
+      "loss": 0.4758,
+      "step": 580
+    },
+    {
+      "epoch": 1.4594257178526842,
+      "grad_norm": 1.1143102645874023,
+      "learning_rate": 2.548339101574664e-05,
+      "loss": 0.4989,
+      "step": 585
+    },
+    {
+      "epoch": 1.4719101123595506,
+      "grad_norm": 1.159244179725647,
+      "learning_rate": 2.5394526188193706e-05,
+      "loss": 0.476,
+      "step": 590
+    },
+    {
+      "epoch": 1.484394506866417,
+      "grad_norm": 1.282088041305542,
+      "learning_rate": 2.5304953890523526e-05,
+      "loss": 0.492,
+      "step": 595
+    },
+    {
+      "epoch": 1.4968789013732833,
+      "grad_norm": 1.136864423751831,
+      "learning_rate": 2.521468021918753e-05,
+      "loss": 0.4168,
+      "step": 600
+    },
+    {
+      "epoch": 1.5093632958801497,
+      "grad_norm": 1.0521190166473389,
+      "learning_rate": 2.5123711318373886e-05,
+      "loss": 0.4764,
+      "step": 605
+    },
+    {
+      "epoch": 1.5218476903870162,
+      "grad_norm": 1.1445989608764648,
+      "learning_rate": 2.5032053379589327e-05,
+      "loss": 0.4669,
+      "step": 610
+    },
+    {
+      "epoch": 1.5343320848938826,
+      "grad_norm": 1.2416304349899292,
+      "learning_rate": 2.493971264123776e-05,
+      "loss": 0.4868,
+      "step": 615
+    },
+    {
+      "epoch": 1.546816479400749,
+      "grad_norm": 1.0602422952651978,
+      "learning_rate": 2.484669538819564e-05,
+      "loss": 0.4246,
+      "step": 620
+    },
+    {
+      "epoch": 1.5593008739076155,
+      "grad_norm": 1.0995105504989624,
+      "learning_rate": 2.4753007951384236e-05,
+      "loss": 0.4557,
+      "step": 625
+    },
+    {
+      "epoch": 1.571785268414482,
+      "grad_norm": 1.056087851524353,
+      "learning_rate": 2.4658656707338733e-05,
+      "loss": 0.4376,
+      "step": 630
+    },
+    {
+      "epoch": 1.5842696629213484,
+      "grad_norm": 1.069985270500183,
+      "learning_rate": 2.4563648077774217e-05,
+      "loss": 0.4328,
+      "step": 635
+    },
+    {
+      "epoch": 1.5967540574282149,
+      "grad_norm": 1.0622888803482056,
+      "learning_rate": 2.446798852914862e-05,
+      "loss": 0.4295,
+      "step": 640
+    },
+    {
+      "epoch": 1.6092384519350813,
+      "grad_norm": 1.1215320825576782,
+      "learning_rate": 2.4371684572222596e-05,
+      "loss": 0.4429,
+      "step": 645
+    },
+    {
+      "epoch": 1.6217228464419475,
+      "grad_norm": 0.973064661026001,
+      "learning_rate": 2.427474276161637e-05,
+      "loss": 0.4127,
+      "step": 650
+    },
+    {
+      "epoch": 1.634207240948814,
+      "grad_norm": 1.140210747718811,
+      "learning_rate": 2.417716969536365e-05,
+      "loss": 0.4128,
+      "step": 655
+    },
+    {
+      "epoch": 1.6466916354556804,
+      "grad_norm": 0.985823929309845,
+      "learning_rate": 2.4078972014462533e-05,
+      "loss": 0.4305,
+      "step": 660
+    },
+    {
+      "epoch": 1.6591760299625467,
+      "grad_norm": 1.1081101894378662,
+      "learning_rate": 2.398015640242351e-05,
+      "loss": 0.4178,
+      "step": 665
+    },
+    {
+      "epoch": 1.671660424469413,
+      "grad_norm": 0.9790671467781067,
+      "learning_rate": 2.3880729584814575e-05,
+      "loss": 0.4562,
+      "step": 670
+    },
+    {
+      "epoch": 1.6841448189762795,
+      "grad_norm": 1.1061886548995972,
+      "learning_rate": 2.3780698328803467e-05,
+      "loss": 0.407,
+      "step": 675
+    },
+    {
+      "epoch": 1.696629213483146,
+      "grad_norm": 0.9817659854888916,
+      "learning_rate": 2.368006944269709e-05,
+      "loss": 0.4274,
+      "step": 680
+    },
+    {
+      "epoch": 1.7091136079900124,
+      "grad_norm": 1.0625079870224,
+      "learning_rate": 2.3578849775478124e-05,
+      "loss": 0.3839,
+      "step": 685
+    },
+    {
+      "epoch": 1.7215980024968789,
+      "grad_norm": 1.0568345785140991,
+      "learning_rate": 2.3477046216338877e-05,
+      "loss": 0.3915,
+      "step": 690
+    },
+    {
+      "epoch": 1.7340823970037453,
+      "grad_norm": 1.054922342300415,
+      "learning_rate": 2.337466569421239e-05,
+      "loss": 0.426,
+      "step": 695
+    },
+    {
+      "epoch": 1.7465667915106118,
+      "grad_norm": 1.109113097190857,
+      "learning_rate": 2.327171517730084e-05,
+      "loss": 0.4058,
+      "step": 700
+    },
+    {
+      "epoch": 1.7590511860174782,
+      "grad_norm": 1.0435470342636108,
+      "learning_rate": 2.3168201672601263e-05,
+      "loss": 0.4703,
+      "step": 705
+    },
+    {
+      "epoch": 1.7715355805243447,
+      "grad_norm": 1.0479152202606201,
+      "learning_rate": 2.306413222542866e-05,
+      "loss": 0.3692,
+      "step": 710
+    },
+    {
+      "epoch": 1.7840199750312111,
+      "grad_norm": 1.0771406888961792,
+      "learning_rate": 2.2959513918936488e-05,
+      "loss": 0.397,
+      "step": 715
+    },
+    {
+      "epoch": 1.7965043695380774,
+      "grad_norm": 1.0804780721664429,
+      "learning_rate": 2.2854353873634536e-05,
+      "loss": 0.3722,
+      "step": 720
+    },
+    {
+      "epoch": 1.8089887640449438,
+      "grad_norm": 1.2521559000015259,
+      "learning_rate": 2.274865924690432e-05,
+      "loss": 0.426,
+      "step": 725
+    },
+    {
+      "epoch": 1.8214731585518102,
+      "grad_norm": 0.9907971024513245,
+      "learning_rate": 2.2642437232511926e-05,
+      "loss": 0.3528,
+      "step": 730
+    },
+    {
+      "epoch": 1.8339575530586767,
+      "grad_norm": 1.2269495725631714,
+      "learning_rate": 2.2535695060118392e-05,
+      "loss": 0.3983,
+      "step": 735
+    },
+    {
+      "epoch": 1.846441947565543,
+      "grad_norm": 0.960803210735321,
+      "learning_rate": 2.242843999478764e-05,
+      "loss": 0.3916,
+      "step": 740
+    },
+    {
+      "epoch": 1.8589263420724094,
+      "grad_norm": 1.123892068862915,
+      "learning_rate": 2.2320679336492e-05,
+      "loss": 0.3705,
+      "step": 745
+    },
+    {
+      "epoch": 1.8714107365792758,
+      "grad_norm": 0.9948912858963013,
+      "learning_rate": 2.221242041961539e-05,
+      "loss": 0.3702,
+      "step": 750
+    },
+    {
+      "epoch": 1.8838951310861423,
+      "grad_norm": 1.0175302028656006,
+      "learning_rate": 2.210367061245408e-05,
+      "loss": 0.3855,
+      "step": 755
+    },
+    {
+      "epoch": 1.8963795255930087,
+      "grad_norm": 0.9982346892356873,
+      "learning_rate": 2.199443731671522e-05,
+      "loss": 0.354,
+      "step": 760
+    },
+    {
+      "epoch": 1.9088639200998752,
+      "grad_norm": 1.0150700807571411,
+      "learning_rate": 2.1884727967013065e-05,
+      "loss": 0.413,
+      "step": 765
+    },
+    {
+      "epoch": 1.9213483146067416,
+      "grad_norm": 1.300173282623291,
+      "learning_rate": 2.1774550030362956e-05,
+      "loss": 0.352,
+      "step": 770
+    },
+    {
+      "epoch": 1.933832709113608,
+      "grad_norm": 0.9992619752883911,
+      "learning_rate": 2.1663911005673092e-05,
+      "loss": 0.4219,
+      "step": 775
+    },
+    {
+      "epoch": 1.9463171036204745,
+      "grad_norm": 1.2440812587738037,
+      "learning_rate": 2.1552818423234175e-05,
+      "loss": 0.3394,
+      "step": 780
+    },
+    {
+      "epoch": 1.958801498127341,
+      "grad_norm": 1.0696502923965454,
+      "learning_rate": 2.1441279844206846e-05,
+      "loss": 0.3864,
+      "step": 785
+    },
+    {
+      "epoch": 1.9712858926342074,
+      "grad_norm": 1.016789436340332,
+      "learning_rate": 2.1329302860107065e-05,
+      "loss": 0.3774,
+      "step": 790
+    },
+    {
+      "epoch": 1.9837702871410736,
+      "grad_norm": 1.07418692111969,
+      "learning_rate": 2.1216895092289457e-05,
+      "loss": 0.3181,
+      "step": 795
+    },
+    {
+      "epoch": 1.99625468164794,
+      "grad_norm": 0.966789186000824,
+      "learning_rate": 2.1104064191428533e-05,
+      "loss": 0.3103,
+      "step": 800
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 2005,
+  "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": 1.2804601097717023e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

22/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab908394fcc15545e42e27e85e35e75bf6a764a2a43772c9d28a145a1ae5c580
+size 7736

22/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)