Upload folder using huggingface_hub

README.md

@@ -1,3 +1,204 @@
----
-license: mit
----
+---
+library_name: peft
+base_model: meta-llama/Llama-2-13b-chat-hf
+---
+
+# 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 0.7.2.dev0

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "meta-llama/Llama-2-13b-chat-hf",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "embed_tokens",
+    "k_proj",
+    "up_proj",
+    "gate_proj",
+    "down_proj",
+    "q_proj",
+    "o_proj",
+    "lm_head",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfdadae59e529c475c0bf5895d8bbe37d7bb98b2225e991eb946a91590fa2387
+size 782986592

global_step349/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:227f508662e219d11f8b82f6dd5ed9dc3490f20c4498e2c8f216d011e55c6da0
+size 382651760

global_step349/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfff29c240990f8509f364047d7879f618215b1944fed3c284120b8be7d596dd
+size 382651760

global_step349/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92b06072360a0ecfea945c87eacc2207200c5da7c152a4fc7f40899bc7eac427
+size 533650

global_step349/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9b0d8cbdc52093ac282e16b065481fa39527a162a316a9660e085b460629aa2
+size 533650

latest

@@ -0,0 +1 @@
+global_step349

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f1e3a7e8971b5372df7dce7381e487c9905f7a3f760026a973dc5e0a1e529b8
+size 14448

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2659044d140b5f9af0de3c9e525b25957369b4a27b31ae5b1d3a5d194f0deb25
+size 14448

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:099f8364752ca7e27f992474edb7aa6d931962d79af42f7de3e8f1afef9e4fc0
+size 1064

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' '  + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

trainer_state.json

@@ -0,0 +1,528 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.995708154506438,
+  "eval_steps": 500,
+  "global_step": 349,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "grad_norm": 3.838202126167235,
+      "learning_rate": 9.997135147120633e-05,
+      "loss": 2.1255,
+      "step": 5
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 2.1177204859561205,
+      "learning_rate": 9.98854387143534e-05,
+      "loss": 1.7295,
+      "step": 10
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 1.4704605855835866,
+      "learning_rate": 9.974236018040474e-05,
+      "loss": 1.4756,
+      "step": 15
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 1.4984947152628136,
+      "learning_rate": 9.954227982894034e-05,
+      "loss": 1.3201,
+      "step": 20
+    },
+    {
+      "epoch": 0.21,
+      "grad_norm": 1.3923423053547814,
+      "learning_rate": 9.928542694026862e-05,
+      "loss": 1.28,
+      "step": 25
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 1.3739605464965448,
+      "learning_rate": 9.897209585268458e-05,
+      "loss": 1.2543,
+      "step": 30
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 1.275550734636967,
+      "learning_rate": 9.86026456251757e-05,
+      "loss": 1.2261,
+      "step": 35
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 1.3278212210523785,
+      "learning_rate": 9.817749962596115e-05,
+      "loss": 1.1967,
+      "step": 40
+    },
+    {
+      "epoch": 0.39,
+      "grad_norm": 1.407416916788669,
+      "learning_rate": 9.769714504733694e-05,
+      "loss": 1.1876,
+      "step": 45
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 1.4541097126165785,
+      "learning_rate": 9.716213234738215e-05,
+      "loss": 1.1845,
+      "step": 50
+    },
+    {
+      "epoch": 0.47,
+      "grad_norm": 1.4706662049445451,
+      "learning_rate": 9.657307461916635e-05,
+      "loss": 1.1681,
+      "step": 55
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 1.3719130683126355,
+      "learning_rate": 9.59306468881811e-05,
+      "loss": 1.1622,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 1.3153375215490886,
+      "learning_rate": 9.52355853388003e-05,
+      "loss": 1.1298,
+      "step": 65
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 1.488930807431193,
+      "learning_rate": 9.448868647065642e-05,
+      "loss": 1.1598,
+      "step": 70
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 1.2649466524851178,
+      "learning_rate": 9.369080618589864e-05,
+      "loss": 1.135,
+      "step": 75
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 1.4161938356640078,
+      "learning_rate": 9.284285880837946e-05,
+      "loss": 1.1325,
+      "step": 80
+    },
+    {
+      "epoch": 0.73,
+      "grad_norm": 1.418651947979571,
+      "learning_rate": 9.194581603589328e-05,
+      "loss": 1.1078,
+      "step": 85
+    },
+    {
+      "epoch": 0.77,
+      "grad_norm": 1.2926106311909347,
+      "learning_rate": 9.100070582666795e-05,
+      "loss": 1.1134,
+      "step": 90
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 1.387359204219206,
+      "learning_rate": 9.000861122138517e-05,
+      "loss": 1.0808,
+      "step": 95
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 1.5961791185416505,
+      "learning_rate": 8.897066910207958e-05,
+      "loss": 1.0832,
+      "step": 100
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 1.476620650207239,
+      "learning_rate": 8.788806888933881e-05,
+      "loss": 1.0843,
+      "step": 105
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 1.270735111137924,
+      "learning_rate": 8.676205117929752e-05,
+      "loss": 1.1029,
+      "step": 110
+    },
+    {
+      "epoch": 0.99,
+      "grad_norm": 1.425141501784203,
+      "learning_rate": 8.559390632198723e-05,
+      "loss": 1.101,
+      "step": 115
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.1002874374389648,
+      "eval_runtime": 69.0354,
+      "eval_samples_per_second": 4.346,
+      "eval_steps_per_second": 0.55,
+      "step": 116
+    },
+    {
+      "epoch": 1.03,
+      "grad_norm": 1.1791383340555186,
+      "learning_rate": 8.438497294267117e-05,
+      "loss": 1.0411,
+      "step": 120
+    },
+    {
+      "epoch": 1.07,
+      "grad_norm": 1.333740137623119,
+      "learning_rate": 8.313663640785839e-05,
+      "loss": 1.0033,
+      "step": 125
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 1.4447991614503835,
+      "learning_rate": 8.185032723775539e-05,
+      "loss": 1.035,
+      "step": 130
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 1.2844125525377348,
+      "learning_rate": 8.052751946697403e-05,
+      "loss": 0.9736,
+      "step": 135
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 1.2700768647818643,
+      "learning_rate": 7.916972895537471e-05,
+      "loss": 1.0296,
+      "step": 140
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 1.3154218666274633,
+      "learning_rate": 7.777851165098012e-05,
+      "loss": 1.0291,
+      "step": 145
+    },
+    {
+      "epoch": 1.29,
+      "grad_norm": 1.1964311130128484,
+      "learning_rate": 7.635546180695038e-05,
+      "loss": 0.9498,
+      "step": 150
+    },
+    {
+      "epoch": 1.33,
+      "grad_norm": 1.5080490468085683,
+      "learning_rate": 7.490221015466279e-05,
+      "loss": 0.9916,
+      "step": 155
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 1.4371536477549405,
+      "learning_rate": 7.342042203498951e-05,
+      "loss": 1.0462,
+      "step": 160
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 1.633026684473331,
+      "learning_rate": 7.191179548991507e-05,
+      "loss": 0.9906,
+      "step": 165
+    },
+    {
+      "epoch": 1.46,
+      "grad_norm": 1.5739785726895548,
+      "learning_rate": 7.037805931668005e-05,
+      "loss": 1.0287,
+      "step": 170
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.429154502412058,
+      "learning_rate": 6.882097108668132e-05,
+      "loss": 1.0154,
+      "step": 175
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 1.1785345522362023,
+      "learning_rate": 6.724231513139852e-05,
+      "loss": 1.0018,
+      "step": 180
+    },
+    {
+      "epoch": 1.59,
+      "grad_norm": 1.6392707933966606,
+      "learning_rate": 6.564390049765528e-05,
+      "loss": 0.9824,
+      "step": 185
+    },
+    {
+      "epoch": 1.63,
+      "grad_norm": 1.5381793424250634,
+      "learning_rate": 6.402755887455792e-05,
+      "loss": 0.9953,
+      "step": 190
+    },
+    {
+      "epoch": 1.67,
+      "grad_norm": 1.2412583778357746,
+      "learning_rate": 6.239514249448767e-05,
+      "loss": 0.9974,
+      "step": 195
+    },
+    {
+      "epoch": 1.72,
+      "grad_norm": 1.5172544424431558,
+      "learning_rate": 6.0748522010551215e-05,
+      "loss": 0.9531,
+      "step": 200
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 1.3293697921192489,
+      "learning_rate": 5.908958435292241e-05,
+      "loss": 0.9842,
+      "step": 205
+    },
+    {
+      "epoch": 1.8,
+      "grad_norm": 1.4362543686797302,
+      "learning_rate": 5.742023056653131e-05,
+      "loss": 0.9993,
+      "step": 210
+    },
+    {
+      "epoch": 1.85,
+      "grad_norm": 1.3658639776213064,
+      "learning_rate": 5.574237363257858e-05,
+      "loss": 0.9936,
+      "step": 215
+    },
+    {
+      "epoch": 1.89,
+      "grad_norm": 1.583719578671925,
+      "learning_rate": 5.4057936276371565e-05,
+      "loss": 0.9686,
+      "step": 220
+    },
+    {
+      "epoch": 1.93,
+      "grad_norm": 1.2596980255346535,
+      "learning_rate": 5.236884876399429e-05,
+      "loss": 1.0095,
+      "step": 225
+    },
+    {
+      "epoch": 1.97,
+      "grad_norm": 1.3013094438190969,
+      "learning_rate": 5.0677046690336096e-05,
+      "loss": 0.999,
+      "step": 230
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.0417044162750244,
+      "eval_runtime": 68.6388,
+      "eval_samples_per_second": 4.371,
+      "eval_steps_per_second": 0.554,
+      "step": 233
+    },
+    {
+      "epoch": 2.02,
+      "grad_norm": 1.2842579006806363,
+      "learning_rate": 4.898446876101379e-05,
+      "loss": 0.9169,
+      "step": 235
+    },
+    {
+      "epoch": 2.06,
+      "grad_norm": 1.3799527209106432,
+      "learning_rate": 4.729305457072913e-05,
+      "loss": 0.9044,
+      "step": 240
+    },
+    {
+      "epoch": 2.1,
+      "grad_norm": 1.332914263778945,
+      "learning_rate": 4.560474238060739e-05,
+      "loss": 0.8905,
+      "step": 245
+    },
+    {
+      "epoch": 2.15,
+      "grad_norm": 1.4483104158765132,
+      "learning_rate": 4.392146689706425e-05,
+      "loss": 0.895,
+      "step": 250
+    },
+    {
+      "epoch": 2.19,
+      "grad_norm": 1.2300017765319582,
+      "learning_rate": 4.224515705474603e-05,
+      "loss": 0.9144,
+      "step": 255
+    },
+    {
+      "epoch": 2.23,
+      "grad_norm": 1.5671642418549678,
+      "learning_rate": 4.057773380608411e-05,
+      "loss": 0.8692,
+      "step": 260
+    },
+    {
+      "epoch": 2.27,
+      "grad_norm": 1.1553793609487584,
+      "learning_rate": 3.892110791999649e-05,
+      "loss": 0.9092,
+      "step": 265
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 1.3993702614119785,
+      "learning_rate": 3.7277177792259114e-05,
+      "loss": 0.8869,
+      "step": 270
+    },
+    {
+      "epoch": 2.36,
+      "grad_norm": 1.512174508677905,
+      "learning_rate": 3.5647827270055945e-05,
+      "loss": 0.9104,
+      "step": 275
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 1.4126799043823064,
+      "learning_rate": 3.403492349320101e-05,
+      "loss": 0.8837,
+      "step": 280
+    },
+    {
+      "epoch": 2.45,
+      "grad_norm": 1.3750228704249976,
+      "learning_rate": 3.244031475450599e-05,
+      "loss": 0.9164,
+      "step": 285
+    },
+    {
+      "epoch": 2.49,
+      "grad_norm": 1.5268920387463116,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 0.8805,
+      "step": 290
+    },
+    {
+      "epoch": 2.53,
+      "grad_norm": 1.4206059480865851,
+      "learning_rate": 2.9313268643646986e-05,
+      "loss": 0.8942,
+      "step": 295
+    },
+    {
+      "epoch": 2.58,
+      "grad_norm": 1.4073831093700344,
+      "learning_rate": 2.7784414682304832e-05,
+      "loss": 0.9289,
+      "step": 300
+    },
+    {
+      "epoch": 2.62,
+      "grad_norm": 1.4564634651821706,
+      "learning_rate": 2.628101847438835e-05,
+      "loss": 0.8864,
+      "step": 305
+    },
+    {
+      "epoch": 2.66,
+      "grad_norm": 1.4217201527416836,
+      "learning_rate": 2.4804802823479613e-05,
+      "loss": 0.8947,
+      "step": 310
+    },
+    {
+      "epoch": 2.7,
+      "grad_norm": 1.5100802994235183,
+      "learning_rate": 2.3357459385841823e-05,
+      "loss": 0.8829,
+      "step": 315
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 1.4578776901613366,
+      "learning_rate": 2.194064673188089e-05,
+      "loss": 0.922,
+      "step": 320
+    },
+    {
+      "epoch": 2.79,
+      "grad_norm": 1.3183332342703558,
+      "learning_rate": 2.055598844552129e-05,
+      "loss": 0.8785,
+      "step": 325
+    },
+    {
+      "epoch": 2.83,
+      "grad_norm": 1.3169540028278115,
+      "learning_rate": 1.920507126367448e-05,
+      "loss": 0.8853,
+      "step": 330
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 1.4997550219337101,
+      "learning_rate": 1.7889443257931737e-05,
+      "loss": 0.8932,
+      "step": 335
+    },
+    {
+      "epoch": 2.92,
+      "grad_norm": 1.2868613064278502,
+      "learning_rate": 1.6610612060565234e-05,
+      "loss": 0.9047,
+      "step": 340
+    },
+    {
+      "epoch": 2.96,
+      "grad_norm": 1.3566421396199244,
+      "learning_rate": 1.5370043136870148e-05,
+      "loss": 0.9106,
+      "step": 345
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 1.0241451263427734,
+      "eval_runtime": 68.638,
+      "eval_samples_per_second": 4.371,
+      "eval_steps_per_second": 0.554,
+      "step": 349
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 464,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "total_flos": 2581092118822912.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efbd64b26de84bf1ed6654abf27d744c683c234fec81c3f0b51801a6ad1acd9a
+size 6072

zero_to_fp32.py

@@ -0,0 +1,587 @@
+#!/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):
+    """
+    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)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+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 _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()
+            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):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _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):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _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):
+    """
+    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``
+
+    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)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    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``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    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("-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)