Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,202 @@
+---
+base_model: Qwen/Qwen2.5-72B-Instruct
+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 0.12.0

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-72B-Instruct",
+  "bias": "none",
+  "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": 16,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "down_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj",
+    "gate_proj",
+    "q_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fe9a98ff979934742812fba760456491933a2331ec2f41758e87a07c17bc976
+size 210685048

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

global_step90/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:ab171d5adc7e392280480b1519dd2f4833d742f7a5d98af3e4a3dfb346e1c196
+size 315805264

global_step90/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96085c229d3e0ba5eae3a5a68f956f839694b90db836565ec8ce9d4136611b93
+size 315805264

global_step90/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99c70fe60994a96aa945650e7a5c20cbe9a15ca0a7c065882bd2c3f92d981697
+size 315805264

global_step90/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48b7550518e3297234d1d4b09e8721cd31363300062916ae1aed7d50b01e6761
+size 315805264

global_step90/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:412948439f14472486c9e02fb1a734f119e21f83f071702c6454abd133d70c09
+size 1113926

global_step90/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:794a00d01423e8bc0fad97caf2180fa36039ccf1e36131c5cb46f9901eb74ccf
+size 1113926

global_step90/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fd2375f17babaaf843d877516733051723dbea0a2b327a2c365a5f2851be197
+size 1113926

global_step90/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9da28fd6fc9d392c372ab942988d9b5405160e7ffd57ade9ef47f6656a6187e
+size 1113926

latest

@@ -0,0 +1 @@
+global_step90

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee97cd82dba4d425fdd8dfdb88d4a43d0d4b1979b5c81ab4a24914fb00d4f332
+size 15024

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91dad95440fb85dc4a31745642117165c1a72173b2e389679ea8c0b2b6fcd7e2
+size 15024

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98698326b023c2af02c94f18726ce52c7f7a6fe290734dd7edbe99bc807fcfa0
+size 15024

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:708e7c6b5bf8a327e688779ebc08830ce249928bcb1ff5c82b1b1d0bf6d2660b
+size 15024

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cc822f1bec298ebdebdeec13e889d39885b756a2342fdf9b4c05d554b19aa5f
+size 1064

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,711 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.8274111675126905,
+  "eval_steps": 15,
+  "global_step": 90,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02030456852791878,
+      "grad_norm": 2.874329090118408,
+      "learning_rate": 1.25e-05,
+      "loss": 3.984,
+      "step": 1
+    },
+    {
+      "epoch": 0.04060913705583756,
+      "grad_norm": 2.71229887008667,
+      "learning_rate": 2.5e-05,
+      "loss": 4.0724,
+      "step": 2
+    },
+    {
+      "epoch": 0.06091370558375635,
+      "grad_norm": 2.8235723972320557,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 4.0758,
+      "step": 3
+    },
+    {
+      "epoch": 0.08121827411167512,
+      "grad_norm": 2.7236318588256836,
+      "learning_rate": 5e-05,
+      "loss": 3.9899,
+      "step": 4
+    },
+    {
+      "epoch": 0.10152284263959391,
+      "grad_norm": 2.6509652137756348,
+      "learning_rate": 6.25e-05,
+      "loss": 4.0496,
+      "step": 5
+    },
+    {
+      "epoch": 0.1218274111675127,
+      "grad_norm": 2.71999454498291,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 3.9312,
+      "step": 6
+    },
+    {
+      "epoch": 0.14213197969543148,
+      "grad_norm": 2.423703193664551,
+      "learning_rate": 8.75e-05,
+      "loss": 3.8856,
+      "step": 7
+    },
+    {
+      "epoch": 0.16243654822335024,
+      "grad_norm": 2.5448386669158936,
+      "learning_rate": 0.0001,
+      "loss": 3.5434,
+      "step": 8
+    },
+    {
+      "epoch": 0.18274111675126903,
+      "grad_norm": 1.8756992816925049,
+      "learning_rate": 9.99872299773906e-05,
+      "loss": 3.0033,
+      "step": 9
+    },
+    {
+      "epoch": 0.20304568527918782,
+      "grad_norm": 0.8611156344413757,
+      "learning_rate": 9.994892643250147e-05,
+      "loss": 2.9103,
+      "step": 10
+    },
+    {
+      "epoch": 0.2233502538071066,
+      "grad_norm": 0.5660148859024048,
+      "learning_rate": 9.988510893081799e-05,
+      "loss": 2.6369,
+      "step": 11
+    },
+    {
+      "epoch": 0.2436548223350254,
+      "grad_norm": 0.5000542402267456,
+      "learning_rate": 9.979581007037776e-05,
+      "loss": 2.7784,
+      "step": 12
+    },
+    {
+      "epoch": 0.2639593908629442,
+      "grad_norm": 0.4509906768798828,
+      "learning_rate": 9.968107546511942e-05,
+      "loss": 2.4779,
+      "step": 13
+    },
+    {
+      "epoch": 0.28426395939086296,
+      "grad_norm": 0.4468601942062378,
+      "learning_rate": 9.95409637215831e-05,
+      "loss": 2.546,
+      "step": 14
+    },
+    {
+      "epoch": 0.30456852791878175,
+      "grad_norm": 0.47582972049713135,
+      "learning_rate": 9.937554640897413e-05,
+      "loss": 2.5675,
+      "step": 15
+    },
+    {
+      "epoch": 0.30456852791878175,
+      "eval_loss": 2.252539873123169,
+      "eval_runtime": 125.5198,
+      "eval_samples_per_second": 0.661,
+      "eval_steps_per_second": 0.167,
+      "step": 15
+    },
+    {
+      "epoch": 0.3248730964467005,
+      "grad_norm": 0.4718024730682373,
+      "learning_rate": 9.918490802260538e-05,
+      "loss": 2.3554,
+      "step": 16
+    },
+    {
+      "epoch": 0.34517766497461927,
+      "grad_norm": 0.4537792205810547,
+      "learning_rate": 9.896914594073703e-05,
+      "loss": 2.3913,
+      "step": 17
+    },
+    {
+      "epoch": 0.36548223350253806,
+      "grad_norm": 0.44835665822029114,
+      "learning_rate": 9.87283703748356e-05,
+      "loss": 2.3686,
+      "step": 18
+    },
+    {
+      "epoch": 0.38578680203045684,
+      "grad_norm": 0.46018558740615845,
+      "learning_rate": 9.846270431327793e-05,
+      "loss": 2.3555,
+      "step": 19
+    },
+    {
+      "epoch": 0.40609137055837563,
+      "grad_norm": 0.4601113200187683,
+      "learning_rate": 9.817228345852851e-05,
+      "loss": 2.3785,
+      "step": 20
+    },
+    {
+      "epoch": 0.4263959390862944,
+      "grad_norm": 0.480472594499588,
+      "learning_rate": 9.785725615782262e-05,
+      "loss": 2.278,
+      "step": 21
+    },
+    {
+      "epoch": 0.4467005076142132,
+      "grad_norm": 0.5582306981086731,
+      "learning_rate": 9.751778332739033e-05,
+      "loss": 2.4215,
+      "step": 22
+    },
+    {
+      "epoch": 0.467005076142132,
+      "grad_norm": 0.547773003578186,
+      "learning_rate": 9.715403837026046e-05,
+      "loss": 2.3028,
+      "step": 23
+    },
+    {
+      "epoch": 0.4873096446700508,
+      "grad_norm": 0.5253167748451233,
+      "learning_rate": 9.676620708768607e-05,
+      "loss": 2.3265,
+      "step": 24
+    },
+    {
+      "epoch": 0.5076142131979695,
+      "grad_norm": 0.5330803394317627,
+      "learning_rate": 9.635448758423702e-05,
+      "loss": 2.1958,
+      "step": 25
+    },
+    {
+      "epoch": 0.5279187817258884,
+      "grad_norm": 0.5347826480865479,
+      "learning_rate": 9.591909016660807e-05,
+      "loss": 2.1515,
+      "step": 26
+    },
+    {
+      "epoch": 0.5482233502538071,
+      "grad_norm": 0.5064857602119446,
+      "learning_rate": 9.546023723619386e-05,
+      "loss": 2.23,
+      "step": 27
+    },
+    {
+      "epoch": 0.5685279187817259,
+      "grad_norm": 0.5688650012016296,
+      "learning_rate": 9.497816317548625e-05,
+      "loss": 2.0931,
+      "step": 28
+    },
+    {
+      "epoch": 0.5888324873096447,
+      "grad_norm": 0.5568444728851318,
+      "learning_rate": 9.447311422835142e-05,
+      "loss": 2.2783,
+      "step": 29
+    },
+    {
+      "epoch": 0.6091370558375635,
+      "grad_norm": 0.5993527770042419,
+      "learning_rate": 9.39453483742483e-05,
+      "loss": 2.2624,
+      "step": 30
+    },
+    {
+      "epoch": 0.6091370558375635,
+      "eval_loss": 2.009953737258911,
+      "eval_runtime": 125.0198,
+      "eval_samples_per_second": 0.664,
+      "eval_steps_per_second": 0.168,
+      "step": 30
+    },
+    {
+      "epoch": 0.6294416243654822,
+      "grad_norm": 0.556568443775177,
+      "learning_rate": 9.339513519645249e-05,
+      "loss": 2.265,
+      "step": 31
+    },
+    {
+      "epoch": 0.649746192893401,
+      "grad_norm": 0.5960446000099182,
+      "learning_rate": 9.282275574435281e-05,
+      "loss": 2.1947,
+      "step": 32
+    },
+    {
+      "epoch": 0.6700507614213198,
+      "grad_norm": 0.5874341130256653,
+      "learning_rate": 9.222850238989103e-05,
+      "loss": 2.2951,
+      "step": 33
+    },
+    {
+      "epoch": 0.6903553299492385,
+      "grad_norm": 0.582109808921814,
+      "learning_rate": 9.161267867821802e-05,
+      "loss": 2.1387,
+      "step": 34
+    },
+    {
+      "epoch": 0.7106598984771574,
+      "grad_norm": 0.6334108114242554,
+      "learning_rate": 9.097559917264267e-05,
+      "loss": 2.2382,
+      "step": 35
+    },
+    {
+      "epoch": 0.7309644670050761,
+      "grad_norm": 0.5571526885032654,
+      "learning_rate": 9.031758929395258e-05,
+      "loss": 2.1604,
+      "step": 36
+    },
+    {
+      "epoch": 0.751269035532995,
+      "grad_norm": 0.5545266270637512,
+      "learning_rate": 8.963898515418884e-05,
+      "loss": 2.0082,
+      "step": 37
+    },
+    {
+      "epoch": 0.7715736040609137,
+      "grad_norm": 0.5715925693511963,
+      "learning_rate": 8.894013338495981e-05,
+      "loss": 2.2066,
+      "step": 38
+    },
+    {
+      "epoch": 0.7918781725888325,
+      "grad_norm": 0.5664649605751038,
+      "learning_rate": 8.82213909603812e-05,
+      "loss": 2.1361,
+      "step": 39
+    },
+    {
+      "epoch": 0.8121827411167513,
+      "grad_norm": 0.5984413623809814,
+      "learning_rate": 8.74831250147335e-05,
+      "loss": 2.0842,
+      "step": 40
+    },
+    {
+      "epoch": 0.8324873096446701,
+      "grad_norm": 0.6048261523246765,
+      "learning_rate": 8.672571265492943e-05,
+      "loss": 2.1793,
+      "step": 41
+    },
+    {
+      "epoch": 0.8527918781725888,
+      "grad_norm": 0.6012648940086365,
+      "learning_rate": 8.594954076788736e-05,
+      "loss": 2.0668,
+      "step": 42
+    },
+    {
+      "epoch": 0.8730964467005076,
+      "grad_norm": 0.6133915781974792,
+      "learning_rate": 8.515500582290914e-05,
+      "loss": 2.2131,
+      "step": 43
+    },
+    {
+      "epoch": 0.8934010152284264,
+      "grad_norm": 0.5678202509880066,
+      "learning_rate": 8.434251366916322e-05,
+      "loss": 2.1902,
+      "step": 44
+    },
+    {
+      "epoch": 0.9137055837563451,
+      "grad_norm": 0.6124091744422913,
+      "learning_rate": 8.351247932837655e-05,
+      "loss": 2.1055,
+      "step": 45
+    },
+    {
+      "epoch": 0.9137055837563451,
+      "eval_loss": 1.9042185544967651,
+      "eval_runtime": 124.2495,
+      "eval_samples_per_second": 0.668,
+      "eval_steps_per_second": 0.169,
+      "step": 45
+    },
+    {
+      "epoch": 0.934010152284264,
+      "grad_norm": 0.6008590459823608,
+      "learning_rate": 8.266532678284103e-05,
+      "loss": 1.9734,
+      "step": 46
+    },
+    {
+      "epoch": 0.9543147208121827,
+      "grad_norm": 0.6411317586898804,
+      "learning_rate": 8.18014887588431e-05,
+      "loss": 2.0015,
+      "step": 47
+    },
+    {
+      "epoch": 0.9746192893401016,
+      "grad_norm": 0.6390472650527954,
+      "learning_rate": 8.092140650562665e-05,
+      "loss": 2.0936,
+      "step": 48
+    },
+    {
+      "epoch": 0.9949238578680203,
+      "grad_norm": 0.6495903730392456,
+      "learning_rate": 8.002552957000254e-05,
+      "loss": 1.9886,
+      "step": 49
+    },
+    {
+      "epoch": 1.015228426395939,
+      "grad_norm": 1.22446870803833,
+      "learning_rate": 7.911431556671968e-05,
+      "loss": 3.395,
+      "step": 50
+    },
+    {
+      "epoch": 1.0355329949238579,
+      "grad_norm": 0.606292188167572,
+      "learning_rate": 7.818822994471505e-05,
+      "loss": 1.8093,
+      "step": 51
+    },
+    {
+      "epoch": 1.0558375634517767,
+      "grad_norm": 0.6382877230644226,
+      "learning_rate": 7.724774574936188e-05,
+      "loss": 2.0305,
+      "step": 52
+    },
+    {
+      "epoch": 1.0761421319796955,
+      "grad_norm": 0.661160945892334,
+      "learning_rate": 7.629334338083773e-05,
+      "loss": 1.9056,
+      "step": 53
+    },
+    {
+      "epoch": 1.0964467005076142,
+      "grad_norm": 0.6464510560035706,
+      "learning_rate": 7.532551034873559e-05,
+      "loss": 1.9847,
+      "step": 54
+    },
+    {
+      "epoch": 1.116751269035533,
+      "grad_norm": 0.6648440361022949,
+      "learning_rate": 7.43447410230435e-05,
+      "loss": 1.9907,
+      "step": 55
+    },
+    {
+      "epoch": 1.1370558375634519,
+      "grad_norm": 0.6609318256378174,
+      "learning_rate": 7.335153638162005e-05,
+      "loss": 1.9298,
+      "step": 56
+    },
+    {
+      "epoch": 1.1573604060913705,
+      "grad_norm": 0.6394674777984619,
+      "learning_rate": 7.234640375429427e-05,
+      "loss": 1.9282,
+      "step": 57
+    },
+    {
+      "epoch": 1.1776649746192893,
+      "grad_norm": 0.7306097149848938,
+      "learning_rate": 7.132985656372126e-05,
+      "loss": 1.9263,
+      "step": 58
+    },
+    {
+      "epoch": 1.1979695431472082,
+      "grad_norm": 0.665061354637146,
+      "learning_rate": 7.030241406312528e-05,
+      "loss": 1.7979,
+      "step": 59
+    },
+    {
+      "epoch": 1.218274111675127,
+      "grad_norm": 0.7927564978599548,
+      "learning_rate": 6.926460107106482e-05,
+      "loss": 2.1949,
+      "step": 60
+    },
+    {
+      "epoch": 1.218274111675127,
+      "eval_loss": 1.8604909181594849,
+      "eval_runtime": 124.3295,
+      "eval_samples_per_second": 0.668,
+      "eval_steps_per_second": 0.169,
+      "step": 60
+    },
+    {
+      "epoch": 1.2385786802030456,
+      "grad_norm": 0.7951006889343262,
+      "learning_rate": 6.821694770335481e-05,
+      "loss": 1.9729,
+      "step": 61
+    },
+    {
+      "epoch": 1.2588832487309645,
+      "grad_norm": 0.7578277587890625,
+      "learning_rate": 6.715998910228296e-05,
+      "loss": 1.7568,
+      "step": 62
+    },
+    {
+      "epoch": 1.2791878172588833,
+      "grad_norm": 0.8395410180091858,
+      "learning_rate": 6.60942651632586e-05,
+      "loss": 2.0653,
+      "step": 63
+    },
+    {
+      "epoch": 1.299492385786802,
+      "grad_norm": 0.7950598001480103,
+      "learning_rate": 6.502032025903355e-05,
+      "loss": 1.7411,
+      "step": 64
+    },
+    {
+      "epoch": 1.3197969543147208,
+      "grad_norm": 0.9455265998840332,
+      "learning_rate": 6.393870296163615e-05,
+      "loss": 2.0562,
+      "step": 65
+    },
+    {
+      "epoch": 1.3401015228426396,
+      "grad_norm": 0.8618624806404114,
+      "learning_rate": 6.284996576216014e-05,
+      "loss": 1.7543,
+      "step": 66
+    },
+    {
+      "epoch": 1.3604060913705585,
+      "grad_norm": 0.8657338619232178,
+      "learning_rate": 6.17546647885516e-05,
+      "loss": 2.0544,
+      "step": 67
+    },
+    {
+      "epoch": 1.380710659898477,
+      "grad_norm": 0.8475260138511658,
+      "learning_rate": 6.065335952153845e-05,
+      "loss": 1.9429,
+      "step": 68
+    },
+    {
+      "epoch": 1.401015228426396,
+      "grad_norm": 0.8515224456787109,
+      "learning_rate": 5.954661250884703e-05,
+      "loss": 1.7757,
+      "step": 69
+    },
+    {
+      "epoch": 1.4213197969543148,
+      "grad_norm": 0.9630141258239746,
+      "learning_rate": 5.8434989077852366e-05,
+      "loss": 2.0841,
+      "step": 70
+    },
+    {
+      "epoch": 1.4416243654822334,
+      "grad_norm": 0.8603659868240356,
+      "learning_rate": 5.7319057046808335e-05,
+      "loss": 1.6943,
+      "step": 71
+    },
+    {
+      "epoch": 1.4619289340101522,
+      "grad_norm": 0.9940032362937927,
+      "learning_rate": 5.619938643480561e-05,
+      "loss": 2.0689,
+      "step": 72
+    },
+    {
+      "epoch": 1.482233502538071,
+      "grad_norm": 0.8981780409812927,
+      "learning_rate": 5.5076549170605406e-05,
+      "loss": 1.811,
+      "step": 73
+    },
+    {
+      "epoch": 1.50253807106599,
+      "grad_norm": 0.8641999363899231,
+      "learning_rate": 5.3951118800497745e-05,
+      "loss": 1.8436,
+      "step": 74
+    },
+    {
+      "epoch": 1.5228426395939088,
+      "grad_norm": 0.8729159832000732,
+      "learning_rate": 5.2823670195333494e-05,
+      "loss": 1.7436,
+      "step": 75
+    },
+    {
+      "epoch": 1.5228426395939088,
+      "eval_loss": 1.8201268911361694,
+      "eval_runtime": 124.1955,
+      "eval_samples_per_second": 0.668,
+      "eval_steps_per_second": 0.169,
+      "step": 75
+    },
+    {
+      "epoch": 1.5431472081218274,
+      "grad_norm": 0.9512869715690613,
+      "learning_rate": 5.169477925687981e-05,
+      "loss": 1.7664,
+      "step": 76
+    },
+    {
+      "epoch": 1.5634517766497462,
+      "grad_norm": 0.9679103493690491,
+      "learning_rate": 5.0565022623648995e-05,
+      "loss": 1.9316,
+      "step": 77
+    },
+    {
+      "epoch": 1.5837563451776648,
+      "grad_norm": 0.9876412153244019,
+      "learning_rate": 4.9434977376351023e-05,
+      "loss": 1.9186,
+      "step": 78
+    },
+    {
+      "epoch": 1.6040609137055837,
+      "grad_norm": 0.9529967904090881,
+      "learning_rate": 4.8305220743120186e-05,
+      "loss": 1.7784,
+      "step": 79
+    },
+    {
+      "epoch": 1.6243654822335025,
+      "grad_norm": 1.0214565992355347,
+      "learning_rate": 4.717632980466651e-05,
+      "loss": 1.8424,
+      "step": 80
+    },
+    {
+      "epoch": 1.6446700507614214,
+      "grad_norm": 0.9737630486488342,
+      "learning_rate": 4.6048881199502267e-05,
+      "loss": 1.8476,
+      "step": 81
+    },
+    {
+      "epoch": 1.6649746192893402,
+      "grad_norm": 1.0335719585418701,
+      "learning_rate": 4.4923450829394605e-05,
+      "loss": 1.9076,
+      "step": 82
+    },
+    {
+      "epoch": 1.6852791878172588,
+      "grad_norm": 1.0101923942565918,
+      "learning_rate": 4.380061356519441e-05,
+      "loss": 1.829,
+      "step": 83
+    },
+    {
+      "epoch": 1.7055837563451777,
+      "grad_norm": 0.9788248538970947,
+      "learning_rate": 4.268094295319167e-05,
+      "loss": 1.7334,
+      "step": 84
+    },
+    {
+      "epoch": 1.7258883248730963,
+      "grad_norm": 1.0588387250900269,
+      "learning_rate": 4.1565010922147646e-05,
+      "loss": 2.0164,
+      "step": 85
+    },
+    {
+      "epoch": 1.7461928934010151,
+      "grad_norm": 0.9743729829788208,
+      "learning_rate": 4.045338749115298e-05,
+      "loss": 1.7956,
+      "step": 86
+    },
+    {
+      "epoch": 1.766497461928934,
+      "grad_norm": 1.0397388935089111,
+      "learning_rate": 3.934664047846156e-05,
+      "loss": 1.7431,
+      "step": 87
+    },
+    {
+      "epoch": 1.7868020304568528,
+      "grad_norm": 1.1108192205429077,
+      "learning_rate": 3.82453352114484e-05,
+      "loss": 1.8768,
+      "step": 88
+    },
+    {
+      "epoch": 1.8071065989847717,
+      "grad_norm": 1.0383998155593872,
+      "learning_rate": 3.715003423783986e-05,
+      "loss": 1.8673,
+      "step": 89
+    },
+    {
+      "epoch": 1.8274111675126905,
+      "grad_norm": 0.9685633182525635,
+      "learning_rate": 3.606129703836385e-05,
+      "loss": 1.812,
+      "step": 90
+    },
+    {
+      "epoch": 1.8274111675126905,
+      "eval_loss": 1.7956775426864624,
+      "eval_runtime": 124.2818,
+      "eval_samples_per_second": 0.668,
+      "eval_steps_per_second": 0.169,
+      "step": 90
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 147,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 15,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 190111214469120.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83206bb0df029e282009b5077b1107ab542c5099dfc8397b86db40204a8cf790
+size 7352

zero_to_fp32.py

@@ -0,0 +1,674 @@
+#!/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 . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import json
+from tqdm import tqdm
+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 tqdm(param_shapes.items(), desc='Gathering Sharded Weights'):
+        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_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               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_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``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
+    """
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        state_dict_split = split_torch_state_dict_into_shards(state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard = {tensor: state_dict[tensor].contiguous() for tensor in tensors}
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard, output_path)
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+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_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    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_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)