Uploading /ephemeral/hossein/output/newData-progressive-yoco-tiny-llama-CDL-20

README.md

@@ -0,0 +1,61 @@
+---
+library_name: transformers
+license: other
+base_model: TinyLlama/TinyLlama_v1.1
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: newData-progressive-yoco-tiny-llama-CDL-20
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# newData-progressive-yoco-tiny-llama-CDL-20
+
+This model is a fine-tuned version of [/ephemeral/hossein/output/newData-progressive-yoco-tiny-llama-CDL-21/checkpoint-50](https://huggingface.co//ephemeral/hossein/output/newData-progressive-yoco-tiny-llama-CDL-21/checkpoint-50) on the alpaca_reformatted, the UltraInteract_sft_reformatted, the reformatted_ultrachat_200k, the reformatted_MathInstruct and the small_slim_pajama datasets.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 58
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 1856
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.005
+- training_steps: 50
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.45.2
+- Pytorch 2.5.1+cu124
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.05959475566150179,
+    "total_flos": 106890730143744.0,
+    "train_loss": 1.9756734085083008,
+    "train_runtime": 1598.4061,
+    "train_samples_per_second": 58.058,
+    "train_steps_per_second": 0.031
+}

checkpoint-50/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "TinyLlama/TinyLlama_v1.1",
+  "architectures": [
+    "ProgressiveYocoLlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "crossDecoder_start_idx": 2,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "mlp_bias": false,
+  "model_type": "progressive_yoco_llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32000
+}

checkpoint-50/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.45.2"
+}

checkpoint-50/latest

@@ -0,0 +1 @@
+global_step50

checkpoint-50/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b53a3e8c75ccf099ac77b0790e3f96f661a4fed2eaf311cea40eea7d2c5a77a0
+size 2198026696

checkpoint-50/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:575119a228f98110923ffa2dedcb50e3317251b26054355d015e0b2240d566f2
+size 15984

checkpoint-50/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0728b56dab7abb5ef8a0d4bae3519c5767c97467bdd886d26bf19cc8599d0312
+size 15984

checkpoint-50/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4e481d4ef1546694da7337f6bb6c658b866dcb79b85deeb477da0d27ebe851e
+size 15984

checkpoint-50/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:353c60be37ea56fc992fca446598ceca5d1fd002aa3bd6dbb9ad740e6f47ebb3
+size 15984

checkpoint-50/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9107fe964ba7205e354084b85210e5a5ea1c98cfd4d38adb9cd3926945dcae4
+size 15984

checkpoint-50/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69d1bb1abee38b92e53f3f23549b642ce0f1edcdccf7b6129847ac61636e96d5
+size 15984

checkpoint-50/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afd5516048e20f36959601574e29e40106085a7d3cdc7bf425ce5e84633490e6
+size 15984

checkpoint-50/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e2c46927fc06939b4c976a01e4b95dec1f8b98ceaea86d31a5d756fc30ff006
+size 15984

checkpoint-50/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9b23fe3d7c8cff984a05b7865b1b491bac693fd2f496a4e0e24b70ad75a25bd
+size 1064

checkpoint-50/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-50/tokenizer.model

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

checkpoint-50/tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "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 %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% 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": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-50/trainer_state.json

@@ -0,0 +1,383 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.05959475566150179,
+  "eval_steps": 50,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0011918951132300357,
+      "grad_norm": 1.8323841453219294,
+      "learning_rate": 2e-05,
+      "loss": 2.1039,
+      "step": 1
+    },
+    {
+      "epoch": 0.0023837902264600714,
+      "grad_norm": 1.837920719204094,
+      "learning_rate": 1.9979453927503366e-05,
+      "loss": 2.1015,
+      "step": 2
+    },
+    {
+      "epoch": 0.003575685339690107,
+      "grad_norm": 0.9173452411852326,
+      "learning_rate": 1.991790013823246e-05,
+      "loss": 2.0311,
+      "step": 3
+    },
+    {
+      "epoch": 0.004767580452920143,
+      "grad_norm": 1.265146161082013,
+      "learning_rate": 1.9815591569910654e-05,
+      "loss": 2.0187,
+      "step": 4
+    },
+    {
+      "epoch": 0.0059594755661501785,
+      "grad_norm": 1.0055189392399428,
+      "learning_rate": 1.9672948630390296e-05,
+      "loss": 2.0142,
+      "step": 5
+    },
+    {
+      "epoch": 0.007151370679380214,
+      "grad_norm": 1.6116133535057522,
+      "learning_rate": 1.949055747010669e-05,
+      "loss": 2.047,
+      "step": 6
+    },
+    {
+      "epoch": 0.00834326579261025,
+      "grad_norm": 1.0164569935770877,
+      "learning_rate": 1.926916757346022e-05,
+      "loss": 2.0114,
+      "step": 7
+    },
+    {
+      "epoch": 0.009535160905840286,
+      "grad_norm": 0.8088386058968114,
+      "learning_rate": 1.900968867902419e-05,
+      "loss": 1.979,
+      "step": 8
+    },
+    {
+      "epoch": 0.010727056019070322,
+      "grad_norm": 0.7238192809262382,
+      "learning_rate": 1.8713187041233896e-05,
+      "loss": 1.9969,
+      "step": 9
+    },
+    {
+      "epoch": 0.011918951132300357,
+      "grad_norm": 0.5757195660436086,
+      "learning_rate": 1.8380881048918406e-05,
+      "loss": 1.9719,
+      "step": 10
+    },
+    {
+      "epoch": 0.013110846245530394,
+      "grad_norm": 0.6136131835485481,
+      "learning_rate": 1.8014136218679566e-05,
+      "loss": 1.9736,
+      "step": 11
+    },
+    {
+      "epoch": 0.014302741358760428,
+      "grad_norm": 0.6462531692000065,
+      "learning_rate": 1.7614459583691346e-05,
+      "loss": 1.9625,
+      "step": 12
+    },
+    {
+      "epoch": 0.015494636471990465,
+      "grad_norm": 0.5160196738824037,
+      "learning_rate": 1.7183493500977277e-05,
+      "loss": 1.9524,
+      "step": 13
+    },
+    {
+      "epoch": 0.0166865315852205,
+      "grad_norm": 0.48508132154258404,
+      "learning_rate": 1.672300890261317e-05,
+      "loss": 1.9584,
+      "step": 14
+    },
+    {
+      "epoch": 0.017878426698450536,
+      "grad_norm": 0.5181005310771087,
+      "learning_rate": 1.6234898018587336e-05,
+      "loss": 1.967,
+      "step": 15
+    },
+    {
+      "epoch": 0.01907032181168057,
+      "grad_norm": 0.4150834616993424,
+      "learning_rate": 1.5721166601221697e-05,
+      "loss": 1.9456,
+      "step": 16
+    },
+    {
+      "epoch": 0.02026221692491061,
+      "grad_norm": 0.3379165251881648,
+      "learning_rate": 1.5183925683105254e-05,
+      "loss": 1.9637,
+      "step": 17
+    },
+    {
+      "epoch": 0.021454112038140644,
+      "grad_norm": 0.3484808167274383,
+      "learning_rate": 1.4625382902408356e-05,
+      "loss": 1.9582,
+      "step": 18
+    },
+    {
+      "epoch": 0.02264600715137068,
+      "grad_norm": 0.35126742165373703,
+      "learning_rate": 1.4047833431223938e-05,
+      "loss": 1.9716,
+      "step": 19
+    },
+    {
+      "epoch": 0.023837902264600714,
+      "grad_norm": 0.2995298854480071,
+      "learning_rate": 1.3453650544213078e-05,
+      "loss": 1.9804,
+      "step": 20
+    },
+    {
+      "epoch": 0.025029797377830752,
+      "grad_norm": 0.2812553357334388,
+      "learning_rate": 1.2845275866310325e-05,
+      "loss": 1.943,
+      "step": 21
+    },
+    {
+      "epoch": 0.026221692491060787,
+      "grad_norm": 0.29784459167074245,
+      "learning_rate": 1.2225209339563144e-05,
+      "loss": 1.9517,
+      "step": 22
+    },
+    {
+      "epoch": 0.027413587604290822,
+      "grad_norm": 0.28759384963571305,
+      "learning_rate": 1.1595998950333794e-05,
+      "loss": 1.9676,
+      "step": 23
+    },
+    {
+      "epoch": 0.028605482717520857,
+      "grad_norm": 0.26154732718281476,
+      "learning_rate": 1.0960230259076819e-05,
+      "loss": 1.9585,
+      "step": 24
+    },
+    {
+      "epoch": 0.029797377830750895,
+      "grad_norm": 0.2417322199141499,
+      "learning_rate": 1.0320515775716556e-05,
+      "loss": 1.9658,
+      "step": 25
+    },
+    {
+      "epoch": 0.03098927294398093,
+      "grad_norm": 0.2647008811703257,
+      "learning_rate": 9.67948422428345e-06,
+      "loss": 1.9677,
+      "step": 26
+    },
+    {
+      "epoch": 0.03218116805721097,
+      "grad_norm": 0.28156330533090274,
+      "learning_rate": 9.039769740923183e-06,
+      "loss": 1.971,
+      "step": 27
+    },
+    {
+      "epoch": 0.033373063170441,
+      "grad_norm": 0.23451944041424483,
+      "learning_rate": 8.404001049666211e-06,
+      "loss": 1.9777,
+      "step": 28
+    },
+    {
+      "epoch": 0.03456495828367104,
+      "grad_norm": 0.20425845712936946,
+      "learning_rate": 7.774790660436857e-06,
+      "loss": 1.9455,
+      "step": 29
+    },
+    {
+      "epoch": 0.03575685339690107,
+      "grad_norm": 0.20003632743257047,
+      "learning_rate": 7.154724133689677e-06,
+      "loss": 1.9219,
+      "step": 30
+    },
+    {
+      "epoch": 0.03694874851013111,
+      "grad_norm": 0.22248700808509062,
+      "learning_rate": 6.546349455786926e-06,
+      "loss": 1.9595,
+      "step": 31
+    },
+    {
+      "epoch": 0.03814064362336114,
+      "grad_norm": 0.22141566358189815,
+      "learning_rate": 5.952166568776062e-06,
+      "loss": 1.9736,
+      "step": 32
+    },
+    {
+      "epoch": 0.03933253873659118,
+      "grad_norm": 0.20350762417692123,
+      "learning_rate": 5.37461709759165e-06,
+      "loss": 1.9491,
+      "step": 33
+    },
+    {
+      "epoch": 0.04052443384982122,
+      "grad_norm": 0.18362905257668707,
+      "learning_rate": 4.81607431689475e-06,
+      "loss": 1.9877,
+      "step": 34
+    },
+    {
+      "epoch": 0.041716328963051254,
+      "grad_norm": 0.18209143534810343,
+      "learning_rate": 4.278833398778306e-06,
+      "loss": 1.9711,
+      "step": 35
+    },
+    {
+      "epoch": 0.04290822407628129,
+      "grad_norm": 0.1870043994840097,
+      "learning_rate": 3.7651019814126656e-06,
+      "loss": 1.9997,
+      "step": 36
+    },
+    {
+      "epoch": 0.04410011918951132,
+      "grad_norm": 0.18563803773188292,
+      "learning_rate": 3.2769910973868314e-06,
+      "loss": 1.9611,
+      "step": 37
+    },
+    {
+      "epoch": 0.04529201430274136,
+      "grad_norm": 0.1803488628139512,
+      "learning_rate": 2.8165064990227255e-06,
+      "loss": 1.9779,
+      "step": 38
+    },
+    {
+      "epoch": 0.04648390941597139,
+      "grad_norm": 0.18280901111654344,
+      "learning_rate": 2.3855404163086558e-06,
+      "loss": 1.9753,
+      "step": 39
+    },
+    {
+      "epoch": 0.04767580452920143,
+      "grad_norm": 0.18003744910780412,
+      "learning_rate": 1.9858637813204352e-06,
+      "loss": 1.9327,
+      "step": 40
+    },
+    {
+      "epoch": 0.04886769964243146,
+      "grad_norm": 0.1751954707202772,
+      "learning_rate": 1.6191189510815942e-06,
+      "loss": 1.9578,
+      "step": 41
+    },
+    {
+      "epoch": 0.050059594755661505,
+      "grad_norm": 0.17152593814056669,
+      "learning_rate": 1.286812958766106e-06,
+      "loss": 1.9176,
+      "step": 42
+    },
+    {
+      "epoch": 0.05125148986889154,
+      "grad_norm": 0.16959349088820902,
+      "learning_rate": 9.903113209758098e-07,
+      "loss": 1.9505,
+      "step": 43
+    },
+    {
+      "epoch": 0.052443384982121574,
+      "grad_norm": 0.16660058312767648,
+      "learning_rate": 7.308324265397837e-07,
+      "loss": 1.9683,
+      "step": 44
+    },
+    {
+      "epoch": 0.05363528009535161,
+      "grad_norm": 0.17674746764132468,
+      "learning_rate": 5.094425298933136e-07,
+      "loss": 1.9643,
+      "step": 45
+    },
+    {
+      "epoch": 0.054827175208581644,
+      "grad_norm": 0.16795370746145422,
+      "learning_rate": 3.2705136960970554e-07,
+      "loss": 1.9812,
+      "step": 46
+    },
+    {
+      "epoch": 0.05601907032181168,
+      "grad_norm": 0.17056366756389463,
+      "learning_rate": 1.844084300893456e-07,
+      "loss": 1.959,
+      "step": 47
+    },
+    {
+      "epoch": 0.057210965435041714,
+      "grad_norm": 0.16170892183194993,
+      "learning_rate": 8.209986176753947e-08,
+      "loss": 1.9624,
+      "step": 48
+    },
+    {
+      "epoch": 0.058402860548271755,
+      "grad_norm": 0.17173937225335953,
+      "learning_rate": 2.054607249663665e-08,
+      "loss": 1.9578,
+      "step": 49
+    },
+    {
+      "epoch": 0.05959475566150179,
+      "grad_norm": 0.17021831986392572,
+      "learning_rate": 0.0,
+      "loss": 1.9973,
+      "step": 50
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 106890730143744.0,
+  "train_batch_size": 58,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-50/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3125fe88bf614ded0206db12e56072f9ec4fb9fe65a19870757576de3b6144d0
+size 7224

checkpoint-50/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)

config.json

@@ -0,0 +1 @@
+{"_name_or_path": "TinyLlama/TinyLlama_v1.1", "architectures": ["ProgressiveYocoLlamaForCausalLM"], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "crossDecoder_start_idx": 2, "eos_token_id": 2, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 5632, "max_position_embeddings": 2048, "mlp_bias": false, "model_type": "progressive_yoco_llama", "num_attention_heads": 32, "num_hidden_layers": 22, "num_key_value_heads": 4, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000.0, "tie_word_embeddings": false, "torch_dtype": "bfloat16", "transformers_version": "4.45.2", "use_cache": false, "vocab_size": 32000}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.45.2"
+}

runs/Nov29_08-38-26_creative-turing-2/events.out.tfevents.1732869726.creative-turing-2.2655059.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc212a41ff0a75fb942cf1b1cd272298b9cd0bf5b2d9f38aaad2abc3e89a573b
+size 16158

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "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 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "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 %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% 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": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.05959475566150179,
+    "total_flos": 106890730143744.0,
+    "train_loss": 1.9756734085083008,
+    "train_runtime": 1598.4061,
+    "train_samples_per_second": 58.058,
+    "train_steps_per_second": 0.031
+}

trainer_log.jsonl

@@ -0,0 +1,51 @@
+{"current_steps": 1, "total_steps": 50, "loss": 2.1039, "lr": 2e-05, "epoch": 0.0011918951132300357, "percentage": 2.0, "elapsed_time": "0:00:35", "remaining_time": "0:28:50"}
+{"current_steps": 2, "total_steps": 50, "loss": 2.1015, "lr": 1.9979453927503366e-05, "epoch": 0.0023837902264600714, "percentage": 4.0, "elapsed_time": "0:01:06", "remaining_time": "0:26:41"}
+{"current_steps": 3, "total_steps": 50, "loss": 2.0311, "lr": 1.991790013823246e-05, "epoch": 0.003575685339690107, "percentage": 6.0, "elapsed_time": "0:01:38", "remaining_time": "0:25:40"}
+{"current_steps": 4, "total_steps": 50, "loss": 2.0187, "lr": 1.9815591569910654e-05, "epoch": 0.004767580452920143, "percentage": 8.0, "elapsed_time": "0:02:10", "remaining_time": "0:24:55"}
+{"current_steps": 5, "total_steps": 50, "loss": 2.0142, "lr": 1.9672948630390296e-05, "epoch": 0.0059594755661501785, "percentage": 10.0, "elapsed_time": "0:02:41", "remaining_time": "0:24:15"}
+{"current_steps": 6, "total_steps": 50, "loss": 2.047, "lr": 1.949055747010669e-05, "epoch": 0.007151370679380214, "percentage": 12.0, "elapsed_time": "0:03:13", "remaining_time": "0:23:39"}
+{"current_steps": 7, "total_steps": 50, "loss": 2.0114, "lr": 1.926916757346022e-05, "epoch": 0.00834326579261025, "percentage": 14.0, "elapsed_time": "0:03:45", "remaining_time": "0:23:03"}
+{"current_steps": 8, "total_steps": 50, "loss": 1.979, "lr": 1.900968867902419e-05, "epoch": 0.009535160905840286, "percentage": 16.0, "elapsed_time": "0:04:17", "remaining_time": "0:22:29"}
+{"current_steps": 9, "total_steps": 50, "loss": 1.9969, "lr": 1.8713187041233896e-05, "epoch": 0.010727056019070322, "percentage": 18.0, "elapsed_time": "0:04:48", "remaining_time": "0:21:55"}
+{"current_steps": 10, "total_steps": 50, "loss": 1.9719, "lr": 1.8380881048918406e-05, "epoch": 0.011918951132300357, "percentage": 20.0, "elapsed_time": "0:05:20", "remaining_time": "0:21:22"}
+{"current_steps": 11, "total_steps": 50, "loss": 1.9736, "lr": 1.8014136218679566e-05, "epoch": 0.013110846245530394, "percentage": 22.0, "elapsed_time": "0:05:52", "remaining_time": "0:20:48"}
+{"current_steps": 12, "total_steps": 50, "loss": 1.9625, "lr": 1.7614459583691346e-05, "epoch": 0.014302741358760428, "percentage": 24.0, "elapsed_time": "0:06:23", "remaining_time": "0:20:15"}
+{"current_steps": 13, "total_steps": 50, "loss": 1.9524, "lr": 1.7183493500977277e-05, "epoch": 0.015494636471990465, "percentage": 26.0, "elapsed_time": "0:06:55", "remaining_time": "0:19:43"}
+{"current_steps": 14, "total_steps": 50, "loss": 1.9584, "lr": 1.672300890261317e-05, "epoch": 0.0166865315852205, "percentage": 28.0, "elapsed_time": "0:07:27", "remaining_time": "0:19:10"}
+{"current_steps": 15, "total_steps": 50, "loss": 1.967, "lr": 1.6234898018587336e-05, "epoch": 0.017878426698450536, "percentage": 30.0, "elapsed_time": "0:07:59", "remaining_time": "0:18:38"}
+{"current_steps": 16, "total_steps": 50, "loss": 1.9456, "lr": 1.5721166601221697e-05, "epoch": 0.01907032181168057, "percentage": 32.0, "elapsed_time": "0:08:31", "remaining_time": "0:18:05"}
+{"current_steps": 17, "total_steps": 50, "loss": 1.9637, "lr": 1.5183925683105254e-05, "epoch": 0.02026221692491061, "percentage": 34.0, "elapsed_time": "0:09:02", "remaining_time": "0:17:33"}
+{"current_steps": 18, "total_steps": 50, "loss": 1.9582, "lr": 1.4625382902408356e-05, "epoch": 0.021454112038140644, "percentage": 36.0, "elapsed_time": "0:09:34", "remaining_time": "0:17:01"}
+{"current_steps": 19, "total_steps": 50, "loss": 1.9716, "lr": 1.4047833431223938e-05, "epoch": 0.02264600715137068, "percentage": 38.0, "elapsed_time": "0:10:06", "remaining_time": "0:16:29"}
+{"current_steps": 20, "total_steps": 50, "loss": 1.9804, "lr": 1.3453650544213078e-05, "epoch": 0.023837902264600714, "percentage": 40.0, "elapsed_time": "0:10:38", "remaining_time": "0:15:57"}
+{"current_steps": 21, "total_steps": 50, "loss": 1.943, "lr": 1.2845275866310325e-05, "epoch": 0.025029797377830752, "percentage": 42.0, "elapsed_time": "0:11:09", "remaining_time": "0:15:25"}
+{"current_steps": 22, "total_steps": 50, "loss": 1.9517, "lr": 1.2225209339563144e-05, "epoch": 0.026221692491060787, "percentage": 44.0, "elapsed_time": "0:11:41", "remaining_time": "0:14:53"}
+{"current_steps": 23, "total_steps": 50, "loss": 1.9676, "lr": 1.1595998950333794e-05, "epoch": 0.027413587604290822, "percentage": 46.0, "elapsed_time": "0:12:13", "remaining_time": "0:14:20"}
+{"current_steps": 24, "total_steps": 50, "loss": 1.9585, "lr": 1.0960230259076819e-05, "epoch": 0.028605482717520857, "percentage": 48.0, "elapsed_time": "0:12:45", "remaining_time": "0:13:48"}
+{"current_steps": 25, "total_steps": 50, "loss": 1.9658, "lr": 1.0320515775716556e-05, "epoch": 0.029797377830750895, "percentage": 50.0, "elapsed_time": "0:13:16", "remaining_time": "0:13:16"}
+{"current_steps": 26, "total_steps": 50, "loss": 1.9677, "lr": 9.67948422428345e-06, "epoch": 0.03098927294398093, "percentage": 52.0, "elapsed_time": "0:13:48", "remaining_time": "0:12:44"}
+{"current_steps": 27, "total_steps": 50, "loss": 1.971, "lr": 9.039769740923183e-06, "epoch": 0.03218116805721097, "percentage": 54.0, "elapsed_time": "0:14:20", "remaining_time": "0:12:13"}
+{"current_steps": 28, "total_steps": 50, "loss": 1.9777, "lr": 8.404001049666211e-06, "epoch": 0.033373063170441, "percentage": 56.0, "elapsed_time": "0:14:52", "remaining_time": "0:11:41"}
+{"current_steps": 29, "total_steps": 50, "loss": 1.9455, "lr": 7.774790660436857e-06, "epoch": 0.03456495828367104, "percentage": 58.0, "elapsed_time": "0:15:24", "remaining_time": "0:11:09"}
+{"current_steps": 30, "total_steps": 50, "loss": 1.9219, "lr": 7.154724133689677e-06, "epoch": 0.03575685339690107, "percentage": 60.0, "elapsed_time": "0:15:55", "remaining_time": "0:10:37"}
+{"current_steps": 31, "total_steps": 50, "loss": 1.9595, "lr": 6.546349455786926e-06, "epoch": 0.03694874851013111, "percentage": 62.0, "elapsed_time": "0:16:27", "remaining_time": "0:10:05"}
+{"current_steps": 32, "total_steps": 50, "loss": 1.9736, "lr": 5.952166568776062e-06, "epoch": 0.03814064362336114, "percentage": 64.0, "elapsed_time": "0:16:59", "remaining_time": "0:09:33"}
+{"current_steps": 33, "total_steps": 50, "loss": 1.9491, "lr": 5.37461709759165e-06, "epoch": 0.03933253873659118, "percentage": 66.0, "elapsed_time": "0:17:31", "remaining_time": "0:09:01"}
+{"current_steps": 34, "total_steps": 50, "loss": 1.9877, "lr": 4.81607431689475e-06, "epoch": 0.04052443384982122, "percentage": 68.0, "elapsed_time": "0:18:02", "remaining_time": "0:08:29"}
+{"current_steps": 35, "total_steps": 50, "loss": 1.9711, "lr": 4.278833398778306e-06, "epoch": 0.041716328963051254, "percentage": 70.0, "elapsed_time": "0:18:34", "remaining_time": "0:07:57"}
+{"current_steps": 36, "total_steps": 50, "loss": 1.9997, "lr": 3.7651019814126656e-06, "epoch": 0.04290822407628129, "percentage": 72.0, "elapsed_time": "0:19:06", "remaining_time": "0:07:25"}
+{"current_steps": 37, "total_steps": 50, "loss": 1.9611, "lr": 3.2769910973868314e-06, "epoch": 0.04410011918951132, "percentage": 74.0, "elapsed_time": "0:19:38", "remaining_time": "0:06:53"}
+{"current_steps": 38, "total_steps": 50, "loss": 1.9779, "lr": 2.8165064990227255e-06, "epoch": 0.04529201430274136, "percentage": 76.0, "elapsed_time": "0:20:09", "remaining_time": "0:06:22"}
+{"current_steps": 39, "total_steps": 50, "loss": 1.9753, "lr": 2.3855404163086558e-06, "epoch": 0.04648390941597139, "percentage": 78.0, "elapsed_time": "0:20:41", "remaining_time": "0:05:50"}
+{"current_steps": 40, "total_steps": 50, "loss": 1.9327, "lr": 1.9858637813204352e-06, "epoch": 0.04767580452920143, "percentage": 80.0, "elapsed_time": "0:21:13", "remaining_time": "0:05:18"}
+{"current_steps": 41, "total_steps": 50, "loss": 1.9578, "lr": 1.6191189510815942e-06, "epoch": 0.04886769964243146, "percentage": 82.0, "elapsed_time": "0:21:45", "remaining_time": "0:04:46"}
+{"current_steps": 42, "total_steps": 50, "loss": 1.9176, "lr": 1.286812958766106e-06, "epoch": 0.050059594755661505, "percentage": 84.0, "elapsed_time": "0:22:16", "remaining_time": "0:04:14"}
+{"current_steps": 43, "total_steps": 50, "loss": 1.9505, "lr": 9.903113209758098e-07, "epoch": 0.05125148986889154, "percentage": 86.0, "elapsed_time": "0:22:48", "remaining_time": "0:03:42"}
+{"current_steps": 44, "total_steps": 50, "loss": 1.9683, "lr": 7.308324265397837e-07, "epoch": 0.052443384982121574, "percentage": 88.0, "elapsed_time": "0:23:20", "remaining_time": "0:03:10"}
+{"current_steps": 45, "total_steps": 50, "loss": 1.9643, "lr": 5.094425298933136e-07, "epoch": 0.05363528009535161, "percentage": 90.0, "elapsed_time": "0:23:52", "remaining_time": "0:02:39"}
+{"current_steps": 46, "total_steps": 50, "loss": 1.9812, "lr": 3.2705136960970554e-07, "epoch": 0.054827175208581644, "percentage": 92.0, "elapsed_time": "0:24:23", "remaining_time": "0:02:07"}
+{"current_steps": 47, "total_steps": 50, "loss": 1.959, "lr": 1.844084300893456e-07, "epoch": 0.05601907032181168, "percentage": 94.0, "elapsed_time": "0:24:55", "remaining_time": "0:01:35"}
+{"current_steps": 48, "total_steps": 50, "loss": 1.9624, "lr": 8.209986176753947e-08, "epoch": 0.057210965435041714, "percentage": 96.0, "elapsed_time": "0:25:27", "remaining_time": "0:01:03"}
+{"current_steps": 49, "total_steps": 50, "loss": 1.9578, "lr": 2.054607249663665e-08, "epoch": 0.058402860548271755, "percentage": 98.0, "elapsed_time": "0:25:59", "remaining_time": "0:00:31"}
+{"current_steps": 50, "total_steps": 50, "loss": 1.9973, "lr": 0.0, "epoch": 0.05959475566150179, "percentage": 100.0, "elapsed_time": "0:26:30", "remaining_time": "0:00:00"}
+{"current_steps": 50, "total_steps": 50, "epoch": 0.05959475566150179, "percentage": 100.0, "elapsed_time": "0:26:38", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,392 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.05959475566150179,
+  "eval_steps": 50,
+  "global_step": 50,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0011918951132300357,
+      "grad_norm": 1.8323841453219294,
+      "learning_rate": 2e-05,
+      "loss": 2.1039,
+      "step": 1
+    },
+    {
+      "epoch": 0.0023837902264600714,
+      "grad_norm": 1.837920719204094,
+      "learning_rate": 1.9979453927503366e-05,
+      "loss": 2.1015,
+      "step": 2
+    },
+    {
+      "epoch": 0.003575685339690107,
+      "grad_norm": 0.9173452411852326,
+      "learning_rate": 1.991790013823246e-05,
+      "loss": 2.0311,
+      "step": 3
+    },
+    {
+      "epoch": 0.004767580452920143,
+      "grad_norm": 1.265146161082013,
+      "learning_rate": 1.9815591569910654e-05,
+      "loss": 2.0187,
+      "step": 4
+    },
+    {
+      "epoch": 0.0059594755661501785,
+      "grad_norm": 1.0055189392399428,
+      "learning_rate": 1.9672948630390296e-05,
+      "loss": 2.0142,
+      "step": 5
+    },
+    {
+      "epoch": 0.007151370679380214,
+      "grad_norm": 1.6116133535057522,
+      "learning_rate": 1.949055747010669e-05,
+      "loss": 2.047,
+      "step": 6
+    },
+    {
+      "epoch": 0.00834326579261025,
+      "grad_norm": 1.0164569935770877,
+      "learning_rate": 1.926916757346022e-05,
+      "loss": 2.0114,
+      "step": 7
+    },
+    {
+      "epoch": 0.009535160905840286,
+      "grad_norm": 0.8088386058968114,
+      "learning_rate": 1.900968867902419e-05,
+      "loss": 1.979,
+      "step": 8
+    },
+    {
+      "epoch": 0.010727056019070322,
+      "grad_norm": 0.7238192809262382,
+      "learning_rate": 1.8713187041233896e-05,
+      "loss": 1.9969,
+      "step": 9
+    },
+    {
+      "epoch": 0.011918951132300357,
+      "grad_norm": 0.5757195660436086,
+      "learning_rate": 1.8380881048918406e-05,
+      "loss": 1.9719,
+      "step": 10
+    },
+    {
+      "epoch": 0.013110846245530394,
+      "grad_norm": 0.6136131835485481,
+      "learning_rate": 1.8014136218679566e-05,
+      "loss": 1.9736,
+      "step": 11
+    },
+    {
+      "epoch": 0.014302741358760428,
+      "grad_norm": 0.6462531692000065,
+      "learning_rate": 1.7614459583691346e-05,
+      "loss": 1.9625,
+      "step": 12
+    },
+    {
+      "epoch": 0.015494636471990465,
+      "grad_norm": 0.5160196738824037,
+      "learning_rate": 1.7183493500977277e-05,
+      "loss": 1.9524,
+      "step": 13
+    },
+    {
+      "epoch": 0.0166865315852205,
+      "grad_norm": 0.48508132154258404,
+      "learning_rate": 1.672300890261317e-05,
+      "loss": 1.9584,
+      "step": 14
+    },
+    {
+      "epoch": 0.017878426698450536,
+      "grad_norm": 0.5181005310771087,
+      "learning_rate": 1.6234898018587336e-05,
+      "loss": 1.967,
+      "step": 15
+    },
+    {
+      "epoch": 0.01907032181168057,
+      "grad_norm": 0.4150834616993424,
+      "learning_rate": 1.5721166601221697e-05,
+      "loss": 1.9456,
+      "step": 16
+    },
+    {
+      "epoch": 0.02026221692491061,
+      "grad_norm": 0.3379165251881648,
+      "learning_rate": 1.5183925683105254e-05,
+      "loss": 1.9637,
+      "step": 17
+    },
+    {
+      "epoch": 0.021454112038140644,
+      "grad_norm": 0.3484808167274383,
+      "learning_rate": 1.4625382902408356e-05,
+      "loss": 1.9582,
+      "step": 18
+    },
+    {
+      "epoch": 0.02264600715137068,
+      "grad_norm": 0.35126742165373703,
+      "learning_rate": 1.4047833431223938e-05,
+      "loss": 1.9716,
+      "step": 19
+    },
+    {
+      "epoch": 0.023837902264600714,
+      "grad_norm": 0.2995298854480071,
+      "learning_rate": 1.3453650544213078e-05,
+      "loss": 1.9804,
+      "step": 20
+    },
+    {
+      "epoch": 0.025029797377830752,
+      "grad_norm": 0.2812553357334388,
+      "learning_rate": 1.2845275866310325e-05,
+      "loss": 1.943,
+      "step": 21
+    },
+    {
+      "epoch": 0.026221692491060787,
+      "grad_norm": 0.29784459167074245,
+      "learning_rate": 1.2225209339563144e-05,
+      "loss": 1.9517,
+      "step": 22
+    },
+    {
+      "epoch": 0.027413587604290822,
+      "grad_norm": 0.28759384963571305,
+      "learning_rate": 1.1595998950333794e-05,
+      "loss": 1.9676,
+      "step": 23
+    },
+    {
+      "epoch": 0.028605482717520857,
+      "grad_norm": 0.26154732718281476,
+      "learning_rate": 1.0960230259076819e-05,
+      "loss": 1.9585,
+      "step": 24
+    },
+    {
+      "epoch": 0.029797377830750895,
+      "grad_norm": 0.2417322199141499,
+      "learning_rate": 1.0320515775716556e-05,
+      "loss": 1.9658,
+      "step": 25
+    },
+    {
+      "epoch": 0.03098927294398093,
+      "grad_norm": 0.2647008811703257,
+      "learning_rate": 9.67948422428345e-06,
+      "loss": 1.9677,
+      "step": 26
+    },
+    {
+      "epoch": 0.03218116805721097,
+      "grad_norm": 0.28156330533090274,
+      "learning_rate": 9.039769740923183e-06,
+      "loss": 1.971,
+      "step": 27
+    },
+    {
+      "epoch": 0.033373063170441,
+      "grad_norm": 0.23451944041424483,
+      "learning_rate": 8.404001049666211e-06,
+      "loss": 1.9777,
+      "step": 28
+    },
+    {
+      "epoch": 0.03456495828367104,
+      "grad_norm": 0.20425845712936946,
+      "learning_rate": 7.774790660436857e-06,
+      "loss": 1.9455,
+      "step": 29
+    },
+    {
+      "epoch": 0.03575685339690107,
+      "grad_norm": 0.20003632743257047,
+      "learning_rate": 7.154724133689677e-06,
+      "loss": 1.9219,
+      "step": 30
+    },
+    {
+      "epoch": 0.03694874851013111,
+      "grad_norm": 0.22248700808509062,
+      "learning_rate": 6.546349455786926e-06,
+      "loss": 1.9595,
+      "step": 31
+    },
+    {
+      "epoch": 0.03814064362336114,
+      "grad_norm": 0.22141566358189815,
+      "learning_rate": 5.952166568776062e-06,
+      "loss": 1.9736,
+      "step": 32
+    },
+    {
+      "epoch": 0.03933253873659118,
+      "grad_norm": 0.20350762417692123,
+      "learning_rate": 5.37461709759165e-06,
+      "loss": 1.9491,
+      "step": 33
+    },
+    {
+      "epoch": 0.04052443384982122,
+      "grad_norm": 0.18362905257668707,
+      "learning_rate": 4.81607431689475e-06,
+      "loss": 1.9877,
+      "step": 34
+    },
+    {
+      "epoch": 0.041716328963051254,
+      "grad_norm": 0.18209143534810343,
+      "learning_rate": 4.278833398778306e-06,
+      "loss": 1.9711,
+      "step": 35
+    },
+    {
+      "epoch": 0.04290822407628129,
+      "grad_norm": 0.1870043994840097,
+      "learning_rate": 3.7651019814126656e-06,
+      "loss": 1.9997,
+      "step": 36
+    },
+    {
+      "epoch": 0.04410011918951132,
+      "grad_norm": 0.18563803773188292,
+      "learning_rate": 3.2769910973868314e-06,
+      "loss": 1.9611,
+      "step": 37
+    },
+    {
+      "epoch": 0.04529201430274136,
+      "grad_norm": 0.1803488628139512,
+      "learning_rate": 2.8165064990227255e-06,
+      "loss": 1.9779,
+      "step": 38
+    },
+    {
+      "epoch": 0.04648390941597139,
+      "grad_norm": 0.18280901111654344,
+      "learning_rate": 2.3855404163086558e-06,
+      "loss": 1.9753,
+      "step": 39
+    },
+    {
+      "epoch": 0.04767580452920143,
+      "grad_norm": 0.18003744910780412,
+      "learning_rate": 1.9858637813204352e-06,
+      "loss": 1.9327,
+      "step": 40
+    },
+    {
+      "epoch": 0.04886769964243146,
+      "grad_norm": 0.1751954707202772,
+      "learning_rate": 1.6191189510815942e-06,
+      "loss": 1.9578,
+      "step": 41
+    },
+    {
+      "epoch": 0.050059594755661505,
+      "grad_norm": 0.17152593814056669,
+      "learning_rate": 1.286812958766106e-06,
+      "loss": 1.9176,
+      "step": 42
+    },
+    {
+      "epoch": 0.05125148986889154,
+      "grad_norm": 0.16959349088820902,
+      "learning_rate": 9.903113209758098e-07,
+      "loss": 1.9505,
+      "step": 43
+    },
+    {
+      "epoch": 0.052443384982121574,
+      "grad_norm": 0.16660058312767648,
+      "learning_rate": 7.308324265397837e-07,
+      "loss": 1.9683,
+      "step": 44
+    },
+    {
+      "epoch": 0.05363528009535161,
+      "grad_norm": 0.17674746764132468,
+      "learning_rate": 5.094425298933136e-07,
+      "loss": 1.9643,
+      "step": 45
+    },
+    {
+      "epoch": 0.054827175208581644,
+      "grad_norm": 0.16795370746145422,
+      "learning_rate": 3.2705136960970554e-07,
+      "loss": 1.9812,
+      "step": 46
+    },
+    {
+      "epoch": 0.05601907032181168,
+      "grad_norm": 0.17056366756389463,
+      "learning_rate": 1.844084300893456e-07,
+      "loss": 1.959,
+      "step": 47
+    },
+    {
+      "epoch": 0.057210965435041714,
+      "grad_norm": 0.16170892183194993,
+      "learning_rate": 8.209986176753947e-08,
+      "loss": 1.9624,
+      "step": 48
+    },
+    {
+      "epoch": 0.058402860548271755,
+      "grad_norm": 0.17173937225335953,
+      "learning_rate": 2.054607249663665e-08,
+      "loss": 1.9578,
+      "step": 49
+    },
+    {
+      "epoch": 0.05959475566150179,
+      "grad_norm": 0.17021831986392572,
+      "learning_rate": 0.0,
+      "loss": 1.9973,
+      "step": 50
+    },
+    {
+      "epoch": 0.05959475566150179,
+      "step": 50,
+      "total_flos": 106890730143744.0,
+      "train_loss": 1.9756734085083008,
+      "train_runtime": 1598.4061,
+      "train_samples_per_second": 58.058,
+      "train_steps_per_second": 0.031
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 50,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 50,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 106890730143744.0,
+  "train_batch_size": 58,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3125fe88bf614ded0206db12e56072f9ec4fb9fe65a19870757576de3b6144d0
+size 7224