Upload model (hf_t5) to HuggingFace Hub.

checkpoint checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "google/t5-v1_1-base",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "classifier_dropout": 0.0,
+  "d_ff": 2048,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
+  "dropout_rate": 0.0,
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_bf16": true,
+  "is_encoder_decoder": true,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 32104
+}

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.39.3"
+}

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/global_step16000/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e7a9a17b056a141842d1550b46f0b38f209b0348c40c6209e885aed088f0f97
+size 742632752

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/global_step16000/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:e8c7933a379508c19e0f8b99444de7e97a22d4702fb8794fb89dba3c55d9b066
+size 742635056

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/global_step16000/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3712b1eac981e65176da5afbeac32377dbd7dcff4ec9034f111a0df65902e434
+size 742635440

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/global_step16000/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42f90c48ff95369a06aac1adfaab4c83c2307e180d7928d14b65a8f94220af84
+size 742633136

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/global_step16000/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d4b3aa74b78da01afd472e326f95dcafbd825bde3dd7b3536868fa30942ba64
+size 495181624

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/latest

@@ -0,0 +1 @@
+global_step16000

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec44296a496533ba282a617538b938e7a28b01a4a3de7d5a4ed08de78a7eff45
+size 593739520

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03cac535790d2b13dcdc1e4ac58be793407427483ec7d7721f84508e01451bc1
+size 14960

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:098f3fed4ab092d48659ef4c4398ac0ff9a4617711aface757826da42645fa1e
+size 14960

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4009c906e3a9a0249cea7ba8c83eca6eb034e8e1ab0b053994f5a3379fc5199a
+size 14960

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0d645b143b8aa43fd224d3dd37de7b13af86b5f9dab588e01a2fa9a43274f6c
+size 14960

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49bba319b540cc504eaeae73d020447cefb6dfbf468d231b9d78b6b946483146
+size 1064

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/special_tokens_map.json

@@ -0,0 +1,125 @@
+{
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
+size 791656

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/tokenizer_config.json

@@ -0,0 +1,940 @@
+{
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "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": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<extra_id_99>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<extra_id_98>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<extra_id_97>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<extra_id_96>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<extra_id_95>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<extra_id_94>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<extra_id_93>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<extra_id_92>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<extra_id_91>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<extra_id_90>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<extra_id_89>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32011": {
+      "content": "<extra_id_88>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32012": {
+      "content": "<extra_id_87>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32013": {
+      "content": "<extra_id_86>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32014": {
+      "content": "<extra_id_85>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32015": {
+      "content": "<extra_id_84>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32016": {
+      "content": "<extra_id_83>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32017": {
+      "content": "<extra_id_82>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32018": {
+      "content": "<extra_id_81>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32019": {
+      "content": "<extra_id_80>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32020": {
+      "content": "<extra_id_79>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32021": {
+      "content": "<extra_id_78>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32022": {
+      "content": "<extra_id_77>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32023": {
+      "content": "<extra_id_76>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32024": {
+      "content": "<extra_id_75>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32025": {
+      "content": "<extra_id_74>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32026": {
+      "content": "<extra_id_73>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32027": {
+      "content": "<extra_id_72>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32028": {
+      "content": "<extra_id_71>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32029": {
+      "content": "<extra_id_70>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32030": {
+      "content": "<extra_id_69>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32031": {
+      "content": "<extra_id_68>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32032": {
+      "content": "<extra_id_67>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32033": {
+      "content": "<extra_id_66>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32034": {
+      "content": "<extra_id_65>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32035": {
+      "content": "<extra_id_64>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32036": {
+      "content": "<extra_id_63>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32037": {
+      "content": "<extra_id_62>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32038": {
+      "content": "<extra_id_61>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32039": {
+      "content": "<extra_id_60>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32040": {
+      "content": "<extra_id_59>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32041": {
+      "content": "<extra_id_58>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32042": {
+      "content": "<extra_id_57>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32043": {
+      "content": "<extra_id_56>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32044": {
+      "content": "<extra_id_55>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32045": {
+      "content": "<extra_id_54>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32046": {
+      "content": "<extra_id_53>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32047": {
+      "content": "<extra_id_52>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32048": {
+      "content": "<extra_id_51>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32049": {
+      "content": "<extra_id_50>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32050": {
+      "content": "<extra_id_49>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32051": {
+      "content": "<extra_id_48>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32052": {
+      "content": "<extra_id_47>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32053": {
+      "content": "<extra_id_46>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32054": {
+      "content": "<extra_id_45>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32055": {
+      "content": "<extra_id_44>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32056": {
+      "content": "<extra_id_43>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32057": {
+      "content": "<extra_id_42>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32058": {
+      "content": "<extra_id_41>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32059": {
+      "content": "<extra_id_40>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32060": {
+      "content": "<extra_id_39>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32061": {
+      "content": "<extra_id_38>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32062": {
+      "content": "<extra_id_37>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32063": {
+      "content": "<extra_id_36>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32064": {
+      "content": "<extra_id_35>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32065": {
+      "content": "<extra_id_34>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32066": {
+      "content": "<extra_id_33>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32067": {
+      "content": "<extra_id_32>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32068": {
+      "content": "<extra_id_31>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32069": {
+      "content": "<extra_id_30>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32070": {
+      "content": "<extra_id_29>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32071": {
+      "content": "<extra_id_28>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32072": {
+      "content": "<extra_id_27>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32073": {
+      "content": "<extra_id_26>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32074": {
+      "content": "<extra_id_25>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32075": {
+      "content": "<extra_id_24>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32076": {
+      "content": "<extra_id_23>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32077": {
+      "content": "<extra_id_22>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32078": {
+      "content": "<extra_id_21>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32079": {
+      "content": "<extra_id_20>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32080": {
+      "content": "<extra_id_19>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32081": {
+      "content": "<extra_id_18>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32082": {
+      "content": "<extra_id_17>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32083": {
+      "content": "<extra_id_16>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32084": {
+      "content": "<extra_id_15>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32085": {
+      "content": "<extra_id_14>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32086": {
+      "content": "<extra_id_13>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32087": {
+      "content": "<extra_id_12>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32088": {
+      "content": "<extra_id_11>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32089": {
+      "content": "<extra_id_10>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32090": {
+      "content": "<extra_id_9>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32091": {
+      "content": "<extra_id_8>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32092": {
+      "content": "<extra_id_7>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32093": {
+      "content": "<extra_id_6>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32094": {
+      "content": "<extra_id_5>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32095": {
+      "content": "<extra_id_4>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32096": {
+      "content": "<extra_id_3>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32097": {
+      "content": "<extra_id_2>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32098": {
+      "content": "<extra_id_1>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32099": {
+      "content": "<extra_id_0>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<extra_id_0>",
+    "<extra_id_1>",
+    "<extra_id_2>",
+    "<extra_id_3>",
+    "<extra_id_4>",
+    "<extra_id_5>",
+    "<extra_id_6>",
+    "<extra_id_7>",
+    "<extra_id_8>",
+    "<extra_id_9>",
+    "<extra_id_10>",
+    "<extra_id_11>",
+    "<extra_id_12>",
+    "<extra_id_13>",
+    "<extra_id_14>",
+    "<extra_id_15>",
+    "<extra_id_16>",
+    "<extra_id_17>",
+    "<extra_id_18>",
+    "<extra_id_19>",
+    "<extra_id_20>",
+    "<extra_id_21>",
+    "<extra_id_22>",
+    "<extra_id_23>",
+    "<extra_id_24>",
+    "<extra_id_25>",
+    "<extra_id_26>",
+    "<extra_id_27>",
+    "<extra_id_28>",
+    "<extra_id_29>",
+    "<extra_id_30>",
+    "<extra_id_31>",
+    "<extra_id_32>",
+    "<extra_id_33>",
+    "<extra_id_34>",
+    "<extra_id_35>",
+    "<extra_id_36>",
+    "<extra_id_37>",
+    "<extra_id_38>",
+    "<extra_id_39>",
+    "<extra_id_40>",
+    "<extra_id_41>",
+    "<extra_id_42>",
+    "<extra_id_43>",
+    "<extra_id_44>",
+    "<extra_id_45>",
+    "<extra_id_46>",
+    "<extra_id_47>",
+    "<extra_id_48>",
+    "<extra_id_49>",
+    "<extra_id_50>",
+    "<extra_id_51>",
+    "<extra_id_52>",
+    "<extra_id_53>",
+    "<extra_id_54>",
+    "<extra_id_55>",
+    "<extra_id_56>",
+    "<extra_id_57>",
+    "<extra_id_58>",
+    "<extra_id_59>",
+    "<extra_id_60>",
+    "<extra_id_61>",
+    "<extra_id_62>",
+    "<extra_id_63>",
+    "<extra_id_64>",
+    "<extra_id_65>",
+    "<extra_id_66>",
+    "<extra_id_67>",
+    "<extra_id_68>",
+    "<extra_id_69>",
+    "<extra_id_70>",
+    "<extra_id_71>",
+    "<extra_id_72>",
+    "<extra_id_73>",
+    "<extra_id_74>",
+    "<extra_id_75>",
+    "<extra_id_76>",
+    "<extra_id_77>",
+    "<extra_id_78>",
+    "<extra_id_79>",
+    "<extra_id_80>",
+    "<extra_id_81>",
+    "<extra_id_82>",
+    "<extra_id_83>",
+    "<extra_id_84>",
+    "<extra_id_85>",
+    "<extra_id_86>",
+    "<extra_id_87>",
+    "<extra_id_88>",
+    "<extra_id_89>",
+    "<extra_id_90>",
+    "<extra_id_91>",
+    "<extra_id_92>",
+    "<extra_id_93>",
+    "<extra_id_94>",
+    "<extra_id_95>",
+    "<extra_id_96>",
+    "<extra_id_97>",
+    "<extra_id_98>",
+    "<extra_id_99>"
+  ],
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "extra_ids": 100,
+  "legacy": true,
+  "model_max_length": 1000000000,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "T5Tokenizer",
+  "unk_token": "<unk>"
+}

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/trainer_state.json

@@ -0,0 +1,1533 @@
+{
+  "best_metric": 0.5056542754173279,
+  "best_model_checkpoint": "checkpoints//pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-14000",
+  "epoch": 0.016,
+  "eval_steps": 1000,
+  "global_step": 16000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "grad_norm": 43460.96179331515,
+      "learning_rate": 1e-08,
+      "loss": 15.3906,
+      "num_input_tokens_seen": 102400,
+      "step": 1
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 863.6828438075519,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 14.9414,
+      "num_input_tokens_seen": 10240000,
+      "step": 100
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 133.93767179833242,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 14.2755,
+      "num_input_tokens_seen": 20480000,
+      "step": 200
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 63.96658072546188,
+      "learning_rate": 3e-06,
+      "loss": 12.6284,
+      "num_input_tokens_seen": 30720000,
+      "step": 300
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 122.50072793549663,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 9.2201,
+      "num_input_tokens_seen": 40960000,
+      "step": 400
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 10.484723012864677,
+      "learning_rate": 5e-06,
+      "loss": 3.51,
+      "num_input_tokens_seen": 51200000,
+      "step": 500
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 3.2509195695753146,
+      "learning_rate": 6e-06,
+      "loss": 1.3629,
+      "num_input_tokens_seen": 61440000,
+      "step": 600
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.7280951813088964,
+      "learning_rate": 7.000000000000001e-06,
+      "loss": 1.1216,
+      "num_input_tokens_seen": 71680000,
+      "step": 700
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.8267415609909531,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.0623,
+      "num_input_tokens_seen": 81920000,
+      "step": 800
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.36043112995458126,
+      "learning_rate": 9e-06,
+      "loss": 0.9816,
+      "num_input_tokens_seen": 92160000,
+      "step": 900
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3071054349582017,
+      "learning_rate": 1e-05,
+      "loss": 0.9622,
+      "num_input_tokens_seen": 102400000,
+      "step": 1000
+    },
+    {
+      "epoch": 0.0,
+      "eval_average_loss_on_non_sentence_tokens": 0.8788086175918579,
+      "eval_loss": 0.8788086175918579,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.0885,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8510264799772181,
+      "eval_runtime": 81.4892,
+      "eval_samples_per_second": 61.358,
+      "eval_steps_per_second": 0.16,
+      "num_input_tokens_seen": 102400000,
+      "step": 1000
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.2502056334106625,
+      "learning_rate": 1.1000000000000001e-05,
+      "loss": 0.9492,
+      "num_input_tokens_seen": 112640000,
+      "step": 1100
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.4131447950777185,
+      "learning_rate": 1.2e-05,
+      "loss": 0.9381,
+      "num_input_tokens_seen": 122880000,
+      "step": 1200
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.35032463430441246,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 0.9119,
+      "num_input_tokens_seen": 133120000,
+      "step": 1300
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.23087998248198688,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 0.9064,
+      "num_input_tokens_seen": 143360000,
+      "step": 1400
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.2710276748538426,
+      "learning_rate": 1.5e-05,
+      "loss": 0.886,
+      "num_input_tokens_seen": 153600000,
+      "step": 1500
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.26455585507258395,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.8637,
+      "num_input_tokens_seen": 163840000,
+      "step": 1600
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.35915501744739176,
+      "learning_rate": 1.7000000000000003e-05,
+      "loss": 0.8878,
+      "num_input_tokens_seen": 174080000,
+      "step": 1700
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.20870754142081574,
+      "learning_rate": 1.8e-05,
+      "loss": 0.8736,
+      "num_input_tokens_seen": 184320000,
+      "step": 1800
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.28713761080519334,
+      "learning_rate": 1.9e-05,
+      "loss": 0.8764,
+      "num_input_tokens_seen": 194560000,
+      "step": 1900
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.29572729316826546,
+      "learning_rate": 2e-05,
+      "loss": 0.8632,
+      "num_input_tokens_seen": 204800000,
+      "step": 2000
+    },
+    {
+      "epoch": 0.0,
+      "eval_average_loss_on_non_sentence_tokens": 0.7859668135643005,
+      "eval_loss": 0.7859668135643005,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.3642,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8615504149217942,
+      "eval_runtime": 79.4566,
+      "eval_samples_per_second": 62.927,
+      "eval_steps_per_second": 0.164,
+      "num_input_tokens_seen": 204800000,
+      "step": 2000
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.2836766066769867,
+      "learning_rate": 2.1e-05,
+      "loss": 0.8504,
+      "num_input_tokens_seen": 215040000,
+      "step": 2100
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3004986314599583,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.8612,
+      "num_input_tokens_seen": 225280000,
+      "step": 2200
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.2607008654822589,
+      "learning_rate": 2.3000000000000003e-05,
+      "loss": 0.854,
+      "num_input_tokens_seen": 235520000,
+      "step": 2300
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.4084304742274508,
+      "learning_rate": 2.4e-05,
+      "loss": 0.812,
+      "num_input_tokens_seen": 245760000,
+      "step": 2400
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3350486028966764,
+      "learning_rate": 2.5e-05,
+      "loss": 0.8396,
+      "num_input_tokens_seen": 256000000,
+      "step": 2500
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.2794394029400218,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 0.8261,
+      "num_input_tokens_seen": 266240000,
+      "step": 2600
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.31900263666540146,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.8291,
+      "num_input_tokens_seen": 276480000,
+      "step": 2700
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3415219809134491,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.8287,
+      "num_input_tokens_seen": 286720000,
+      "step": 2800
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.42162229775558907,
+      "learning_rate": 2.9e-05,
+      "loss": 0.8211,
+      "num_input_tokens_seen": 296960000,
+      "step": 2900
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.6230432635833446,
+      "learning_rate": 3e-05,
+      "loss": 0.8057,
+      "num_input_tokens_seen": 307200000,
+      "step": 3000
+    },
+    {
+      "epoch": 0.0,
+      "eval_average_loss_on_non_sentence_tokens": 0.7231347560882568,
+      "eval_loss": 0.7231347560882568,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 76.01565,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8692755743839554,
+      "eval_runtime": 80.2947,
+      "eval_samples_per_second": 62.271,
+      "eval_steps_per_second": 0.162,
+      "num_input_tokens_seen": 307200000,
+      "step": 3000
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.4419432049494865,
+      "learning_rate": 3.1e-05,
+      "loss": 0.804,
+      "num_input_tokens_seen": 317440000,
+      "step": 3100
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.37573353728190156,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.7838,
+      "num_input_tokens_seen": 327680000,
+      "step": 3200
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.30248504338648025,
+      "learning_rate": 3.3e-05,
+      "loss": 0.7943,
+      "num_input_tokens_seen": 337920000,
+      "step": 3300
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.5238509466902019,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.7828,
+      "num_input_tokens_seen": 348160000,
+      "step": 3400
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3629820011592176,
+      "learning_rate": 3.5e-05,
+      "loss": 0.7769,
+      "num_input_tokens_seen": 358400000,
+      "step": 3500
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3793351901469991,
+      "learning_rate": 3.6e-05,
+      "loss": 0.7797,
+      "num_input_tokens_seen": 368640000,
+      "step": 3600
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3965386085813541,
+      "learning_rate": 3.7e-05,
+      "loss": 0.76,
+      "num_input_tokens_seen": 378880000,
+      "step": 3700
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.7172234328832077,
+      "learning_rate": 3.8e-05,
+      "loss": 0.7512,
+      "num_input_tokens_seen": 389120000,
+      "step": 3800
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3292733714629288,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.7608,
+      "num_input_tokens_seen": 399360000,
+      "step": 3900
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3719515773722568,
+      "learning_rate": 4e-05,
+      "loss": 0.7581,
+      "num_input_tokens_seen": 409600000,
+      "step": 4000
+    },
+    {
+      "epoch": 0.0,
+      "eval_average_loss_on_non_sentence_tokens": 0.6934961080551147,
+      "eval_loss": 0.6934961080551147,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.9202,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8731403154127814,
+      "eval_runtime": 79.8616,
+      "eval_samples_per_second": 62.608,
+      "eval_steps_per_second": 0.163,
+      "num_input_tokens_seen": 409600000,
+      "step": 4000
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.4763329687962473,
+      "learning_rate": 4.1e-05,
+      "loss": 0.773,
+      "num_input_tokens_seen": 419840000,
+      "step": 4100
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.30681374394935307,
+      "learning_rate": 4.2e-05,
+      "loss": 0.7449,
+      "num_input_tokens_seen": 430080000,
+      "step": 4200
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.3103964759300168,
+      "learning_rate": 4.3e-05,
+      "loss": 0.7488,
+      "num_input_tokens_seen": 440320000,
+      "step": 4300
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.92009994176494,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.7597,
+      "num_input_tokens_seen": 450560000,
+      "step": 4400
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.6241253812854839,
+      "learning_rate": 4.5e-05,
+      "loss": 0.7496,
+      "num_input_tokens_seen": 460800000,
+      "step": 4500
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.4993639267754847,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.7474,
+      "num_input_tokens_seen": 471040000,
+      "step": 4600
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.46628951587896234,
+      "learning_rate": 4.7e-05,
+      "loss": 0.7426,
+      "num_input_tokens_seen": 481280000,
+      "step": 4700
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.7179158594950166,
+      "learning_rate": 4.8e-05,
+      "loss": 0.7161,
+      "num_input_tokens_seen": 491520000,
+      "step": 4800
+    },
+    {
+      "epoch": 0.0,
+      "grad_norm": 0.37892254076672666,
+      "learning_rate": 4.9e-05,
+      "loss": 0.7307,
+      "num_input_tokens_seen": 501760000,
+      "step": 4900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3050243948244417,
+      "learning_rate": 5e-05,
+      "loss": 0.7153,
+      "num_input_tokens_seen": 512000000,
+      "step": 5000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.6558203101158142,
+      "eval_loss": 0.6558203101158142,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 76.01175,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8783376766170906,
+      "eval_runtime": 79.5555,
+      "eval_samples_per_second": 62.849,
+      "eval_steps_per_second": 0.163,
+      "num_input_tokens_seen": 512000000,
+      "step": 5000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5280712337324736,
+      "learning_rate": 5.1000000000000006e-05,
+      "loss": 0.7274,
+      "num_input_tokens_seen": 522240000,
+      "step": 5100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5395533496916384,
+      "learning_rate": 5.2000000000000004e-05,
+      "loss": 0.7267,
+      "num_input_tokens_seen": 532480000,
+      "step": 5200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3549010648426737,
+      "learning_rate": 5.300000000000001e-05,
+      "loss": 0.7361,
+      "num_input_tokens_seen": 542720000,
+      "step": 5300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4072753639161728,
+      "learning_rate": 5.4000000000000005e-05,
+      "loss": 0.7026,
+      "num_input_tokens_seen": 552960000,
+      "step": 5400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3433693815993015,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 0.7234,
+      "num_input_tokens_seen": 563200000,
+      "step": 5500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.644273082006586,
+      "learning_rate": 5.6000000000000006e-05,
+      "loss": 0.7188,
+      "num_input_tokens_seen": 573440000,
+      "step": 5600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4252838589276199,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 0.7032,
+      "num_input_tokens_seen": 583680000,
+      "step": 5700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 1.0592894854394406,
+      "learning_rate": 5.8e-05,
+      "loss": 0.6996,
+      "num_input_tokens_seen": 593920000,
+      "step": 5800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6904084068806473,
+      "learning_rate": 5.9e-05,
+      "loss": 0.6872,
+      "num_input_tokens_seen": 604160000,
+      "step": 5900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.44791648600448986,
+      "learning_rate": 6e-05,
+      "loss": 0.6936,
+      "num_input_tokens_seen": 614400000,
+      "step": 6000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.6245507597923279,
+      "eval_loss": 0.6245507597923279,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.7185,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8813006847462675,
+      "eval_runtime": 79.938,
+      "eval_samples_per_second": 62.549,
+      "eval_steps_per_second": 0.163,
+      "num_input_tokens_seen": 614400000,
+      "step": 6000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4594815341525536,
+      "learning_rate": 6.1e-05,
+      "loss": 0.6735,
+      "num_input_tokens_seen": 624640000,
+      "step": 6100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.8405483838445511,
+      "learning_rate": 6.2e-05,
+      "loss": 0.6799,
+      "num_input_tokens_seen": 634880000,
+      "step": 6200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6336202646486054,
+      "learning_rate": 6.3e-05,
+      "loss": 0.6815,
+      "num_input_tokens_seen": 645120000,
+      "step": 6300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6430228992710968,
+      "learning_rate": 6.400000000000001e-05,
+      "loss": 0.6583,
+      "num_input_tokens_seen": 655360000,
+      "step": 6400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.8111510449497575,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 0.6668,
+      "num_input_tokens_seen": 665600000,
+      "step": 6500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6436069681468451,
+      "learning_rate": 6.6e-05,
+      "loss": 0.6554,
+      "num_input_tokens_seen": 675840000,
+      "step": 6600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6739816739483769,
+      "learning_rate": 6.7e-05,
+      "loss": 0.6414,
+      "num_input_tokens_seen": 686080000,
+      "step": 6700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7557762652302368,
+      "learning_rate": 6.800000000000001e-05,
+      "loss": 0.6466,
+      "num_input_tokens_seen": 696320000,
+      "step": 6800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5748098146998661,
+      "learning_rate": 6.9e-05,
+      "loss": 0.646,
+      "num_input_tokens_seen": 706560000,
+      "step": 6900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6292390120724913,
+      "learning_rate": 7e-05,
+      "loss": 0.6313,
+      "num_input_tokens_seen": 716800000,
+      "step": 7000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5793261528015137,
+      "eval_loss": 0.5793261528015137,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.8407,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8874100728229607,
+      "eval_runtime": 78.5261,
+      "eval_samples_per_second": 63.673,
+      "eval_steps_per_second": 0.166,
+      "num_input_tokens_seen": 716800000,
+      "step": 7000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5088852921346764,
+      "learning_rate": 7.1e-05,
+      "loss": 0.6415,
+      "num_input_tokens_seen": 727040000,
+      "step": 7100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5802094410441135,
+      "learning_rate": 7.2e-05,
+      "loss": 0.6322,
+      "num_input_tokens_seen": 737280000,
+      "step": 7200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7783010042444823,
+      "learning_rate": 7.3e-05,
+      "loss": 0.6245,
+      "num_input_tokens_seen": 747520000,
+      "step": 7300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7662907061258049,
+      "learning_rate": 7.4e-05,
+      "loss": 0.6121,
+      "num_input_tokens_seen": 757760000,
+      "step": 7400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7354690437648583,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.6281,
+      "num_input_tokens_seen": 768000000,
+      "step": 7500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 1.4254502630785943,
+      "learning_rate": 7.6e-05,
+      "loss": 0.6115,
+      "num_input_tokens_seen": 778240000,
+      "step": 7600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5225208925446124,
+      "learning_rate": 7.7e-05,
+      "loss": 0.6206,
+      "num_input_tokens_seen": 788480000,
+      "step": 7700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3900621555389805,
+      "learning_rate": 7.800000000000001e-05,
+      "loss": 0.623,
+      "num_input_tokens_seen": 798720000,
+      "step": 7800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4951271868717405,
+      "learning_rate": 7.900000000000001e-05,
+      "loss": 0.6189,
+      "num_input_tokens_seen": 808960000,
+      "step": 7900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4848361358265983,
+      "learning_rate": 8e-05,
+      "loss": 0.6197,
+      "num_input_tokens_seen": 819200000,
+      "step": 8000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5578222870826721,
+      "eval_loss": 0.5578222870826721,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.8388,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8900220791825759,
+      "eval_runtime": 79.6116,
+      "eval_samples_per_second": 62.805,
+      "eval_steps_per_second": 0.163,
+      "num_input_tokens_seen": 819200000,
+      "step": 8000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5858893883344134,
+      "learning_rate": 8.1e-05,
+      "loss": 0.6118,
+      "num_input_tokens_seen": 829440000,
+      "step": 8100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6249727004764817,
+      "learning_rate": 8.2e-05,
+      "loss": 0.6182,
+      "num_input_tokens_seen": 839680000,
+      "step": 8200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.36091241377808103,
+      "learning_rate": 8.3e-05,
+      "loss": 0.6164,
+      "num_input_tokens_seen": 849920000,
+      "step": 8300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.46039504119810887,
+      "learning_rate": 8.4e-05,
+      "loss": 0.6176,
+      "num_input_tokens_seen": 860160000,
+      "step": 8400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5636374972640031,
+      "learning_rate": 8.5e-05,
+      "loss": 0.6112,
+      "num_input_tokens_seen": 870400000,
+      "step": 8500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.26723681909344243,
+      "learning_rate": 8.6e-05,
+      "loss": 0.62,
+      "num_input_tokens_seen": 880640000,
+      "step": 8600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6910037577177741,
+      "learning_rate": 8.7e-05,
+      "loss": 0.6108,
+      "num_input_tokens_seen": 890880000,
+      "step": 8700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6869737496691483,
+      "learning_rate": 8.800000000000001e-05,
+      "loss": 0.5994,
+      "num_input_tokens_seen": 901120000,
+      "step": 8800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7455802068821363,
+      "learning_rate": 8.900000000000001e-05,
+      "loss": 0.6018,
+      "num_input_tokens_seen": 911360000,
+      "step": 8900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6632968078514041,
+      "learning_rate": 9e-05,
+      "loss": 0.6063,
+      "num_input_tokens_seen": 921600000,
+      "step": 9000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5403027534484863,
+      "eval_loss": 0.5403027534484863,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.83535,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8922244388490125,
+      "eval_runtime": 77.9848,
+      "eval_samples_per_second": 64.115,
+      "eval_steps_per_second": 0.167,
+      "num_input_tokens_seen": 921600000,
+      "step": 9000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5725768352987969,
+      "learning_rate": 9.1e-05,
+      "loss": 0.6041,
+      "num_input_tokens_seen": 931840000,
+      "step": 9100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4551368642725825,
+      "learning_rate": 9.200000000000001e-05,
+      "loss": 0.594,
+      "num_input_tokens_seen": 942080000,
+      "step": 9200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6533702380800677,
+      "learning_rate": 9.300000000000001e-05,
+      "loss": 0.5936,
+      "num_input_tokens_seen": 952320000,
+      "step": 9300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.9831065527948913,
+      "learning_rate": 9.4e-05,
+      "loss": 0.6019,
+      "num_input_tokens_seen": 962560000,
+      "step": 9400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4541015953145989,
+      "learning_rate": 9.5e-05,
+      "loss": 0.5984,
+      "num_input_tokens_seen": 972800000,
+      "step": 9500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.48129006101400823,
+      "learning_rate": 9.6e-05,
+      "loss": 0.5987,
+      "num_input_tokens_seen": 983040000,
+      "step": 9600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4345752501475107,
+      "learning_rate": 9.7e-05,
+      "loss": 0.5931,
+      "num_input_tokens_seen": 993280000,
+      "step": 9700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4507511208182315,
+      "learning_rate": 9.8e-05,
+      "loss": 0.5953,
+      "num_input_tokens_seen": 1003520000,
+      "step": 9800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.39394850230781386,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 0.5809,
+      "num_input_tokens_seen": 1013760000,
+      "step": 9900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.8214923366681415,
+      "learning_rate": 0.0001,
+      "loss": 0.5913,
+      "num_input_tokens_seen": 1024000000,
+      "step": 10000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5407128930091858,
+      "eval_loss": 0.5407128930091858,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.7806,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8921869038417058,
+      "eval_runtime": 78.3053,
+      "eval_samples_per_second": 63.853,
+      "eval_steps_per_second": 0.166,
+      "num_input_tokens_seen": 1024000000,
+      "step": 10000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5530163749888272,
+      "learning_rate": 9.950371902099894e-05,
+      "loss": 0.5912,
+      "num_input_tokens_seen": 1034240000,
+      "step": 10100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.35133718051823265,
+      "learning_rate": 9.901475429766743e-05,
+      "loss": 0.5977,
+      "num_input_tokens_seen": 1044480000,
+      "step": 10200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.42324033313410125,
+      "learning_rate": 9.853292781642932e-05,
+      "loss": 0.5878,
+      "num_input_tokens_seen": 1054720000,
+      "step": 10300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.765109336182901,
+      "learning_rate": 9.805806756909202e-05,
+      "loss": 0.5766,
+      "num_input_tokens_seen": 1064960000,
+      "step": 10400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3981751906169302,
+      "learning_rate": 9.759000729485331e-05,
+      "loss": 0.5863,
+      "num_input_tokens_seen": 1075200000,
+      "step": 10500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6650572898932402,
+      "learning_rate": 9.712858623572642e-05,
+      "loss": 0.5836,
+      "num_input_tokens_seen": 1085440000,
+      "step": 10600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3922744259495626,
+      "learning_rate": 9.667364890456637e-05,
+      "loss": 0.5869,
+      "num_input_tokens_seen": 1095680000,
+      "step": 10700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.8388418386979233,
+      "learning_rate": 9.622504486493762e-05,
+      "loss": 0.5776,
+      "num_input_tokens_seen": 1105920000,
+      "step": 10800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3629415164325868,
+      "learning_rate": 9.578262852211513e-05,
+      "loss": 0.5829,
+      "num_input_tokens_seen": 1116160000,
+      "step": 10900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.710732252235504,
+      "learning_rate": 9.534625892455923e-05,
+      "loss": 0.5889,
+      "num_input_tokens_seen": 1126400000,
+      "step": 11000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5240722894668579,
+      "eval_loss": 0.5240722894668579,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.9179,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8951130934236704,
+      "eval_runtime": 79.2009,
+      "eval_samples_per_second": 63.131,
+      "eval_steps_per_second": 0.164,
+      "num_input_tokens_seen": 1126400000,
+      "step": 11000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.9997340086993262,
+      "learning_rate": 9.49157995752499e-05,
+      "loss": 0.5851,
+      "num_input_tokens_seen": 1136640000,
+      "step": 11100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.29653408528420816,
+      "learning_rate": 9.44911182523068e-05,
+      "loss": 0.5701,
+      "num_input_tokens_seen": 1146880000,
+      "step": 11200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5193442388117125,
+      "learning_rate": 9.407208683835973e-05,
+      "loss": 0.5652,
+      "num_input_tokens_seen": 1157120000,
+      "step": 11300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4733544072168943,
+      "learning_rate": 9.36585811581694e-05,
+      "loss": 0.5764,
+      "num_input_tokens_seen": 1167360000,
+      "step": 11400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.33229163001946194,
+      "learning_rate": 9.325048082403138e-05,
+      "loss": 0.5711,
+      "num_input_tokens_seen": 1177600000,
+      "step": 11500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.40506432895103284,
+      "learning_rate": 9.284766908852594e-05,
+      "loss": 0.5735,
+      "num_input_tokens_seen": 1187840000,
+      "step": 11600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.34822471091276047,
+      "learning_rate": 9.245003270420487e-05,
+      "loss": 0.581,
+      "num_input_tokens_seen": 1198080000,
+      "step": 11700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3848654525529098,
+      "learning_rate": 9.205746178983236e-05,
+      "loss": 0.5841,
+      "num_input_tokens_seen": 1208320000,
+      "step": 11800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6372302990169364,
+      "learning_rate": 9.166984970282113e-05,
+      "loss": 0.5788,
+      "num_input_tokens_seen": 1218560000,
+      "step": 11900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5061331048864363,
+      "learning_rate": 9.12870929175277e-05,
+      "loss": 0.5658,
+      "num_input_tokens_seen": 1228800000,
+      "step": 12000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5278515815734863,
+      "eval_loss": 0.5278515815734863,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.2487,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8946204735462583,
+      "eval_runtime": 78.9561,
+      "eval_samples_per_second": 63.326,
+      "eval_steps_per_second": 0.165,
+      "num_input_tokens_seen": 1228800000,
+      "step": 12000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4136071130130777,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 0.5672,
+      "num_input_tokens_seen": 1239040000,
+      "step": 12100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.40986780287289215,
+      "learning_rate": 9.053574604251853e-05,
+      "loss": 0.5734,
+      "num_input_tokens_seen": 1249280000,
+      "step": 12200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3400074575461829,
+      "learning_rate": 9.016696346674324e-05,
+      "loss": 0.5662,
+      "num_input_tokens_seen": 1259520000,
+      "step": 12300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.2995299282452748,
+      "learning_rate": 8.980265101338745e-05,
+      "loss": 0.5644,
+      "num_input_tokens_seen": 1269760000,
+      "step": 12400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.49184292130037205,
+      "learning_rate": 8.944271909999159e-05,
+      "loss": 0.5666,
+      "num_input_tokens_seen": 1280000000,
+      "step": 12500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3063195363723024,
+      "learning_rate": 8.90870806374748e-05,
+      "loss": 0.5575,
+      "num_input_tokens_seen": 1290240000,
+      "step": 12600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5232424870939063,
+      "learning_rate": 8.873565094161139e-05,
+      "loss": 0.5677,
+      "num_input_tokens_seen": 1300480000,
+      "step": 12700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.25960379772756,
+      "learning_rate": 8.838834764831845e-05,
+      "loss": 0.5643,
+      "num_input_tokens_seen": 1310720000,
+      "step": 12800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4290554860535722,
+      "learning_rate": 8.804509063256238e-05,
+      "loss": 0.5606,
+      "num_input_tokens_seen": 1320960000,
+      "step": 12900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.46106836918077254,
+      "learning_rate": 8.770580193070293e-05,
+      "loss": 0.5701,
+      "num_input_tokens_seen": 1331200000,
+      "step": 13000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5180273652076721,
+      "eval_loss": 0.5180273652076721,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 74.5892,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8969406717030101,
+      "eval_runtime": 78.5382,
+      "eval_samples_per_second": 63.663,
+      "eval_steps_per_second": 0.166,
+      "num_input_tokens_seen": 1331200000,
+      "step": 13000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3079362879346589,
+      "learning_rate": 8.737040566610379e-05,
+      "loss": 0.5698,
+      "num_input_tokens_seen": 1341440000,
+      "step": 13100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.34218364630667303,
+      "learning_rate": 8.703882797784892e-05,
+      "loss": 0.5671,
+      "num_input_tokens_seen": 1351680000,
+      "step": 13200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.9023826887968611,
+      "learning_rate": 8.6710996952412e-05,
+      "loss": 0.5711,
+      "num_input_tokens_seen": 1361920000,
+      "step": 13300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3135519679754609,
+      "learning_rate": 8.638684255813601e-05,
+      "loss": 0.573,
+      "num_input_tokens_seen": 1372160000,
+      "step": 13400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3585506656419919,
+      "learning_rate": 8.606629658238704e-05,
+      "loss": 0.5586,
+      "num_input_tokens_seen": 1382400000,
+      "step": 13500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4652392924422838,
+      "learning_rate": 8.574929257125441e-05,
+      "loss": 0.5656,
+      "num_input_tokens_seen": 1392640000,
+      "step": 13600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.2871497573479709,
+      "learning_rate": 8.54357657716761e-05,
+      "loss": 0.555,
+      "num_input_tokens_seen": 1402880000,
+      "step": 13700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.3314747747253342,
+      "learning_rate": 8.512565307587486e-05,
+      "loss": 0.5581,
+      "num_input_tokens_seen": 1413120000,
+      "step": 13800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.37101291478786336,
+      "learning_rate": 8.481889296799709e-05,
+      "loss": 0.5565,
+      "num_input_tokens_seen": 1423360000,
+      "step": 13900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.7724364092340954,
+      "learning_rate": 8.451542547285167e-05,
+      "loss": 0.5547,
+      "num_input_tokens_seen": 1433600000,
+      "step": 14000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5056542754173279,
+      "eval_loss": 0.5056542754173279,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.1399,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.898289469401122,
+      "eval_runtime": 79.3513,
+      "eval_samples_per_second": 63.011,
+      "eval_steps_per_second": 0.164,
+      "num_input_tokens_seen": 1433600000,
+      "step": 14000
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.2540379769406795,
+      "learning_rate": 8.42151921066519e-05,
+      "loss": 0.5665,
+      "num_input_tokens_seen": 1443840000,
+      "step": 14100
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.2932621408688775,
+      "learning_rate": 8.391813582966892e-05,
+      "loss": 0.5598,
+      "num_input_tokens_seen": 1454080000,
+      "step": 14200
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.48877245800197655,
+      "learning_rate": 8.362420100070908e-05,
+      "loss": 0.5537,
+      "num_input_tokens_seen": 1464320000,
+      "step": 14300
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.6541690758541313,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 0.5729,
+      "num_input_tokens_seen": 1474560000,
+      "step": 14400
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.48472735448325754,
+      "learning_rate": 8.304547985373997e-05,
+      "loss": 0.5461,
+      "num_input_tokens_seen": 1484800000,
+      "step": 14500
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.4397966525675985,
+      "learning_rate": 8.27605888602368e-05,
+      "loss": 0.5717,
+      "num_input_tokens_seen": 1495040000,
+      "step": 14600
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.27255208873195974,
+      "learning_rate": 8.247860988423225e-05,
+      "loss": 0.5693,
+      "num_input_tokens_seen": 1505280000,
+      "step": 14700
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5406951988266505,
+      "learning_rate": 8.219949365267866e-05,
+      "loss": 0.5682,
+      "num_input_tokens_seen": 1515520000,
+      "step": 14800
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.5836951008207953,
+      "learning_rate": 8.192319205190405e-05,
+      "loss": 0.5632,
+      "num_input_tokens_seen": 1525760000,
+      "step": 14900
+    },
+    {
+      "epoch": 0.01,
+      "grad_norm": 0.38210496912794795,
+      "learning_rate": 8.164965809277262e-05,
+      "loss": 0.5606,
+      "num_input_tokens_seen": 1536000000,
+      "step": 15000
+    },
+    {
+      "epoch": 0.01,
+      "eval_average_loss_on_non_sentence_tokens": 0.5040527582168579,
+      "eval_loss": 0.5040527582168579,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 75.84535,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8993130757289856,
+      "eval_runtime": 79.0335,
+      "eval_samples_per_second": 63.264,
+      "eval_steps_per_second": 0.164,
+      "num_input_tokens_seen": 1536000000,
+      "step": 15000
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.5319226158246073,
+      "learning_rate": 8.137884587711596e-05,
+      "loss": 0.556,
+      "num_input_tokens_seen": 1546240000,
+      "step": 15100
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.45977183218681883,
+      "learning_rate": 8.111071056538128e-05,
+      "loss": 0.5529,
+      "num_input_tokens_seen": 1556480000,
+      "step": 15200
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.28993334914794955,
+      "learning_rate": 8.084520834544432e-05,
+      "loss": 0.5671,
+      "num_input_tokens_seen": 1566720000,
+      "step": 15300
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.23815085016697934,
+      "learning_rate": 8.058229640253802e-05,
+      "loss": 0.5432,
+      "num_input_tokens_seen": 1576960000,
+      "step": 15400
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.29220323367264955,
+      "learning_rate": 8.032193289024989e-05,
+      "loss": 0.56,
+      "num_input_tokens_seen": 1587200000,
+      "step": 15500
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.4509226866368844,
+      "learning_rate": 8.006407690254358e-05,
+      "loss": 0.5559,
+      "num_input_tokens_seen": 1597440000,
+      "step": 15600
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.5510756470119066,
+      "learning_rate": 7.980868844676222e-05,
+      "loss": 0.5502,
+      "num_input_tokens_seen": 1607680000,
+      "step": 15700
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.4242689766840494,
+      "learning_rate": 7.955572841757301e-05,
+      "loss": 0.5473,
+      "num_input_tokens_seen": 1617920000,
+      "step": 15800
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.2960137749918929,
+      "learning_rate": 7.930515857181441e-05,
+      "loss": 0.5633,
+      "num_input_tokens_seen": 1628160000,
+      "step": 15900
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 0.4031487716644737,
+      "learning_rate": 7.905694150420948e-05,
+      "loss": 0.5596,
+      "num_input_tokens_seen": 1638400000,
+      "step": 16000
+    },
+    {
+      "epoch": 0.02,
+      "eval_average_loss_on_non_sentence_tokens": 0.5138280987739563,
+      "eval_loss": 0.5138280987739563,
+      "eval_non_padding_tokens_in_labels": 512.0,
+      "eval_num_sentinel_tokens_in_labels": 73.3444,
+      "eval_padding_tokens_in_labels": 0.0,
+      "eval_reconstruction_accuracy": 0.8969269411729885,
+      "eval_runtime": 79.0049,
+      "eval_samples_per_second": 63.287,
+      "eval_steps_per_second": 0.165,
+      "num_input_tokens_seen": 1638400000,
+      "step": 16000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 1000000,
+  "num_input_tokens_seen": 1638400000,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 2000,
+  "total_flos": 2.1910423467353702e+18,
+  "train_batch_size": 25,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62543e926cc4dcf54914bf3306e7ed31b0797271b9748242dd7283fb264fd8d9
+size 6584

checkpoints/pre_train/from_pretrained/hf_t5/allenai/c4_en/lr_0_0001_inverse_sqrt_bsz_200_shuffle_p_0_5/2024-04-03_20-31/checkpoint-16000/zero_to_fp32.py

@@ -0,0 +1,592 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _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):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file, tag=args.tag)