Upload 11 files

.gitattributes

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

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "FacebookAI/xlm-roberta-base",
+  "architectures": [
+    "XLMRobertaForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.38.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a61ac13ef5ef3a5f75c24a1664e28af273e5291d866c49f451af6733635528eb
+size 1109857804

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d9f80d0034c6cb095410be8523af7f232bc3acda83df85ee1ee5905804736b6
+size 2219833914

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75a2ead7bd552e501b543286091db3131f817d9120ceb538c0e159ed5a58a58f
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4180c5204a8c5ad54cd30fb9b4bfe263d18c184f014d3154ecda2fc2694f1155
+size 1064

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9a4661b2cb6b8a1007906509fe18cbfbc03062a086102bf7b80cfedb80f16c37
+size 17082854

tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,1385 @@
+{
+  "best_metric": 0.26576361060142517,
+  "best_model_checkpoint": "./checkpoint-xlm/checkpoint-62000",
+  "epoch": 2.74482025854436,
+  "eval_steps": 1000,
+  "global_step": 62000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "grad_norm": 14.701729774475098,
+      "learning_rate": 1.9873510587163855e-05,
+      "loss": 0.7955,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_LOC_f1": 0.5905623116100197,
+      "eval_ORG_f1": 0.5314018889950122,
+      "eval_PER_f1": 0.6970927767045186,
+      "eval_loss": 0.5589519143104553,
+      "eval_overall_accuracy": 0.8305196709829297,
+      "eval_overall_f1": 0.6098834954730151,
+      "eval_overall_precision": 0.6090804388676729,
+      "eval_overall_recall": 0.6106886724922678,
+      "eval_runtime": 423.4055,
+      "eval_samples_per_second": 155.17,
+      "eval_steps_per_second": 9.7,
+      "step": 1000
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 32.162567138671875,
+      "learning_rate": 1.974702117432771e-05,
+      "loss": 0.4418,
+      "step": 2000
+    },
+    {
+      "epoch": 0.09,
+      "eval_LOC_f1": 0.726466851182444,
+      "eval_ORG_f1": 0.6100225052728323,
+      "eval_PER_f1": 0.7472875356145855,
+      "eval_loss": 0.4735046327114105,
+      "eval_overall_accuracy": 0.8612480637309139,
+      "eval_overall_f1": 0.6999315297088808,
+      "eval_overall_precision": 0.7133899577070636,
+      "eval_overall_recall": 0.6869714980413346,
+      "eval_runtime": 1200.8027,
+      "eval_samples_per_second": 54.713,
+      "eval_steps_per_second": 0.428,
+      "step": 2000
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 22.541187286376953,
+      "learning_rate": 1.9620531761491565e-05,
+      "loss": 0.4571,
+      "step": 3000
+    },
+    {
+      "epoch": 0.13,
+      "eval_LOC_f1": 0.7147521203733278,
+      "eval_ORG_f1": 0.6152306241925535,
+      "eval_PER_f1": 0.7587303433967385,
+      "eval_loss": 0.4161168932914734,
+      "eval_overall_accuracy": 0.8662245123763159,
+      "eval_overall_f1": 0.7014490640563852,
+      "eval_overall_precision": 0.6885331216944984,
+      "eval_overall_recall": 0.7148588410104012,
+      "eval_runtime": 1210.8088,
+      "eval_samples_per_second": 54.261,
+      "eval_steps_per_second": 0.425,
+      "step": 3000
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 11.889817237854004,
+      "learning_rate": 1.949404234865542e-05,
+      "loss": 0.4338,
+      "step": 4000
+    },
+    {
+      "epoch": 0.18,
+      "eval_LOC_f1": 0.7620754026697686,
+      "eval_ORG_f1": 0.6140874035989717,
+      "eval_PER_f1": 0.760585513488345,
+      "eval_loss": 0.39507248997688293,
+      "eval_overall_accuracy": 0.8734290140043626,
+      "eval_overall_f1": 0.7179785677035143,
+      "eval_overall_precision": 0.715343099460293,
+      "eval_overall_recall": 0.7206335269485343,
+      "eval_runtime": 1168.4178,
+      "eval_samples_per_second": 56.23,
+      "eval_steps_per_second": 0.44,
+      "step": 4000
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 1.7912464141845703,
+      "learning_rate": 1.9367552935819272e-05,
+      "loss": 0.411,
+      "step": 5000
+    },
+    {
+      "epoch": 0.22,
+      "eval_LOC_f1": 0.7642403132070866,
+      "eval_ORG_f1": 0.6378310214375789,
+      "eval_PER_f1": 0.7455119879889838,
+      "eval_loss": 0.4034789800643921,
+      "eval_overall_accuracy": 0.871553757152341,
+      "eval_overall_f1": 0.721039469679929,
+      "eval_overall_precision": 0.7065764223847311,
+      "eval_overall_recall": 0.7361069836552749,
+      "eval_runtime": 1151.9834,
+      "eval_samples_per_second": 57.032,
+      "eval_steps_per_second": 0.446,
+      "step": 5000
+    },
+    {
+      "epoch": 0.27,
+      "grad_norm": 16.268198013305664,
+      "learning_rate": 1.924106352298313e-05,
+      "loss": 0.3996,
+      "step": 6000
+    },
+    {
+      "epoch": 0.27,
+      "eval_LOC_f1": 0.7602170744129683,
+      "eval_ORG_f1": 0.6418268648032774,
+      "eval_PER_f1": 0.7751195247983773,
+      "eval_loss": 0.3775835335254669,
+      "eval_overall_accuracy": 0.8816552334587298,
+      "eval_overall_f1": 0.7327942917159862,
+      "eval_overall_precision": 0.7236376806822298,
+      "eval_overall_recall": 0.7421856004322572,
+      "eval_runtime": 1142.8454,
+      "eval_samples_per_second": 57.488,
+      "eval_steps_per_second": 0.45,
+      "step": 6000
+    },
+    {
+      "epoch": 0.31,
+      "grad_norm": 31.32206153869629,
+      "learning_rate": 1.9114574110146982e-05,
+      "loss": 0.3887,
+      "step": 7000
+    },
+    {
+      "epoch": 0.31,
+      "eval_LOC_f1": 0.7745706910097646,
+      "eval_ORG_f1": 0.6498818525519849,
+      "eval_PER_f1": 0.7943746756616502,
+      "eval_loss": 0.3745666742324829,
+      "eval_overall_accuracy": 0.8846546328201562,
+      "eval_overall_f1": 0.7445416212235937,
+      "eval_overall_precision": 0.7664073337814453,
+      "eval_overall_recall": 0.7238889639335404,
+      "eval_runtime": 1105.9439,
+      "eval_samples_per_second": 59.406,
+      "eval_steps_per_second": 0.465,
+      "step": 7000
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 18.666685104370117,
+      "learning_rate": 1.8988084697310836e-05,
+      "loss": 0.3702,
+      "step": 8000
+    },
+    {
+      "epoch": 0.35,
+      "eval_LOC_f1": 0.7867664013429133,
+      "eval_ORG_f1": 0.6550611335848372,
+      "eval_PER_f1": 0.7935211181949948,
+      "eval_loss": 0.3557104468345642,
+      "eval_overall_accuracy": 0.8899326652546391,
+      "eval_overall_f1": 0.7497458350842472,
+      "eval_overall_precision": 0.7523873653281097,
+      "eval_overall_recall": 0.7471227880588951,
+      "eval_runtime": 1133.29,
+      "eval_samples_per_second": 57.973,
+      "eval_steps_per_second": 0.454,
+      "step": 8000
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 17.36050796508789,
+      "learning_rate": 1.8861595284474693e-05,
+      "loss": 0.3653,
+      "step": 9000
+    },
+    {
+      "epoch": 0.4,
+      "eval_LOC_f1": 0.7880055788005579,
+      "eval_ORG_f1": 0.6528712687433267,
+      "eval_PER_f1": 0.807534509589927,
+      "eval_loss": 0.359466552734375,
+      "eval_overall_accuracy": 0.8921916985426611,
+      "eval_overall_f1": 0.7549862974573792,
+      "eval_overall_precision": 0.7601919943031648,
+      "eval_overall_recall": 0.749851411589896,
+      "eval_runtime": 1138.9812,
+      "eval_samples_per_second": 57.683,
+      "eval_steps_per_second": 0.451,
+      "step": 9000
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 7.423336982727051,
+      "learning_rate": 1.8735105871638546e-05,
+      "loss": 0.3659,
+      "step": 10000
+    },
+    {
+      "epoch": 0.44,
+      "eval_LOC_f1": 0.7842564019273456,
+      "eval_ORG_f1": 0.6720731037545135,
+      "eval_PER_f1": 0.8074153213042102,
+      "eval_loss": 0.332335889339447,
+      "eval_overall_accuracy": 0.8897644864540195,
+      "eval_overall_f1": 0.7598549199807725,
+      "eval_overall_precision": 0.7563691603482544,
+      "eval_overall_recall": 0.763372956909361,
+      "eval_runtime": 1162.0455,
+      "eval_samples_per_second": 56.538,
+      "eval_steps_per_second": 0.442,
+      "step": 10000
+    },
+    {
+      "epoch": 0.49,
+      "grad_norm": 4.525887966156006,
+      "learning_rate": 1.86086164588024e-05,
+      "loss": 0.3503,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "eval_LOC_f1": 0.7864019140558769,
+      "eval_ORG_f1": 0.6835087245907537,
+      "eval_PER_f1": 0.8168448012608974,
+      "eval_loss": 0.34435194730758667,
+      "eval_overall_accuracy": 0.8948016312079158,
+      "eval_overall_f1": 0.7659673455352116,
+      "eval_overall_precision": 0.7646757897429339,
+      "eval_overall_recall": 0.7672632716466298,
+      "eval_runtime": 1177.3569,
+      "eval_samples_per_second": 55.803,
+      "eval_steps_per_second": 0.437,
+      "step": 11000
+    },
+    {
+      "epoch": 0.53,
+      "grad_norm": 2.3115265369415283,
+      "learning_rate": 1.8482127045966253e-05,
+      "loss": 0.3386,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "eval_LOC_f1": 0.7956286696262351,
+      "eval_ORG_f1": 0.6888199916080004,
+      "eval_PER_f1": 0.8247360603749347,
+      "eval_loss": 0.33748894929885864,
+      "eval_overall_accuracy": 0.8961179780608858,
+      "eval_overall_f1": 0.7745517686964949,
+      "eval_overall_precision": 0.7718298146310058,
+      "eval_overall_recall": 0.7772929893286505,
+      "eval_runtime": 1149.1534,
+      "eval_samples_per_second": 57.173,
+      "eval_steps_per_second": 0.447,
+      "step": 12000
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 21.933504104614258,
+      "learning_rate": 1.8355637633130106e-05,
+      "loss": 0.3398,
+      "step": 13000
+    },
+    {
+      "epoch": 0.58,
+      "eval_LOC_f1": 0.7985453119289768,
+      "eval_ORG_f1": 0.6847335352987636,
+      "eval_PER_f1": 0.8177605881902427,
+      "eval_loss": 0.3362252414226532,
+      "eval_overall_accuracy": 0.8956772990231721,
+      "eval_overall_f1": 0.7731774789536882,
+      "eval_overall_precision": 0.759676178805616,
+      "eval_overall_recall": 0.7871673645819263,
+      "eval_runtime": 1138.7931,
+      "eval_samples_per_second": 57.693,
+      "eval_steps_per_second": 0.451,
+      "step": 13000
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 36.83283615112305,
+      "learning_rate": 1.8229148220293963e-05,
+      "loss": 0.3335,
+      "step": 14000
+    },
+    {
+      "epoch": 0.62,
+      "eval_LOC_f1": 0.790251739825188,
+      "eval_ORG_f1": 0.6715311174812253,
+      "eval_PER_f1": 0.826087807168571,
+      "eval_loss": 0.3248044550418854,
+      "eval_overall_accuracy": 0.89719723074005,
+      "eval_overall_f1": 0.7691362656847152,
+      "eval_overall_precision": 0.7564545519858401,
+      "eval_overall_recall": 0.7822504390112117,
+      "eval_runtime": 1135.913,
+      "eval_samples_per_second": 57.839,
+      "eval_steps_per_second": 0.452,
+      "step": 14000
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 9.774023056030273,
+      "learning_rate": 1.8102658807457817e-05,
+      "loss": 0.3502,
+      "step": 15000
+    },
+    {
+      "epoch": 0.66,
+      "eval_LOC_f1": 0.8089101520697128,
+      "eval_ORG_f1": 0.6878634314938933,
+      "eval_PER_f1": 0.8301438926543143,
+      "eval_loss": 0.3092363178730011,
+      "eval_overall_accuracy": 0.8991053646508393,
+      "eval_overall_f1": 0.7814600123162602,
+      "eval_overall_precision": 0.7745336093198525,
+      "eval_overall_recall": 0.7885114142915034,
+      "eval_runtime": 1106.1286,
+      "eval_samples_per_second": 59.396,
+      "eval_steps_per_second": 0.465,
+      "step": 15000
+    },
+    {
+      "epoch": 0.71,
+      "grad_norm": 2.795013427734375,
+      "learning_rate": 1.7976169394621674e-05,
+      "loss": 0.3324,
+      "step": 16000
+    },
+    {
+      "epoch": 0.71,
+      "eval_LOC_f1": 0.8186463753520208,
+      "eval_ORG_f1": 0.7147083782482054,
+      "eval_PER_f1": 0.8387758290691769,
+      "eval_loss": 0.3244176208972931,
+      "eval_overall_accuracy": 0.9025454430499794,
+      "eval_overall_f1": 0.7950573365134548,
+      "eval_overall_precision": 0.7931229164426282,
+      "eval_overall_recall": 0.7970012157233554,
+      "eval_runtime": 1140.7225,
+      "eval_samples_per_second": 57.595,
+      "eval_steps_per_second": 0.451,
+      "step": 16000
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 6.714651107788086,
+      "learning_rate": 1.7849679981785527e-05,
+      "loss": 0.3286,
+      "step": 17000
+    },
+    {
+      "epoch": 0.75,
+      "eval_LOC_f1": 0.8141723531609012,
+      "eval_ORG_f1": 0.7138781751375037,
+      "eval_PER_f1": 0.8335467609813607,
+      "eval_loss": 0.32237449288368225,
+      "eval_overall_accuracy": 0.900314228811684,
+      "eval_overall_f1": 0.7909030362654712,
+      "eval_overall_precision": 0.7917837149954986,
+      "eval_overall_recall": 0.790024314467108,
+      "eval_runtime": 1162.7611,
+      "eval_samples_per_second": 56.503,
+      "eval_steps_per_second": 0.442,
+      "step": 17000
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 10.458712577819824,
+      "learning_rate": 1.772319056894938e-05,
+      "loss": 0.3324,
+      "step": 18000
+    },
+    {
+      "epoch": 0.8,
+      "eval_LOC_f1": 0.8134620950977556,
+      "eval_ORG_f1": 0.701657121225073,
+      "eval_PER_f1": 0.8402019239186469,
+      "eval_loss": 0.3189184069633484,
+      "eval_overall_accuracy": 0.9007409983245345,
+      "eval_overall_f1": 0.792014971092267,
+      "eval_overall_precision": 0.7852093276424006,
+      "eval_overall_recall": 0.7989396190733487,
+      "eval_runtime": 1164.2796,
+      "eval_samples_per_second": 56.43,
+      "eval_steps_per_second": 0.441,
+      "step": 18000
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 9.315893173217773,
+      "learning_rate": 1.7596701156113234e-05,
+      "loss": 0.3217,
+      "step": 19000
+    },
+    {
+      "epoch": 0.84,
+      "eval_LOC_f1": 0.8004060309598606,
+      "eval_ORG_f1": 0.7067041425030539,
+      "eval_PER_f1": 0.8399988176524489,
+      "eval_loss": 0.33454081416130066,
+      "eval_overall_accuracy": 0.8964947997344546,
+      "eval_overall_f1": 0.7874678014335652,
+      "eval_overall_precision": 0.7771836007130125,
+      "eval_overall_recall": 0.7980278265568013,
+      "eval_runtime": 1143.6223,
+      "eval_samples_per_second": 57.449,
+      "eval_steps_per_second": 0.449,
+      "step": 19000
+    },
+    {
+      "epoch": 0.89,
+      "grad_norm": 15.540427207946777,
+      "learning_rate": 1.747021174327709e-05,
+      "loss": 0.315,
+      "step": 20000
+    },
+    {
+      "epoch": 0.89,
+      "eval_LOC_f1": 0.8207858780210457,
+      "eval_ORG_f1": 0.7015760412476689,
+      "eval_PER_f1": 0.8415452051584938,
+      "eval_loss": 0.3044009804725647,
+      "eval_overall_accuracy": 0.9041500964182974,
+      "eval_overall_f1": 0.7948658900027092,
+      "eval_overall_precision": 0.7971335416383644,
+      "eval_overall_recall": 0.792611103606646,
+      "eval_runtime": 1139.9409,
+      "eval_samples_per_second": 57.635,
+      "eval_steps_per_second": 0.451,
+      "step": 20000
+    },
+    {
+      "epoch": 0.93,
+      "grad_norm": 23.639232635498047,
+      "learning_rate": 1.7343722330440944e-05,
+      "loss": 0.31,
+      "step": 21000
+    },
+    {
+      "epoch": 0.93,
+      "eval_LOC_f1": 0.816508239958933,
+      "eval_ORG_f1": 0.7147606382978723,
+      "eval_PER_f1": 0.8431382131302962,
+      "eval_loss": 0.31776291131973267,
+      "eval_overall_accuracy": 0.9023190971453862,
+      "eval_overall_f1": 0.7969938869469123,
+      "eval_overall_precision": 0.7880406435895235,
+      "eval_overall_recall": 0.8061529109820343,
+      "eval_runtime": 1142.8251,
+      "eval_samples_per_second": 57.489,
+      "eval_steps_per_second": 0.45,
+      "step": 21000
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 4.997385501861572,
+      "learning_rate": 1.7217232917604798e-05,
+      "loss": 0.3211,
+      "step": 22000
+    },
+    {
+      "epoch": 0.97,
+      "eval_LOC_f1": 0.8216905726486368,
+      "eval_ORG_f1": 0.7168825837058662,
+      "eval_PER_f1": 0.8431281995853436,
+      "eval_loss": 0.30595797300338745,
+      "eval_overall_accuracy": 0.9043296557392596,
+      "eval_overall_f1": 0.798254501680293,
+      "eval_overall_precision": 0.7904963740521209,
+      "eval_overall_recall": 0.8061664190193165,
+      "eval_runtime": 1106.5669,
+      "eval_samples_per_second": 59.373,
+      "eval_steps_per_second": 0.464,
+      "step": 22000
+    },
+    {
+      "epoch": 1.02,
+      "grad_norm": 28.196109771728516,
+      "learning_rate": 1.709074350476865e-05,
+      "loss": 0.295,
+      "step": 23000
+    },
+    {
+      "epoch": 1.02,
+      "eval_LOC_f1": 0.8204851561325467,
+      "eval_ORG_f1": 0.7065405753329702,
+      "eval_PER_f1": 0.8443901962698954,
+      "eval_loss": 0.31898969411849976,
+      "eval_overall_accuracy": 0.9047918313153985,
+      "eval_overall_f1": 0.7976778330942538,
+      "eval_overall_precision": 0.7972607360928381,
+      "eval_overall_recall": 0.7980953667432122,
+      "eval_runtime": 1135.3077,
+      "eval_samples_per_second": 57.87,
+      "eval_steps_per_second": 0.453,
+      "step": 23000
+    },
+    {
+      "epoch": 1.06,
+      "grad_norm": 19.183074951171875,
+      "learning_rate": 1.6964254091932504e-05,
+      "loss": 0.2802,
+      "step": 24000
+    },
+    {
+      "epoch": 1.06,
+      "eval_LOC_f1": 0.8284965702953003,
+      "eval_ORG_f1": 0.7225369919460572,
+      "eval_PER_f1": 0.8405619568962218,
+      "eval_loss": 0.3125617206096649,
+      "eval_overall_accuracy": 0.9055676034520912,
+      "eval_overall_f1": 0.802513357212895,
+      "eval_overall_precision": 0.7926840517070985,
+      "eval_overall_recall": 0.8125894907469945,
+      "eval_runtime": 1161.4292,
+      "eval_samples_per_second": 56.568,
+      "eval_steps_per_second": 0.443,
+      "step": 24000
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 6.878443717956543,
+      "learning_rate": 1.683776467909636e-05,
+      "loss": 0.264,
+      "step": 25000
+    },
+    {
+      "epoch": 1.11,
+      "eval_LOC_f1": 0.8208462732919255,
+      "eval_ORG_f1": 0.7125242718446602,
+      "eval_PER_f1": 0.8417305324641859,
+      "eval_loss": 0.31377699971199036,
+      "eval_overall_accuracy": 0.9023981285366548,
+      "eval_overall_f1": 0.7984267322416735,
+      "eval_overall_precision": 0.7869028800808076,
+      "eval_overall_recall": 0.8102931244090233,
+      "eval_runtime": 1172.0666,
+      "eval_samples_per_second": 56.055,
+      "eval_steps_per_second": 0.439,
+      "step": 25000
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 0.35791274905204773,
+      "learning_rate": 1.6711275266260215e-05,
+      "loss": 0.2745,
+      "step": 26000
+    },
+    {
+      "epoch": 1.15,
+      "eval_LOC_f1": 0.8062705631894718,
+      "eval_ORG_f1": 0.7220110474396811,
+      "eval_PER_f1": 0.8478144130598216,
+      "eval_loss": 0.32361114025115967,
+      "eval_overall_accuracy": 0.9060771978629912,
+      "eval_overall_f1": 0.7961038875365302,
+      "eval_overall_precision": 0.7780019856363707,
+      "eval_overall_recall": 0.815068215588275,
+      "eval_runtime": 1158.6471,
+      "eval_samples_per_second": 56.704,
+      "eval_steps_per_second": 0.444,
+      "step": 26000
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 3.953798294067383,
+      "learning_rate": 1.658478585342407e-05,
+      "loss": 0.2694,
+      "step": 27000
+    },
+    {
+      "epoch": 1.2,
+      "eval_LOC_f1": 0.8275287422548819,
+      "eval_ORG_f1": 0.7351823270599618,
+      "eval_PER_f1": 0.8573874707236907,
+      "eval_loss": 0.31383150815963745,
+      "eval_overall_accuracy": 0.9091986216925363,
+      "eval_overall_f1": 0.8109861146610988,
+      "eval_overall_precision": 0.7973993589785434,
+      "eval_overall_recall": 0.8250439011211671,
+      "eval_runtime": 1141.8213,
+      "eval_samples_per_second": 57.54,
+      "eval_steps_per_second": 0.45,
+      "step": 27000
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 26.60283660888672,
+      "learning_rate": 1.6458296440587925e-05,
+      "loss": 0.2718,
+      "step": 28000
+    },
+    {
+      "epoch": 1.24,
+      "eval_LOC_f1": 0.8314516716439654,
+      "eval_ORG_f1": 0.7274378529428888,
+      "eval_PER_f1": 0.8476108935940972,
+      "eval_loss": 0.2855270504951477,
+      "eval_overall_accuracy": 0.9096057914203521,
+      "eval_overall_f1": 0.8072466863328752,
+      "eval_overall_precision": 0.7978009076038204,
+      "eval_overall_recall": 0.8169188166959341,
+      "eval_runtime": 1139.8571,
+      "eval_samples_per_second": 57.639,
+      "eval_steps_per_second": 0.451,
+      "step": 28000
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 3.3331353664398193,
+      "learning_rate": 1.633180702775178e-05,
+      "loss": 0.2756,
+      "step": 29000
+    },
+    {
+      "epoch": 1.28,
+      "eval_LOC_f1": 0.8289485514934634,
+      "eval_ORG_f1": 0.7433670842629141,
+      "eval_PER_f1": 0.8584255314903678,
+      "eval_loss": 0.3032143712043762,
+      "eval_overall_accuracy": 0.9096652230265861,
+      "eval_overall_f1": 0.813805795355685,
+      "eval_overall_precision": 0.8064124668435013,
+      "eval_overall_recall": 0.8213359448872078,
+      "eval_runtime": 1133.0881,
+      "eval_samples_per_second": 57.983,
+      "eval_steps_per_second": 0.454,
+      "step": 29000
+    },
+    {
+      "epoch": 1.33,
+      "grad_norm": 8.788959503173828,
+      "learning_rate": 1.6205317614915632e-05,
+      "loss": 0.2729,
+      "step": 30000
+    },
+    {
+      "epoch": 1.33,
+      "eval_LOC_f1": 0.8293964198974544,
+      "eval_ORG_f1": 0.7276638026769536,
+      "eval_PER_f1": 0.840676871652676,
+      "eval_loss": 0.3058653771877289,
+      "eval_overall_accuracy": 0.9050023709417381,
+      "eval_overall_f1": 0.8045148879389981,
+      "eval_overall_precision": 0.7890921844858829,
+      "eval_overall_recall": 0.8205524787248413,
+      "eval_runtime": 1144.4868,
+      "eval_samples_per_second": 57.406,
+      "eval_steps_per_second": 0.449,
+      "step": 30000
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 3.506653070449829,
+      "learning_rate": 1.607882820207949e-05,
+      "loss": 0.2617,
+      "step": 31000
+    },
+    {
+      "epoch": 1.37,
+      "eval_LOC_f1": 0.8289568604818929,
+      "eval_ORG_f1": 0.7348994804608087,
+      "eval_PER_f1": 0.8568195366413063,
+      "eval_loss": 0.29605427384376526,
+      "eval_overall_accuracy": 0.9115885309644991,
+      "eval_overall_f1": 0.8102905743459481,
+      "eval_overall_precision": 0.804122408529488,
+      "eval_overall_recall": 0.8165540996893151,
+      "eval_runtime": 1167.0926,
+      "eval_samples_per_second": 56.294,
+      "eval_steps_per_second": 0.44,
+      "step": 31000
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 7.686177730560303,
+      "learning_rate": 1.5952338789243342e-05,
+      "loss": 0.2563,
+      "step": 32000
+    },
+    {
+      "epoch": 1.42,
+      "eval_LOC_f1": 0.8341975970139766,
+      "eval_ORG_f1": 0.7425018373560738,
+      "eval_PER_f1": 0.8481429974617561,
+      "eval_loss": 0.2977641522884369,
+      "eval_overall_accuracy": 0.9102247652767679,
+      "eval_overall_f1": 0.8126805977345553,
+      "eval_overall_precision": 0.8058246625188412,
+      "eval_overall_recall": 0.8196541942455762,
+      "eval_runtime": 1172.5301,
+      "eval_samples_per_second": 56.033,
+      "eval_steps_per_second": 0.438,
+      "step": 32000
+    },
+    {
+      "epoch": 1.46,
+      "grad_norm": 20.15245246887207,
+      "learning_rate": 1.5825849376407196e-05,
+      "loss": 0.2709,
+      "step": 33000
+    },
+    {
+      "epoch": 1.46,
+      "eval_LOC_f1": 0.8421714021714021,
+      "eval_ORG_f1": 0.7459342165372317,
+      "eval_PER_f1": 0.8616692197460096,
+      "eval_loss": 0.287436842918396,
+      "eval_overall_accuracy": 0.91337590490943,
+      "eval_overall_f1": 0.820439775061454,
+      "eval_overall_precision": 0.8180981801087905,
+      "eval_overall_recall": 0.8227948129136836,
+      "eval_runtime": 1126.2379,
+      "eval_samples_per_second": 58.336,
+      "eval_steps_per_second": 0.456,
+      "step": 33000
+    },
+    {
+      "epoch": 1.51,
+      "grad_norm": 15.89538860321045,
+      "learning_rate": 1.569935996357105e-05,
+      "loss": 0.2622,
+      "step": 34000
+    },
+    {
+      "epoch": 1.51,
+      "eval_LOC_f1": 0.8303916383552803,
+      "eval_ORG_f1": 0.7346764810911061,
+      "eval_PER_f1": 0.8459822750330616,
+      "eval_loss": 0.29386088252067566,
+      "eval_overall_accuracy": 0.9097006290898745,
+      "eval_overall_f1": 0.807937429511435,
+      "eval_overall_precision": 0.7984159225503515,
+      "eval_overall_recall": 0.8176887748210185,
+      "eval_runtime": 1137.1401,
+      "eval_samples_per_second": 57.777,
+      "eval_steps_per_second": 0.452,
+      "step": 34000
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 8.729211807250977,
+      "learning_rate": 1.5572870550734906e-05,
+      "loss": 0.2509,
+      "step": 35000
+    },
+    {
+      "epoch": 1.55,
+      "eval_LOC_f1": 0.8372501808127878,
+      "eval_ORG_f1": 0.7348823687270776,
+      "eval_PER_f1": 0.8613693584255022,
+      "eval_loss": 0.299064964056015,
+      "eval_overall_accuracy": 0.9102475263174533,
+      "eval_overall_f1": 0.8149951120244199,
+      "eval_overall_precision": 0.8026696708104638,
+      "eval_overall_recall": 0.8277049844657571,
+      "eval_runtime": 1160.5705,
+      "eval_samples_per_second": 56.61,
+      "eval_steps_per_second": 0.443,
+      "step": 35000
+    },
+    {
+      "epoch": 1.59,
+      "grad_norm": 10.565467834472656,
+      "learning_rate": 1.544638113789876e-05,
+      "loss": 0.2612,
+      "step": 36000
+    },
+    {
+      "epoch": 1.59,
+      "eval_LOC_f1": 0.8352879238980465,
+      "eval_ORG_f1": 0.7413166030187306,
+      "eval_PER_f1": 0.8542469914709663,
+      "eval_loss": 0.326786607503891,
+      "eval_overall_accuracy": 0.9090159011159232,
+      "eval_overall_f1": 0.8142790666755302,
+      "eval_overall_precision": 0.8016492146596859,
+      "eval_overall_recall": 0.8273132513845738,
+      "eval_runtime": 1136.8694,
+      "eval_samples_per_second": 57.79,
+      "eval_steps_per_second": 0.452,
+      "step": 36000
+    },
+    {
+      "epoch": 1.64,
+      "grad_norm": 2.5293238162994385,
+      "learning_rate": 1.5319891725062616e-05,
+      "loss": 0.2471,
+      "step": 37000
+    },
+    {
+      "epoch": 1.64,
+      "eval_LOC_f1": 0.8327791444765844,
+      "eval_ORG_f1": 0.7424818446110567,
+      "eval_PER_f1": 0.8635365346946043,
+      "eval_loss": 0.2980642318725586,
+      "eval_overall_accuracy": 0.9123959156576993,
+      "eval_overall_f1": 0.8174172435155008,
+      "eval_overall_precision": 0.814958779773893,
+      "eval_overall_recall": 0.8198905848980144,
+      "eval_runtime": 1118.4252,
+      "eval_samples_per_second": 58.743,
+      "eval_steps_per_second": 0.46,
+      "step": 37000
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 7.829558372497559,
+      "learning_rate": 1.5193402312226468e-05,
+      "loss": 0.2618,
+      "step": 38000
+    },
+    {
+      "epoch": 1.68,
+      "eval_LOC_f1": 0.839616830189353,
+      "eval_ORG_f1": 0.7407022341759993,
+      "eval_PER_f1": 0.8689252708731676,
+      "eval_loss": 0.28679007291793823,
+      "eval_overall_accuracy": 0.9135636834950842,
+      "eval_overall_f1": 0.8203098518533288,
+      "eval_overall_precision": 0.807435659239294,
+      "eval_overall_recall": 0.8336012427394299,
+      "eval_runtime": 1154.975,
+      "eval_samples_per_second": 56.884,
+      "eval_steps_per_second": 0.445,
+      "step": 38000
+    },
+    {
+      "epoch": 1.73,
+      "grad_norm": 2.377560615539551,
+      "learning_rate": 1.5066912899390323e-05,
+      "loss": 0.2691,
+      "step": 39000
+    },
+    {
+      "epoch": 1.73,
+      "eval_LOC_f1": 0.8439257798232158,
+      "eval_ORG_f1": 0.7472704832431407,
+      "eval_PER_f1": 0.8600562382634247,
+      "eval_loss": 0.2865625023841858,
+      "eval_overall_accuracy": 0.9137938229064585,
+      "eval_overall_f1": 0.8220520446598079,
+      "eval_overall_precision": 0.82085350824618,
+      "eval_overall_recall": 0.8232540861812778,
+      "eval_runtime": 1172.051,
+      "eval_samples_per_second": 56.056,
+      "eval_steps_per_second": 0.439,
+      "step": 39000
+    },
+    {
+      "epoch": 1.77,
+      "grad_norm": 18.09626579284668,
+      "learning_rate": 1.4940423486554176e-05,
+      "loss": 0.2528,
+      "step": 40000
+    },
+    {
+      "epoch": 1.77,
+      "eval_LOC_f1": 0.8390521604627537,
+      "eval_ORG_f1": 0.744894884454025,
+      "eval_PER_f1": 0.8507444298075191,
+      "eval_loss": 0.282395601272583,
+      "eval_overall_accuracy": 0.9139050991053647,
+      "eval_overall_f1": 0.816164112375524,
+      "eval_overall_precision": 0.802982967542908,
+      "eval_overall_recall": 0.8297852222072133,
+      "eval_runtime": 1169.9185,
+      "eval_samples_per_second": 56.158,
+      "eval_steps_per_second": 0.439,
+      "step": 40000
+    },
+    {
+      "epoch": 1.82,
+      "grad_norm": 20.97409439086914,
+      "learning_rate": 1.481393407371803e-05,
+      "loss": 0.2626,
+      "step": 41000
+    },
+    {
+      "epoch": 1.82,
+      "eval_LOC_f1": 0.84465632972083,
+      "eval_ORG_f1": 0.7437936213446418,
+      "eval_PER_f1": 0.8681983071342201,
+      "eval_loss": 0.28616106510162354,
+      "eval_overall_accuracy": 0.9150418866373724,
+      "eval_overall_f1": 0.8221516066157305,
+      "eval_overall_precision": 0.8086165341833169,
+      "eval_overall_recall": 0.8361475077671214,
+      "eval_runtime": 1122.7328,
+      "eval_samples_per_second": 58.518,
+      "eval_steps_per_second": 0.458,
+      "step": 41000
+    },
+    {
+      "epoch": 1.86,
+      "grad_norm": 8.107739448547363,
+      "learning_rate": 1.4687444660881885e-05,
+      "loss": 0.2641,
+      "step": 42000
+    },
+    {
+      "epoch": 1.86,
+      "eval_LOC_f1": 0.8411461369445316,
+      "eval_ORG_f1": 0.7528591481234324,
+      "eval_PER_f1": 0.860931649302524,
+      "eval_loss": 0.2878096401691437,
+      "eval_overall_accuracy": 0.9135719027597762,
+      "eval_overall_f1": 0.822460655710519,
+      "eval_overall_precision": 0.8080289354491837,
+      "eval_overall_recall": 0.8374172632716467,
+      "eval_runtime": 1135.0969,
+      "eval_samples_per_second": 57.881,
+      "eval_steps_per_second": 0.453,
+      "step": 42000
+    },
+    {
+      "epoch": 1.9,
+      "grad_norm": 7.000519275665283,
+      "learning_rate": 1.456095524804574e-05,
+      "loss": 0.2437,
+      "step": 43000
+    },
+    {
+      "epoch": 1.9,
+      "eval_LOC_f1": 0.8380676328502416,
+      "eval_ORG_f1": 0.7540926707930814,
+      "eval_PER_f1": 0.8668607569521133,
+      "eval_loss": 0.29815977811813354,
+      "eval_overall_accuracy": 0.9138791768090285,
+      "eval_overall_f1": 0.8234281998699545,
+      "eval_overall_precision": 0.8131919517897718,
+      "eval_overall_recall": 0.8339254356342024,
+      "eval_runtime": 1138.8201,
+      "eval_samples_per_second": 57.691,
+      "eval_steps_per_second": 0.451,
+      "step": 43000
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 21.582551956176758,
+      "learning_rate": 1.4434465835209595e-05,
+      "loss": 0.2512,
+      "step": 44000
+    },
+    {
+      "epoch": 1.95,
+      "eval_LOC_f1": 0.8458274398868458,
+      "eval_ORG_f1": 0.7543260552125373,
+      "eval_PER_f1": 0.8698453095796715,
+      "eval_loss": 0.29138705134391785,
+      "eval_overall_accuracy": 0.9158928966585528,
+      "eval_overall_f1": 0.8271313828477359,
+      "eval_overall_precision": 0.8262422580014692,
+      "eval_overall_recall": 0.8280224233418885,
+      "eval_runtime": 1149.345,
+      "eval_samples_per_second": 57.163,
+      "eval_steps_per_second": 0.447,
+      "step": 44000
+    },
+    {
+      "epoch": 1.99,
+      "grad_norm": 6.298182010650635,
+      "learning_rate": 1.4307976422373449e-05,
+      "loss": 0.2382,
+      "step": 45000
+    },
+    {
+      "epoch": 1.99,
+      "eval_LOC_f1": 0.8426425710908074,
+      "eval_ORG_f1": 0.75806763945369,
+      "eval_PER_f1": 0.8686175216669002,
+      "eval_loss": 0.2709099054336548,
+      "eval_overall_accuracy": 0.917617677741599,
+      "eval_overall_f1": 0.8263731673134973,
+      "eval_overall_precision": 0.8243957196246605,
+      "eval_overall_recall": 0.828360124273943,
+      "eval_runtime": 1185.4869,
+      "eval_samples_per_second": 55.42,
+      "eval_steps_per_second": 0.434,
+      "step": 45000
+    },
+    {
+      "epoch": 2.04,
+      "grad_norm": 109.3368148803711,
+      "learning_rate": 1.4181487009537302e-05,
+      "loss": 0.224,
+      "step": 46000
+    },
+    {
+      "epoch": 2.04,
+      "eval_LOC_f1": 0.8326199042341798,
+      "eval_ORG_f1": 0.7524188296585729,
+      "eval_PER_f1": 0.8686386257826296,
+      "eval_loss": 0.27940934896469116,
+      "eval_overall_accuracy": 0.9156703442607403,
+      "eval_overall_f1": 0.8215930348490961,
+      "eval_overall_precision": 0.8113848768054376,
+      "eval_overall_recall": 0.8320613264892611,
+      "eval_runtime": 1151.7704,
+      "eval_samples_per_second": 57.043,
+      "eval_steps_per_second": 0.446,
+      "step": 46000
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 5.203092098236084,
+      "learning_rate": 1.4054997596701157e-05,
+      "loss": 0.2107,
+      "step": 47000
+    },
+    {
+      "epoch": 2.08,
+      "eval_LOC_f1": 0.849728086288491,
+      "eval_ORG_f1": 0.7564509355606366,
+      "eval_PER_f1": 0.8641517093397512,
+      "eval_loss": 0.30060428380966187,
+      "eval_overall_accuracy": 0.9154243985711125,
+      "eval_overall_f1": 0.8280268803929288,
+      "eval_overall_precision": 0.8182579564489112,
+      "eval_overall_recall": 0.8380318789679859,
+      "eval_runtime": 1142.4349,
+      "eval_samples_per_second": 57.509,
+      "eval_steps_per_second": 0.45,
+      "step": 47000
+    },
+    {
+      "epoch": 2.13,
+      "grad_norm": 7.594598770141602,
+      "learning_rate": 1.3928508183865012e-05,
+      "loss": 0.2141,
+      "step": 48000
+    },
+    {
+      "epoch": 2.13,
+      "eval_LOC_f1": 0.8461600770414638,
+      "eval_ORG_f1": 0.7533498536716638,
+      "eval_PER_f1": 0.8708999200465901,
+      "eval_loss": 0.29842332005500793,
+      "eval_overall_accuracy": 0.9150633831757974,
+      "eval_overall_f1": 0.8270264355887414,
+      "eval_overall_precision": 0.8122651278811522,
+      "eval_overall_recall": 0.8423341888423612,
+      "eval_runtime": 1138.7795,
+      "eval_samples_per_second": 57.693,
+      "eval_steps_per_second": 0.451,
+      "step": 48000
+    },
+    {
+      "epoch": 2.17,
+      "grad_norm": 22.177534103393555,
+      "learning_rate": 1.3802018771028867e-05,
+      "loss": 0.209,
+      "step": 49000
+    },
+    {
+      "epoch": 2.17,
+      "eval_LOC_f1": 0.8327727180910393,
+      "eval_ORG_f1": 0.7525920197211012,
+      "eval_PER_f1": 0.8653763775847968,
+      "eval_loss": 0.29675909876823425,
+      "eval_overall_accuracy": 0.9130692631113078,
+      "eval_overall_f1": 0.8217937881890739,
+      "eval_overall_precision": 0.8130664427888652,
+      "eval_overall_recall": 0.8307105227610428,
+      "eval_runtime": 1124.5682,
+      "eval_samples_per_second": 58.422,
+      "eval_steps_per_second": 0.457,
+      "step": 49000
+    },
+    {
+      "epoch": 2.21,
+      "grad_norm": 12.588418960571289,
+      "learning_rate": 1.3675529358192721e-05,
+      "loss": 0.2146,
+      "step": 50000
+    },
+    {
+      "epoch": 2.21,
+      "eval_LOC_f1": 0.8454975253010267,
+      "eval_ORG_f1": 0.7527126724109324,
+      "eval_PER_f1": 0.862459641913707,
+      "eval_loss": 0.32881608605384827,
+      "eval_overall_accuracy": 0.9147478898618532,
+      "eval_overall_f1": 0.8233820181778908,
+      "eval_overall_precision": 0.8104674364593896,
+      "eval_overall_recall": 0.8367148453329731,
+      "eval_runtime": 1149.76,
+      "eval_samples_per_second": 57.142,
+      "eval_steps_per_second": 0.447,
+      "step": 50000
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 31.024621963500977,
+      "learning_rate": 1.3549039945356574e-05,
+      "loss": 0.2193,
+      "step": 51000
+    },
+    {
+      "epoch": 2.26,
+      "eval_LOC_f1": 0.8356590369572263,
+      "eval_ORG_f1": 0.7558741537236161,
+      "eval_PER_f1": 0.8684495133221907,
+      "eval_loss": 0.29472604393959045,
+      "eval_overall_accuracy": 0.9149875130401796,
+      "eval_overall_f1": 0.8246923664019925,
+      "eval_overall_precision": 0.8175440692364873,
+      "eval_overall_recall": 0.8319667702282858,
+      "eval_runtime": 1163.9713,
+      "eval_samples_per_second": 56.445,
+      "eval_steps_per_second": 0.442,
+      "step": 51000
+    },
+    {
+      "epoch": 2.3,
+      "grad_norm": 53.772743225097656,
+      "learning_rate": 1.3422550532520428e-05,
+      "loss": 0.2194,
+      "step": 52000
+    },
+    {
+      "epoch": 2.3,
+      "eval_LOC_f1": 0.8537082537082539,
+      "eval_ORG_f1": 0.7621269943445675,
+      "eval_PER_f1": 0.8700663432022973,
+      "eval_loss": 0.301561564207077,
+      "eval_overall_accuracy": 0.9174033446084785,
+      "eval_overall_f1": 0.8325074863365964,
+      "eval_overall_precision": 0.8276285576974421,
+      "eval_overall_recall": 0.837444279346211,
+      "eval_runtime": 1169.1445,
+      "eval_samples_per_second": 56.195,
+      "eval_steps_per_second": 0.44,
+      "step": 52000
+    },
+    {
+      "epoch": 2.35,
+      "grad_norm": 6.144921779632568,
+      "learning_rate": 1.3296061119684283e-05,
+      "loss": 0.2077,
+      "step": 53000
+    },
+    {
+      "epoch": 2.35,
+      "eval_LOC_f1": 0.8536625787205835,
+      "eval_ORG_f1": 0.7566537335783813,
+      "eval_PER_f1": 0.8741282228588362,
+      "eval_loss": 0.30693191289901733,
+      "eval_overall_accuracy": 0.917192804982139,
+      "eval_overall_f1": 0.8319194061505832,
+      "eval_overall_precision": 0.8267368336801665,
+      "eval_overall_recall": 0.8371673645819262,
+      "eval_runtime": 1124.2075,
+      "eval_samples_per_second": 58.441,
+      "eval_steps_per_second": 0.457,
+      "step": 53000
+    },
+    {
+      "epoch": 2.39,
+      "grad_norm": 21.740257263183594,
+      "learning_rate": 1.3169571706848138e-05,
+      "loss": 0.2149,
+      "step": 54000
+    },
+    {
+      "epoch": 2.39,
+      "eval_LOC_f1": 0.8524622644907958,
+      "eval_ORG_f1": 0.7579534927354388,
+      "eval_PER_f1": 0.8705574261129816,
+      "eval_loss": 0.2857012152671814,
+      "eval_overall_accuracy": 0.9183801726045585,
+      "eval_overall_f1": 0.8315770745282607,
+      "eval_overall_precision": 0.826673430022626,
+      "eval_overall_recall": 0.8365392408483048,
+      "eval_runtime": 1136.8592,
+      "eval_samples_per_second": 57.791,
+      "eval_steps_per_second": 0.452,
+      "step": 54000
+    },
+    {
+      "epoch": 2.43,
+      "grad_norm": 16.476070404052734,
+      "learning_rate": 1.3043082294011993e-05,
+      "loss": 0.2117,
+      "step": 55000
+    },
+    {
+      "epoch": 2.43,
+      "eval_LOC_f1": 0.8431090256300877,
+      "eval_ORG_f1": 0.7585622008457642,
+      "eval_PER_f1": 0.8652493280417632,
+      "eval_loss": 0.2959577441215515,
+      "eval_overall_accuracy": 0.9159529605159169,
+      "eval_overall_f1": 0.8265480330308926,
+      "eval_overall_precision": 0.8116499671217536,
+      "eval_overall_recall": 0.8420032419289477,
+      "eval_runtime": 1145.489,
+      "eval_samples_per_second": 57.355,
+      "eval_steps_per_second": 0.449,
+      "step": 55000
+    },
+    {
+      "epoch": 2.48,
+      "grad_norm": 3.9474117755889893,
+      "learning_rate": 1.2916592881175847e-05,
+      "loss": 0.2161,
+      "step": 56000
+    },
+    {
+      "epoch": 2.48,
+      "eval_LOC_f1": 0.8557258137637138,
+      "eval_ORG_f1": 0.7627416919286533,
+      "eval_PER_f1": 0.872469800848841,
+      "eval_loss": 0.2923972010612488,
+      "eval_overall_accuracy": 0.9190674295830303,
+      "eval_overall_f1": 0.8336308087845794,
+      "eval_overall_precision": 0.8328917295362149,
+      "eval_overall_recall": 0.8343712008645144,
+      "eval_runtime": 1130.1794,
+      "eval_samples_per_second": 58.132,
+      "eval_steps_per_second": 0.455,
+      "step": 56000
+    },
+    {
+      "epoch": 2.52,
+      "grad_norm": 3.9784624576568604,
+      "learning_rate": 1.27901034683397e-05,
+      "loss": 0.2103,
+      "step": 57000
+    },
+    {
+      "epoch": 2.52,
+      "eval_LOC_f1": 0.8558928976628091,
+      "eval_ORG_f1": 0.7694271394444705,
+      "eval_PER_f1": 0.8743142094796486,
+      "eval_loss": 0.2938244640827179,
+      "eval_overall_accuracy": 0.9200885151582209,
+      "eval_overall_f1": 0.8371521928845412,
+      "eval_overall_precision": 0.8385192914933121,
+      "eval_overall_recall": 0.8357895447791436,
+      "eval_runtime": 1145.4202,
+      "eval_samples_per_second": 57.359,
+      "eval_steps_per_second": 0.449,
+      "step": 57000
+    },
+    {
+      "epoch": 2.57,
+      "grad_norm": 3.8758695125579834,
+      "learning_rate": 1.2663614055503555e-05,
+      "loss": 0.2177,
+      "step": 58000
+    },
+    {
+      "epoch": 2.57,
+      "eval_LOC_f1": 0.8549010401618029,
+      "eval_ORG_f1": 0.7626243802166277,
+      "eval_PER_f1": 0.8736514259492496,
+      "eval_loss": 0.29625701904296875,
+      "eval_overall_accuracy": 0.9183770113489078,
+      "eval_overall_f1": 0.8341873429804383,
+      "eval_overall_precision": 0.8190024340401161,
+      "eval_overall_recall": 0.8499459678508713,
+      "eval_runtime": 773.8476,
+      "eval_samples_per_second": 84.9,
+      "eval_steps_per_second": 0.664,
+      "step": 58000
+    },
+    {
+      "epoch": 2.61,
+      "grad_norm": 17.358938217163086,
+      "learning_rate": 1.2537124642667409e-05,
+      "loss": 0.2103,
+      "step": 59000
+    },
+    {
+      "epoch": 2.61,
+      "eval_LOC_f1": 0.8570758757495153,
+      "eval_ORG_f1": 0.7695653174402283,
+      "eval_PER_f1": 0.8721403453317,
+      "eval_loss": 0.3069661259651184,
+      "eval_overall_accuracy": 0.9178124110896848,
+      "eval_overall_f1": 0.8367424052431827,
+      "eval_overall_precision": 0.82861021888142,
+      "eval_overall_recall": 0.8450357962987978,
+      "eval_runtime": 774.0633,
+      "eval_samples_per_second": 84.877,
+      "eval_steps_per_second": 0.664,
+      "step": 59000
+    },
+    {
+      "epoch": 2.66,
+      "grad_norm": 14.227778434753418,
+      "learning_rate": 1.2410635229831265e-05,
+      "loss": 0.2097,
+      "step": 60000
+    },
+    {
+      "epoch": 2.66,
+      "eval_LOC_f1": 0.8508951924685824,
+      "eval_ORG_f1": 0.764511452594955,
+      "eval_PER_f1": 0.8779140735997439,
+      "eval_loss": 0.2908080518245697,
+      "eval_overall_accuracy": 0.9176454967913256,
+      "eval_overall_f1": 0.8350538165366573,
+      "eval_overall_precision": 0.8251374767575266,
+      "eval_overall_recall": 0.8452114007834661,
+      "eval_runtime": 809.1979,
+      "eval_samples_per_second": 81.192,
+      "eval_steps_per_second": 0.635,
+      "step": 60000
+    },
+    {
+      "epoch": 2.7,
+      "grad_norm": 12.757182121276855,
+      "learning_rate": 1.2284145816995119e-05,
+      "loss": 0.2136,
+      "step": 61000
+    },
+    {
+      "epoch": 2.7,
+      "eval_LOC_f1": 0.8522066339221911,
+      "eval_ORG_f1": 0.7652064716500914,
+      "eval_PER_f1": 0.8736700693279988,
+      "eval_loss": 0.27922523021698,
+      "eval_overall_accuracy": 0.9190965131350173,
+      "eval_overall_f1": 0.8337437566493549,
+      "eval_overall_precision": 0.8185486232681032,
+      "eval_overall_recall": 0.8495137106578414,
+      "eval_runtime": 777.5103,
+      "eval_samples_per_second": 84.5,
+      "eval_steps_per_second": 0.661,
+      "step": 61000
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 2.6560215950012207,
+      "learning_rate": 1.2157656404158972e-05,
+      "loss": 0.2027,
+      "step": 62000
+    },
+    {
+      "epoch": 2.74,
+      "eval_LOC_f1": 0.8569138059936766,
+      "eval_ORG_f1": 0.7663184297393547,
+      "eval_PER_f1": 0.8801295038334722,
+      "eval_loss": 0.26576361060142517,
+      "eval_overall_accuracy": 0.9209066481206335,
+      "eval_overall_f1": 0.8389562164534314,
+      "eval_overall_precision": 0.8386447998920159,
+      "eval_overall_recall": 0.8392678643793057,
+      "eval_runtime": 773.3284,
+      "eval_samples_per_second": 84.957,
+      "eval_steps_per_second": 0.665,
+      "step": 62000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 158116,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 7,
+  "save_steps": 500,
+  "total_flos": 2.0466275035151616e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4346872abe35c783c70b6e579ad7ad19d9dff98da06133a0d0781d4180ab6dc
+size 4856