Upload 11 files

config.json

@@ -10,24 +10,24 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "O",
-    "1": "B-PER",
-    "2": "I-PER",
-    "3": "B-ORG",
-    "4": "I-ORG",
-    "5": "B-LOC",
-    "6": "I-LOC"
+    "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": {
-    "O": 0,
-    "B-PER": 1,
-    "I-PER": 2,
-    "B-ORG": 3,
-    "I-ORG": 4,
-    "B-LOC": 5,
-    "I-LOC": 6
+    "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-12,
   "max_position_embeddings": 512,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3464507543bcd5a63144ad9e7c40083a605e6ae28e76daa26b9df5a09404d6c
+oid sha256:13aae2a832fcc9a396a3d5b345e7240ec562d07e7044b6689f124e8871053e04
 size 709096284

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c4762c65f673b48b6e30a37ae48167d77dc5e9c0f1ef957bfd9c54a2b27f232
+oid sha256:ffd8b3f7f6f848ed76978c04c531e580dfaa7e4f8acac2247bfa36c8d866ce66
 size 1418312250

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e173b028a7b49b7598e1d0be461b9d1381eae9728de0257e5a2a3230ba0a5ad5
+oid sha256:99ecc6ce06f99c747d7eccd8c8e5ba6a5a7d3824113e1e8d26da659a5de2e704
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e9d47ac0c4a6583ddb94f81f2a71d6ba82f5603d2acd75906e08ceb02509c00
+oid sha256:c01948096ca09d6133b37c1d4832dd98f343fc0f51ab59a285b9840e309e1d5c
 size 1064

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 0.9518328315919957,
+  "best_metric": 0.257554292678833,
+  "best_model_checkpoint": "./checkpoint/checkpoint-45000",
+  "epoch": 1.9922082521692934,
   "eval_steps": 1000,
-  "global_step": 21500,
+  "global_step": 45000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -469,6 +469,534 @@
       "eval_samples_per_second": 76.309,
       "eval_steps_per_second": 0.15,
       "step": 21000
+    },
+    {
+      "epoch": 0.97,
+      "grad_norm": 9.784720420837402,
+      "learning_rate": 1.7217232917604798e-05,
+      "loss": 0.2862,
+      "step": 22000
+    },
+    {
+      "epoch": 0.97,
+      "eval_LOC_f1": 0.8464706167336854,
+      "eval_ORG_f1": 0.7516536234342064,
+      "eval_PER_f1": 0.8579948502396776,
+      "eval_loss": 0.2908647060394287,
+      "eval_overall_accuracy": 0.9129029665003394,
+      "eval_overall_f1": 0.822899354947748,
+      "eval_overall_precision": 0.8222161241750046,
+      "eval_overall_recall": 0.823583722143129,
+      "eval_runtime": 998.6892,
+      "eval_samples_per_second": 65.786,
+      "eval_steps_per_second": 0.515,
+      "step": 22000
+    },
+    {
+      "epoch": 1.02,
+      "grad_norm": 10.310678482055664,
+      "learning_rate": 1.709074350476865e-05,
+      "loss": 0.2578,
+      "step": 23000
+    },
+    {
+      "epoch": 1.02,
+      "eval_LOC_f1": 0.841074796336982,
+      "eval_ORG_f1": 0.7516604978958576,
+      "eval_PER_f1": 0.8534542846463892,
+      "eval_loss": 0.31625884771347046,
+      "eval_overall_accuracy": 0.9131046730418835,
+      "eval_overall_f1": 0.8203140861475114,
+      "eval_overall_precision": 0.8292370287689482,
+      "eval_overall_recall": 0.811581128511622,
+      "eval_runtime": 992.6607,
+      "eval_samples_per_second": 66.186,
+      "eval_steps_per_second": 0.518,
+      "step": 23000
+    },
+    {
+      "epoch": 1.06,
+      "grad_norm": 12.428755760192871,
+      "learning_rate": 1.6964254091932504e-05,
+      "loss": 0.241,
+      "step": 24000
+    },
+    {
+      "epoch": 1.06,
+      "eval_LOC_f1": 0.8458245540438758,
+      "eval_ORG_f1": 0.7417449698379979,
+      "eval_PER_f1": 0.8449764637383674,
+      "eval_loss": 0.31306418776512146,
+      "eval_overall_accuracy": 0.910274082719034,
+      "eval_overall_f1": 0.8166761890407407,
+      "eval_overall_precision": 0.8123452830057111,
+      "eval_overall_recall": 0.8210535218414784,
+      "eval_runtime": 1000.0329,
+      "eval_samples_per_second": 65.698,
+      "eval_steps_per_second": 0.514,
+      "step": 24000
+    },
+    {
+      "epoch": 1.11,
+      "grad_norm": 7.654708385467529,
+      "learning_rate": 1.683776467909636e-05,
+      "loss": 0.2273,
+      "step": 25000
+    },
+    {
+      "epoch": 1.11,
+      "eval_LOC_f1": 0.8447393652873104,
+      "eval_ORG_f1": 0.755644780920822,
+      "eval_PER_f1": 0.8620841122450023,
+      "eval_loss": 0.29408279061317444,
+      "eval_overall_accuracy": 0.9140618487334913,
+      "eval_overall_f1": 0.8249119505874042,
+      "eval_overall_precision": 0.818763755528247,
+      "eval_overall_recall": 0.8311531793129696,
+      "eval_runtime": 941.7518,
+      "eval_samples_per_second": 69.764,
+      "eval_steps_per_second": 0.546,
+      "step": 25000
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 0.1430547833442688,
+      "learning_rate": 1.6711275266260215e-05,
+      "loss": 0.2512,
+      "step": 26000
+    },
+    {
+      "epoch": 1.15,
+      "eval_LOC_f1": 0.8500541479262808,
+      "eval_ORG_f1": 0.7624169717495258,
+      "eval_PER_f1": 0.865798904312724,
+      "eval_loss": 0.2906274199485779,
+      "eval_overall_accuracy": 0.9169281656762454,
+      "eval_overall_f1": 0.8295221228947797,
+      "eval_overall_precision": 0.8304302989207211,
+      "eval_overall_recall": 0.8286159310996152,
+      "eval_runtime": 948.7716,
+      "eval_samples_per_second": 69.247,
+      "eval_steps_per_second": 0.542,
+      "step": 26000
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 2.0707778930664062,
+      "learning_rate": 1.658478585342407e-05,
+      "loss": 0.2355,
+      "step": 27000
+    },
+    {
+      "epoch": 1.2,
+      "eval_LOC_f1": 0.8528026151331602,
+      "eval_ORG_f1": 0.7654581692212784,
+      "eval_PER_f1": 0.8630401626842907,
+      "eval_loss": 0.30151626467704773,
+      "eval_overall_accuracy": 0.9169624036869872,
+      "eval_overall_f1": 0.8306308694863459,
+      "eval_overall_precision": 0.8223685119434121,
+      "eval_overall_recall": 0.8390609362445908,
+      "eval_runtime": 985.7403,
+      "eval_samples_per_second": 66.65,
+      "eval_steps_per_second": 0.521,
+      "step": 27000
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 7.580867290496826,
+      "learning_rate": 1.6458296440587925e-05,
+      "loss": 0.2395,
+      "step": 28000
+    },
+    {
+      "epoch": 1.24,
+      "eval_LOC_f1": 0.8430476991285736,
+      "eval_ORG_f1": 0.7574625494543183,
+      "eval_PER_f1": 0.8637680570098303,
+      "eval_loss": 0.3095172941684723,
+      "eval_overall_accuracy": 0.914303747722428,
+      "eval_overall_f1": 0.8256623551811597,
+      "eval_overall_precision": 0.8230774618293879,
+      "eval_overall_recall": 0.8282635355144271,
+      "eval_runtime": 999.1758,
+      "eval_samples_per_second": 65.754,
+      "eval_steps_per_second": 0.514,
+      "step": 28000
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.15234589576721191,
+      "learning_rate": 1.633180702775178e-05,
+      "loss": 0.2322,
+      "step": 29000
+    },
+    {
+      "epoch": 1.28,
+      "eval_LOC_f1": 0.8442432749878247,
+      "eval_ORG_f1": 0.7601534050352662,
+      "eval_PER_f1": 0.8612064440586263,
+      "eval_loss": 0.2927219569683075,
+      "eval_overall_accuracy": 0.9170986114253731,
+      "eval_overall_f1": 0.8258450109270158,
+      "eval_overall_precision": 0.8182069728832319,
+      "eval_overall_recall": 0.8336269963209901,
+      "eval_runtime": 972.0367,
+      "eval_samples_per_second": 67.59,
+      "eval_steps_per_second": 0.529,
+      "step": 29000
+    },
+    {
+      "epoch": 1.33,
+      "grad_norm": 12.916611671447754,
+      "learning_rate": 1.6205317614915632e-05,
+      "loss": 0.2433,
+      "step": 30000
+    },
+    {
+      "epoch": 1.33,
+      "eval_LOC_f1": 0.8444784033582501,
+      "eval_ORG_f1": 0.7596891611525098,
+      "eval_PER_f1": 0.8642486197465865,
+      "eval_loss": 0.2888515293598175,
+      "eval_overall_accuracy": 0.91348501268295,
+      "eval_overall_f1": 0.8271813480146815,
+      "eval_overall_precision": 0.8160067477678877,
+      "eval_overall_recall": 0.8386662531891801,
+      "eval_runtime": 961.788,
+      "eval_samples_per_second": 68.31,
+      "eval_steps_per_second": 0.534,
+      "step": 30000
+    },
+    {
+      "epoch": 1.37,
+      "grad_norm": 0.4655471444129944,
+      "learning_rate": 1.607882820207949e-05,
+      "loss": 0.2358,
+      "step": 31000
+    },
+    {
+      "epoch": 1.37,
+      "eval_LOC_f1": 0.8478889967668045,
+      "eval_ORG_f1": 0.7620526592328799,
+      "eval_PER_f1": 0.8645434620748959,
+      "eval_loss": 0.29194697737693787,
+      "eval_overall_accuracy": 0.9150346548213073,
+      "eval_overall_f1": 0.8278141773862643,
+      "eval_overall_precision": 0.8183533624875177,
+      "eval_overall_recall": 0.8374962998463555,
+      "eval_runtime": 937.9473,
+      "eval_samples_per_second": 70.047,
+      "eval_steps_per_second": 0.548,
+      "step": 31000
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 4.9739909172058105,
+      "learning_rate": 1.5952338789243342e-05,
+      "loss": 0.2258,
+      "step": 32000
+    },
+    {
+      "epoch": 1.42,
+      "eval_LOC_f1": 0.8552193079338859,
+      "eval_ORG_f1": 0.763369485049633,
+      "eval_PER_f1": 0.8649409287399143,
+      "eval_loss": 0.2976300120353699,
+      "eval_overall_accuracy": 0.9165649450405498,
+      "eval_overall_f1": 0.8319835859514232,
+      "eval_overall_precision": 0.8266665738956187,
+      "eval_overall_recall": 0.8373694374356878,
+      "eval_runtime": 961.624,
+      "eval_samples_per_second": 68.322,
+      "eval_steps_per_second": 0.535,
+      "step": 32000
+    },
+    {
+      "epoch": 1.46,
+      "grad_norm": 13.10745620727539,
+      "learning_rate": 1.5825849376407196e-05,
+      "loss": 0.2372,
+      "step": 33000
+    },
+    {
+      "epoch": 1.46,
+      "eval_LOC_f1": 0.8430370470368156,
+      "eval_ORG_f1": 0.7732836874254703,
+      "eval_PER_f1": 0.8708081138557172,
+      "eval_loss": 0.2847268581390381,
+      "eval_overall_accuracy": 0.91798879970466,
+      "eval_overall_f1": 0.8324051096948847,
+      "eval_overall_precision": 0.8326281644453847,
+      "eval_overall_recall": 0.8321821744217188,
+      "eval_runtime": 1011.399,
+      "eval_samples_per_second": 64.96,
+      "eval_steps_per_second": 0.508,
+      "step": 33000
+    },
+    {
+      "epoch": 1.51,
+      "grad_norm": 2.8712925910949707,
+      "learning_rate": 1.569935996357105e-05,
+      "loss": 0.2358,
+      "step": 34000
+    },
+    {
+      "epoch": 1.51,
+      "eval_LOC_f1": 0.8512200643745139,
+      "eval_ORG_f1": 0.7646914589293013,
+      "eval_PER_f1": 0.8725994049229104,
+      "eval_loss": 0.28364238142967224,
+      "eval_overall_accuracy": 0.9192392314013171,
+      "eval_overall_f1": 0.8323193111236589,
+      "eval_overall_precision": 0.8335312495582227,
+      "eval_overall_recall": 0.831110891842747,
+      "eval_runtime": 994.3994,
+      "eval_samples_per_second": 66.07,
+      "eval_steps_per_second": 0.517,
+      "step": 34000
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 1.0189175605773926,
+      "learning_rate": 1.5572870550734906e-05,
+      "loss": 0.2314,
+      "step": 35000
+    },
+    {
+      "epoch": 1.55,
+      "eval_LOC_f1": 0.8507849855779469,
+      "eval_ORG_f1": 0.7650251812241418,
+      "eval_PER_f1": 0.8689379418372184,
+      "eval_loss": 0.285095751285553,
+      "eval_overall_accuracy": 0.9186356003858475,
+      "eval_overall_f1": 0.8310027409855706,
+      "eval_overall_precision": 0.8214160303503928,
+      "eval_overall_recall": 0.8408158662588275,
+      "eval_runtime": 935.5326,
+      "eval_samples_per_second": 70.227,
+      "eval_steps_per_second": 0.549,
+      "step": 35000
+    },
+    {
+      "epoch": 1.59,
+      "grad_norm": 3.663295269012451,
+      "learning_rate": 1.544638113789876e-05,
+      "loss": 0.234,
+      "step": 36000
+    },
+    {
+      "epoch": 1.59,
+      "eval_LOC_f1": 0.8537102746066029,
+      "eval_ORG_f1": 0.7690944951636334,
+      "eval_PER_f1": 0.8700584783258904,
+      "eval_loss": 0.28537243604660034,
+      "eval_overall_accuracy": 0.9194431708566052,
+      "eval_overall_f1": 0.8341569064491033,
+      "eval_overall_precision": 0.8295923488735222,
+      "eval_overall_recall": 0.8387719718647365,
+      "eval_runtime": 945.3448,
+      "eval_samples_per_second": 69.498,
+      "eval_steps_per_second": 0.544,
+      "step": 36000
+    },
+    {
+      "epoch": 1.64,
+      "grad_norm": 2.8524346351623535,
+      "learning_rate": 1.5319891725062616e-05,
+      "loss": 0.218,
+      "step": 37000
+    },
+    {
+      "epoch": 1.64,
+      "eval_LOC_f1": 0.8480498985496354,
+      "eval_ORG_f1": 0.7720144752714113,
+      "eval_PER_f1": 0.8681549995333838,
+      "eval_loss": 0.28636589646339417,
+      "eval_overall_accuracy": 0.9189809577115909,
+      "eval_overall_f1": 0.8333121549691502,
+      "eval_overall_precision": 0.8346791871137447,
+      "eval_overall_recall": 0.8319495933354947,
+      "eval_runtime": 971.8405,
+      "eval_samples_per_second": 67.604,
+      "eval_steps_per_second": 0.529,
+      "step": 37000
+    },
+    {
+      "epoch": 1.68,
+      "grad_norm": 4.014327049255371,
+      "learning_rate": 1.5193402312226468e-05,
+      "loss": 0.239,
+      "step": 38000
+    },
+    {
+      "epoch": 1.68,
+      "eval_LOC_f1": 0.856354181853526,
+      "eval_ORG_f1": 0.7607628933739787,
+      "eval_PER_f1": 0.8713080831892713,
+      "eval_loss": 0.2887238562107086,
+      "eval_overall_accuracy": 0.9170405556680282,
+      "eval_overall_f1": 0.8318997253753843,
+      "eval_overall_precision": 0.8238473394292507,
+      "eval_overall_recall": 0.8401110750884513,
+      "eval_runtime": 984.8035,
+      "eval_samples_per_second": 66.714,
+      "eval_steps_per_second": 0.522,
+      "step": 38000
+    },
+    {
+      "epoch": 1.73,
+      "grad_norm": 48.57558059692383,
+      "learning_rate": 1.5066912899390323e-05,
+      "loss": 0.2374,
+      "step": 39000
+    },
+    {
+      "epoch": 1.73,
+      "eval_LOC_f1": 0.8629891838741397,
+      "eval_ORG_f1": 0.7734306355559895,
+      "eval_PER_f1": 0.8760364502093424,
+      "eval_loss": 0.28808096051216125,
+      "eval_overall_accuracy": 0.921347101975682,
+      "eval_overall_f1": 0.8414042377584563,
+      "eval_overall_precision": 0.8495036133499992,
+      "eval_overall_recall": 0.8334578464400998,
+      "eval_runtime": 1002.3252,
+      "eval_samples_per_second": 65.548,
+      "eval_steps_per_second": 0.513,
+      "step": 39000
+    },
+    {
+      "epoch": 1.77,
+      "grad_norm": 18.295597076416016,
+      "learning_rate": 1.4940423486554176e-05,
+      "loss": 0.2327,
+      "step": 40000
+    },
+    {
+      "epoch": 1.77,
+      "eval_LOC_f1": 0.8568252173162995,
+      "eval_ORG_f1": 0.765257865392274,
+      "eval_PER_f1": 0.8650392170611746,
+      "eval_loss": 0.27475544810295105,
+      "eval_overall_accuracy": 0.918402633051887,
+      "eval_overall_f1": 0.8338476462277082,
+      "eval_overall_precision": 0.8316554025631678,
+      "eval_overall_recall": 0.8360514779470842,
+      "eval_runtime": 966.7516,
+      "eval_samples_per_second": 67.96,
+      "eval_steps_per_second": 0.532,
+      "step": 40000
+    },
+    {
+      "epoch": 1.82,
+      "grad_norm": 8.551658630371094,
+      "learning_rate": 1.481393407371803e-05,
+      "loss": 0.2266,
+      "step": 41000
+    },
+    {
+      "epoch": 1.82,
+      "eval_LOC_f1": 0.8626095376422174,
+      "eval_ORG_f1": 0.7735328126106142,
+      "eval_PER_f1": 0.8753750051370567,
+      "eval_loss": 0.27126726508140564,
+      "eval_overall_accuracy": 0.9233812863964941,
+      "eval_overall_f1": 0.8405072872606979,
+      "eval_overall_precision": 0.835852936051577,
+      "eval_overall_recall": 0.845213763161975,
+      "eval_runtime": 949.0619,
+      "eval_samples_per_second": 69.226,
+      "eval_steps_per_second": 0.542,
+      "step": 41000
+    },
+    {
+      "epoch": 1.86,
+      "grad_norm": 15.628961563110352,
+      "learning_rate": 1.4687444660881885e-05,
+      "loss": 0.2321,
+      "step": 42000
+    },
+    {
+      "epoch": 1.86,
+      "eval_LOC_f1": 0.8591221352011853,
+      "eval_ORG_f1": 0.7785703957198631,
+      "eval_PER_f1": 0.8756674763500703,
+      "eval_loss": 0.2676946520805359,
+      "eval_overall_accuracy": 0.9217490264496077,
+      "eval_overall_f1": 0.8412990217616726,
+      "eval_overall_precision": 0.8391781494337506,
+      "eval_overall_recall": 0.8434306415009233,
+      "eval_runtime": 933.8514,
+      "eval_samples_per_second": 70.354,
+      "eval_steps_per_second": 0.55,
+      "step": 42000
+    },
+    {
+      "epoch": 1.9,
+      "grad_norm": 3.160637140274048,
+      "learning_rate": 1.456095524804574e-05,
+      "loss": 0.2103,
+      "step": 43000
+    },
+    {
+      "epoch": 1.9,
+      "eval_LOC_f1": 0.8572030651340997,
+      "eval_ORG_f1": 0.7727868545480081,
+      "eval_PER_f1": 0.8757355631042593,
+      "eval_loss": 0.2812298536300659,
+      "eval_overall_accuracy": 0.9198480725488561,
+      "eval_overall_f1": 0.8381508761238307,
+      "eval_overall_precision": 0.8330536854344057,
+      "eval_overall_recall": 0.8433108270019594,
+      "eval_runtime": 959.7072,
+      "eval_samples_per_second": 68.458,
+      "eval_steps_per_second": 0.536,
+      "step": 43000
+    },
+    {
+      "epoch": 1.95,
+      "grad_norm": 19.63943862915039,
+      "learning_rate": 1.4434465835209595e-05,
+      "loss": 0.2299,
+      "step": 44000
+    },
+    {
+      "epoch": 1.95,
+      "eval_LOC_f1": 0.8625562915195548,
+      "eval_ORG_f1": 0.7738310591448732,
+      "eval_PER_f1": 0.8727216526711657,
+      "eval_loss": 0.2698276937007904,
+      "eval_overall_accuracy": 0.9231393874075574,
+      "eval_overall_f1": 0.8398996262454074,
+      "eval_overall_precision": 0.8411791735887738,
+      "eval_overall_recall": 0.8386239657189575,
+      "eval_runtime": 959.292,
+      "eval_samples_per_second": 68.488,
+      "eval_steps_per_second": 0.536,
+      "step": 44000
+    },
+    {
+      "epoch": 1.99,
+      "grad_norm": 5.957523345947266,
+      "learning_rate": 1.4307976422373449e-05,
+      "loss": 0.217,
+      "step": 45000
+    },
+    {
+      "epoch": 1.99,
+      "eval_LOC_f1": 0.8548825805462975,
+      "eval_ORG_f1": 0.7795103416137936,
+      "eval_PER_f1": 0.8763479817569583,
+      "eval_loss": 0.257554292678833,
+      "eval_overall_accuracy": 0.9223258625001488,
+      "eval_overall_f1": 0.8396542612618684,
+      "eval_overall_precision": 0.8451246242708534,
+      "eval_overall_recall": 0.834254260462625,
+      "eval_runtime": 986.5064,
+      "eval_samples_per_second": 66.599,
+      "eval_steps_per_second": 0.521,
+      "step": 45000
     }
   ],
   "logging_steps": 1000,
@@ -476,7 +1004,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 7,
   "save_steps": 500,
-  "total_flos": 5913608888301312.0,
+  "total_flos": 1.2387920637624576e+16,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0f7a60270f3ee223660c001ddc4e29051fd77c829918010af9f92d6bd976d2e
+oid sha256:044ffe2888cc55ab35cc5b7f4e9520b0a9a678ac5d8c8ae4f6fa63b6dfcbe6f5
 size 4856