Training in progress, epoch 1, checkpoint

checkpoint-555/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf60d2f658351ec678f15a2b665ef64085fd8bfc89f669cd514c38acfa2a2a01
+oid sha256:91b0eaf62f6feb29f8e6c91e27f4c943c7c5cb4f86aeb4b64c72f034a0e92bdd
 size 437958648

checkpoint-555/trainer_state.json

@@ -8,411 +8,404 @@
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 0,
-      "eval_f1": 0.6562770562770562,
-      "eval_loss": 0.7476533651351929,
-      "eval_precision": 0.48976954555244456,
-      "eval_recall": 0.9943156974202011,
-      "eval_runtime": 16.4688,
-      "eval_samples_per_second": 283.749,
-      "eval_steps_per_second": 35.522,
-      "step": 0
-    },
     {
       "epoch": 0.018026137899954935,
-      "grad_norm": 14.20566463470459,
+      "grad_norm": 13.032843589782715,
       "learning_rate": 4.981949458483755e-05,
-      "loss": 0.6876,
+      "loss": 0.6633,
       "step": 10
     },
     {
       "epoch": 0.03605227579990987,
-      "grad_norm": 27.79998016357422,
+      "grad_norm": 35.63669204711914,
       "learning_rate": 4.963898916967509e-05,
-      "loss": 0.5988,
+      "loss": 0.5651,
       "step": 20
     },
     {
       "epoch": 0.054078413699864804,
-      "grad_norm": 39.63682556152344,
+      "grad_norm": 16.95796012878418,
       "learning_rate": 4.945848375451264e-05,
-      "loss": 0.5677,
+      "loss": 0.5359,
       "step": 30
     },
     {
       "epoch": 0.07210455159981974,
-      "grad_norm": 38.49867630004883,
+      "grad_norm": 14.045697212219238,
       "learning_rate": 4.927797833935018e-05,
-      "loss": 0.5922,
+      "loss": 0.5421,
       "step": 40
     },
     {
       "epoch": 0.09013068949977468,
-      "grad_norm": 52.28614044189453,
+      "grad_norm": 43.34794235229492,
       "learning_rate": 4.909747292418773e-05,
-      "loss": 0.5506,
+      "loss": 0.5222,
       "step": 50
     },
     {
       "epoch": 0.10815682739972961,
-      "grad_norm": 29.180490493774414,
+      "grad_norm": 32.69561767578125,
       "learning_rate": 4.891696750902527e-05,
-      "loss": 0.5173,
+      "loss": 0.4589,
       "step": 60
     },
     {
       "epoch": 0.12618296529968454,
-      "grad_norm": 33.55237579345703,
+      "grad_norm": 29.39051055908203,
       "learning_rate": 4.873646209386282e-05,
-      "loss": 0.4931,
+      "loss": 0.4858,
       "step": 70
     },
     {
       "epoch": 0.14420910319963948,
-      "grad_norm": 32.063228607177734,
+      "grad_norm": 30.52906608581543,
       "learning_rate": 4.855595667870036e-05,
-      "loss": 0.483,
+      "loss": 0.4418,
       "step": 80
     },
     {
       "epoch": 0.16223524109959442,
-      "grad_norm": 24.234926223754883,
+      "grad_norm": 15.103797912597656,
       "learning_rate": 4.837545126353791e-05,
-      "loss": 0.47,
+      "loss": 0.4713,
       "step": 90
     },
     {
       "epoch": 0.18026137899954936,
-      "grad_norm": 33.66775894165039,
+      "grad_norm": 40.57420349121094,
       "learning_rate": 4.819494584837546e-05,
-      "loss": 0.4564,
+      "loss": 0.468,
       "step": 100
     },
     {
       "epoch": 0.19828751689950427,
-      "grad_norm": 53.322574615478516,
+      "grad_norm": 37.67963409423828,
       "learning_rate": 4.8014440433213e-05,
-      "loss": 0.5083,
+      "loss": 0.492,
       "step": 110
     },
     {
       "epoch": 0.21631365479945922,
-      "grad_norm": 29.443470001220703,
+      "grad_norm": 30.945823669433594,
       "learning_rate": 4.783393501805055e-05,
-      "loss": 0.463,
+      "loss": 0.4633,
       "step": 120
     },
     {
       "epoch": 0.23433979269941416,
-      "grad_norm": 33.13877487182617,
+      "grad_norm": 40.709590911865234,
       "learning_rate": 4.765342960288809e-05,
-      "loss": 0.4626,
+      "loss": 0.4392,
       "step": 130
     },
     {
       "epoch": 0.25236593059936907,
-      "grad_norm": 25.509536743164062,
+      "grad_norm": 19.698150634765625,
       "learning_rate": 4.747292418772563e-05,
-      "loss": 0.5113,
+      "loss": 0.4707,
       "step": 140
     },
     {
       "epoch": 0.270392068499324,
-      "grad_norm": 21.015031814575195,
+      "grad_norm": 21.289947509765625,
       "learning_rate": 4.7292418772563177e-05,
-      "loss": 0.4611,
+      "loss": 0.4576,
       "step": 150
     },
     {
       "epoch": 0.28841820639927895,
-      "grad_norm": 28.079387664794922,
+      "grad_norm": 26.911935806274414,
       "learning_rate": 4.711191335740072e-05,
-      "loss": 0.4783,
+      "loss": 0.4583,
       "step": 160
     },
     {
       "epoch": 0.3064443442992339,
-      "grad_norm": 19.978599548339844,
+      "grad_norm": 19.308853149414062,
       "learning_rate": 4.693140794223827e-05,
-      "loss": 0.5026,
+      "loss": 0.4937,
       "step": 170
     },
     {
       "epoch": 0.32447048219918884,
-      "grad_norm": 29.316265106201172,
+      "grad_norm": 34.61503219604492,
       "learning_rate": 4.675090252707581e-05,
-      "loss": 0.4906,
+      "loss": 0.4836,
       "step": 180
     },
     {
       "epoch": 0.3424966200991438,
-      "grad_norm": 18.79282569885254,
+      "grad_norm": 16.551424026489258,
       "learning_rate": 4.657039711191336e-05,
-      "loss": 0.5016,
+      "loss": 0.4919,
       "step": 190
     },
     {
       "epoch": 0.3605227579990987,
-      "grad_norm": 18.66801643371582,
+      "grad_norm": 16.011377334594727,
       "learning_rate": 4.63898916967509e-05,
-      "loss": 0.4502,
+      "loss": 0.4518,
       "step": 200
     },
     {
       "epoch": 0.3785488958990536,
-      "grad_norm": 27.929332733154297,
+      "grad_norm": 22.71525764465332,
       "learning_rate": 4.620938628158845e-05,
-      "loss": 0.4302,
+      "loss": 0.4221,
       "step": 210
     },
     {
       "epoch": 0.39657503379900855,
-      "grad_norm": 20.541879653930664,
+      "grad_norm": 22.2515811920166,
       "learning_rate": 4.602888086642599e-05,
-      "loss": 0.4496,
+      "loss": 0.4286,
       "step": 220
     },
     {
       "epoch": 0.4146011716989635,
-      "grad_norm": 19.034873962402344,
+      "grad_norm": 18.233482360839844,
       "learning_rate": 4.584837545126354e-05,
-      "loss": 0.4541,
+      "loss": 0.4442,
       "step": 230
     },
     {
       "epoch": 0.43262730959891843,
-      "grad_norm": 32.077945709228516,
+      "grad_norm": 33.190948486328125,
       "learning_rate": 4.566787003610109e-05,
-      "loss": 0.4562,
+      "loss": 0.4584,
       "step": 240
     },
     {
       "epoch": 0.45065344749887337,
-      "grad_norm": 25.475997924804688,
+      "grad_norm": 29.945205688476562,
       "learning_rate": 4.548736462093863e-05,
-      "loss": 0.4456,
+      "loss": 0.4339,
       "step": 250
     },
     {
       "epoch": 0.4686795853988283,
-      "grad_norm": 17.132720947265625,
+      "grad_norm": 17.004419326782227,
       "learning_rate": 4.530685920577618e-05,
-      "loss": 0.4629,
+      "loss": 0.4504,
       "step": 260
     },
     {
       "epoch": 0.48670572329878325,
-      "grad_norm": 22.313941955566406,
+      "grad_norm": 23.36432647705078,
       "learning_rate": 4.5126353790613716e-05,
-      "loss": 0.4444,
+      "loss": 0.4405,
       "step": 270
     },
     {
       "epoch": 0.5047318611987381,
-      "grad_norm": 22.373477935791016,
+      "grad_norm": 29.156505584716797,
       "learning_rate": 4.494584837545127e-05,
-      "loss": 0.4404,
+      "loss": 0.4319,
       "step": 280
     },
     {
       "epoch": 0.5227579990986931,
-      "grad_norm": 16.72430992126465,
+      "grad_norm": 16.330259323120117,
       "learning_rate": 4.4765342960288806e-05,
-      "loss": 0.4611,
+      "loss": 0.453,
       "step": 290
     },
     {
       "epoch": 0.540784136998648,
-      "grad_norm": 27.037927627563477,
+      "grad_norm": 21.66246223449707,
       "learning_rate": 4.458483754512636e-05,
-      "loss": 0.4272,
+      "loss": 0.416,
       "step": 300
     },
     {
       "epoch": 0.558810274898603,
-      "grad_norm": 31.637985229492188,
+      "grad_norm": 22.177268981933594,
       "learning_rate": 4.44043321299639e-05,
-      "loss": 0.448,
+      "loss": 0.4456,
       "step": 310
     },
     {
       "epoch": 0.5768364127985579,
-      "grad_norm": 35.010459899902344,
+      "grad_norm": 38.450069427490234,
       "learning_rate": 4.422382671480145e-05,
-      "loss": 0.4215,
+      "loss": 0.4216,
       "step": 320
     },
     {
       "epoch": 0.5948625506985128,
-      "grad_norm": 26.25156021118164,
+      "grad_norm": 28.213655471801758,
       "learning_rate": 4.404332129963899e-05,
-      "loss": 0.4396,
+      "loss": 0.4392,
       "step": 330
     },
     {
       "epoch": 0.6128886885984678,
-      "grad_norm": 19.4619083404541,
+      "grad_norm": 19.354459762573242,
       "learning_rate": 4.386281588447654e-05,
-      "loss": 0.4378,
+      "loss": 0.4303,
       "step": 340
     },
     {
       "epoch": 0.6309148264984227,
-      "grad_norm": 19.16460609436035,
+      "grad_norm": 20.212663650512695,
       "learning_rate": 4.368231046931408e-05,
-      "loss": 0.4366,
+      "loss": 0.4426,
       "step": 350
     },
     {
       "epoch": 0.6489409643983777,
-      "grad_norm": 30.745384216308594,
+      "grad_norm": 34.96459197998047,
       "learning_rate": 4.350180505415163e-05,
-      "loss": 0.4062,
+      "loss": 0.4146,
       "step": 360
     },
     {
       "epoch": 0.6669671022983326,
-      "grad_norm": 16.873245239257812,
+      "grad_norm": 17.65025520324707,
       "learning_rate": 4.332129963898917e-05,
-      "loss": 0.439,
+      "loss": 0.4506,
       "step": 370
     },
     {
       "epoch": 0.6849932401982876,
-      "grad_norm": 25.771486282348633,
+      "grad_norm": 18.796554565429688,
       "learning_rate": 4.314079422382672e-05,
-      "loss": 0.4053,
+      "loss": 0.3765,
       "step": 380
     },
     {
       "epoch": 0.7030193780982424,
-      "grad_norm": 14.48513412475586,
+      "grad_norm": 21.690357208251953,
       "learning_rate": 4.296028880866426e-05,
-      "loss": 0.4164,
+      "loss": 0.4158,
       "step": 390
     },
     {
       "epoch": 0.7210455159981974,
-      "grad_norm": 34.993465423583984,
+      "grad_norm": 39.8782844543457,
       "learning_rate": 4.277978339350181e-05,
-      "loss": 0.4438,
+      "loss": 0.4602,
       "step": 400
     },
     {
       "epoch": 0.7390716538981523,
-      "grad_norm": 22.576608657836914,
+      "grad_norm": 26.3798885345459,
       "learning_rate": 4.259927797833935e-05,
-      "loss": 0.4518,
+      "loss": 0.4512,
       "step": 410
     },
     {
       "epoch": 0.7570977917981072,
-      "grad_norm": 19.088132858276367,
+      "grad_norm": 18.073617935180664,
       "learning_rate": 4.24187725631769e-05,
-      "loss": 0.4495,
+      "loss": 0.4347,
       "step": 420
     },
     {
       "epoch": 0.7751239296980622,
-      "grad_norm": 35.2998046875,
+      "grad_norm": 30.09633445739746,
       "learning_rate": 4.223826714801444e-05,
-      "loss": 0.3897,
+      "loss": 0.3858,
       "step": 430
     },
     {
       "epoch": 0.7931500675980171,
-      "grad_norm": 20.395572662353516,
+      "grad_norm": 17.02672576904297,
       "learning_rate": 4.205776173285199e-05,
-      "loss": 0.4818,
+      "loss": 0.4815,
       "step": 440
     },
     {
       "epoch": 0.8111762054979721,
-      "grad_norm": 21.174386978149414,
+      "grad_norm": 19.149808883666992,
       "learning_rate": 4.187725631768953e-05,
-      "loss": 0.4103,
+      "loss": 0.4196,
       "step": 450
     },
     {
       "epoch": 0.829202343397927,
-      "grad_norm": 32.97265625,
+      "grad_norm": 28.140705108642578,
       "learning_rate": 4.169675090252708e-05,
-      "loss": 0.4036,
+      "loss": 0.4085,
       "step": 460
     },
     {
       "epoch": 0.847228481297882,
-      "grad_norm": 23.78690528869629,
+      "grad_norm": 21.819814682006836,
       "learning_rate": 4.151624548736462e-05,
-      "loss": 0.4079,
+      "loss": 0.3986,
       "step": 470
     },
     {
       "epoch": 0.8652546191978369,
-      "grad_norm": 17.087247848510742,
+      "grad_norm": 15.050581932067871,
       "learning_rate": 4.1335740072202167e-05,
-      "loss": 0.4371,
+      "loss": 0.4267,
       "step": 480
     },
     {
       "epoch": 0.8832807570977917,
-      "grad_norm": 13.810160636901855,
+      "grad_norm": 14.929478645324707,
       "learning_rate": 4.115523465703972e-05,
-      "loss": 0.4249,
+      "loss": 0.4315,
       "step": 490
     },
     {
       "epoch": 0.9013068949977467,
-      "grad_norm": 20.197996139526367,
+      "grad_norm": 20.087432861328125,
       "learning_rate": 4.0974729241877256e-05,
-      "loss": 0.3991,
+      "loss": 0.3886,
       "step": 500
     },
     {
       "epoch": 0.9193330328977016,
-      "grad_norm": 33.1153564453125,
+      "grad_norm": 32.96128463745117,
       "learning_rate": 4.079422382671481e-05,
-      "loss": 0.3994,
+      "loss": 0.4046,
       "step": 510
     },
     {
       "epoch": 0.9373591707976566,
-      "grad_norm": 18.20081901550293,
+      "grad_norm": 19.673940658569336,
       "learning_rate": 4.0613718411552346e-05,
-      "loss": 0.4176,
+      "loss": 0.4127,
       "step": 520
     },
     {
       "epoch": 0.9553853086976115,
-      "grad_norm": 20.09193992614746,
+      "grad_norm": 16.12790298461914,
       "learning_rate": 4.043321299638989e-05,
-      "loss": 0.4297,
+      "loss": 0.4227,
       "step": 530
     },
     {
       "epoch": 0.9734114465975665,
-      "grad_norm": 15.939692497253418,
+      "grad_norm": 20.4881649017334,
       "learning_rate": 4.0252707581227436e-05,
-      "loss": 0.4309,
+      "loss": 0.4223,
       "step": 540
     },
     {
       "epoch": 0.9914375844975214,
-      "grad_norm": 24.811601638793945,
+      "grad_norm": 17.96915054321289,
       "learning_rate": 4.007220216606498e-05,
-      "loss": 0.3908,
+      "loss": 0.395,
       "step": 550
     },
     {
       "epoch": 1.0,
-      "eval_f1": 0.8367177786987153,
-      "eval_loss": 0.37827184796333313,
-      "eval_precision": 0.7951949586451359,
-      "eval_recall": 0.8828159160472234,
-      "eval_runtime": 15.9582,
-      "eval_samples_per_second": 292.827,
-      "eval_steps_per_second": 36.658,
+      "eval_f1": 0.8363481591828048,
+      "eval_fn": 322,
+      "eval_fp": 447,
+      "eval_loss": 0.372616171836853,
+      "eval_precision": 0.8146766169154229,
+      "eval_recall": 0.8592041976388282,
+      "eval_runtime": 12.0589,
+      "eval_samples_per_second": 387.513,
+      "eval_steps_per_second": 48.512,
+      "eval_tn": 1939,
+      "eval_tp": 1965,
       "step": 555
     }
   ],

checkpoint-555/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92aaf17ba6b51dd0179acd28630a512a049c53024ba01863d7232c815d6018dd
+oid sha256:7e924750e5e12e6bb9372df333907b4d7f42d9dd2da839a2407f8f861ab40940
 size 5432