Training in progress, epoch 0

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.142857142857143,
-    "eval_accuracy": 0.7804878048780488,
-    "eval_loss": 1.077329158782959,
-    "eval_runtime": 5.6706,
-    "eval_samples_per_second": 14.461,
-    "eval_steps_per_second": 3.703
+    "eval_accuracy": 0.8028169014084507,
+    "eval_loss": 1.103053331375122,
+    "eval_runtime": 4.9334,
+    "eval_samples_per_second": 14.392,
+    "eval_steps_per_second": 3.649
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59a9352d6e9e3f904cadf0c8ec4c5f4d9e71257187590e4be3a56269450864ca
+oid sha256:7f247e39984064745b1af387be3ee18ee964433ae6589b206412a2a943db465f
 size 344952716

runs/Aug11_13-45-55_prod3/events.out.tfevents.1754892098.prod3.753305.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3a5d08465f532175aa78449cfb120674ef931279786f1bb4e3ee3678fe60c7b
+size 411

runs/Aug11_14-13-04_prod3/events.out.tfevents.1754892793.prod3.788163.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1930c512d73dd86d2a2093b52f45f1ecce9826625f4c33bbee6eebe2f0e986d4
+size 5487

runs/Aug11_14-14-16_prod3/events.out.tfevents.1754892863.prod3.789915.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32787120c0ace084a08491616cd320f007d6602d74558f7b38598207c052e965
+size 6839

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.142857142857143,
-    "eval_accuracy": 0.7804878048780488,
-    "eval_loss": 1.077329158782959,
-    "eval_runtime": 5.6706,
-    "eval_samples_per_second": 14.461,
-    "eval_steps_per_second": 3.703
+    "eval_accuracy": 0.8028169014084507,
+    "eval_loss": 1.103053331375122,
+    "eval_runtime": 4.9334,
+    "eval_samples_per_second": 14.392,
+    "eval_steps_per_second": 3.649
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_global_step": 230,
-  "best_metric": 0.7804878048780488,
-  "best_model_checkpoint": "./Models/Matcha_clips_224_fintuned_5/checkpoint-230",
+  "best_global_step": 460,
+  "best_metric": 0.782051282051282,
+  "best_model_checkpoint": "./Models/Matcha_clips_224_fintuned_5/checkpoint-460",
   "epoch": 6.142857142857143,
   "eval_steps": 500,
   "global_step": 805,
@@ -11,363 +11,363 @@
   "log_history": [
     {
       "epoch": 0.024844720496894408,
-      "grad_norm": 11.596962928771973,
+      "grad_norm": 11.83416748046875,
       "learning_rate": 1.1728395061728396e-05,
-      "loss": 1.6871,
+      "loss": 1.7545,
       "step": 20
     },
     {
       "epoch": 0.049689440993788817,
-      "grad_norm": 15.900735855102539,
+      "grad_norm": 14.771072387695312,
       "learning_rate": 2.4074074074074074e-05,
-      "loss": 0.9615,
+      "loss": 1.0393,
       "step": 40
     },
     {
       "epoch": 0.07453416149068323,
-      "grad_norm": 4.670846939086914,
+      "grad_norm": 3.698537826538086,
       "learning_rate": 3.6419753086419754e-05,
-      "loss": 0.5851,
+      "loss": 0.6203,
       "step": 60
     },
     {
       "epoch": 0.09937888198757763,
-      "grad_norm": 6.628495216369629,
+      "grad_norm": 1.2401633262634277,
       "learning_rate": 4.876543209876544e-05,
-      "loss": 0.3508,
+      "loss": 0.3541,
       "step": 80
     },
     {
       "epoch": 0.12422360248447205,
-      "grad_norm": 10.711517333984375,
+      "grad_norm": 27.550607681274414,
       "learning_rate": 4.875690607734807e-05,
-      "loss": 0.3922,
+      "loss": 0.4705,
       "step": 100
     },
     {
       "epoch": 0.14285714285714285,
-      "eval_accuracy": 0.7682926829268293,
-      "eval_loss": 0.996225118637085,
-      "eval_runtime": 6.4019,
-      "eval_samples_per_second": 12.809,
-      "eval_steps_per_second": 3.28,
+      "eval_accuracy": 0.7317073170731707,
+      "eval_loss": 1.0046272277832031,
+      "eval_runtime": 6.2101,
+      "eval_samples_per_second": 13.204,
+      "eval_steps_per_second": 3.382,
       "step": 115
     },
     {
       "epoch": 1.0062111801242235,
-      "grad_norm": 0.29712462425231934,
+      "grad_norm": 0.25678861141204834,
       "learning_rate": 4.737569060773481e-05,
-      "loss": 0.1851,
+      "loss": 0.224,
       "step": 120
     },
     {
       "epoch": 1.031055900621118,
-      "grad_norm": 3.168236494064331,
+      "grad_norm": 0.9883336424827576,
       "learning_rate": 4.599447513812155e-05,
-      "loss": 0.1939,
+      "loss": 0.1258,
       "step": 140
     },
     {
       "epoch": 1.0559006211180124,
-      "grad_norm": 0.13498322665691376,
+      "grad_norm": 0.18707028031349182,
       "learning_rate": 4.461325966850829e-05,
-      "loss": 0.3082,
+      "loss": 0.2609,
       "step": 160
     },
     {
       "epoch": 1.0807453416149069,
-      "grad_norm": 0.22494107484817505,
+      "grad_norm": 0.08376210927963257,
       "learning_rate": 4.323204419889503e-05,
-      "loss": 0.0559,
+      "loss": 0.049,
       "step": 180
     },
     {
       "epoch": 1.1055900621118013,
-      "grad_norm": 0.5178938508033752,
+      "grad_norm": 0.08791894465684891,
       "learning_rate": 4.1850828729281773e-05,
-      "loss": 0.2148,
+      "loss": 0.1955,
       "step": 200
     },
     {
       "epoch": 1.1304347826086956,
-      "grad_norm": 0.1849067360162735,
+      "grad_norm": 0.12012392282485962,
       "learning_rate": 4.046961325966851e-05,
-      "loss": 0.208,
+      "loss": 0.1493,
       "step": 220
     },
     {
       "epoch": 1.1428571428571428,
-      "eval_accuracy": 0.7804878048780488,
-      "eval_loss": 1.0773290395736694,
-      "eval_runtime": 6.2229,
-      "eval_samples_per_second": 13.177,
-      "eval_steps_per_second": 3.375,
+      "eval_accuracy": 0.7560975609756098,
+      "eval_loss": 1.226146936416626,
+      "eval_runtime": 6.1393,
+      "eval_samples_per_second": 13.357,
+      "eval_steps_per_second": 3.421,
       "step": 230
     },
     {
       "epoch": 2.012422360248447,
-      "grad_norm": 2.8872368335723877,
+      "grad_norm": 0.09424888342618942,
       "learning_rate": 3.9088397790055245e-05,
-      "loss": 0.1527,
+      "loss": 0.1158,
       "step": 240
     },
     {
       "epoch": 2.0372670807453415,
-      "grad_norm": 0.04912768676877022,
+      "grad_norm": 0.05555570125579834,
       "learning_rate": 3.770718232044199e-05,
-      "loss": 0.0707,
+      "loss": 0.0235,
       "step": 260
     },
     {
       "epoch": 2.062111801242236,
-      "grad_norm": 0.0381060354411602,
+      "grad_norm": 0.02632896415889263,
       "learning_rate": 3.632596685082873e-05,
-      "loss": 0.0086,
+      "loss": 0.0072,
       "step": 280
     },
     {
       "epoch": 2.0869565217391304,
-      "grad_norm": 42.93180465698242,
+      "grad_norm": 1.1457473039627075,
       "learning_rate": 3.4944751381215476e-05,
-      "loss": 0.022,
+      "loss": 0.0063,
       "step": 300
     },
     {
       "epoch": 2.111801242236025,
-      "grad_norm": 9.570910453796387,
+      "grad_norm": 1.7765306234359741,
       "learning_rate": 3.3563535911602215e-05,
-      "loss": 0.0579,
+      "loss": 0.0898,
       "step": 320
     },
     {
       "epoch": 2.1366459627329193,
-      "grad_norm": 0.1904195100069046,
+      "grad_norm": 0.04399551451206207,
       "learning_rate": 3.218232044198895e-05,
-      "loss": 0.0868,
+      "loss": 0.0965,
       "step": 340
     },
     {
       "epoch": 2.142857142857143,
-      "eval_accuracy": 0.7317073170731707,
-      "eval_loss": 1.2267667055130005,
-      "eval_runtime": 6.118,
-      "eval_samples_per_second": 13.403,
-      "eval_steps_per_second": 3.432,
+      "eval_accuracy": 0.7692307692307693,
+      "eval_loss": 1.1611318588256836,
+      "eval_runtime": 5.888,
+      "eval_samples_per_second": 13.247,
+      "eval_steps_per_second": 3.397,
       "step": 345
     },
     {
       "epoch": 3.018633540372671,
-      "grad_norm": 0.011533539742231369,
+      "grad_norm": 0.022856663912534714,
       "learning_rate": 3.0801104972375693e-05,
-      "loss": 0.0023,
+      "loss": 0.0181,
       "step": 360
     },
     {
       "epoch": 3.0434782608695654,
-      "grad_norm": 0.018733657896518707,
+      "grad_norm": 0.019706225022673607,
       "learning_rate": 2.9419889502762433e-05,
-      "loss": 0.0034,
+      "loss": 0.002,
       "step": 380
     },
     {
       "epoch": 3.0683229813664594,
-      "grad_norm": 0.02364545315504074,
+      "grad_norm": 0.028400592505931854,
       "learning_rate": 2.8038674033149172e-05,
-      "loss": 0.0041,
+      "loss": 0.0045,
       "step": 400
     },
     {
       "epoch": 3.093167701863354,
-      "grad_norm": 0.011309165507555008,
+      "grad_norm": 0.01030923891812563,
       "learning_rate": 2.6657458563535914e-05,
-      "loss": 0.0014,
+      "loss": 0.0011,
       "step": 420
     },
     {
       "epoch": 3.1180124223602483,
-      "grad_norm": 0.01748155988752842,
+      "grad_norm": 0.10795993357896805,
       "learning_rate": 2.5276243093922653e-05,
-      "loss": 0.0042,
+      "loss": 0.1063,
       "step": 440
     },
     {
       "epoch": 3.142857142857143,
-      "grad_norm": 0.025904180482029915,
+      "grad_norm": 0.020210983231663704,
       "learning_rate": 2.3895027624309393e-05,
-      "loss": 0.0032,
+      "loss": 0.0018,
       "step": 460
     },
     {
       "epoch": 3.142857142857143,
-      "eval_accuracy": 0.7317073170731707,
-      "eval_loss": 1.3057693243026733,
-      "eval_runtime": 6.2762,
-      "eval_samples_per_second": 13.065,
-      "eval_steps_per_second": 3.346,
+      "eval_accuracy": 0.782051282051282,
+      "eval_loss": 1.2064043283462524,
+      "eval_runtime": 5.8576,
+      "eval_samples_per_second": 13.316,
+      "eval_steps_per_second": 3.414,
       "step": 460
     },
     {
       "epoch": 4.024844720496894,
-      "grad_norm": 0.04104848951101303,
+      "grad_norm": 0.029298782348632812,
       "learning_rate": 2.2513812154696135e-05,
       "loss": 0.0012,
       "step": 480
     },
     {
       "epoch": 4.049689440993789,
-      "grad_norm": 0.013835583813488483,
+      "grad_norm": 0.014873038977384567,
       "learning_rate": 2.1132596685082874e-05,
       "loss": 0.0009,
       "step": 500
     },
     {
       "epoch": 4.074534161490683,
-      "grad_norm": 0.016223575919866562,
+      "grad_norm": 0.011718147434294224,
       "learning_rate": 1.9751381215469613e-05,
-      "loss": 0.001,
+      "loss": 0.0008,
       "step": 520
     },
     {
       "epoch": 4.099378881987578,
-      "grad_norm": 0.010322828777134418,
+      "grad_norm": 0.010744371451437473,
       "learning_rate": 1.8370165745856356e-05,
-      "loss": 0.0008,
+      "loss": 0.0012,
       "step": 540
     },
     {
       "epoch": 4.124223602484472,
-      "grad_norm": 0.012116617523133755,
+      "grad_norm": 0.014341841451823711,
       "learning_rate": 1.6988950276243095e-05,
-      "loss": 0.0014,
+      "loss": 0.012,
       "step": 560
     },
     {
       "epoch": 4.142857142857143,
-      "eval_accuracy": 0.7317073170731707,
-      "eval_loss": 1.3161958456039429,
-      "eval_runtime": 6.3153,
-      "eval_samples_per_second": 12.984,
-      "eval_steps_per_second": 3.325,
+      "eval_accuracy": 0.782051282051282,
+      "eval_loss": 1.1255377531051636,
+      "eval_runtime": 6.0116,
+      "eval_samples_per_second": 12.975,
+      "eval_steps_per_second": 3.327,
       "step": 575
     },
     {
       "epoch": 5.0062111801242235,
-      "grad_norm": 0.012491249479353428,
+      "grad_norm": 0.016358228400349617,
       "learning_rate": 1.5607734806629834e-05,
-      "loss": 0.0008,
+      "loss": 0.0007,
       "step": 580
     },
     {
       "epoch": 5.031055900621118,
-      "grad_norm": 0.040033698081970215,
+      "grad_norm": 0.03938188776373863,
       "learning_rate": 1.4226519337016575e-05,
       "loss": 0.0007,
       "step": 600
     },
     {
       "epoch": 5.055900621118012,
-      "grad_norm": 0.010547437705099583,
+      "grad_norm": 0.00977695919573307,
       "learning_rate": 1.2845303867403316e-05,
-      "loss": 0.0006,
+      "loss": 0.0007,
       "step": 620
     },
     {
       "epoch": 5.080745341614906,
-      "grad_norm": 0.016931477934122086,
+      "grad_norm": 0.014728990383446217,
       "learning_rate": 1.1464088397790055e-05,
-      "loss": 0.0009,
+      "loss": 0.0008,
       "step": 640
     },
     {
       "epoch": 5.105590062111801,
-      "grad_norm": 0.021733667701482773,
+      "grad_norm": 0.016683587804436684,
       "learning_rate": 1.0082872928176797e-05,
       "loss": 0.0006,
       "step": 660
     },
     {
       "epoch": 5.130434782608695,
-      "grad_norm": 0.053632985800504684,
+      "grad_norm": 0.038530658930540085,
       "learning_rate": 8.701657458563537e-06,
-      "loss": 0.0007,
+      "loss": 0.0009,
       "step": 680
     },
     {
       "epoch": 5.142857142857143,
-      "eval_accuracy": 0.7317073170731707,
-      "eval_loss": 1.3207590579986572,
-      "eval_runtime": 6.2982,
-      "eval_samples_per_second": 13.02,
-      "eval_steps_per_second": 3.334,
+      "eval_accuracy": 0.782051282051282,
+      "eval_loss": 1.0889217853546143,
+      "eval_runtime": 6.0113,
+      "eval_samples_per_second": 12.976,
+      "eval_steps_per_second": 3.327,
       "step": 690
     },
     {
       "epoch": 6.012422360248447,
-      "grad_norm": 0.01116255484521389,
+      "grad_norm": 0.007233364041894674,
       "learning_rate": 7.320441988950276e-06,
       "loss": 0.0006,
       "step": 700
     },
     {
       "epoch": 6.037267080745342,
-      "grad_norm": 0.01857682503759861,
+      "grad_norm": 0.01591545157134533,
       "learning_rate": 5.939226519337017e-06,
       "loss": 0.0006,
       "step": 720
     },
     {
       "epoch": 6.062111801242236,
-      "grad_norm": 0.010129285044968128,
+      "grad_norm": 0.010462045669555664,
       "learning_rate": 4.5580110497237574e-06,
       "loss": 0.0006,
       "step": 740
     },
     {
       "epoch": 6.086956521739131,
-      "grad_norm": 0.007020850665867329,
+      "grad_norm": 0.006039341911673546,
       "learning_rate": 3.1767955801104974e-06,
       "loss": 0.0006,
       "step": 760
     },
     {
       "epoch": 6.111801242236025,
-      "grad_norm": 0.009550134651362896,
+      "grad_norm": 0.010201127268373966,
       "learning_rate": 1.7955801104972376e-06,
       "loss": 0.0006,
       "step": 780
     },
     {
       "epoch": 6.136645962732919,
-      "grad_norm": 0.013940293341875076,
+      "grad_norm": 0.013513385318219662,
       "learning_rate": 4.143646408839779e-07,
       "loss": 0.0007,
       "step": 800
     },
     {
       "epoch": 6.142857142857143,
-      "eval_accuracy": 0.7317073170731707,
-      "eval_loss": 1.3218822479248047,
-      "eval_runtime": 6.2549,
-      "eval_samples_per_second": 13.11,
-      "eval_steps_per_second": 3.357,
+      "eval_accuracy": 0.7702702702702703,
+      "eval_loss": 1.14006769657135,
+      "eval_runtime": 5.5871,
+      "eval_samples_per_second": 13.245,
+      "eval_steps_per_second": 3.401,
       "step": 805
     },
     {
       "epoch": 6.142857142857143,
       "step": 805,
       "total_flos": 4.012506890622075e+18,
-      "train_loss": 0.1384738716829036,
-      "train_runtime": 848.1252,
-      "train_samples_per_second": 3.797,
-      "train_steps_per_second": 0.949
+      "train_loss": 0.1426203187803813,
+      "train_runtime": 889.4005,
+      "train_samples_per_second": 3.62,
+      "train_steps_per_second": 0.905
     },
     {
       "epoch": 6.142857142857143,
-      "eval_accuracy": 0.7804878048780488,
-      "eval_loss": 1.077329158782959,
-      "eval_runtime": 5.6706,
-      "eval_samples_per_second": 14.461,
-      "eval_steps_per_second": 3.703,
+      "eval_accuracy": 0.8028169014084507,
+      "eval_loss": 1.103053331375122,
+      "eval_runtime": 4.9334,
+      "eval_samples_per_second": 14.392,
+      "eval_steps_per_second": 3.649,
       "step": 805
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:acfa06c9a4d7a0d20f474dc00057d5fe4ef6648d6c36d35f2ad4dfe9c51a8a7b
+oid sha256:4cadfece94be4587a2a84d50343b8e2befa8b9d8af2d646cdedf5a702a909129
 size 5368