Upload 4 files

config.json

@@ -9,33 +9,23 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "round",
-    "1": "vneck",
-    "2": "collared",
-    "3": "straight",
-    "4": "highneck",
-    "5": "hoodie",
-    "6": "henley",
-    "7": "halter",
-    "8": "sweetheart",
-    "9": "polo",
-    "10": "asymmetrical"
+    "0": "Crop",
+    "1": "Regular",
+    "2": "Hip",
+    "3": "Knee",
+    "4": "Maxi",
+    "5": "Bodysuit"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "asymmetrical": 10,
-    "collared": 2,
-    "halter": 7,
-    "henley": 6,
-    "highneck": 4,
-    "hoodie": 5,
-    "polo": 9,
-    "round": 0,
-    "straight": 3,
-    "sweetheart": 8,
-    "vneck": 1
+    "Bodysuit": 5,
+    "Crop": 0,
+    "Hip": 2,
+    "Knee": 3,
+    "Maxi": 4,
+    "Regular": 1
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aeb3ba6832a1f8ea587c9e935a64a1337fb16be967c27b95e4e24aef18a4ce37
-size 343251660
+oid sha256:955e38edf95dd693c962e91cc18b87fede8c26c0d08cec97a97957f150c24477
+size 343236280

trainer_state.json

@@ -1,838 +1,214 @@
 {
-  "best_metric": 0.7836363636363637,
-  "best_model_checkpoint": "/content/drive/MyDrive/autoTaggingProject/ViT/General/Features/NeckLine/Results/model_2024-10-16_test/checkpoint-5192",
-  "epoch": 11.0,
+  "best_metric": 0.8563636363636363,
+  "best_model_checkpoint": "/content/drive/MyDrive/autoTaggingProject/ViT/General/Features/TopLength/Results/model_2024-10-16_test/checkpoint-1925",
+  "epoch": 7.0,
   "eval_steps": 500,
-  "global_step": 5192,
+  "global_step": 1925,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1059322033898305,
-      "grad_norm": 1.528609037399292,
-      "learning_rate": 9.964689265536724e-06,
-      "loss": 2.3665,
-      "step": 50
-    },
-    {
-      "epoch": 0.211864406779661,
-      "grad_norm": 1.4840689897537231,
-      "learning_rate": 9.929378531073447e-06,
-      "loss": 2.2771,
+      "epoch": 0.36363636363636365,
+      "grad_norm": 1.4969431161880493,
+      "learning_rate": 9.81818181818182e-06,
+      "loss": 1.6534,
       "step": 100
     },
     {
-      "epoch": 0.3177966101694915,
-      "grad_norm": 1.6797162294387817,
-      "learning_rate": 9.89406779661017e-06,
-      "loss": 2.1584,
-      "step": 150
-    },
-    {
-      "epoch": 0.423728813559322,
-      "grad_norm": 1.5869511365890503,
-      "learning_rate": 9.858757062146892e-06,
-      "loss": 2.0337,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 2.2577221393585205,
+      "learning_rate": 9.636363636363638e-06,
+      "loss": 1.317,
       "step": 200
     },
     {
-      "epoch": 0.5296610169491526,
-      "grad_norm": 1.625272512435913,
-      "learning_rate": 9.823446327683617e-06,
-      "loss": 1.945,
-      "step": 250
-    },
-    {
-      "epoch": 0.635593220338983,
-      "grad_norm": 1.918320655822754,
-      "learning_rate": 9.788135593220339e-06,
-      "loss": 1.8575,
+      "epoch": 1.0,
+      "eval_accuracy": 0.7218181818181818,
+      "eval_loss": 1.0995656251907349,
+      "eval_runtime": 12.6938,
+      "eval_samples_per_second": 43.328,
+      "eval_steps_per_second": 1.418,
+      "step": 275
+    },
+    {
+      "epoch": 1.0909090909090908,
+      "grad_norm": 2.7553977966308594,
+      "learning_rate": 9.454545454545456e-06,
+      "loss": 1.0529,
       "step": 300
     },
     {
-      "epoch": 0.7415254237288136,
-      "grad_norm": 2.73588490486145,
-      "learning_rate": 9.752824858757062e-06,
-      "loss": 1.795,
-      "step": 350
-    },
-    {
-      "epoch": 0.847457627118644,
-      "grad_norm": 1.978835105895996,
-      "learning_rate": 9.717514124293787e-06,
-      "loss": 1.7206,
+      "epoch": 1.4545454545454546,
+      "grad_norm": 2.5488719940185547,
+      "learning_rate": 9.272727272727273e-06,
+      "loss": 0.9003,
       "step": 400
     },
     {
-      "epoch": 0.9533898305084746,
-      "grad_norm": 2.827810764312744,
-      "learning_rate": 9.682203389830509e-06,
-      "loss": 1.643,
-      "step": 450
-    },
-    {
-      "epoch": 1.0,
-      "eval_accuracy": 0.5754545454545454,
-      "eval_loss": 1.6228725910186768,
-      "eval_runtime": 14.283,
-      "eval_samples_per_second": 77.014,
-      "eval_steps_per_second": 2.45,
-      "step": 472
-    },
-    {
-      "epoch": 1.0593220338983051,
-      "grad_norm": 2.191129207611084,
-      "learning_rate": 9.646892655367232e-06,
-      "loss": 1.5876,
+      "epoch": 1.8181818181818183,
+      "grad_norm": 2.936871290206909,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 0.7955,
       "step": 500
     },
     {
-      "epoch": 1.1652542372881356,
-      "grad_norm": 2.16697359085083,
-      "learning_rate": 9.611581920903955e-06,
-      "loss": 1.5416,
+      "epoch": 2.0,
+      "eval_accuracy": 0.7890909090909091,
+      "eval_loss": 0.8251153826713562,
+      "eval_runtime": 11.8903,
+      "eval_samples_per_second": 46.256,
+      "eval_steps_per_second": 1.514,
       "step": 550
     },
     {
-      "epoch": 1.271186440677966,
-      "grad_norm": 2.1197969913482666,
-      "learning_rate": 9.576271186440679e-06,
-      "loss": 1.5015,
+      "epoch": 2.1818181818181817,
+      "grad_norm": 3.6827473640441895,
+      "learning_rate": 8.90909090909091e-06,
+      "loss": 0.7386,
       "step": 600
     },
     {
-      "epoch": 1.3771186440677967,
-      "grad_norm": 2.465144634246826,
-      "learning_rate": 9.540960451977402e-06,
-      "loss": 1.4279,
-      "step": 650
-    },
-    {
-      "epoch": 1.4830508474576272,
-      "grad_norm": 1.8128849267959595,
-      "learning_rate": 9.505649717514125e-06,
-      "loss": 1.4298,
+      "epoch": 2.5454545454545454,
+      "grad_norm": 2.6154849529266357,
+      "learning_rate": 8.727272727272728e-06,
+      "loss": 0.6758,
       "step": 700
     },
     {
-      "epoch": 1.5889830508474576,
-      "grad_norm": 2.8659541606903076,
-      "learning_rate": 9.470338983050848e-06,
-      "loss": 1.3816,
-      "step": 750
-    },
-    {
-      "epoch": 1.694915254237288,
-      "grad_norm": 2.582030773162842,
-      "learning_rate": 9.435028248587572e-06,
-      "loss": 1.3254,
+      "epoch": 2.909090909090909,
+      "grad_norm": 6.043990135192871,
+      "learning_rate": 8.545454545454546e-06,
+      "loss": 0.6518,
       "step": 800
     },
     {
-      "epoch": 1.8008474576271185,
-      "grad_norm": 2.6557815074920654,
-      "learning_rate": 9.399717514124295e-06,
-      "loss": 1.2937,
-      "step": 850
-    },
-    {
-      "epoch": 1.9067796610169492,
-      "grad_norm": 3.4831998348236084,
-      "learning_rate": 9.364406779661017e-06,
-      "loss": 1.2511,
+      "epoch": 3.0,
+      "eval_accuracy": 0.82,
+      "eval_loss": 0.7125250101089478,
+      "eval_runtime": 11.5548,
+      "eval_samples_per_second": 47.599,
+      "eval_steps_per_second": 1.558,
+      "step": 825
+    },
+    {
+      "epoch": 3.2727272727272725,
+      "grad_norm": 1.676995873451233,
+      "learning_rate": 8.363636363636365e-06,
+      "loss": 0.6034,
       "step": 900
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.6927272727272727,
-      "eval_loss": 1.3029816150665283,
-      "eval_runtime": 14.6712,
-      "eval_samples_per_second": 74.977,
-      "eval_steps_per_second": 2.386,
-      "step": 944
-    },
-    {
-      "epoch": 2.01271186440678,
-      "grad_norm": 2.9632270336151123,
-      "learning_rate": 9.329096045197742e-06,
-      "loss": 1.2574,
-      "step": 950
-    },
-    {
-      "epoch": 2.1186440677966103,
-      "grad_norm": 4.297321796417236,
-      "learning_rate": 9.293785310734465e-06,
-      "loss": 1.213,
+      "epoch": 3.6363636363636362,
+      "grad_norm": 2.6665966510772705,
+      "learning_rate": 8.181818181818183e-06,
+      "loss": 0.5865,
       "step": 1000
     },
     {
-      "epoch": 2.2245762711864407,
-      "grad_norm": 5.289111614227295,
-      "learning_rate": 9.258474576271187e-06,
-      "loss": 1.2127,
-      "step": 1050
-    },
-    {
-      "epoch": 2.330508474576271,
-      "grad_norm": 5.046447277069092,
-      "learning_rate": 9.22316384180791e-06,
-      "loss": 1.1403,
+      "epoch": 4.0,
+      "grad_norm": 3.419482469558716,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.5531,
       "step": 1100
     },
     {
-      "epoch": 2.4364406779661016,
-      "grad_norm": 3.322784662246704,
-      "learning_rate": 9.187853107344633e-06,
-      "loss": 1.1244,
-      "step": 1150
+      "epoch": 4.0,
+      "eval_accuracy": 0.8218181818181818,
+      "eval_loss": 0.5728441476821899,
+      "eval_runtime": 11.88,
+      "eval_samples_per_second": 46.296,
+      "eval_steps_per_second": 1.515,
+      "step": 1100
     },
     {
-      "epoch": 2.542372881355932,
-      "grad_norm": 2.8034584522247314,
-      "learning_rate": 9.152542372881356e-06,
-      "loss": 1.1052,
+      "epoch": 4.363636363636363,
+      "grad_norm": 2.151674747467041,
+      "learning_rate": 7.81818181818182e-06,
+      "loss": 0.5151,
       "step": 1200
     },
     {
-      "epoch": 2.648305084745763,
-      "grad_norm": 2.822523593902588,
-      "learning_rate": 9.11723163841808e-06,
-      "loss": 1.0929,
-      "step": 1250
-    },
-    {
-      "epoch": 2.7542372881355934,
-      "grad_norm": 4.488712787628174,
-      "learning_rate": 9.081920903954803e-06,
-      "loss": 1.0991,
+      "epoch": 4.7272727272727275,
+      "grad_norm": 4.103756427764893,
+      "learning_rate": 7.63818181818182e-06,
+      "loss": 0.5118,
       "step": 1300
     },
     {
-      "epoch": 2.860169491525424,
-      "grad_norm": 4.986589431762695,
-      "learning_rate": 9.046610169491526e-06,
-      "loss": 1.0464,
-      "step": 1350
-    },
-    {
-      "epoch": 2.9661016949152543,
-      "grad_norm": 4.071073532104492,
-      "learning_rate": 9.01129943502825e-06,
-      "loss": 1.0532,
+      "epoch": 5.0,
+      "eval_accuracy": 0.84,
+      "eval_loss": 0.631469190120697,
+      "eval_runtime": 11.4374,
+      "eval_samples_per_second": 48.088,
+      "eval_steps_per_second": 1.574,
+      "step": 1375
+    },
+    {
+      "epoch": 5.090909090909091,
+      "grad_norm": 2.0776937007904053,
+      "learning_rate": 7.456363636363637e-06,
+      "loss": 0.4878,
       "step": 1400
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.7209090909090909,
-      "eval_loss": 1.0850567817687988,
-      "eval_runtime": 17.8307,
-      "eval_samples_per_second": 61.691,
-      "eval_steps_per_second": 1.963,
-      "step": 1416
-    },
-    {
-      "epoch": 3.0720338983050848,
-      "grad_norm": 4.750606536865234,
-      "learning_rate": 8.975988700564973e-06,
-      "loss": 1.0453,
-      "step": 1450
-    },
-    {
-      "epoch": 3.1779661016949152,
-      "grad_norm": 3.7604198455810547,
-      "learning_rate": 8.940677966101694e-06,
-      "loss": 1.0231,
+      "epoch": 5.454545454545454,
+      "grad_norm": 4.111904621124268,
+      "learning_rate": 7.274545454545456e-06,
+      "loss": 0.4802,
       "step": 1500
     },
     {
-      "epoch": 3.2838983050847457,
-      "grad_norm": 5.222371578216553,
-      "learning_rate": 8.90536723163842e-06,
-      "loss": 0.9812,
-      "step": 1550
-    },
-    {
-      "epoch": 3.389830508474576,
-      "grad_norm": 6.1730570793151855,
-      "learning_rate": 8.870056497175143e-06,
-      "loss": 0.9771,
+      "epoch": 5.818181818181818,
+      "grad_norm": 3.490161657333374,
+      "learning_rate": 7.092727272727273e-06,
+      "loss": 0.4713,
       "step": 1600
     },
     {
-      "epoch": 3.4957627118644066,
-      "grad_norm": 5.553199291229248,
-      "learning_rate": 8.834745762711864e-06,
-      "loss": 0.9682,
+      "epoch": 6.0,
+      "eval_accuracy": 0.84,
+      "eval_loss": 0.4893127381801605,
+      "eval_runtime": 11.1344,
+      "eval_samples_per_second": 49.396,
+      "eval_steps_per_second": 1.617,
       "step": 1650
     },
     {
-      "epoch": 3.601694915254237,
-      "grad_norm": 4.2731451988220215,
-      "learning_rate": 8.79943502824859e-06,
-      "loss": 1.0006,
+      "epoch": 6.181818181818182,
+      "grad_norm": 3.1339528560638428,
+      "learning_rate": 6.910909090909092e-06,
+      "loss": 0.4365,
       "step": 1700
     },
     {
-      "epoch": 3.707627118644068,
-      "grad_norm": 3.617053985595703,
-      "learning_rate": 8.764124293785311e-06,
-      "loss": 0.9154,
-      "step": 1750
-    },
-    {
-      "epoch": 3.8135593220338984,
-      "grad_norm": 2.5833144187927246,
-      "learning_rate": 8.728813559322034e-06,
-      "loss": 0.935,
+      "epoch": 6.545454545454545,
+      "grad_norm": 3.686084032058716,
+      "learning_rate": 6.72909090909091e-06,
+      "loss": 0.4155,
       "step": 1800
     },
     {
-      "epoch": 3.919491525423729,
-      "grad_norm": 3.1606085300445557,
-      "learning_rate": 8.693502824858758e-06,
-      "loss": 0.9216,
-      "step": 1850
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.74,
-      "eval_loss": 0.9795950651168823,
-      "eval_runtime": 16.0161,
-      "eval_samples_per_second": 68.681,
-      "eval_steps_per_second": 2.185,
-      "step": 1888
-    },
-    {
-      "epoch": 4.02542372881356,
-      "grad_norm": 3.959052801132202,
-      "learning_rate": 8.65819209039548e-06,
-      "loss": 0.9288,
+      "epoch": 6.909090909090909,
+      "grad_norm": 3.6479032039642334,
+      "learning_rate": 6.5472727272727275e-06,
+      "loss": 0.4251,
       "step": 1900
     },
-    {
-      "epoch": 4.13135593220339,
-      "grad_norm": 3.6853768825531006,
-      "learning_rate": 8.622881355932204e-06,
-      "loss": 0.879,
-      "step": 1950
-    },
-    {
-      "epoch": 4.237288135593221,
-      "grad_norm": 3.1763620376586914,
-      "learning_rate": 8.587570621468927e-06,
-      "loss": 0.8912,
-      "step": 2000
-    },
-    {
-      "epoch": 4.343220338983051,
-      "grad_norm": 4.019489288330078,
-      "learning_rate": 8.55225988700565e-06,
-      "loss": 0.9077,
-      "step": 2050
-    },
-    {
-      "epoch": 4.4491525423728815,
-      "grad_norm": 4.087663650512695,
-      "learning_rate": 8.516949152542372e-06,
-      "loss": 0.8812,
-      "step": 2100
-    },
-    {
-      "epoch": 4.555084745762712,
-      "grad_norm": 3.6994051933288574,
-      "learning_rate": 8.481638418079097e-06,
-      "loss": 0.8344,
-      "step": 2150
-    },
-    {
-      "epoch": 4.661016949152542,
-      "grad_norm": 2.284302234649658,
-      "learning_rate": 8.44632768361582e-06,
-      "loss": 0.8501,
-      "step": 2200
-    },
-    {
-      "epoch": 4.766949152542373,
-      "grad_norm": 4.348343372344971,
-      "learning_rate": 8.411016949152542e-06,
-      "loss": 0.8712,
-      "step": 2250
-    },
-    {
-      "epoch": 4.872881355932203,
-      "grad_norm": 4.996354103088379,
-      "learning_rate": 8.375706214689267e-06,
-      "loss": 0.8264,
-      "step": 2300
-    },
-    {
-      "epoch": 4.978813559322034,
-      "grad_norm": 3.136770486831665,
-      "learning_rate": 8.340395480225989e-06,
-      "loss": 0.843,
-      "step": 2350
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.7390909090909091,
-      "eval_loss": 0.9454855918884277,
-      "eval_runtime": 15.0504,
-      "eval_samples_per_second": 73.088,
-      "eval_steps_per_second": 2.326,
-      "step": 2360
-    },
-    {
-      "epoch": 5.084745762711864,
-      "grad_norm": 6.956203937530518,
-      "learning_rate": 8.305084745762712e-06,
-      "loss": 0.8094,
-      "step": 2400
-    },
-    {
-      "epoch": 5.190677966101695,
-      "grad_norm": 6.91636323928833,
-      "learning_rate": 8.269774011299437e-06,
-      "loss": 0.8301,
-      "step": 2450
-    },
-    {
-      "epoch": 5.296610169491525,
-      "grad_norm": 2.561798334121704,
-      "learning_rate": 8.234463276836159e-06,
-      "loss": 0.8562,
-      "step": 2500
-    },
-    {
-      "epoch": 5.4025423728813555,
-      "grad_norm": 4.503079891204834,
-      "learning_rate": 8.199152542372882e-06,
-      "loss": 0.7487,
-      "step": 2550
-    },
-    {
-      "epoch": 5.508474576271187,
-      "grad_norm": 3.560302257537842,
-      "learning_rate": 8.163841807909605e-06,
-      "loss": 0.8222,
-      "step": 2600
-    },
-    {
-      "epoch": 5.614406779661017,
-      "grad_norm": 6.565722465515137,
-      "learning_rate": 8.128531073446328e-06,
-      "loss": 0.7917,
-      "step": 2650
-    },
-    {
-      "epoch": 5.720338983050848,
-      "grad_norm": 7.790140151977539,
-      "learning_rate": 8.093220338983052e-06,
-      "loss": 0.7764,
-      "step": 2700
-    },
-    {
-      "epoch": 5.826271186440678,
-      "grad_norm": 4.017592430114746,
-      "learning_rate": 8.057909604519775e-06,
-      "loss": 0.7718,
-      "step": 2750
-    },
-    {
-      "epoch": 5.932203389830509,
-      "grad_norm": 6.110499382019043,
-      "learning_rate": 8.022598870056498e-06,
-      "loss": 0.7445,
-      "step": 2800
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.7527272727272727,
-      "eval_loss": 0.8885732889175415,
-      "eval_runtime": 15.0594,
-      "eval_samples_per_second": 73.044,
-      "eval_steps_per_second": 2.324,
-      "step": 2832
-    },
-    {
-      "epoch": 6.038135593220339,
-      "grad_norm": 4.600073337554932,
-      "learning_rate": 7.987288135593222e-06,
-      "loss": 0.7767,
-      "step": 2850
-    },
-    {
-      "epoch": 6.1440677966101696,
-      "grad_norm": 5.356403827667236,
-      "learning_rate": 7.951977401129945e-06,
-      "loss": 0.708,
-      "step": 2900
-    },
-    {
-      "epoch": 6.25,
-      "grad_norm": 5.1397223472595215,
-      "learning_rate": 7.917372881355932e-06,
-      "loss": 0.7397,
-      "step": 2950
-    },
-    {
-      "epoch": 6.3559322033898304,
-      "grad_norm": 6.384206771850586,
-      "learning_rate": 7.882062146892657e-06,
-      "loss": 0.7523,
-      "step": 3000
-    },
-    {
-      "epoch": 6.461864406779661,
-      "grad_norm": 5.545274257659912,
-      "learning_rate": 7.846751412429378e-06,
-      "loss": 0.7011,
-      "step": 3050
-    },
-    {
-      "epoch": 6.567796610169491,
-      "grad_norm": 9.406649589538574,
-      "learning_rate": 7.811440677966102e-06,
-      "loss": 0.7975,
-      "step": 3100
-    },
-    {
-      "epoch": 6.673728813559322,
-      "grad_norm": 7.81419563293457,
-      "learning_rate": 7.776129943502827e-06,
-      "loss": 0.7385,
-      "step": 3150
-    },
-    {
-      "epoch": 6.779661016949152,
-      "grad_norm": 3.415956974029541,
-      "learning_rate": 7.740819209039548e-06,
-      "loss": 0.7356,
-      "step": 3200
-    },
-    {
-      "epoch": 6.885593220338983,
-      "grad_norm": 5.487062931060791,
-      "learning_rate": 7.705508474576271e-06,
-      "loss": 0.7121,
-      "step": 3250
-    },
-    {
-      "epoch": 6.991525423728813,
-      "grad_norm": 5.682718276977539,
-      "learning_rate": 7.670197740112995e-06,
-      "loss": 0.7191,
-      "step": 3300
-    },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7545454545454545,
-      "eval_loss": 0.8240677118301392,
-      "eval_runtime": 14.7678,
-      "eval_samples_per_second": 74.486,
-      "eval_steps_per_second": 2.37,
-      "step": 3304
-    },
-    {
-      "epoch": 7.0974576271186445,
-      "grad_norm": 8.319087028503418,
-      "learning_rate": 7.634887005649718e-06,
-      "loss": 0.7096,
-      "step": 3350
-    },
-    {
-      "epoch": 7.203389830508475,
-      "grad_norm": 5.857816696166992,
-      "learning_rate": 7.599576271186442e-06,
-      "loss": 0.7006,
-      "step": 3400
-    },
-    {
-      "epoch": 7.309322033898305,
-      "grad_norm": 4.400519371032715,
-      "learning_rate": 7.564265536723165e-06,
-      "loss": 0.706,
-      "step": 3450
-    },
-    {
-      "epoch": 7.415254237288136,
-      "grad_norm": 4.573615550994873,
-      "learning_rate": 7.528954802259888e-06,
-      "loss": 0.6543,
-      "step": 3500
-    },
-    {
-      "epoch": 7.521186440677966,
-      "grad_norm": 7.545746803283691,
-      "learning_rate": 7.49364406779661e-06,
-      "loss": 0.6498,
-      "step": 3550
-    },
-    {
-      "epoch": 7.627118644067797,
-      "grad_norm": 6.38883638381958,
-      "learning_rate": 7.4583333333333345e-06,
-      "loss": 0.6849,
-      "step": 3600
-    },
-    {
-      "epoch": 7.733050847457627,
-      "grad_norm": 4.496486186981201,
-      "learning_rate": 7.423022598870057e-06,
-      "loss": 0.6741,
-      "step": 3650
-    },
-    {
-      "epoch": 7.838983050847458,
-      "grad_norm": 5.2381792068481445,
-      "learning_rate": 7.38771186440678e-06,
-      "loss": 0.6441,
-      "step": 3700
-    },
-    {
-      "epoch": 7.944915254237288,
-      "grad_norm": 6.047347068786621,
-      "learning_rate": 7.3524011299435035e-06,
-      "loss": 0.7089,
-      "step": 3750
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.7718181818181818,
-      "eval_loss": 0.845079779624939,
-      "eval_runtime": 14.9091,
-      "eval_samples_per_second": 73.78,
-      "eval_steps_per_second": 2.348,
-      "step": 3776
-    },
-    {
-      "epoch": 8.05084745762712,
-      "grad_norm": 5.383782386779785,
-      "learning_rate": 7.317090395480226e-06,
-      "loss": 0.6375,
-      "step": 3800
-    },
-    {
-      "epoch": 8.15677966101695,
-      "grad_norm": 7.663337707519531,
-      "learning_rate": 7.28177966101695e-06,
-      "loss": 0.6101,
-      "step": 3850
-    },
-    {
-      "epoch": 8.26271186440678,
-      "grad_norm": 4.593461036682129,
-      "learning_rate": 7.2464689265536725e-06,
-      "loss": 0.6356,
-      "step": 3900
-    },
-    {
-      "epoch": 8.36864406779661,
-      "grad_norm": 7.878734111785889,
-      "learning_rate": 7.211158192090396e-06,
-      "loss": 0.6412,
-      "step": 3950
-    },
-    {
-      "epoch": 8.474576271186441,
-      "grad_norm": 3.501059055328369,
-      "learning_rate": 7.17584745762712e-06,
-      "loss": 0.6296,
-      "step": 4000
-    },
-    {
-      "epoch": 8.580508474576272,
-      "grad_norm": 3.8199708461761475,
-      "learning_rate": 7.140536723163842e-06,
-      "loss": 0.654,
-      "step": 4050
-    },
-    {
-      "epoch": 8.686440677966102,
-      "grad_norm": 6.42057466506958,
-      "learning_rate": 7.105225988700566e-06,
-      "loss": 0.6378,
-      "step": 4100
-    },
-    {
-      "epoch": 8.792372881355933,
-      "grad_norm": 6.310295104980469,
-      "learning_rate": 7.069915254237288e-06,
-      "loss": 0.6235,
-      "step": 4150
-    },
-    {
-      "epoch": 8.898305084745763,
-      "grad_norm": 4.627810478210449,
-      "learning_rate": 7.034604519774012e-06,
-      "loss": 0.6631,
-      "step": 4200
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.77,
-      "eval_loss": 0.8626542687416077,
-      "eval_runtime": 15.548,
-      "eval_samples_per_second": 70.749,
-      "eval_steps_per_second": 2.251,
-      "step": 4248
-    },
-    {
-      "epoch": 9.004237288135593,
-      "grad_norm": 5.442898273468018,
-      "learning_rate": 6.999293785310735e-06,
-      "loss": 0.6358,
-      "step": 4250
-    },
-    {
-      "epoch": 9.110169491525424,
-      "grad_norm": 5.042696475982666,
-      "learning_rate": 6.963983050847458e-06,
-      "loss": 0.6183,
-      "step": 4300
-    },
-    {
-      "epoch": 9.216101694915254,
-      "grad_norm": 5.006898403167725,
-      "learning_rate": 6.928672316384182e-06,
-      "loss": 0.6438,
-      "step": 4350
-    },
-    {
-      "epoch": 9.322033898305085,
-      "grad_norm": 6.093140125274658,
-      "learning_rate": 6.893361581920905e-06,
-      "loss": 0.5826,
-      "step": 4400
-    },
-    {
-      "epoch": 9.427966101694915,
-      "grad_norm": 4.637847900390625,
-      "learning_rate": 6.858050847457628e-06,
-      "loss": 0.5559,
-      "step": 4450
-    },
-    {
-      "epoch": 9.533898305084746,
-      "grad_norm": 2.860111951828003,
-      "learning_rate": 6.82274011299435e-06,
-      "loss": 0.5577,
-      "step": 4500
-    },
-    {
-      "epoch": 9.639830508474576,
-      "grad_norm": 10.876856803894043,
-      "learning_rate": 6.7874293785310745e-06,
-      "loss": 0.6233,
-      "step": 4550
-    },
-    {
-      "epoch": 9.745762711864407,
-      "grad_norm": 5.635727882385254,
-      "learning_rate": 6.752118644067798e-06,
-      "loss": 0.5703,
-      "step": 4600
-    },
-    {
-      "epoch": 9.851694915254237,
-      "grad_norm": 6.9388532638549805,
-      "learning_rate": 6.71680790960452e-06,
-      "loss": 0.6323,
-      "step": 4650
-    },
-    {
-      "epoch": 9.957627118644067,
-      "grad_norm": 7.485644340515137,
-      "learning_rate": 6.6814971751412435e-06,
-      "loss": 0.6021,
-      "step": 4700
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.7772727272727272,
-      "eval_loss": 0.8030957579612732,
-      "eval_runtime": 14.6023,
-      "eval_samples_per_second": 75.331,
-      "eval_steps_per_second": 2.397,
-      "step": 4720
-    },
-    {
-      "epoch": 10.063559322033898,
-      "grad_norm": 13.398885726928711,
-      "learning_rate": 6.646186440677966e-06,
-      "loss": 0.5698,
-      "step": 4750
-    },
-    {
-      "epoch": 10.169491525423728,
-      "grad_norm": 7.821059226989746,
-      "learning_rate": 6.61087570621469e-06,
-      "loss": 0.5364,
-      "step": 4800
-    },
-    {
-      "epoch": 10.275423728813559,
-      "grad_norm": 7.535600662231445,
-      "learning_rate": 6.576271186440678e-06,
-      "loss": 0.5959,
-      "step": 4850
-    },
-    {
-      "epoch": 10.38135593220339,
-      "grad_norm": 8.605109214782715,
-      "learning_rate": 6.540960451977402e-06,
-      "loss": 0.5837,
-      "step": 4900
-    },
-    {
-      "epoch": 10.48728813559322,
-      "grad_norm": 5.240401268005371,
-      "learning_rate": 6.505649717514125e-06,
-      "loss": 0.5796,
-      "step": 4950
-    },
-    {
-      "epoch": 10.59322033898305,
-      "grad_norm": 6.025055408477783,
-      "learning_rate": 6.4703389830508476e-06,
-      "loss": 0.5749,
-      "step": 5000
-    },
-    {
-      "epoch": 10.69915254237288,
-      "grad_norm": 6.605931282043457,
-      "learning_rate": 6.435028248587572e-06,
-      "loss": 0.5185,
-      "step": 5050
-    },
-    {
-      "epoch": 10.805084745762711,
-      "grad_norm": 3.71102237701416,
-      "learning_rate": 6.399717514124294e-06,
-      "loss": 0.5289,
-      "step": 5100
-    },
-    {
-      "epoch": 10.911016949152543,
-      "grad_norm": 4.973482608795166,
-      "learning_rate": 6.3644067796610174e-06,
-      "loss": 0.5691,
-      "step": 5150
-    },
-    {
-      "epoch": 11.0,
-      "eval_accuracy": 0.7836363636363637,
-      "eval_loss": 0.8530685305595398,
-      "eval_runtime": 14.8369,
-      "eval_samples_per_second": 74.14,
-      "eval_steps_per_second": 2.359,
-      "step": 5192
+      "eval_accuracy": 0.8563636363636363,
+      "eval_loss": 0.48278242349624634,
+      "eval_runtime": 11.2502,
+      "eval_samples_per_second": 48.888,
+      "eval_steps_per_second": 1.6,
+      "step": 1925
     }
   ],
-  "logging_steps": 50,
-  "max_steps": 14160,
+  "logging_steps": 100,
+  "max_steps": 5500,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 30,
+  "num_train_epochs": 20,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -846,7 +222,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.286478541133356e+19,
+  "total_flos": 4.773677693976576e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null