End of training

README.md

@@ -5,6 +5,8 @@ license: apache-2.0
 metrics:
 - accuracy
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 model-index:
 - name: vit-base-beans
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0804
-- Accuracy: 0.9699
+- Loss: 0.0466
+- Accuracy: 0.9774
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_accuracy": 0.8045112781954887,
-    "eval_loss": 0.8347006440162659,
-    "eval_runtime": 4.3276,
-    "eval_samples_per_second": 30.733,
-    "eval_steps_per_second": 3.928,
+    "eval_accuracy": 0.9774436090225563,
+    "eval_loss": 0.04662612825632095,
+    "eval_runtime": 8.4011,
+    "eval_samples_per_second": 15.831,
+    "eval_steps_per_second": 2.024,
     "total_flos": 4.0477911662555136e+17,
-    "train_loss": 0.9380592258159931,
-    "train_runtime": 422.3657,
-    "train_samples_per_second": 12.241,
-    "train_steps_per_second": 1.539
+    "train_loss": 0.18834018239608177,
+    "train_runtime": 471.6062,
+    "train_samples_per_second": 10.963,
+    "train_steps_per_second": 1.378
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "eval_accuracy": 0.8045112781954887,
-    "eval_loss": 0.8347006440162659,
-    "eval_runtime": 4.3276,
-    "eval_samples_per_second": 30.733,
-    "eval_steps_per_second": 3.928
+    "eval_accuracy": 0.9774436090225563,
+    "eval_loss": 0.04662612825632095,
+    "eval_runtime": 8.4011,
+    "eval_samples_per_second": 15.831,
+    "eval_steps_per_second": 2.024
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.0477911662555136e+17,
-    "train_loss": 0.9380592258159931,
-    "train_runtime": 422.3657,
-    "train_samples_per_second": 12.241,
-    "train_steps_per_second": 1.539
+    "train_loss": 0.18834018239608177,
+    "train_runtime": 471.6062,
+    "train_samples_per_second": 10.963,
+    "train_steps_per_second": 1.378
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8347006440162659,
-  "best_model_checkpoint": "./beans_lora_outputs/checkpoint-650",
+  "best_metric": 0.04662612825632095,
+  "best_model_checkpoint": "./beans_lora_outputs/checkpoint-520",
   "epoch": 5.0,
   "eval_steps": 500,
   "global_step": 650,
@@ -10,512 +10,512 @@
   "log_history": [
     {
       "epoch": 0.07692307692307693,
-      "grad_norm": 1.496038794517517,
-      "learning_rate": 1.9692307692307696e-05,
-      "loss": 1.0741,
+      "grad_norm": 1.4020828008651733,
+      "learning_rate": 0.0004923076923076924,
+      "loss": 1.0336,
       "step": 10
     },
     {
       "epoch": 0.15384615384615385,
-      "grad_norm": 1.14773690700531,
-      "learning_rate": 1.9384615384615386e-05,
-      "loss": 1.0878,
+      "grad_norm": 1.1858539581298828,
+      "learning_rate": 0.0004846153846153846,
+      "loss": 0.9408,
       "step": 20
     },
     {
       "epoch": 0.23076923076923078,
-      "grad_norm": 1.61867356300354,
-      "learning_rate": 1.907692307692308e-05,
-      "loss": 1.08,
+      "grad_norm": 1.314568281173706,
+      "learning_rate": 0.000476923076923077,
+      "loss": 0.7742,
       "step": 30
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 2.2048051357269287,
-      "learning_rate": 1.876923076923077e-05,
-      "loss": 1.049,
+      "grad_norm": 1.8166171312332153,
+      "learning_rate": 0.00046923076923076926,
+      "loss": 0.5557,
       "step": 40
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 1.3817859888076782,
-      "learning_rate": 1.8461538461538465e-05,
-      "loss": 1.0672,
+      "grad_norm": 0.892727792263031,
+      "learning_rate": 0.0004615384615384616,
+      "loss": 0.5459,
       "step": 50
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 1.385685682296753,
-      "learning_rate": 1.8153846153846155e-05,
-      "loss": 1.0604,
+      "grad_norm": 1.4540048837661743,
+      "learning_rate": 0.00045384615384615385,
+      "loss": 0.4776,
       "step": 60
     },
     {
       "epoch": 0.5384615384615384,
-      "grad_norm": 1.9848027229309082,
-      "learning_rate": 1.784615384615385e-05,
-      "loss": 1.0445,
+      "grad_norm": 0.7993093729019165,
+      "learning_rate": 0.0004461538461538462,
+      "loss": 0.4206,
       "step": 70
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 1.7842564582824707,
-      "learning_rate": 1.753846153846154e-05,
-      "loss": 1.0431,
+      "grad_norm": 0.4129832088947296,
+      "learning_rate": 0.00043846153846153845,
+      "loss": 0.2951,
       "step": 80
     },
     {
       "epoch": 0.6923076923076923,
-      "grad_norm": 1.729790449142456,
-      "learning_rate": 1.7230769230769234e-05,
-      "loss": 1.0511,
+      "grad_norm": 1.1914254426956177,
+      "learning_rate": 0.00043076923076923083,
+      "loss": 0.2664,
       "step": 90
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 2.3279037475585938,
-      "learning_rate": 1.6923076923076924e-05,
-      "loss": 1.0556,
+      "grad_norm": 2.3125526905059814,
+      "learning_rate": 0.0004230769230769231,
+      "loss": 0.3194,
       "step": 100
     },
     {
       "epoch": 0.8461538461538461,
-      "grad_norm": 1.5423173904418945,
-      "learning_rate": 1.6615384615384618e-05,
-      "loss": 1.0575,
+      "grad_norm": 1.2275660037994385,
+      "learning_rate": 0.0004153846153846154,
+      "loss": 0.2889,
       "step": 110
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 1.8890392780303955,
-      "learning_rate": 1.630769230769231e-05,
-      "loss": 1.0304,
+      "grad_norm": 0.6353504657745361,
+      "learning_rate": 0.0004076923076923077,
+      "loss": 0.1575,
       "step": 120
     },
     {
       "epoch": 1.0,
-      "grad_norm": 2.8158857822418213,
-      "learning_rate": 1.6000000000000003e-05,
-      "loss": 1.0299,
+      "grad_norm": 3.177952527999878,
+      "learning_rate": 0.0004,
+      "loss": 0.2167,
       "step": 130
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6240601503759399,
-      "eval_loss": 1.0291587114334106,
-      "eval_runtime": 5.0639,
-      "eval_samples_per_second": 26.264,
-      "eval_steps_per_second": 3.357,
+      "eval_accuracy": 0.9473684210526315,
+      "eval_loss": 0.17138947546482086,
+      "eval_runtime": 9.2462,
+      "eval_samples_per_second": 14.384,
+      "eval_steps_per_second": 1.839,
       "step": 130
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 1.5042554140090942,
-      "learning_rate": 1.5692307692307693e-05,
-      "loss": 1.0233,
+      "grad_norm": 0.3688104748725891,
+      "learning_rate": 0.0003923076923076923,
+      "loss": 0.1432,
       "step": 140
     },
     {
       "epoch": 1.1538461538461537,
-      "grad_norm": 1.574145793914795,
-      "learning_rate": 1.5384615384615387e-05,
-      "loss": 1.0102,
+      "grad_norm": 1.1805006265640259,
+      "learning_rate": 0.00038461538461538467,
+      "loss": 0.2113,
       "step": 150
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 1.3951938152313232,
-      "learning_rate": 1.5076923076923078e-05,
-      "loss": 1.0171,
+      "grad_norm": 0.634524405002594,
+      "learning_rate": 0.00037692307692307694,
+      "loss": 0.1851,
       "step": 160
     },
     {
       "epoch": 1.3076923076923077,
-      "grad_norm": 1.6762858629226685,
-      "learning_rate": 1.4769230769230772e-05,
-      "loss": 1.0067,
+      "grad_norm": 0.15019208192825317,
+      "learning_rate": 0.00036923076923076927,
+      "loss": 0.1933,
       "step": 170
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 1.5464376211166382,
-      "learning_rate": 1.4461538461538462e-05,
-      "loss": 0.9966,
+      "grad_norm": 0.3095666766166687,
+      "learning_rate": 0.00036153846153846154,
+      "loss": 0.1281,
       "step": 180
     },
     {
       "epoch": 1.4615384615384617,
-      "grad_norm": 1.553270697593689,
-      "learning_rate": 1.4153846153846156e-05,
-      "loss": 0.998,
+      "grad_norm": 1.1930674314498901,
+      "learning_rate": 0.00035384615384615386,
+      "loss": 0.1348,
       "step": 190
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.3178517818450928,
-      "learning_rate": 1.3846153846153847e-05,
-      "loss": 1.0004,
+      "grad_norm": 0.1358138620853424,
+      "learning_rate": 0.00034615384615384613,
+      "loss": 0.2277,
       "step": 200
     },
     {
       "epoch": 1.6153846153846154,
-      "grad_norm": 1.414166808128357,
-      "learning_rate": 1.353846153846154e-05,
-      "loss": 0.9889,
+      "grad_norm": 2.295457363128662,
+      "learning_rate": 0.0003384615384615385,
+      "loss": 0.1802,
       "step": 210
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 2.7710893154144287,
-      "learning_rate": 1.3230769230769231e-05,
-      "loss": 0.9705,
+      "grad_norm": 0.1559237241744995,
+      "learning_rate": 0.0003307692307692308,
+      "loss": 0.1972,
       "step": 220
     },
     {
       "epoch": 1.7692307692307692,
-      "grad_norm": 1.6936691999435425,
-      "learning_rate": 1.2923076923076925e-05,
-      "loss": 0.9814,
+      "grad_norm": 0.8495343923568726,
+      "learning_rate": 0.0003230769230769231,
+      "loss": 0.176,
       "step": 230
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 1.4454400539398193,
-      "learning_rate": 1.2615384615384616e-05,
-      "loss": 0.9935,
+      "grad_norm": 1.684859037399292,
+      "learning_rate": 0.0003153846153846154,
+      "loss": 0.2912,
       "step": 240
     },
     {
       "epoch": 1.9230769230769231,
-      "grad_norm": 1.6399688720703125,
-      "learning_rate": 1.230769230769231e-05,
-      "loss": 0.965,
+      "grad_norm": 0.2597804367542267,
+      "learning_rate": 0.0003076923076923077,
+      "loss": 0.1231,
       "step": 250
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.0613536834716797,
-      "learning_rate": 1.2e-05,
-      "loss": 0.9616,
+      "grad_norm": 0.09072303026914597,
+      "learning_rate": 0.0003,
+      "loss": 0.1288,
       "step": 260
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7443609022556391,
-      "eval_loss": 0.9589495062828064,
-      "eval_runtime": 4.3534,
-      "eval_samples_per_second": 30.551,
-      "eval_steps_per_second": 3.905,
+      "eval_accuracy": 0.9548872180451128,
+      "eval_loss": 0.13066871464252472,
+      "eval_runtime": 7.0796,
+      "eval_samples_per_second": 18.786,
+      "eval_steps_per_second": 2.401,
       "step": 260
     },
     {
       "epoch": 2.076923076923077,
-      "grad_norm": 1.2324800491333008,
-      "learning_rate": 1.1692307692307694e-05,
-      "loss": 0.9474,
+      "grad_norm": 0.09045865386724472,
+      "learning_rate": 0.00029230769230769235,
+      "loss": 0.0485,
       "step": 270
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 1.842686414718628,
-      "learning_rate": 1.1384615384615385e-05,
-      "loss": 0.9555,
+      "grad_norm": 0.0802936926484108,
+      "learning_rate": 0.0002846153846153846,
+      "loss": 0.1315,
       "step": 280
     },
     {
       "epoch": 2.230769230769231,
-      "grad_norm": 1.8932175636291504,
-      "learning_rate": 1.1076923076923079e-05,
-      "loss": 0.9411,
+      "grad_norm": 2.443058967590332,
+      "learning_rate": 0.00027692307692307695,
+      "loss": 0.1178,
       "step": 290
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.4860458374023438,
-      "learning_rate": 1.076923076923077e-05,
-      "loss": 0.931,
+      "grad_norm": 2.5117151737213135,
+      "learning_rate": 0.0002692307692307692,
+      "loss": 0.1197,
       "step": 300
     },
     {
       "epoch": 2.3846153846153846,
-      "grad_norm": 1.520959496498108,
-      "learning_rate": 1.0461538461538463e-05,
-      "loss": 0.9442,
+      "grad_norm": 0.6276033520698547,
+      "learning_rate": 0.00026153846153846154,
+      "loss": 0.1327,
       "step": 310
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 1.3432276248931885,
-      "learning_rate": 1.0153846153846154e-05,
-      "loss": 0.9292,
+      "grad_norm": 0.07618015259504318,
+      "learning_rate": 0.0002538461538461538,
+      "loss": 0.1751,
       "step": 320
     },
     {
       "epoch": 2.5384615384615383,
-      "grad_norm": 1.2196266651153564,
-      "learning_rate": 9.846153846153848e-06,
-      "loss": 0.9271,
+      "grad_norm": 1.933332920074463,
+      "learning_rate": 0.0002461538461538462,
+      "loss": 0.1409,
       "step": 330
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 1.521013855934143,
-      "learning_rate": 9.53846153846154e-06,
-      "loss": 0.917,
+      "grad_norm": 1.7085862159729004,
+      "learning_rate": 0.0002384615384615385,
+      "loss": 0.076,
       "step": 340
     },
     {
       "epoch": 2.6923076923076925,
-      "grad_norm": 1.5005229711532593,
-      "learning_rate": 9.230769230769232e-06,
-      "loss": 0.9225,
+      "grad_norm": 0.08163443952798843,
+      "learning_rate": 0.0002307692307692308,
+      "loss": 0.1995,
       "step": 350
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 1.3230764865875244,
-      "learning_rate": 8.923076923076925e-06,
-      "loss": 0.9115,
+      "grad_norm": 3.548895835876465,
+      "learning_rate": 0.0002230769230769231,
+      "loss": 0.1168,
       "step": 360
     },
     {
       "epoch": 2.8461538461538463,
-      "grad_norm": 1.4843106269836426,
-      "learning_rate": 8.615384615384617e-06,
-      "loss": 0.9077,
+      "grad_norm": 0.2048652619123459,
+      "learning_rate": 0.00021538461538461541,
+      "loss": 0.1259,
       "step": 370
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 1.432364583015442,
-      "learning_rate": 8.307692307692309e-06,
-      "loss": 0.901,
+      "grad_norm": 0.46744510531425476,
+      "learning_rate": 0.0002076923076923077,
+      "loss": 0.0786,
       "step": 380
     },
     {
       "epoch": 3.0,
-      "grad_norm": 2.669428586959839,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 0.8783,
+      "grad_norm": 0.05968974158167839,
+      "learning_rate": 0.0002,
+      "loss": 0.1056,
       "step": 390
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7894736842105263,
-      "eval_loss": 0.8891889452934265,
-      "eval_runtime": 17.6473,
-      "eval_samples_per_second": 7.537,
-      "eval_steps_per_second": 0.963,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.06689254194498062,
+      "eval_runtime": 7.9823,
+      "eval_samples_per_second": 16.662,
+      "eval_steps_per_second": 2.13,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 1.462823748588562,
-      "learning_rate": 7.692307692307694e-06,
-      "loss": 0.9124,
+      "grad_norm": 1.8612762689590454,
+      "learning_rate": 0.00019230769230769233,
+      "loss": 0.1471,
       "step": 400
     },
     {
       "epoch": 3.1538461538461537,
-      "grad_norm": 1.2650939226150513,
-      "learning_rate": 7.384615384615386e-06,
-      "loss": 0.9026,
+      "grad_norm": 3.9264113903045654,
+      "learning_rate": 0.00018461538461538463,
+      "loss": 0.1647,
       "step": 410
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 1.514074683189392,
-      "learning_rate": 7.076923076923078e-06,
-      "loss": 0.9024,
+      "grad_norm": 1.6901596784591675,
+      "learning_rate": 0.00017692307692307693,
+      "loss": 0.0862,
       "step": 420
     },
     {
       "epoch": 3.3076923076923075,
-      "grad_norm": 2.043466091156006,
-      "learning_rate": 6.76923076923077e-06,
-      "loss": 0.8674,
+      "grad_norm": 0.046002600342035294,
+      "learning_rate": 0.00016923076923076926,
+      "loss": 0.0178,
       "step": 430
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 1.8109254837036133,
-      "learning_rate": 6.461538461538463e-06,
-      "loss": 0.8879,
+      "grad_norm": 1.302810549736023,
+      "learning_rate": 0.00016153846153846155,
+      "loss": 0.1012,
       "step": 440
     },
     {
       "epoch": 3.4615384615384617,
-      "grad_norm": 1.5040030479431152,
-      "learning_rate": 6.153846153846155e-06,
-      "loss": 0.8623,
+      "grad_norm": 1.406225323677063,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 0.0446,
       "step": 450
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 1.7636866569519043,
-      "learning_rate": 5.846153846153847e-06,
-      "loss": 0.8787,
+      "grad_norm": 3.55471134185791,
+      "learning_rate": 0.00014615384615384618,
+      "loss": 0.0553,
       "step": 460
     },
     {
       "epoch": 3.6153846153846154,
-      "grad_norm": 1.368589162826538,
-      "learning_rate": 5.538461538461539e-06,
-      "loss": 0.8492,
+      "grad_norm": 3.552142858505249,
+      "learning_rate": 0.00013846153846153847,
+      "loss": 0.0779,
       "step": 470
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 2.1044576168060303,
-      "learning_rate": 5.230769230769232e-06,
-      "loss": 0.8737,
+      "grad_norm": 0.5659548044204712,
+      "learning_rate": 0.00013076923076923077,
+      "loss": 0.0464,
       "step": 480
     },
     {
       "epoch": 3.769230769230769,
-      "grad_norm": 1.4423880577087402,
-      "learning_rate": 4.923076923076924e-06,
-      "loss": 0.8443,
+      "grad_norm": 0.04047946259379387,
+      "learning_rate": 0.0001230769230769231,
+      "loss": 0.0512,
       "step": 490
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 1.2733564376831055,
-      "learning_rate": 4.615384615384616e-06,
-      "loss": 0.8557,
+      "grad_norm": 0.4618072211742401,
+      "learning_rate": 0.0001153846153846154,
+      "loss": 0.1398,
       "step": 500
     },
     {
       "epoch": 3.9230769230769234,
-      "grad_norm": 1.2562549114227295,
-      "learning_rate": 4.307692307692308e-06,
-      "loss": 0.8574,
+      "grad_norm": 0.030209071934223175,
+      "learning_rate": 0.00010769230769230771,
+      "loss": 0.113,
       "step": 510
     },
     {
       "epoch": 4.0,
-      "grad_norm": 3.079272747039795,
-      "learning_rate": 4.000000000000001e-06,
-      "loss": 0.836,
+      "grad_norm": 0.0347941592335701,
+      "learning_rate": 0.0001,
+      "loss": 0.0772,
       "step": 520
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8045112781954887,
-      "eval_loss": 0.8480122685432434,
-      "eval_runtime": 5.8399,
-      "eval_samples_per_second": 22.774,
-      "eval_steps_per_second": 2.911,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.04662612825632095,
+      "eval_runtime": 6.9929,
+      "eval_samples_per_second": 19.019,
+      "eval_steps_per_second": 2.431,
       "step": 520
     },
     {
       "epoch": 4.076923076923077,
-      "grad_norm": 2.445186138153076,
-      "learning_rate": 3.692307692307693e-06,
-      "loss": 0.8466,
+      "grad_norm": 0.3986341953277588,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0794,
       "step": 530
     },
     {
       "epoch": 4.153846153846154,
-      "grad_norm": 1.5983587503433228,
-      "learning_rate": 3.384615384615385e-06,
-      "loss": 0.8842,
+      "grad_norm": 0.538861870765686,
+      "learning_rate": 8.461538461538463e-05,
+      "loss": 0.1024,
       "step": 540
     },
     {
       "epoch": 4.230769230769231,
-      "grad_norm": 2.1697800159454346,
-      "learning_rate": 3.0769230769230774e-06,
-      "loss": 0.8777,
+      "grad_norm": 0.10060963779687881,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 0.1196,
       "step": 550
     },
     {
       "epoch": 4.3076923076923075,
-      "grad_norm": 1.2998322248458862,
-      "learning_rate": 2.7692307692307697e-06,
-      "loss": 0.8483,
+      "grad_norm": 0.4288579225540161,
+      "learning_rate": 6.923076923076924e-05,
+      "loss": 0.0233,
       "step": 560
     },
     {
       "epoch": 4.384615384615385,
-      "grad_norm": 1.4422998428344727,
-      "learning_rate": 2.461538461538462e-06,
-      "loss": 0.8451,
+      "grad_norm": 0.06229620799422264,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": 0.0887,
       "step": 570
     },
     {
       "epoch": 4.461538461538462,
-      "grad_norm": 1.6491024494171143,
-      "learning_rate": 2.153846153846154e-06,
-      "loss": 0.8165,
+      "grad_norm": 0.06911199539899826,
+      "learning_rate": 5.3846153846153853e-05,
+      "loss": 0.055,
       "step": 580
     },
     {
       "epoch": 4.538461538461538,
-      "grad_norm": 1.5623419284820557,
-      "learning_rate": 1.8461538461538465e-06,
-      "loss": 0.8344,
+      "grad_norm": 0.31500548124313354,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": 0.0385,
       "step": 590
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 1.266867756843567,
-      "learning_rate": 1.5384615384615387e-06,
-      "loss": 0.8581,
+      "grad_norm": 0.1616271734237671,
+      "learning_rate": 3.846153846153846e-05,
+      "loss": 0.1278,
       "step": 600
     },
     {
       "epoch": 4.6923076923076925,
-      "grad_norm": 2.4471514225006104,
-      "learning_rate": 1.230769230769231e-06,
-      "loss": 0.8628,
+      "grad_norm": 2.4755680561065674,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": 0.0742,
       "step": 610
     },
     {
       "epoch": 4.769230769230769,
-      "grad_norm": 1.3094156980514526,
-      "learning_rate": 9.230769230769232e-07,
-      "loss": 0.8131,
+      "grad_norm": 0.2346724271774292,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.0398,
       "step": 620
     },
     {
       "epoch": 4.846153846153846,
-      "grad_norm": 1.4745099544525146,
-      "learning_rate": 6.153846153846155e-07,
-      "loss": 0.8315,
+      "grad_norm": 0.09390363842248917,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0235,
       "step": 630
     },
     {
       "epoch": 4.923076923076923,
-      "grad_norm": 2.0481581687927246,
-      "learning_rate": 3.0769230769230774e-07,
-      "loss": 0.8188,
+      "grad_norm": 0.0435369536280632,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 0.0719,
       "step": 640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 3.112938404083252,
+      "grad_norm": 0.09755070507526398,
       "learning_rate": 0.0,
-      "loss": 0.8494,
+      "loss": 0.0941,
       "step": 650
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8045112781954887,
-      "eval_loss": 0.8347006440162659,
-      "eval_runtime": 8.0546,
-      "eval_samples_per_second": 16.512,
-      "eval_steps_per_second": 2.111,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.08044340461492538,
+      "eval_runtime": 9.1805,
+      "eval_samples_per_second": 14.487,
+      "eval_steps_per_second": 1.852,
       "step": 650
     },
     {
       "epoch": 5.0,
       "step": 650,
       "total_flos": 4.0477911662555136e+17,
-      "train_loss": 0.9380592258159931,
-      "train_runtime": 422.3657,
-      "train_samples_per_second": 12.241,
-      "train_steps_per_second": 1.539
+      "train_loss": 0.18834018239608177,
+      "train_runtime": 471.6062,
+      "train_samples_per_second": 10.963,
+      "train_steps_per_second": 1.378
     }
   ],
   "logging_steps": 10,