Training in progress, epoch 0

all_results.json

@@ -1,13 +1,8 @@
 {
     "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9368950832365467,
-    "eval_loss": 0.16935844719409943,
-    "eval_runtime": 25.4894,
-    "eval_samples_per_second": 101.336,
-    "eval_steps_per_second": 6.356,
     "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.4196511161327362,
-    "train_runtime": 246.9642,
-    "train_samples_per_second": 52.295,
-    "train_steps_per_second": 0.81
+    "train_loss": 0.39835033178329465,
+    "train_runtime": 251.5945,
+    "train_samples_per_second": 51.333,
+    "train_steps_per_second": 0.795
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e53b624541f7868df76c5b2cc852ac8b512907c1f715fcb4b52ad96d5b7bbc11
+oid sha256:d510575276b2c476fb72573199b83878061dfb3b0ae921867444fb78a92d0838
 size 110348984

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9368950832365467,
-    "eval_loss": 0.16935844719409943,
-    "eval_runtime": 25.4894,
-    "eval_samples_per_second": 101.336,
-    "eval_steps_per_second": 6.356
+    "total_flos": 3.1727957353537536e+17,
+    "train_loss": 0.39835033178329465,
+    "train_runtime": 251.5945,
+    "train_samples_per_second": 51.333,
+    "train_steps_per_second": 0.795
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9547038327526133,
+  "best_metric": 0.9442508710801394,
   "best_model_checkpoint": "brain-tumor-classification/checkpoint-200",
   "epoch": 4.938271604938271,
   "eval_steps": 500,
@@ -8,226 +8,94 @@
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 0.24691358024691357,
-      "grad_norm": 8.300398826599121,
-      "learning_rate": 2.5e-05,
-      "loss": 1.3599,
-      "step": 10
-    },
-    {
-      "epoch": 0.49382716049382713,
-      "grad_norm": 11.234930992126465,
-      "learning_rate": 5e-05,
-      "loss": 1.0545,
-      "step": 20
-    },
-    {
-      "epoch": 0.7407407407407407,
-      "grad_norm": 12.957039833068848,
-      "learning_rate": 4.722222222222222e-05,
-      "loss": 0.7446,
-      "step": 30
-    },
     {
       "epoch": 0.9876543209876543,
-      "grad_norm": 13.77059268951416,
+      "grad_norm": 26.882118225097656,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.6001,
+      "loss": 0.87,
       "step": 40
     },
     {
       "epoch": 0.9876543209876543,
-      "eval_accuracy": 0.8466898954703833,
-      "eval_loss": 0.4058869779109955,
-      "eval_runtime": 2.3913,
-      "eval_samples_per_second": 120.016,
-      "eval_steps_per_second": 7.527,
+      "eval_accuracy": 0.8710801393728222,
+      "eval_loss": 0.3874738812446594,
+      "eval_runtime": 4.2257,
+      "eval_samples_per_second": 67.917,
+      "eval_steps_per_second": 4.26,
       "step": 40
     },
-    {
-      "epoch": 1.2345679012345678,
-      "grad_norm": 12.874049186706543,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.4598,
-      "step": 50
-    },
-    {
-      "epoch": 1.4814814814814814,
-      "grad_norm": 15.899703979492188,
-      "learning_rate": 3.888888888888889e-05,
-      "loss": 0.4362,
-      "step": 60
-    },
-    {
-      "epoch": 1.7283950617283952,
-      "grad_norm": 14.211153030395508,
-      "learning_rate": 3.611111111111111e-05,
-      "loss": 0.4013,
-      "step": 70
-    },
-    {
-      "epoch": 1.9753086419753085,
-      "grad_norm": 14.76016902923584,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.3615,
-      "step": 80
-    },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9128919860627178,
-      "eval_loss": 0.2405817061662674,
-      "eval_runtime": 4.5171,
-      "eval_samples_per_second": 63.537,
-      "eval_steps_per_second": 3.985,
+      "grad_norm": 6.630888938903809,
+      "learning_rate": 3.3055555555555553e-05,
+      "loss": 0.426,
       "step": 81
     },
     {
-      "epoch": 2.2222222222222223,
-      "grad_norm": 10.028989791870117,
-      "learning_rate": 3.055555555555556e-05,
-      "loss": 0.3202,
-      "step": 90
-    },
-    {
-      "epoch": 2.4691358024691357,
-      "grad_norm": 8.175374984741211,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.2769,
-      "step": 100
-    },
-    {
-      "epoch": 2.7160493827160495,
-      "grad_norm": 7.94071102142334,
-      "learning_rate": 2.5e-05,
-      "loss": 0.2848,
-      "step": 110
-    },
-    {
-      "epoch": 2.962962962962963,
-      "grad_norm": 14.77879524230957,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.2619,
-      "step": 120
+      "epoch": 2.0,
+      "eval_accuracy": 0.9024390243902439,
+      "eval_loss": 0.22352387011051178,
+      "eval_runtime": 2.3991,
+      "eval_samples_per_second": 119.628,
+      "eval_steps_per_second": 7.503,
+      "step": 81
     },
     {
       "epoch": 2.9876543209876543,
-      "eval_accuracy": 0.9372822299651568,
-      "eval_loss": 0.18160930275917053,
-      "eval_runtime": 2.3852,
-      "eval_samples_per_second": 120.323,
-      "eval_steps_per_second": 7.546,
+      "grad_norm": 9.56891918182373,
+      "learning_rate": 2.1944444444444445e-05,
+      "loss": 0.292,
       "step": 121
     },
     {
-      "epoch": 3.2098765432098766,
-      "grad_norm": 5.213917255401611,
-      "learning_rate": 1.9444444444444445e-05,
-      "loss": 0.2809,
-      "step": 130
-    },
-    {
-      "epoch": 3.45679012345679,
-      "grad_norm": 9.913806915283203,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.2239,
-      "step": 140
-    },
-    {
-      "epoch": 3.7037037037037037,
-      "grad_norm": 9.552595138549805,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.2331,
-      "step": 150
-    },
-    {
-      "epoch": 3.950617283950617,
-      "grad_norm": 13.436369895935059,
-      "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.2262,
-      "step": 160
+      "epoch": 2.9876543209876543,
+      "eval_accuracy": 0.926829268292683,
+      "eval_loss": 0.2056380957365036,
+      "eval_runtime": 2.3874,
+      "eval_samples_per_second": 120.214,
+      "eval_steps_per_second": 7.54,
+      "step": 121
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9442508710801394,
-      "eval_loss": 0.14013153314590454,
-      "eval_runtime": 2.3881,
-      "eval_samples_per_second": 120.178,
-      "eval_steps_per_second": 7.537,
+      "grad_norm": 7.695461273193359,
+      "learning_rate": 1.0555555555555555e-05,
+      "loss": 0.217,
       "step": 162
     },
     {
-      "epoch": 4.197530864197531,
-      "grad_norm": 5.0693039894104,
-      "learning_rate": 8.333333333333334e-06,
-      "loss": 0.2334,
-      "step": 170
-    },
-    {
-      "epoch": 4.444444444444445,
-      "grad_norm": 9.17518424987793,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.2374,
-      "step": 180
-    },
-    {
-      "epoch": 4.6913580246913575,
-      "grad_norm": 7.441468715667725,
-      "learning_rate": 2.777777777777778e-06,
-      "loss": 0.2173,
-      "step": 190
+      "epoch": 4.0,
+      "eval_accuracy": 0.9337979094076655,
+      "eval_loss": 0.1578059196472168,
+      "eval_runtime": 2.6136,
+      "eval_samples_per_second": 109.811,
+      "eval_steps_per_second": 6.887,
+      "step": 162
     },
     {
       "epoch": 4.938271604938271,
-      "grad_norm": 3.2817890644073486,
+      "grad_norm": 7.951013088226318,
       "learning_rate": 0.0,
-      "loss": 0.179,
+      "loss": 0.1797,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9547038327526133,
-      "eval_loss": 0.12142720818519592,
-      "eval_runtime": 3.0543,
-      "eval_samples_per_second": 93.966,
-      "eval_steps_per_second": 5.893,
+      "eval_accuracy": 0.9442508710801394,
+      "eval_loss": 0.12732850015163422,
+      "eval_runtime": 2.9064,
+      "eval_samples_per_second": 98.748,
+      "eval_steps_per_second": 6.193,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
       "step": 200,
       "total_flos": 3.1727957353537536e+17,
-      "train_loss": 0.4196511161327362,
-      "train_runtime": 246.9642,
-      "train_samples_per_second": 52.295,
-      "train_steps_per_second": 0.81
-    },
-    {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9547038327526133,
-      "eval_loss": 0.12142720818519592,
-      "eval_runtime": 2.5284,
-      "eval_samples_per_second": 113.511,
-      "eval_steps_per_second": 7.119,
-      "step": 200
-    },
-    {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9407665505226481,
-      "eval_loss": 0.16431476175785065,
-      "eval_runtime": 26.8879,
-      "eval_samples_per_second": 96.066,
-      "eval_steps_per_second": 6.025,
-      "step": 200
-    },
-    {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9368950832365467,
-      "eval_loss": 0.16935844719409943,
-      "eval_runtime": 25.4894,
-      "eval_samples_per_second": 101.336,
-      "eval_steps_per_second": 6.356,
-      "step": 200
+      "train_loss": 0.39835033178329465,
+      "train_runtime": 251.5945,
+      "train_samples_per_second": 51.333,
+      "train_steps_per_second": 0.795
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42348b886db9325751586ea83556554c60b6c99df997bad6c5e12f0fd8faf6a9
+oid sha256:af61a708b20246126c56c562b0ad58295f13221d185c96d039aa4cea207f9cbb
 size 5176