Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9156020131629887,
-    "eval_loss": 0.22556395828723907,
-    "eval_runtime": 22.7546,
-    "eval_samples_per_second": 113.516,
-    "eval_steps_per_second": 3.56,
-    "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.4883076953887939,
-    "train_runtime": 265.3223,
-    "train_samples_per_second": 48.677,
-    "train_steps_per_second": 0.377
+    "epoch": 9.876543209876543,
+    "eval_accuracy": 0.9721254355400697,
+    "eval_loss": 0.07933783531188965,
+    "eval_runtime": 2.8381,
+    "eval_samples_per_second": 101.125,
+    "eval_steps_per_second": 3.171,
+    "total_flos": 6.343354306682266e+17,
+    "train_loss": 0.2855896496772766,
+    "train_runtime": 464.8772,
+    "train_samples_per_second": 55.563,
+    "train_steps_per_second": 0.43
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9372822299651568,
-    "eval_loss": 0.1715514212846756,
-    "eval_runtime": 2.7579,
-    "eval_samples_per_second": 104.066,
-    "eval_steps_per_second": 3.263
+    "epoch": 9.876543209876543,
+    "eval_accuracy": 0.9721254355400697,
+    "eval_loss": 0.07933783531188965,
+    "eval_runtime": 2.8381,
+    "eval_samples_per_second": 101.125,
+    "eval_steps_per_second": 3.171
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3352dce405fa7c3c45c175e3e1842a8d57b38370adbce10cc4b4e93a0b69c011
+oid sha256:449c359d823337a57fae41d2b957610a181ccb985b72a6108f12fa811ee9fe48
 size 110348984

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9156020131629887,
-    "eval_loss": 0.22556395828723907,
-    "eval_runtime": 22.7546,
-    "eval_samples_per_second": 113.516,
-    "eval_steps_per_second": 3.56
+    "epoch": 9.876543209876543,
+    "eval_accuracy": 0.9519938056523423,
+    "eval_loss": 0.1284445822238922,
+    "eval_runtime": 22.1837,
+    "eval_samples_per_second": 116.437,
+    "eval_steps_per_second": 3.651
 }

trainer_state.json

@@ -1,160 +1,223 @@
 {
-  "best_metric": 0.9372822299651568,
-  "best_model_checkpoint": "brain-tumor-classification/checkpoint-81",
-  "epoch": 4.938271604938271,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.876543209876543,
   "eval_steps": 500,
-  "global_step": 100,
+  "global_step": 200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
-    {
-      "epoch": 0.49382716049382713,
-      "grad_norm": 11.951680183410645,
-      "learning_rate": 5e-05,
-      "loss": 1.2238,
-      "step": 10
-    },
     {
       "epoch": 0.9876543209876543,
-      "grad_norm": 11.546021461486816,
-      "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.8509,
+      "grad_norm": 8.28109359741211,
+      "learning_rate": 4.5e-05,
+      "loss": 0.8062,
       "step": 20
     },
     {
       "epoch": 0.9876543209876543,
-      "eval_accuracy": 0.8466898954703833,
-      "eval_loss": 0.5305207967758179,
-      "eval_runtime": 2.7821,
-      "eval_samples_per_second": 103.16,
-      "eval_steps_per_second": 3.235,
+      "eval_accuracy": 0.818815331010453,
+      "eval_loss": 0.4432358145713806,
+      "eval_runtime": 2.3752,
+      "eval_samples_per_second": 120.833,
+      "eval_steps_per_second": 3.789,
       "step": 20
     },
-    {
-      "epoch": 1.4814814814814814,
-      "grad_norm": 13.172475814819336,
-      "learning_rate": 3.888888888888889e-05,
-      "loss": 0.5721,
-      "step": 30
-    },
     {
       "epoch": 1.9753086419753085,
-      "grad_norm": 7.819632530212402,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.4478,
+      "grad_norm": 9.780616760253906,
+      "learning_rate": 4e-05,
+      "loss": 0.4153,
       "step": 40
     },
     {
       "epoch": 1.9753086419753085,
-      "eval_accuracy": 0.9094076655052264,
-      "eval_loss": 0.3092304468154907,
-      "eval_runtime": 2.5037,
-      "eval_samples_per_second": 114.631,
-      "eval_steps_per_second": 3.595,
+      "eval_accuracy": 0.8536585365853658,
+      "eval_loss": 0.3407208323478699,
+      "eval_runtime": 2.4373,
+      "eval_samples_per_second": 117.752,
+      "eval_steps_per_second": 3.693,
       "step": 40
     },
-    {
-      "epoch": 2.4691358024691357,
-      "grad_norm": 10.437914848327637,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.3668,
-      "step": 50
-    },
     {
       "epoch": 2.962962962962963,
-      "grad_norm": 9.35677433013916,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.3313,
+      "grad_norm": 6.261844635009766,
+      "learning_rate": 3.5e-05,
+      "loss": 0.3213,
       "step": 60
     },
     {
       "epoch": 2.962962962962963,
-      "eval_accuracy": 0.9233449477351916,
-      "eval_loss": 0.24220147728919983,
-      "eval_runtime": 2.4542,
-      "eval_samples_per_second": 116.944,
-      "eval_steps_per_second": 3.667,
+      "eval_accuracy": 0.9372822299651568,
+      "eval_loss": 0.1876000016927719,
+      "eval_runtime": 2.4811,
+      "eval_samples_per_second": 115.675,
+      "eval_steps_per_second": 3.627,
       "step": 60
     },
     {
-      "epoch": 3.45679012345679,
-      "grad_norm": 8.183242797851562,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.2979,
-      "step": 70
-    },
-    {
-      "epoch": 3.950617283950617,
-      "grad_norm": 8.208118438720703,
-      "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.2777,
-      "step": 80
+      "epoch": 4.0,
+      "grad_norm": 10.462350845336914,
+      "learning_rate": 2.975e-05,
+      "loss": 0.2633,
+      "step": 81
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9372822299651568,
-      "eval_loss": 0.1715514212846756,
-      "eval_runtime": 2.7133,
-      "eval_samples_per_second": 105.776,
-      "eval_steps_per_second": 3.317,
+      "eval_accuracy": 0.9442508710801394,
+      "eval_loss": 0.15536989271640778,
+      "eval_runtime": 2.8656,
+      "eval_samples_per_second": 100.154,
+      "eval_steps_per_second": 3.141,
       "step": 81
     },
     {
-      "epoch": 4.444444444444445,
-      "grad_norm": 8.672185897827148,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.2684,
-      "step": 90
+      "epoch": 4.987654320987654,
+      "grad_norm": 6.831620693206787,
+      "learning_rate": 2.4750000000000002e-05,
+      "loss": 0.2201,
+      "step": 101
     },
     {
-      "epoch": 4.938271604938271,
-      "grad_norm": 10.231292724609375,
-      "learning_rate": 0.0,
-      "loss": 0.2465,
-      "step": 100
+      "epoch": 4.987654320987654,
+      "eval_accuracy": 0.9547038327526133,
+      "eval_loss": 0.13280798494815826,
+      "eval_runtime": 2.4786,
+      "eval_samples_per_second": 115.792,
+      "eval_steps_per_second": 3.631,
+      "step": 101
     },
     {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9372822299651568,
-      "eval_loss": 0.16363371908664703,
-      "eval_runtime": 2.7577,
-      "eval_samples_per_second": 104.074,
-      "eval_steps_per_second": 3.264,
-      "step": 100
+      "epoch": 5.9753086419753085,
+      "grad_norm": 8.320969581604004,
+      "learning_rate": 1.9750000000000002e-05,
+      "loss": 0.2087,
+      "step": 121
     },
     {
-      "epoch": 4.938271604938271,
-      "step": 100,
-      "total_flos": 3.1727957353537536e+17,
-      "train_loss": 0.4883076953887939,
-      "train_runtime": 265.3223,
-      "train_samples_per_second": 48.677,
-      "train_steps_per_second": 0.377
+      "epoch": 5.9753086419753085,
+      "eval_accuracy": 0.9721254355400697,
+      "eval_loss": 0.08554696291685104,
+      "eval_runtime": 2.6901,
+      "eval_samples_per_second": 106.686,
+      "eval_steps_per_second": 3.346,
+      "step": 121
     },
     {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9372822299651568,
-      "eval_loss": 0.1715514212846756,
-      "eval_runtime": 2.7579,
-      "eval_samples_per_second": 104.066,
-      "eval_steps_per_second": 3.263,
-      "step": 100
-    },
-    {
-      "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9156020131629887,
-      "eval_loss": 0.22556395828723907,
-      "eval_runtime": 22.7546,
-      "eval_samples_per_second": 113.516,
-      "eval_steps_per_second": 3.56,
-      "step": 100
+      "epoch": 6.962962962962963,
+      "grad_norm": 5.462257385253906,
+      "learning_rate": 1.475e-05,
+      "loss": 0.1797,
+      "step": 141
+    },
+    {
+      "epoch": 6.962962962962963,
+      "eval_accuracy": 0.9442508710801394,
+      "eval_loss": 0.12809309363365173,
+      "eval_runtime": 2.4948,
+      "eval_samples_per_second": 115.041,
+      "eval_steps_per_second": 3.608,
+      "step": 141
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.069087982177734,
+      "learning_rate": 9.5e-06,
+      "loss": 0.1478,
+      "step": 162
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9721254355400697,
+      "eval_loss": 0.08397921919822693,
+      "eval_runtime": 2.5152,
+      "eval_samples_per_second": 114.108,
+      "eval_steps_per_second": 3.578,
+      "step": 162
+    },
+    {
+      "epoch": 8.987654320987655,
+      "grad_norm": 12.428985595703125,
+      "learning_rate": 4.5e-06,
+      "loss": 0.1545,
+      "step": 182
+    },
+    {
+      "epoch": 8.987654320987655,
+      "eval_accuracy": 0.9686411149825784,
+      "eval_loss": 0.08367497473955154,
+      "eval_runtime": 2.4429,
+      "eval_samples_per_second": 117.485,
+      "eval_steps_per_second": 3.684,
+      "step": 182
+    },
+    {
+      "epoch": 9.876543209876543,
+      "grad_norm": 2.292888641357422,
+      "learning_rate": 0.0,
+      "loss": 0.1315,
+      "step": 200
+    },
+    {
+      "epoch": 9.876543209876543,
+      "eval_accuracy": 0.9721254355400697,
+      "eval_loss": 0.07933783531188965,
+      "eval_runtime": 2.7805,
+      "eval_samples_per_second": 103.217,
+      "eval_steps_per_second": 3.237,
+      "step": 200
+    },
+    {
+      "epoch": 9.876543209876543,
+      "step": 200,
+      "total_flos": 6.343354306682266e+17,
+      "train_loss": 0.2855896496772766,
+      "train_runtime": 464.8772,
+      "train_samples_per_second": 55.563,
+      "train_steps_per_second": 0.43
+    },
+    {
+      "epoch": 9.876543209876543,
+      "eval_accuracy": 0.9562524196670538,
+      "eval_loss": 0.11971130222082138,
+      "eval_runtime": 24.8934,
+      "eval_samples_per_second": 103.763,
+      "eval_steps_per_second": 3.254,
+      "step": 200
+    },
+    {
+      "epoch": 9.876543209876543,
+      "eval_accuracy": 0.9562524196670538,
+      "eval_loss": 0.12554492056369781,
+      "eval_runtime": 27.9535,
+      "eval_samples_per_second": 92.403,
+      "eval_steps_per_second": 2.898,
+      "step": 200
+    },
+    {
+      "epoch": 9.876543209876543,
+      "eval_accuracy": 0.9721254355400697,
+      "eval_loss": 0.07933783531188965,
+      "eval_runtime": 2.5048,
+      "eval_samples_per_second": 114.582,
+      "eval_steps_per_second": 3.593,
+      "step": 200
+    },
+    {
+      "epoch": 9.876543209876543,
+      "eval_accuracy": 0.9519938056523423,
+      "eval_loss": 0.1284445822238922,
+      "eval_runtime": 22.1837,
+      "eval_samples_per_second": 116.437,
+      "eval_steps_per_second": 3.651,
+      "step": 200
     }
   ],
   "logging_steps": 10,
-  "max_steps": 100,
+  "max_steps": 200,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -168,7 +231,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3.1727957353537536e+17,
+  "total_flos": 6.343354306682266e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c2d94641c05a8dc4c54effcc8b8b87b4826b5db844cbebecb5ec09a94ca76ce
+oid sha256:07f5f871bd369d923261f3adab95a749dfd8062b1c7d23a971796b6ca6ca54c9
 size 5176