End of training

all_results.json

@@ -1,8 +1,13 @@
 {
-    "epoch": 2.6666666666666665,
-    "total_flos": 2.326039595266867e+16,
-    "train_loss": 3.1840518987395357e+30,
-    "train_runtime": 15.2988,
-    "train_samples_per_second": 169.425,
-    "train_steps_per_second": 1.177
+    "epoch": 2.7272727272727275,
+    "eval_accuracy": 0.2857142857142857,
+    "eval_loss": 1.142593873281104e+31,
+    "eval_runtime": 0.2619,
+    "eval_samples_per_second": 293.955,
+    "eval_steps_per_second": 11.453,
+    "total_flos": 1.912117618678579e+16,
+    "train_loss": 1.0846118163935475e+31,
+    "train_runtime": 12.9408,
+    "train_samples_per_second": 160.191,
+    "train_steps_per_second": 1.159
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.090909090909092,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.8444482088088989,
-    "eval_runtime": 0.2618,
-    "eval_samples_per_second": 294.076,
-    "eval_steps_per_second": 11.458
+    "epoch": 2.7272727272727275,
+    "eval_accuracy": 0.2857142857142857,
+    "eval_loss": 1.142593873281104e+31,
+    "eval_runtime": 0.2619,
+    "eval_samples_per_second": 293.955,
+    "eval_steps_per_second": 11.453
 }

runs/Jun06_03-58-56_b446df81307f/events.out.tfevents.1717646359.b446df81307f.157.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d52491658dbabc4595884f12a1c611110549195df62ca7a4d4a8f614a37552d2
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.6666666666666665,
-    "total_flos": 2.326039595266867e+16,
-    "train_loss": 3.1840518987395357e+30,
-    "train_runtime": 15.2988,
-    "train_samples_per_second": 169.425,
-    "train_steps_per_second": 1.177
+    "epoch": 2.7272727272727275,
+    "total_flos": 1.912117618678579e+16,
+    "train_loss": 1.0846118163935475e+31,
+    "train_runtime": 12.9408,
+    "train_samples_per_second": 160.191,
+    "train_steps_per_second": 1.159
 }

trainer_state.json

@@ -1,59 +1,59 @@
 {
-  "best_metric": 0.3958333333333333,
-  "best_model_checkpoint": "resnet-18/checkpoint-6",
-  "epoch": 2.6666666666666665,
+  "best_metric": 0.2857142857142857,
+  "best_model_checkpoint": "resnet-18/checkpoint-5",
+  "epoch": 2.7272727272727275,
   "eval_steps": 500,
-  "global_step": 18,
+  "global_step": 15,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.8888888888888888,
-      "eval_accuracy": 0.3958333333333333,
-      "eval_loss": 2.8539976205603575e+30,
-      "eval_runtime": 0.2601,
-      "eval_samples_per_second": 369.032,
-      "eval_steps_per_second": 11.532,
-      "step": 6
+      "epoch": 0.9090909090909091,
+      "eval_accuracy": 0.2857142857142857,
+      "eval_loss": 1.142593873281104e+31,
+      "eval_runtime": 0.3831,
+      "eval_samples_per_second": 200.981,
+      "eval_steps_per_second": 7.83,
+      "step": 5
     },
     {
-      "epoch": 1.4814814814814814,
-      "grad_norm": 25.099891662597656,
-      "learning_rate": 2.5e-05,
-      "loss": 3.195940475249626e+30,
+      "epoch": 1.8181818181818183,
+      "grad_norm": 19.20919418334961,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": 1.08768806515315e+31,
       "step": 10
     },
     {
-      "epoch": 1.925925925925926,
-      "eval_accuracy": 0.3958333333333333,
-      "eval_loss": 2.8539976205603575e+30,
-      "eval_runtime": 0.2646,
-      "eval_samples_per_second": 362.761,
-      "eval_steps_per_second": 11.336,
-      "step": 13
+      "epoch": 2.0,
+      "eval_accuracy": 0.2857142857142857,
+      "eval_loss": 1.142593873281104e+31,
+      "eval_runtime": 0.2161,
+      "eval_samples_per_second": 356.345,
+      "eval_steps_per_second": 13.884,
+      "step": 11
     },
     {
-      "epoch": 2.6666666666666665,
-      "eval_accuracy": 0.3958333333333333,
-      "eval_loss": 2.8539976205603575e+30,
-      "eval_runtime": 0.2658,
-      "eval_samples_per_second": 361.215,
-      "eval_steps_per_second": 11.288,
-      "step": 18
+      "epoch": 2.7272727272727275,
+      "eval_accuracy": 0.2857142857142857,
+      "eval_loss": 1.142593873281104e+31,
+      "eval_runtime": 0.2381,
+      "eval_samples_per_second": 323.413,
+      "eval_steps_per_second": 12.601,
+      "step": 15
     },
     {
-      "epoch": 2.6666666666666665,
-      "step": 18,
-      "total_flos": 2.326039595266867e+16,
-      "train_loss": 3.1840518987395357e+30,
-      "train_runtime": 15.2988,
-      "train_samples_per_second": 169.425,
-      "train_steps_per_second": 1.177
+      "epoch": 2.7272727272727275,
+      "step": 15,
+      "total_flos": 1.912117618678579e+16,
+      "train_loss": 1.0846118163935475e+31,
+      "train_runtime": 12.9408,
+      "train_samples_per_second": 160.191,
+      "train_steps_per_second": 1.159
     }
   ],
   "logging_steps": 10,
-  "max_steps": 18,
+  "max_steps": 15,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -69,7 +69,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.326039595266867e+16,
+  "total_flos": 1.912117618678579e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null