End of training

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 10.0,
-    "eval_loss": 6395.85986328125,
-    "eval_runtime": 0.6634,
-    "eval_samples_per_second": 12.059,
-    "eval_steps_per_second": 1.507,
-    "train_loss": 6197.140625,
-    "train_runtime": 176.1588,
-    "train_samples_per_second": 1.817,
-    "train_steps_per_second": 0.227
+    "epoch": 20.0,
+    "eval_loss": 6387.95458984375,
+    "eval_runtime": 0.645,
+    "eval_samples_per_second": 12.403,
+    "eval_steps_per_second": 1.55,
+    "train_loss": 6247.897314453125,
+    "train_runtime": 365.3078,
+    "train_samples_per_second": 1.752,
+    "train_steps_per_second": 0.219
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 10.0,
-    "eval_loss": 6395.85986328125,
-    "eval_runtime": 0.6634,
-    "eval_samples_per_second": 12.059,
-    "eval_steps_per_second": 1.507
+    "epoch": 20.0,
+    "eval_loss": 6387.95458984375,
+    "eval_runtime": 0.645,
+    "eval_samples_per_second": 12.403,
+    "eval_steps_per_second": 1.55
 }

runs/Jul19_10-30-23_amy-2-gpu/events.out.tfevents.1689763016.amy-2-gpu.81792.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ec38ee91181dc5c34fae95b01462de2661d022ad8ac0653383d5cfcb4da0a2a
+size 354

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 10.0,
-    "train_loss": 6197.140625,
-    "train_runtime": 176.1588,
-    "train_samples_per_second": 1.817,
-    "train_steps_per_second": 0.227
+    "epoch": 20.0,
+    "train_loss": 6247.897314453125,
+    "train_runtime": 365.3078,
+    "train_samples_per_second": 1.752,
+    "train_steps_per_second": 0.219
 }

trainer_state.json

@@ -1,129 +1,233 @@
 {
-  "best_metric": 6395.85986328125,
-  "best_model_checkpoint": "./coco_outputs/checkpoint-40",
-  "epoch": 10.0,
-  "global_step": 40,
+  "best_metric": 6387.95458984375,
+  "best_model_checkpoint": "./coco_outputs/checkpoint-80",
+  "epoch": 20.0,
+  "global_step": 80,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_loss": 6398.54345703125,
-      "eval_runtime": 0.5491,
-      "eval_samples_per_second": 14.569,
-      "eval_steps_per_second": 1.821,
+      "eval_loss": 6394.3193359375,
+      "eval_runtime": 0.6785,
+      "eval_samples_per_second": 11.791,
+      "eval_steps_per_second": 1.474,
       "step": 4
     },
     {
       "epoch": 2.0,
-      "eval_loss": 6397.9765625,
-      "eval_runtime": 0.5423,
-      "eval_samples_per_second": 14.751,
-      "eval_steps_per_second": 1.844,
+      "eval_loss": 6393.3564453125,
+      "eval_runtime": 0.625,
+      "eval_samples_per_second": 12.799,
+      "eval_steps_per_second": 1.6,
       "step": 8
     },
     {
       "epoch": 2.5,
-      "learning_rate": 1.5000000000000002e-05,
-      "loss": 6175.8828,
+      "learning_rate": 0.000175,
+      "loss": 6173.6062,
       "step": 10
     },
     {
       "epoch": 3.0,
-      "eval_loss": 6397.42529296875,
-      "eval_runtime": 0.5517,
-      "eval_samples_per_second": 14.501,
-      "eval_steps_per_second": 1.813,
+      "eval_loss": 6390.9482421875,
+      "eval_runtime": 0.6116,
+      "eval_samples_per_second": 13.08,
+      "eval_steps_per_second": 1.635,
       "step": 12
     },
     {
       "epoch": 4.0,
-      "eval_loss": 6396.962890625,
-      "eval_runtime": 0.555,
-      "eval_samples_per_second": 14.415,
-      "eval_steps_per_second": 1.802,
+      "eval_loss": 6389.5869140625,
+      "eval_runtime": 0.5503,
+      "eval_samples_per_second": 14.538,
+      "eval_steps_per_second": 1.817,
       "step": 16
     },
     {
       "epoch": 5.0,
-      "learning_rate": 1e-05,
-      "loss": 6292.3258,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 6287.2859,
       "step": 20
     },
     {
       "epoch": 5.0,
-      "eval_loss": 6396.57177734375,
-      "eval_runtime": 0.5448,
-      "eval_samples_per_second": 14.685,
-      "eval_steps_per_second": 1.836,
+      "eval_loss": 6388.72802734375,
+      "eval_runtime": 0.5904,
+      "eval_samples_per_second": 13.55,
+      "eval_steps_per_second": 1.694,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_loss": 6396.27880859375,
-      "eval_runtime": 0.5466,
-      "eval_samples_per_second": 14.636,
-      "eval_steps_per_second": 1.829,
+      "eval_loss": 6388.6220703125,
+      "eval_runtime": 0.5392,
+      "eval_samples_per_second": 14.837,
+      "eval_steps_per_second": 1.855,
       "step": 24
     },
     {
       "epoch": 7.0,
-      "eval_loss": 6396.07666015625,
-      "eval_runtime": 0.5439,
-      "eval_samples_per_second": 14.708,
-      "eval_steps_per_second": 1.839,
+      "eval_loss": 6388.43994140625,
+      "eval_runtime": 0.5365,
+      "eval_samples_per_second": 14.91,
+      "eval_steps_per_second": 1.864,
       "step": 28
     },
     {
       "epoch": 7.5,
-      "learning_rate": 5e-06,
-      "loss": 6230.4844,
+      "learning_rate": 0.000125,
+      "loss": 6224.3344,
       "step": 30
     },
     {
       "epoch": 8.0,
-      "eval_loss": 6395.9599609375,
-      "eval_runtime": 0.5536,
-      "eval_samples_per_second": 14.451,
-      "eval_steps_per_second": 1.806,
+      "eval_loss": 6388.2119140625,
+      "eval_runtime": 0.5405,
+      "eval_samples_per_second": 14.802,
+      "eval_steps_per_second": 1.85,
       "step": 32
     },
     {
       "epoch": 9.0,
-      "eval_loss": 6395.8671875,
-      "eval_runtime": 0.5708,
-      "eval_samples_per_second": 14.015,
-      "eval_steps_per_second": 1.752,
+      "eval_loss": 6388.544921875,
+      "eval_runtime": 0.5344,
+      "eval_samples_per_second": 14.971,
+      "eval_steps_per_second": 1.871,
       "step": 36
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.0,
-      "loss": 6089.8695,
+      "learning_rate": 0.0001,
+      "loss": 6083.1363,
       "step": 40
     },
     {
       "epoch": 10.0,
-      "eval_loss": 6395.85986328125,
-      "eval_runtime": 0.5496,
-      "eval_samples_per_second": 14.557,
-      "eval_steps_per_second": 1.82,
+      "eval_loss": 6388.326171875,
+      "eval_runtime": 0.5236,
+      "eval_samples_per_second": 15.28,
+      "eval_steps_per_second": 1.91,
       "step": 40
     },
     {
-      "epoch": 10.0,
-      "step": 40,
-      "total_flos": 1.530004230144e+17,
-      "train_loss": 6197.140625,
-      "train_runtime": 176.1588,
-      "train_samples_per_second": 1.817,
-      "train_steps_per_second": 0.227
+      "epoch": 11.0,
+      "eval_loss": 6390.20556640625,
+      "eval_runtime": 0.5417,
+      "eval_samples_per_second": 14.767,
+      "eval_steps_per_second": 1.846,
+      "step": 44
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 6388.66064453125,
+      "eval_runtime": 0.5437,
+      "eval_samples_per_second": 14.713,
+      "eval_steps_per_second": 1.839,
+      "step": 48
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 6260.7426,
+      "step": 50
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 6388.1484375,
+      "eval_runtime": 0.5303,
+      "eval_samples_per_second": 15.086,
+      "eval_steps_per_second": 1.886,
+      "step": 52
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 6388.0478515625,
+      "eval_runtime": 0.5365,
+      "eval_samples_per_second": 14.913,
+      "eval_steps_per_second": 1.864,
+      "step": 56
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 5e-05,
+      "loss": 6377.543,
+      "step": 60
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 6388.0302734375,
+      "eval_runtime": 0.5355,
+      "eval_samples_per_second": 14.94,
+      "eval_steps_per_second": 1.867,
+      "step": 60
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 6388.1337890625,
+      "eval_runtime": 0.5423,
+      "eval_samples_per_second": 14.751,
+      "eval_steps_per_second": 1.844,
+      "step": 64
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 6388.0595703125,
+      "eval_runtime": 0.5448,
+      "eval_samples_per_second": 14.684,
+      "eval_steps_per_second": 1.835,
+      "step": 68
+    },
+    {
+      "epoch": 17.5,
+      "learning_rate": 2.5e-05,
+      "loss": 6318.3168,
+      "step": 70
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 6388.01953125,
+      "eval_runtime": 0.5448,
+      "eval_samples_per_second": 14.685,
+      "eval_steps_per_second": 1.836,
+      "step": 72
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 6387.9638671875,
+      "eval_runtime": 0.5362,
+      "eval_samples_per_second": 14.918,
+      "eval_steps_per_second": 1.865,
+      "step": 76
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0,
+      "loss": 6258.2133,
+      "step": 80
+    },
+    {
+      "epoch": 20.0,
+      "eval_loss": 6387.95458984375,
+      "eval_runtime": 0.544,
+      "eval_samples_per_second": 14.705,
+      "eval_steps_per_second": 1.838,
+      "step": 80
+    },
+    {
+      "epoch": 20.0,
+      "step": 80,
+      "total_flos": 3.060008460288e+17,
+      "train_loss": 6247.897314453125,
+      "train_runtime": 365.3078,
+      "train_samples_per_second": 1.752,
+      "train_steps_per_second": 0.219
     }
   ],
-  "max_steps": 40,
-  "num_train_epochs": 10,
-  "total_flos": 1.530004230144e+17,
+  "max_steps": 80,
+  "num_train_epochs": 20,
+  "total_flos": 3.060008460288e+17,
   "trial_name": null,
   "trial_params": null
 }