Training in progress, step 500

run-10/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176

run-10/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176

run-10/checkpoint-2000/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.4022,
-      "eval_samples_per_second": 2210.798,
-      "eval_steps_per_second": 46.355,
+      "eval_runtime": 1.2498,
+      "eval_samples_per_second": 2480.484,
+      "eval_steps_per_second": 52.01,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8451612903225807,
       "eval_loss": 0.13748767971992493,
-      "eval_runtime": 1.4384,
-      "eval_samples_per_second": 2155.104,
-      "eval_steps_per_second": 45.188,
+      "eval_runtime": 1.253,
+      "eval_samples_per_second": 2473.996,
+      "eval_steps_per_second": 51.874,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8974193548387097,
       "eval_loss": 0.07043754309415817,
-      "eval_runtime": 1.3662,
-      "eval_samples_per_second": 2269.06,
-      "eval_steps_per_second": 47.577,
+      "eval_runtime": 1.2561,
+      "eval_samples_per_second": 2467.887,
+      "eval_steps_per_second": 51.746,
       "step": 954
     },
     {
@@ -53,9 +53,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.9145161290322581,
       "eval_loss": 0.05076972395181656,
-      "eval_runtime": 1.3887,
-      "eval_samples_per_second": 2232.374,
-      "eval_steps_per_second": 46.808,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.139,
+      "eval_steps_per_second": 51.709,
       "step": 1272
     },
     {
@@ -69,18 +69,18 @@
       "epoch": 5.0,
       "eval_accuracy": 0.9241935483870968,
       "eval_loss": 0.04163844883441925,
-      "eval_runtime": 1.4597,
-      "eval_samples_per_second": 2123.769,
-      "eval_steps_per_second": 44.531,
+      "eval_runtime": 1.2556,
+      "eval_samples_per_second": 2468.853,
+      "eval_steps_per_second": 51.766,
       "step": 1590
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9283870967741935,
       "eval_loss": 0.037362001836299896,
-      "eval_runtime": 1.3818,
-      "eval_samples_per_second": 2243.399,
-      "eval_steps_per_second": 47.039,
+      "eval_runtime": 1.2547,
+      "eval_samples_per_second": 2470.711,
+      "eval_steps_per_second": 51.805,
       "step": 1908
     },
     {
@@ -108,11 +108,11 @@
       "attributes": {}
     }
   },
-  "total_flos": 722985755140992.0,
+  "total_flos": 603995663965932.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7807938087018197,
+    "alpha": 0.9803719554379057,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-10/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176

run-10/checkpoint-2500/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.4022,
-      "eval_samples_per_second": 2210.798,
-      "eval_steps_per_second": 46.355,
+      "eval_runtime": 1.2498,
+      "eval_samples_per_second": 2480.484,
+      "eval_steps_per_second": 52.01,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8451612903225807,
       "eval_loss": 0.13748767971992493,
-      "eval_runtime": 1.4384,
-      "eval_samples_per_second": 2155.104,
-      "eval_steps_per_second": 45.188,
+      "eval_runtime": 1.253,
+      "eval_samples_per_second": 2473.996,
+      "eval_steps_per_second": 51.874,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8974193548387097,
       "eval_loss": 0.07043754309415817,
-      "eval_runtime": 1.3662,
-      "eval_samples_per_second": 2269.06,
-      "eval_steps_per_second": 47.577,
+      "eval_runtime": 1.2561,
+      "eval_samples_per_second": 2467.887,
+      "eval_steps_per_second": 51.746,
       "step": 954
     },
     {
@@ -53,9 +53,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.9145161290322581,
       "eval_loss": 0.05076972395181656,
-      "eval_runtime": 1.3887,
-      "eval_samples_per_second": 2232.374,
-      "eval_steps_per_second": 46.808,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.139,
+      "eval_steps_per_second": 51.709,
       "step": 1272
     },
     {
@@ -69,18 +69,18 @@
       "epoch": 5.0,
       "eval_accuracy": 0.9241935483870968,
       "eval_loss": 0.04163844883441925,
-      "eval_runtime": 1.4597,
-      "eval_samples_per_second": 2123.769,
-      "eval_steps_per_second": 44.531,
+      "eval_runtime": 1.2556,
+      "eval_samples_per_second": 2468.853,
+      "eval_steps_per_second": 51.766,
       "step": 1590
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9283870967741935,
       "eval_loss": 0.037362001836299896,
-      "eval_runtime": 1.3818,
-      "eval_samples_per_second": 2243.399,
-      "eval_steps_per_second": 47.039,
+      "eval_runtime": 1.2547,
+      "eval_samples_per_second": 2470.711,
+      "eval_steps_per_second": 51.805,
       "step": 1908
     },
     {
@@ -94,9 +94,9 @@
       "epoch": 7.0,
       "eval_accuracy": 0.9319354838709677,
       "eval_loss": 0.03430590778589249,
-      "eval_runtime": 1.3689,
-      "eval_samples_per_second": 2264.53,
-      "eval_steps_per_second": 47.482,
+      "eval_runtime": 1.2578,
+      "eval_samples_per_second": 2464.687,
+      "eval_steps_per_second": 51.679,
       "step": 2226
     },
     {
@@ -124,11 +124,11 @@
       "attributes": {}
     }
   },
-  "total_flos": 853150315276476.0,
+  "total_flos": 734160224101416.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7807938087018197,
+    "alpha": 0.9803719554379057,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-10/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176

run-10/checkpoint-3000/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.4022,
-      "eval_samples_per_second": 2210.798,
-      "eval_steps_per_second": 46.355,
+      "eval_runtime": 1.2498,
+      "eval_samples_per_second": 2480.484,
+      "eval_steps_per_second": 52.01,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8451612903225807,
       "eval_loss": 0.13748767971992493,
-      "eval_runtime": 1.4384,
-      "eval_samples_per_second": 2155.104,
-      "eval_steps_per_second": 45.188,
+      "eval_runtime": 1.253,
+      "eval_samples_per_second": 2473.996,
+      "eval_steps_per_second": 51.874,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8974193548387097,
       "eval_loss": 0.07043754309415817,
-      "eval_runtime": 1.3662,
-      "eval_samples_per_second": 2269.06,
-      "eval_steps_per_second": 47.577,
+      "eval_runtime": 1.2561,
+      "eval_samples_per_second": 2467.887,
+      "eval_steps_per_second": 51.746,
       "step": 954
     },
     {
@@ -53,9 +53,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.9145161290322581,
       "eval_loss": 0.05076972395181656,
-      "eval_runtime": 1.3887,
-      "eval_samples_per_second": 2232.374,
-      "eval_steps_per_second": 46.808,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.139,
+      "eval_steps_per_second": 51.709,
       "step": 1272
     },
     {
@@ -69,18 +69,18 @@
       "epoch": 5.0,
       "eval_accuracy": 0.9241935483870968,
       "eval_loss": 0.04163844883441925,
-      "eval_runtime": 1.4597,
-      "eval_samples_per_second": 2123.769,
-      "eval_steps_per_second": 44.531,
+      "eval_runtime": 1.2556,
+      "eval_samples_per_second": 2468.853,
+      "eval_steps_per_second": 51.766,
       "step": 1590
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9283870967741935,
       "eval_loss": 0.037362001836299896,
-      "eval_runtime": 1.3818,
-      "eval_samples_per_second": 2243.399,
-      "eval_steps_per_second": 47.039,
+      "eval_runtime": 1.2547,
+      "eval_samples_per_second": 2470.711,
+      "eval_steps_per_second": 51.805,
       "step": 1908
     },
     {
@@ -94,9 +94,9 @@
       "epoch": 7.0,
       "eval_accuracy": 0.9319354838709677,
       "eval_loss": 0.03430590778589249,
-      "eval_runtime": 1.3689,
-      "eval_samples_per_second": 2264.53,
-      "eval_steps_per_second": 47.482,
+      "eval_runtime": 1.2578,
+      "eval_samples_per_second": 2464.687,
+      "eval_steps_per_second": 51.679,
       "step": 2226
     },
     {
@@ -110,18 +110,18 @@
       "epoch": 8.0,
       "eval_accuracy": 0.9332258064516129,
       "eval_loss": 0.032527994364500046,
-      "eval_runtime": 1.361,
-      "eval_samples_per_second": 2277.734,
-      "eval_steps_per_second": 47.759,
+      "eval_runtime": 1.2604,
+      "eval_samples_per_second": 2459.579,
+      "eval_steps_per_second": 51.572,
       "step": 2544
     },
     {
       "epoch": 9.0,
       "eval_accuracy": 0.9325806451612904,
       "eval_loss": 0.03162752836942673,
-      "eval_runtime": 1.3732,
-      "eval_samples_per_second": 2257.525,
-      "eval_steps_per_second": 47.335,
+      "eval_runtime": 1.2585,
+      "eval_samples_per_second": 2463.197,
+      "eval_steps_per_second": 51.648,
       "step": 2862
     },
     {
@@ -149,11 +149,11 @@
       "attributes": {}
     }
   },
-  "total_flos": 982733271748452.0,
+  "total_flos": 863743180573392.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7807938087018197,
+    "alpha": 0.9803719554379057,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-10/checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176

run-11/checkpoint-1000/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.3612,
-      "eval_samples_per_second": 2277.395,
-      "eval_steps_per_second": 47.752,
+      "eval_runtime": 1.2636,
+      "eval_samples_per_second": 2453.311,
+      "eval_steps_per_second": 51.44,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8451612903225807,
       "eval_loss": 0.13748767971992493,
-      "eval_runtime": 1.356,
-      "eval_samples_per_second": 2286.155,
-      "eval_steps_per_second": 47.936,
+      "eval_runtime": 1.2607,
+      "eval_samples_per_second": 2458.94,
+      "eval_steps_per_second": 51.558,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8974193548387097,
       "eval_loss": 0.07043754309415817,
-      "eval_runtime": 1.3531,
-      "eval_samples_per_second": 2291.022,
-      "eval_steps_per_second": 48.038,
+      "eval_runtime": 1.2596,
+      "eval_samples_per_second": 2461.115,
+      "eval_steps_per_second": 51.604,
       "step": 954
     },
     {
@@ -71,7 +71,7 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8252766996315095,
+    "alpha": 0.9689147754277356,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-11/checkpoint-1500/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.3612,
-      "eval_samples_per_second": 2277.395,
-      "eval_steps_per_second": 47.752,
+      "eval_runtime": 1.2636,
+      "eval_samples_per_second": 2453.311,
+      "eval_steps_per_second": 51.44,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8451612903225807,
       "eval_loss": 0.13748767971992493,
-      "eval_runtime": 1.356,
-      "eval_samples_per_second": 2286.155,
-      "eval_steps_per_second": 47.936,
+      "eval_runtime": 1.2607,
+      "eval_samples_per_second": 2458.94,
+      "eval_steps_per_second": 51.558,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8974193548387097,
       "eval_loss": 0.07043754309415817,
-      "eval_runtime": 1.3531,
-      "eval_samples_per_second": 2291.022,
-      "eval_steps_per_second": 48.038,
+      "eval_runtime": 1.2596,
+      "eval_samples_per_second": 2461.115,
+      "eval_steps_per_second": 51.604,
       "step": 954
     },
     {
@@ -53,9 +53,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.9145161290322581,
       "eval_loss": 0.05076972395181656,
-      "eval_runtime": 1.3624,
-      "eval_samples_per_second": 2275.392,
-      "eval_steps_per_second": 47.71,
+      "eval_runtime": 1.3318,
+      "eval_samples_per_second": 2327.592,
+      "eval_steps_per_second": 48.804,
       "step": 1272
     },
     {
@@ -87,7 +87,7 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8252766996315095,
+    "alpha": 0.9689147754277356,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-11/checkpoint-500/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.6741935483870968,
       "eval_loss": 0.4054252803325653,
-      "eval_runtime": 1.3612,
-      "eval_samples_per_second": 2277.395,
-      "eval_steps_per_second": 47.752,
+      "eval_runtime": 1.2636,
+      "eval_samples_per_second": 2453.311,
+      "eval_steps_per_second": 51.44,
       "step": 318
     },
     {
@@ -46,7 +46,7 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.8252766996315095,
+    "alpha": 0.9689147754277356,
     "num_train_epochs": 10,
     "temperature": 2
   }

run-11/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53b16b25375514918da67404b3403c850c15a69162ae13812852117b14e664d6
+oid sha256:0c3d81cf1e94136cf69f7afe09173f6fff54ee913216fd1109c21e652a8bd239
 size 5176