Training in progress, step 3180

run-11/checkpoint-3180/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc0c49b035b446b79ee3aa6bfc6a9a9ca4fec5d311003789d4d3f877f5751c24
+oid sha256:5c11c8d8fd5dab07dc0da9a774d4ac9892b28391145c860a4321a6d78b2468e5
 size 268290900

run-11/checkpoint-3180/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83a80751ef055cd23819262d089197b80b233164256279dc2d3939f172fe5f04
+oid sha256:bf56a6a19086ea10089ac80d4b475b67fd4477ea849dc7ec47b9c3add6cf41d2
 size 536643898

run-11/checkpoint-3180/trainer_state.json

@@ -10,153 +10,153 @@
   "log_history": [
     {
       "epoch": 0.9968553459119497,
-      "grad_norm": 0.6589117050170898,
+      "grad_norm": 0.9062672853469849,
       "learning_rate": 1.8006289308176103e-05,
-      "loss": 0.4371,
+      "loss": 0.8042,
       "step": 317
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6058064516129033,
-      "eval_loss": 0.21883252263069153,
-      "eval_runtime": 1.4517,
-      "eval_samples_per_second": 2135.493,
-      "eval_steps_per_second": 44.776,
+      "eval_accuracy": 0.6709677419354839,
+      "eval_loss": 0.4064599573612213,
+      "eval_runtime": 1.3737,
+      "eval_samples_per_second": 2256.722,
+      "eval_steps_per_second": 47.318,
       "step": 318
     },
     {
       "epoch": 1.9937106918238994,
-      "grad_norm": 0.548164427280426,
+      "grad_norm": 0.7205497026443481,
       "learning_rate": 1.6012578616352204e-05,
-      "loss": 0.1798,
+      "loss": 0.3038,
       "step": 634
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8412903225806452,
-      "eval_loss": 0.1014009341597557,
-      "eval_runtime": 1.4514,
-      "eval_samples_per_second": 2135.813,
-      "eval_steps_per_second": 44.783,
+      "eval_accuracy": 0.847741935483871,
+      "eval_loss": 0.1362968385219574,
+      "eval_runtime": 1.392,
+      "eval_samples_per_second": 2227.027,
+      "eval_steps_per_second": 46.696,
       "step": 636
     },
     {
       "epoch": 2.990566037735849,
-      "grad_norm": 0.45646050572395325,
+      "grad_norm": 0.5987477898597717,
       "learning_rate": 1.4018867924528304e-05,
-      "loss": 0.1082,
+      "loss": 0.1395,
       "step": 951
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8845161290322581,
-      "eval_loss": 0.06448112428188324,
-      "eval_runtime": 1.4494,
-      "eval_samples_per_second": 2138.806,
-      "eval_steps_per_second": 44.846,
+      "eval_accuracy": 0.8990322580645161,
+      "eval_loss": 0.07024983316659927,
+      "eval_runtime": 1.4088,
+      "eval_samples_per_second": 2200.388,
+      "eval_steps_per_second": 46.137,
       "step": 954
     },
     {
       "epoch": 3.9874213836477987,
-      "grad_norm": 0.40755772590637207,
+      "grad_norm": 0.5540674924850464,
       "learning_rate": 1.2025157232704403e-05,
-      "loss": 0.0791,
+      "loss": 0.0891,
       "step": 1268
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.905483870967742,
-      "eval_loss": 0.0472751259803772,
-      "eval_runtime": 1.6397,
-      "eval_samples_per_second": 1890.587,
-      "eval_steps_per_second": 39.641,
+      "eval_accuracy": 0.9187096774193548,
+      "eval_loss": 0.04933710768818855,
+      "eval_runtime": 1.6037,
+      "eval_samples_per_second": 1932.992,
+      "eval_steps_per_second": 40.53,
       "step": 1272
     },
     {
       "epoch": 4.984276729559748,
-      "grad_norm": 0.3095701336860657,
+      "grad_norm": 0.42864474654197693,
       "learning_rate": 1.0031446540880504e-05,
-      "loss": 0.0638,
+      "loss": 0.0692,
       "step": 1585
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9141935483870968,
-      "eval_loss": 0.03874792158603668,
-      "eval_runtime": 1.4487,
-      "eval_samples_per_second": 2139.852,
-      "eval_steps_per_second": 44.868,
+      "eval_accuracy": 0.9241935483870968,
+      "eval_loss": 0.04158218950033188,
+      "eval_runtime": 1.4224,
+      "eval_samples_per_second": 2179.482,
+      "eval_steps_per_second": 45.699,
       "step": 1590
     },
     {
       "epoch": 5.981132075471698,
-      "grad_norm": 0.37643545866012573,
+      "grad_norm": 0.4545074999332428,
       "learning_rate": 8.037735849056606e-06,
-      "loss": 0.0551,
+      "loss": 0.0595,
       "step": 1902
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9203225806451613,
-      "eval_loss": 0.03338970988988876,
-      "eval_runtime": 1.4551,
-      "eval_samples_per_second": 2130.403,
-      "eval_steps_per_second": 44.67,
+      "eval_accuracy": 0.9270967741935484,
+      "eval_loss": 0.03682653605937958,
+      "eval_runtime": 1.4223,
+      "eval_samples_per_second": 2179.524,
+      "eval_steps_per_second": 45.7,
       "step": 1908
     },
     {
       "epoch": 6.977987421383648,
-      "grad_norm": 0.2515159249305725,
+      "grad_norm": 0.34796932339668274,
       "learning_rate": 6.044025157232704e-06,
-      "loss": 0.0498,
+      "loss": 0.0538,
       "step": 2219
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9261290322580645,
-      "eval_loss": 0.030625004321336746,
-      "eval_runtime": 1.454,
-      "eval_samples_per_second": 2132.098,
-      "eval_steps_per_second": 44.705,
+      "eval_accuracy": 0.9316129032258065,
+      "eval_loss": 0.03404370695352554,
+      "eval_runtime": 1.6095,
+      "eval_samples_per_second": 1926.117,
+      "eval_steps_per_second": 40.386,
       "step": 2226
     },
     {
       "epoch": 7.9748427672955975,
-      "grad_norm": 0.27480530738830566,
+      "grad_norm": 0.3150351941585541,
       "learning_rate": 4.0503144654088055e-06,
-      "loss": 0.0464,
+      "loss": 0.0503,
       "step": 2536
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9290322580645162,
-      "eval_loss": 0.028516776859760284,
-      "eval_runtime": 1.6404,
-      "eval_samples_per_second": 1889.787,
-      "eval_steps_per_second": 39.625,
+      "eval_accuracy": 0.9341935483870968,
+      "eval_loss": 0.03234480321407318,
+      "eval_runtime": 1.4255,
+      "eval_samples_per_second": 2174.649,
+      "eval_steps_per_second": 45.597,
       "step": 2544
     },
     {
       "epoch": 8.971698113207546,
-      "grad_norm": 0.2700980305671692,
+      "grad_norm": 0.3166097104549408,
       "learning_rate": 2.056603773584906e-06,
-      "loss": 0.0441,
+      "loss": 0.0479,
       "step": 2853
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.93,
-      "eval_loss": 0.027497297152876854,
-      "eval_runtime": 1.4563,
-      "eval_samples_per_second": 2128.743,
-      "eval_steps_per_second": 44.635,
+      "eval_accuracy": 0.9341935483870968,
+      "eval_loss": 0.03134315088391304,
+      "eval_runtime": 1.6252,
+      "eval_samples_per_second": 1907.451,
+      "eval_steps_per_second": 39.995,
       "step": 2862
     },
     {
       "epoch": 9.968553459119496,
-      "grad_norm": 0.2162577509880066,
+      "grad_norm": 0.27577438950538635,
       "learning_rate": 6.289308176100629e-08,
-      "loss": 0.043,
+      "loss": 0.0467,
       "step": 3170
     }
   ],
@@ -177,12 +177,12 @@
       "attributes": {}
     }
   },
-  "total_flos": 825688349609088.0,
+  "total_flos": 825254092458012.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7484592181582544,
+    "alpha": 0.5555856325738124,
     "num_train_epochs": 10,
-    "temperature": 6
+    "temperature": 2
   }
 }

run-11/checkpoint-3180/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45dbabd890b1ef70b86e340f0cb53e8ed1df3376e767b2a2d36bcab0630a9d1e
+oid sha256:d588f08c0e85333fa2a4adcf5ec378e9adea9df69c799ff80677ef0f82a3e48a
 size 5368

runs/Dec19_21-14-18_P920/events.out.tfevents.1734621366.P920.569418.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d0e319f4a1a3a60dcc04eae45c83db588942d0127d1bd6c315fd1d192b99f4
+size 18273