Training in progress, step 2862

run-14/checkpoint-2862/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.47.0",
+  "transformers_version": "4.47.1",
   "vocab_size": 30522
 }

run-14/checkpoint-2862/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47d05a3cbd32567a3b9e171fbdc5129d4057528cfc10fbe31fc89cb834a4d6cb
+oid sha256:c18c4a133656b004cd6c65fc4815e9ef62f1ec3522774d37fb11c5daa779d3e2
 size 268290900

run-14/checkpoint-2862/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5dbade491317b012c196d7b07061ac52d915feee32d9989e27f1bcf239517ad4
-size 536641018
+oid sha256:a81143e1443ee75c0758f59ffc911f4898c4f4e07b40479ce205da7301c876f7
+size 536643898

run-14/checkpoint-2862/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e29353d4954fcf585947caa6a0a3fa59c7379d49f3bc56c6a4bd276fef18ca98
-size 13990
+oid sha256:41b5282f1cde964bf5a382eb03f1cdd87f1c8e2f60e43277b3453f63947f5933
+size 14244

run-14/checkpoint-2862/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c64bfaed33db311081ee0596530a03699ac7659f0cf4f57a488fea34df98816
+oid sha256:2a738824491ff8b261b6db2a842e15bf80355d04a07892a16df2c4715b2ee556
 size 1064

run-14/checkpoint-2862/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-14/checkpoint-2862/trainer_state.json

@@ -10,137 +10,137 @@
   "log_history": [
     {
       "epoch": 0.9968553459119497,
-      "grad_norm": 0.6559529304504395,
+      "grad_norm": 0.9016032814979553,
       "learning_rate": 1.7784765897973445e-05,
-      "loss": 0.4379,
+      "loss": 0.8053,
       "step": 317
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6129032258064516,
-      "eval_loss": 0.21713966131210327,
-      "eval_runtime": 3.155,
-      "eval_samples_per_second": 982.556,
-      "eval_steps_per_second": 20.602,
+      "eval_accuracy": 0.6696774193548387,
+      "eval_loss": 0.40919938683509827,
+      "eval_runtime": 1.3804,
+      "eval_samples_per_second": 2245.687,
+      "eval_steps_per_second": 47.087,
       "step": 318
     },
     {
       "epoch": 1.9937106918238994,
-      "grad_norm": 0.5362987518310547,
+      "grad_norm": 0.7222766876220703,
       "learning_rate": 1.556953179594689e-05,
-      "loss": 0.18,
+      "loss": 0.3073,
       "step": 634
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8319354838709677,
-      "eval_loss": 0.10147035866975784,
-      "eval_runtime": 3.1465,
-      "eval_samples_per_second": 985.215,
-      "eval_steps_per_second": 20.658,
+      "eval_accuracy": 0.844516129032258,
+      "eval_loss": 0.13953416049480438,
+      "eval_runtime": 1.4006,
+      "eval_samples_per_second": 2213.361,
+      "eval_steps_per_second": 46.409,
       "step": 636
     },
     {
       "epoch": 2.990566037735849,
-      "grad_norm": 0.54403156042099,
+      "grad_norm": 0.6133838891983032,
       "learning_rate": 1.3354297693920338e-05,
-      "loss": 0.1094,
+      "loss": 0.1428,
       "step": 951
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.885483870967742,
-      "eval_loss": 0.06534993648529053,
-      "eval_runtime": 3.014,
-      "eval_samples_per_second": 1028.54,
-      "eval_steps_per_second": 21.566,
+      "eval_accuracy": 0.8964516129032258,
+      "eval_loss": 0.0722731500864029,
+      "eval_runtime": 1.4142,
+      "eval_samples_per_second": 2192.112,
+      "eval_steps_per_second": 45.964,
       "step": 954
     },
     {
       "epoch": 3.9874213836477987,
-      "grad_norm": 0.39273932576179504,
+      "grad_norm": 0.5777781009674072,
       "learning_rate": 1.1139063591893781e-05,
-      "loss": 0.0802,
+      "loss": 0.0917,
       "step": 1268
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8987096774193548,
-      "eval_loss": 0.048802707344293594,
-      "eval_runtime": 3.0067,
-      "eval_samples_per_second": 1031.015,
-      "eval_steps_per_second": 21.618,
+      "eval_accuracy": 0.9180645161290323,
+      "eval_loss": 0.050948865711688995,
+      "eval_runtime": 1.4204,
+      "eval_samples_per_second": 2182.53,
+      "eval_steps_per_second": 45.763,
       "step": 1272
     },
     {
       "epoch": 4.984276729559748,
-      "grad_norm": 0.2954351007938385,
+      "grad_norm": 0.4380759596824646,
       "learning_rate": 8.923829489867226e-06,
-      "loss": 0.0656,
+      "loss": 0.0715,
       "step": 1585
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.91,
-      "eval_loss": 0.03991023451089859,
-      "eval_runtime": 3.0701,
-      "eval_samples_per_second": 1009.753,
-      "eval_steps_per_second": 21.172,
+      "eval_accuracy": 0.9232258064516129,
+      "eval_loss": 0.042863838374614716,
+      "eval_runtime": 1.7856,
+      "eval_samples_per_second": 1736.138,
+      "eval_steps_per_second": 36.403,
       "step": 1590
     },
     {
       "epoch": 5.981132075471698,
-      "grad_norm": 0.3142315149307251,
+      "grad_norm": 0.46454158425331116,
       "learning_rate": 6.708595387840672e-06,
-      "loss": 0.0572,
+      "loss": 0.0617,
       "step": 1902
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.03489958494901657,
-      "eval_runtime": 3.1188,
-      "eval_samples_per_second": 993.967,
-      "eval_steps_per_second": 20.841,
+      "eval_accuracy": 0.9261290322580645,
+      "eval_loss": 0.03817891329526901,
+      "eval_runtime": 1.6023,
+      "eval_samples_per_second": 1934.755,
+      "eval_steps_per_second": 40.567,
       "step": 1908
     },
     {
       "epoch": 6.977987421383648,
-      "grad_norm": 0.3454739451408386,
+      "grad_norm": 0.34600403904914856,
       "learning_rate": 4.4933612858141165e-06,
-      "loss": 0.0519,
+      "loss": 0.0562,
       "step": 2219
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9216129032258065,
-      "eval_loss": 0.032138124108314514,
-      "eval_runtime": 3.0316,
-      "eval_samples_per_second": 1022.557,
-      "eval_steps_per_second": 21.441,
+      "eval_accuracy": 0.9303225806451613,
+      "eval_loss": 0.035540465265512466,
+      "eval_runtime": 1.6065,
+      "eval_samples_per_second": 1929.7,
+      "eval_steps_per_second": 40.461,
       "step": 2226
     },
     {
       "epoch": 7.9748427672955975,
-      "grad_norm": 0.3118973672389984,
+      "grad_norm": 0.32331326603889465,
       "learning_rate": 2.2781271837875614e-06,
-      "loss": 0.0489,
+      "loss": 0.053,
       "step": 2536
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9235483870967742,
-      "eval_loss": 0.030378255993127823,
-      "eval_runtime": 3.0391,
-      "eval_samples_per_second": 1020.033,
-      "eval_steps_per_second": 21.388,
+      "eval_accuracy": 0.932258064516129,
+      "eval_loss": 0.034126147627830505,
+      "eval_runtime": 1.5999,
+      "eval_samples_per_second": 1937.602,
+      "eval_steps_per_second": 40.627,
       "step": 2544
     },
     {
       "epoch": 8.971698113207546,
-      "grad_norm": 0.4775363802909851,
+      "grad_norm": 0.3256802558898926,
       "learning_rate": 6.289308176100629e-08,
-      "loss": 0.0473,
+      "loss": 0.0511,
       "step": 2853
     }
   ],
@@ -161,12 +161,12 @@
       "attributes": {}
     }
   },
-  "total_flos": 745151547572796.0,
+  "total_flos": 742838622388212.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7111459864621728,
+    "alpha": 0.6723203881609809,
     "num_train_epochs": 9,
-    "temperature": 6
+    "temperature": 2
   }
 }

run-14/checkpoint-2862/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f3ff8764a73f5bb99b6bed819b401d3148e79010400790f8e5ee34c60b9d089
-size 5432
+oid sha256:5c86f3d5ef41ce4b0c9fef4ceb006e512b2745b2eac6815d060edaed99ccb914
+size 5368

runs/Dec21_12-52-13_P920/events.out.tfevents.1734761794.P920.912596.15

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a06519c93625ded4f87aabfe65a3f7d2786db780c3166d1cabcb3542fbc6a6ba
+size 17738