Upload trainer_state.json

trainer_state.json

@@ -0,0 +1,289 @@
+{
+  "best_metric": 0.95920731151508,
+  "best_model_checkpoint": "/scicore/home/lauerg/cerque0000/models/nlp-job-ads/gbert-large-ft-edu-redux/checkpoint-1400",
+  "epoch": 9.0,
+  "global_step": 2664,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.34,
+      "eval_loss": 0.29267561435699463,
+      "eval_lrap": 0.7186470358938737,
+      "eval_runtime": 22.3251,
+      "eval_samples_per_second": 30.101,
+      "eval_steps_per_second": 3.763,
+      "step": 100
+    },
+    {
+      "epoch": 0.68,
+      "eval_loss": 0.19173942506313324,
+      "eval_lrap": 0.8874944283175044,
+      "eval_runtime": 22.732,
+      "eval_samples_per_second": 29.562,
+      "eval_steps_per_second": 3.695,
+      "step": 200
+    },
+    {
+      "epoch": 1.01,
+      "eval_loss": 0.1740952730178833,
+      "eval_lrap": 0.9086508545541496,
+      "eval_runtime": 22.6942,
+      "eval_samples_per_second": 29.611,
+      "eval_steps_per_second": 3.701,
+      "step": 300
+    },
+    {
+      "epoch": 1.35,
+      "eval_loss": 0.19585496187210083,
+      "eval_lrap": 0.9103675474295027,
+      "eval_runtime": 23.3607,
+      "eval_samples_per_second": 28.766,
+      "eval_steps_per_second": 3.596,
+      "step": 400
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.193243243243243e-05,
+      "loss": 0.2242,
+      "step": 500
+    },
+    {
+      "epoch": 1.69,
+      "eval_loss": 0.1878644824028015,
+      "eval_lrap": 0.9258567343567102,
+      "eval_runtime": 22.7639,
+      "eval_samples_per_second": 29.52,
+      "eval_steps_per_second": 3.69,
+      "step": 500
+    },
+    {
+      "epoch": 2.03,
+      "eval_loss": 0.16258566081523895,
+      "eval_lrap": 0.9308544571355302,
+      "eval_runtime": 23.3248,
+      "eval_samples_per_second": 28.81,
+      "eval_steps_per_second": 3.601,
+      "step": 600
+    },
+    {
+      "epoch": 2.36,
+      "eval_loss": 0.15402589738368988,
+      "eval_lrap": 0.9306379874972384,
+      "eval_runtime": 22.759,
+      "eval_samples_per_second": 29.527,
+      "eval_steps_per_second": 3.691,
+      "step": 700
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 0.1420990228652954,
+      "eval_lrap": 0.9364025033506702,
+      "eval_runtime": 23.3898,
+      "eval_samples_per_second": 28.73,
+      "eval_steps_per_second": 3.591,
+      "step": 800
+    },
+    {
+      "epoch": 3.04,
+      "eval_loss": 0.14033490419387817,
+      "eval_lrap": 0.9166296066698628,
+      "eval_runtime": 22.7034,
+      "eval_samples_per_second": 29.599,
+      "eval_steps_per_second": 3.7,
+      "step": 900
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.6864864864864864e-05,
+      "loss": 0.1316,
+      "step": 1000
+    },
+    {
+      "epoch": 3.38,
+      "eval_loss": 0.13963457942008972,
+      "eval_lrap": 0.9372152596624732,
+      "eval_runtime": 23.3136,
+      "eval_samples_per_second": 28.824,
+      "eval_steps_per_second": 3.603,
+      "step": 1000
+    },
+    {
+      "epoch": 3.72,
+      "eval_loss": 0.1491432934999466,
+      "eval_lrap": 0.9427461184278115,
+      "eval_runtime": 22.6955,
+      "eval_samples_per_second": 29.609,
+      "eval_steps_per_second": 3.701,
+      "step": 1100
+    },
+    {
+      "epoch": 4.05,
+      "eval_loss": 0.14092977344989777,
+      "eval_lrap": 0.9405212229801448,
+      "eval_runtime": 22.7532,
+      "eval_samples_per_second": 29.534,
+      "eval_steps_per_second": 3.692,
+      "step": 1200
+    },
+    {
+      "epoch": 4.39,
+      "eval_loss": 0.13760367035865784,
+      "eval_lrap": 0.930910014527521,
+      "eval_runtime": 22.795,
+      "eval_samples_per_second": 29.48,
+      "eval_steps_per_second": 3.685,
+      "step": 1300
+    },
+    {
+      "epoch": 4.73,
+      "eval_loss": 0.13160455226898193,
+      "eval_lrap": 0.95920731151508,
+      "eval_runtime": 22.6631,
+      "eval_samples_per_second": 29.652,
+      "eval_steps_per_second": 3.706,
+      "step": 1400
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 1.1797297297297297e-05,
+      "loss": 0.0757,
+      "step": 1500
+    },
+    {
+      "epoch": 5.07,
+      "eval_loss": 0.12442280352115631,
+      "eval_lrap": 0.9517553489380385,
+      "eval_runtime": 23.4283,
+      "eval_samples_per_second": 28.683,
+      "eval_steps_per_second": 3.585,
+      "step": 1500
+    },
+    {
+      "epoch": 5.41,
+      "eval_loss": 0.12935081124305725,
+      "eval_lrap": 0.9447551279048475,
+      "eval_runtime": 23.3876,
+      "eval_samples_per_second": 28.733,
+      "eval_steps_per_second": 3.592,
+      "step": 1600
+    },
+    {
+      "epoch": 5.74,
+      "eval_loss": 0.13968029618263245,
+      "eval_lrap": 0.9520049098544898,
+      "eval_runtime": 23.3148,
+      "eval_samples_per_second": 28.823,
+      "eval_steps_per_second": 3.603,
+      "step": 1700
+    },
+    {
+      "epoch": 6.08,
+      "eval_loss": 0.1321108192205429,
+      "eval_lrap": 0.9493008222774945,
+      "eval_runtime": 22.7141,
+      "eval_samples_per_second": 29.585,
+      "eval_steps_per_second": 3.698,
+      "step": 1800
+    },
+    {
+      "epoch": 6.42,
+      "eval_loss": 0.1369735449552536,
+      "eval_lrap": 0.9438046186531343,
+      "eval_runtime": 22.7987,
+      "eval_samples_per_second": 29.475,
+      "eval_steps_per_second": 3.684,
+      "step": 1900
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 6.729729729729729e-06,
+      "loss": 0.04,
+      "step": 2000
+    },
+    {
+      "epoch": 6.76,
+      "eval_loss": 0.13185060024261475,
+      "eval_lrap": 0.9582819439486873,
+      "eval_runtime": 22.7045,
+      "eval_samples_per_second": 29.598,
+      "eval_steps_per_second": 3.7,
+      "step": 2000
+    },
+    {
+      "epoch": 7.09,
+      "eval_loss": 0.14138825237751007,
+      "eval_lrap": 0.9504156521289109,
+      "eval_runtime": 23.4267,
+      "eval_samples_per_second": 28.685,
+      "eval_steps_per_second": 3.586,
+      "step": 2100
+    },
+    {
+      "epoch": 7.43,
+      "eval_loss": 0.14279799163341522,
+      "eval_lrap": 0.9578134572863043,
+      "eval_runtime": 22.7296,
+      "eval_samples_per_second": 29.565,
+      "eval_steps_per_second": 3.696,
+      "step": 2200
+    },
+    {
+      "epoch": 7.77,
+      "eval_loss": 0.14730122685432434,
+      "eval_lrap": 0.9416550938849539,
+      "eval_runtime": 22.7712,
+      "eval_samples_per_second": 29.511,
+      "eval_steps_per_second": 3.689,
+      "step": 2300
+    },
+    {
+      "epoch": 8.11,
+      "eval_loss": 0.14928565919399261,
+      "eval_lrap": 0.9514275899253632,
+      "eval_runtime": 23.4136,
+      "eval_samples_per_second": 28.701,
+      "eval_steps_per_second": 3.588,
+      "step": 2400
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 1.662162162162162e-06,
+      "loss": 0.0204,
+      "step": 2500
+    },
+    {
+      "epoch": 8.45,
+      "eval_loss": 0.14889824390411377,
+      "eval_lrap": 0.955463596740468,
+      "eval_runtime": 23.4183,
+      "eval_samples_per_second": 28.695,
+      "eval_steps_per_second": 3.587,
+      "step": 2500
+    },
+    {
+      "epoch": 8.78,
+      "eval_loss": 0.14893263578414917,
+      "eval_lrap": 0.9515173886897298,
+      "eval_runtime": 23.311,
+      "eval_samples_per_second": 28.828,
+      "eval_steps_per_second": 3.603,
+      "step": 2600
+    },
+    {
+      "epoch": 9.0,
+      "step": 2664,
+      "total_flos": 1.9853269227408384e+16,
+      "train_loss": 0.09327275080007834,
+      "train_runtime": 2835.2925,
+      "train_samples_per_second": 7.514,
+      "train_steps_per_second": 0.94
+    }
+  ],
+  "max_steps": 2664,
+  "num_train_epochs": 9,
+  "total_flos": 1.9853269227408384e+16,
+  "trial_name": null,
+  "trial_params": null
+}