Training in progress, epoch 9

checkpoint-1467/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8016f04445628708c33bf649e49067d8ba01bd9c9f7fb7ec23bf16e84302415b
-size 2681490814
+oid sha256:27ff37e113e06eff1db241a9c3d07e590a00e217cfcbb8577767c193b7941dd1
+size 2681485310

checkpoint-1467/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7633807fb9e36e6b1a07c28414b0d317873a57fcff768c36a39a20807d0f3b4f
+oid sha256:aa59b62075ac20c4333b546c0b025606a618273ec6b78f95fe165738d3a0f2af
 size 1340743917

checkpoint-1467/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:219e774be4c8030937df1faf8420e42aff6b143155dd664192f9426967cb7e7d
+oid sha256:e36b0f788cfae254f69b987374e8eb7dea9d5f493b9011eb51e4135cc476bfda
 size 14503

checkpoint-1467/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:604e5dda890598891dfe18e4a7a828eac1f5718acc7a3382899d2cbac39d14db
+oid sha256:96405c1f878387bb6d62cce12f08fc3133dfd93159844cc7dba9f524f0c58020
 size 623

checkpoint-1467/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.2981628179550171,
-  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-978",
+  "best_metric": 0.29900965094566345,
+  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-1467",
   "epoch": 8.997323135755257,
   "global_step": 1467,
   "is_hyper_param_search": false,
@@ -9,152 +9,152 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 9.499999999999999e-07,
-      "loss": 0.4384,
+      "learning_rate": 9e-07,
+      "loss": 0.439,
       "step": 163
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7444061962134251,
-      "eval_f1": 0.7308261375858633,
-      "eval_loss": 0.39615127444267273,
-      "eval_runtime": 9.0599,
-      "eval_samples_per_second": 256.514,
-      "eval_steps_per_second": 32.119,
+      "eval_accuracy": 0.745697074010327,
+      "eval_f1": 0.732240056847258,
+      "eval_loss": 0.3982622027397156,
+      "eval_runtime": 8.9237,
+      "eval_samples_per_second": 260.431,
+      "eval_steps_per_second": 32.61,
       "step": 163
     },
     {
       "epoch": 2.0,
-      "learning_rate": 9e-07,
-      "loss": 0.3447,
+      "learning_rate": 8e-07,
+      "loss": 0.3465,
       "step": 326
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.76592082616179,
-      "eval_f1": 0.7552159046464709,
-      "eval_loss": 0.3409559428691864,
-      "eval_runtime": 9.4378,
-      "eval_samples_per_second": 246.244,
-      "eval_steps_per_second": 30.833,
+      "eval_accuracy": 0.7620481927710844,
+      "eval_f1": 0.750740067157349,
+      "eval_loss": 0.3448249399662018,
+      "eval_runtime": 9.0059,
+      "eval_samples_per_second": 258.053,
+      "eval_steps_per_second": 32.312,
       "step": 326
     },
     {
       "epoch": 3.0,
-      "learning_rate": 8.499999999999999e-07,
-      "loss": 0.3057,
+      "learning_rate": 7e-07,
+      "loss": 0.3089,
       "step": 489
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7749569707401033,
-      "eval_f1": 0.768808341108185,
-      "eval_loss": 0.32338443398475647,
-      "eval_runtime": 9.0418,
-      "eval_samples_per_second": 257.028,
-      "eval_steps_per_second": 32.184,
+      "eval_accuracy": 0.7693631669535284,
+      "eval_f1": 0.7633651185887134,
+      "eval_loss": 0.3303545117378235,
+      "eval_runtime": 9.3737,
+      "eval_samples_per_second": 247.927,
+      "eval_steps_per_second": 31.044,
       "step": 489
     },
     {
       "epoch": 4.0,
-      "learning_rate": 8e-07,
-      "loss": 0.287,
+      "learning_rate": 6e-07,
+      "loss": 0.2916,
       "step": 652
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7857142857142857,
-      "eval_f1": 0.7778970154753132,
-      "eval_loss": 0.3068828284740448,
-      "eval_runtime": 9.3662,
-      "eval_samples_per_second": 248.128,
-      "eval_steps_per_second": 31.069,
+      "eval_accuracy": 0.7839931153184165,
+      "eval_f1": 0.7736843738845695,
+      "eval_loss": 0.30892524123191833,
+      "eval_runtime": 9.0737,
+      "eval_samples_per_second": 256.125,
+      "eval_steps_per_second": 32.071,
       "step": 652
     },
     {
       "epoch": 5.0,
-      "learning_rate": 7.5e-07,
-      "loss": 0.2742,
+      "learning_rate": 5e-07,
+      "loss": 0.2826,
       "step": 815
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.7887263339070568,
-      "eval_f1": 0.7821763089027973,
-      "eval_loss": 0.30303624272346497,
-      "eval_runtime": 8.908,
-      "eval_samples_per_second": 260.89,
-      "eval_steps_per_second": 32.667,
+      "eval_accuracy": 0.7839931153184165,
+      "eval_f1": 0.7744514636272033,
+      "eval_loss": 0.3055116832256317,
+      "eval_runtime": 9.1262,
+      "eval_samples_per_second": 254.651,
+      "eval_steps_per_second": 31.886,
       "step": 815
     },
     {
       "epoch": 6.0,
-      "learning_rate": 7e-07,
-      "loss": 0.2676,
+      "learning_rate": 4e-07,
+      "loss": 0.2747,
       "step": 978
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.7938898450946644,
-      "eval_f1": 0.7850614050415754,
-      "eval_loss": 0.2981628179550171,
-      "eval_runtime": 9.0189,
-      "eval_samples_per_second": 257.681,
-      "eval_steps_per_second": 32.266,
+      "eval_accuracy": 0.7857142857142857,
+      "eval_f1": 0.7771340101366444,
+      "eval_loss": 0.30407363176345825,
+      "eval_runtime": 9.1656,
+      "eval_samples_per_second": 253.557,
+      "eval_steps_per_second": 31.749,
       "step": 978
     },
     {
       "epoch": 7.0,
-      "learning_rate": 6.5e-07,
-      "loss": 0.2585,
+      "learning_rate": 3e-07,
+      "loss": 0.2678,
       "step": 1141
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7908777969018933,
-      "eval_f1": 0.7821955847641968,
-      "eval_loss": 0.3001907467842102,
-      "eval_runtime": 8.8872,
-      "eval_samples_per_second": 261.499,
-      "eval_steps_per_second": 32.744,
+      "eval_accuracy": 0.7865748709122203,
+      "eval_f1": 0.7768983029852594,
+      "eval_loss": 0.3031200170516968,
+      "eval_runtime": 9.6529,
+      "eval_samples_per_second": 240.756,
+      "eval_steps_per_second": 30.146,
       "step": 1141
     },
     {
       "epoch": 8.0,
-      "learning_rate": 6e-07,
-      "loss": 0.2526,
+      "learning_rate": 2e-07,
+      "loss": 0.2641,
       "step": 1304
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.7943201376936316,
-      "eval_f1": 0.7876461988304093,
-      "eval_loss": 0.30516260862350464,
-      "eval_runtime": 9.0042,
-      "eval_samples_per_second": 258.103,
-      "eval_steps_per_second": 32.318,
+      "eval_accuracy": 0.7874354561101549,
+      "eval_f1": 0.7772050769924619,
+      "eval_loss": 0.3010457754135132,
+      "eval_runtime": 9.1465,
+      "eval_samples_per_second": 254.087,
+      "eval_steps_per_second": 31.816,
       "step": 1304
     },
     {
       "epoch": 9.0,
-      "learning_rate": 5.5e-07,
-      "loss": 0.2479,
+      "learning_rate": 1e-07,
+      "loss": 0.2631,
       "step": 1467
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7938898450946644,
-      "eval_f1": 0.784665589216992,
-      "eval_loss": 0.2996860444545746,
-      "eval_runtime": 9.1157,
-      "eval_samples_per_second": 254.944,
-      "eval_steps_per_second": 31.923,
+      "eval_accuracy": 0.7917383820998278,
+      "eval_f1": 0.7830140707046156,
+      "eval_loss": 0.29900965094566345,
+      "eval_runtime": 9.9577,
+      "eval_samples_per_second": 233.387,
+      "eval_steps_per_second": 29.224,
       "step": 1467
     }
   ],
-  "max_steps": 3260,
-  "num_train_epochs": 20,
-  "total_flos": 2.7397418508888e+16,
+  "max_steps": 1630,
+  "num_train_epochs": 10,
+  "total_flos": 2.739683604981792e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-1467/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50dde5eeea306f54118173d342686475ad9209b6c2cac103f7b114d5f582dc36
+oid sha256:cd48e93c542d5f8f840918341d20dc98e6dbd60ec7052cf9f5610075d1655eaf
 size 3119

checkpoint-1630/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f10ac8aa6914a68e479931a864f18ba10f4c2c3365275928cdcaec362c14888
-size 2681490814
+oid sha256:58949addab61a91ba7273887e0d81fb845296dc8d00ee4a1d5f426047707bd55
+size 2681485310

checkpoint-1630/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3f456ad6fff0711c705ef4a02fde64a2461f640a2ebaf6d8128cf037da6c194
+oid sha256:a94b56e1da42f00703387b7c987adc79cef5d650801e77d14a802f4e92301a9c
 size 1340743917

checkpoint-1630/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19de93fa7ce45dde7d1a60c4c97b9e4ec495777424b4e2a14fb06079d8444157
+oid sha256:d895ad69969472b320a4216795557386f8649e29cb20bf0cdadbc0088844bdcb
 size 14503

checkpoint-1630/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c924e180775c2b50e4679ec9d5d5d33fb3eed968a176cbb01236411034beab21
+oid sha256:9c3e7a1b845fb89890dfefde14b31ecc3084465b9c0621dd07e0fdee41ef59fc
 size 623

checkpoint-1630/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.2981628179550171,
-  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-978",
+  "best_metric": 0.29900965094566345,
+  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-1467",
   "epoch": 9.997323135755257,
   "global_step": 1630,
   "is_hyper_param_search": false,
@@ -9,168 +9,168 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 9.499999999999999e-07,
-      "loss": 0.4384,
+      "learning_rate": 9e-07,
+      "loss": 0.439,
       "step": 163
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7444061962134251,
-      "eval_f1": 0.7308261375858633,
-      "eval_loss": 0.39615127444267273,
-      "eval_runtime": 9.0599,
-      "eval_samples_per_second": 256.514,
-      "eval_steps_per_second": 32.119,
+      "eval_accuracy": 0.745697074010327,
+      "eval_f1": 0.732240056847258,
+      "eval_loss": 0.3982622027397156,
+      "eval_runtime": 8.9237,
+      "eval_samples_per_second": 260.431,
+      "eval_steps_per_second": 32.61,
       "step": 163
     },
     {
       "epoch": 2.0,
-      "learning_rate": 9e-07,
-      "loss": 0.3447,
+      "learning_rate": 8e-07,
+      "loss": 0.3465,
       "step": 326
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.76592082616179,
-      "eval_f1": 0.7552159046464709,
-      "eval_loss": 0.3409559428691864,
-      "eval_runtime": 9.4378,
-      "eval_samples_per_second": 246.244,
-      "eval_steps_per_second": 30.833,
+      "eval_accuracy": 0.7620481927710844,
+      "eval_f1": 0.750740067157349,
+      "eval_loss": 0.3448249399662018,
+      "eval_runtime": 9.0059,
+      "eval_samples_per_second": 258.053,
+      "eval_steps_per_second": 32.312,
       "step": 326
     },
     {
       "epoch": 3.0,
-      "learning_rate": 8.499999999999999e-07,
-      "loss": 0.3057,
+      "learning_rate": 7e-07,
+      "loss": 0.3089,
       "step": 489
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7749569707401033,
-      "eval_f1": 0.768808341108185,
-      "eval_loss": 0.32338443398475647,
-      "eval_runtime": 9.0418,
-      "eval_samples_per_second": 257.028,
-      "eval_steps_per_second": 32.184,
+      "eval_accuracy": 0.7693631669535284,
+      "eval_f1": 0.7633651185887134,
+      "eval_loss": 0.3303545117378235,
+      "eval_runtime": 9.3737,
+      "eval_samples_per_second": 247.927,
+      "eval_steps_per_second": 31.044,
       "step": 489
     },
     {
       "epoch": 4.0,
-      "learning_rate": 8e-07,
-      "loss": 0.287,
+      "learning_rate": 6e-07,
+      "loss": 0.2916,
       "step": 652
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7857142857142857,
-      "eval_f1": 0.7778970154753132,
-      "eval_loss": 0.3068828284740448,
-      "eval_runtime": 9.3662,
-      "eval_samples_per_second": 248.128,
-      "eval_steps_per_second": 31.069,
+      "eval_accuracy": 0.7839931153184165,
+      "eval_f1": 0.7736843738845695,
+      "eval_loss": 0.30892524123191833,
+      "eval_runtime": 9.0737,
+      "eval_samples_per_second": 256.125,
+      "eval_steps_per_second": 32.071,
       "step": 652
     },
     {
       "epoch": 5.0,
-      "learning_rate": 7.5e-07,
-      "loss": 0.2742,
+      "learning_rate": 5e-07,
+      "loss": 0.2826,
       "step": 815
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.7887263339070568,
-      "eval_f1": 0.7821763089027973,
-      "eval_loss": 0.30303624272346497,
-      "eval_runtime": 8.908,
-      "eval_samples_per_second": 260.89,
-      "eval_steps_per_second": 32.667,
+      "eval_accuracy": 0.7839931153184165,
+      "eval_f1": 0.7744514636272033,
+      "eval_loss": 0.3055116832256317,
+      "eval_runtime": 9.1262,
+      "eval_samples_per_second": 254.651,
+      "eval_steps_per_second": 31.886,
       "step": 815
     },
     {
       "epoch": 6.0,
-      "learning_rate": 7e-07,
-      "loss": 0.2676,
+      "learning_rate": 4e-07,
+      "loss": 0.2747,
       "step": 978
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.7938898450946644,
-      "eval_f1": 0.7850614050415754,
-      "eval_loss": 0.2981628179550171,
-      "eval_runtime": 9.0189,
-      "eval_samples_per_second": 257.681,
-      "eval_steps_per_second": 32.266,
+      "eval_accuracy": 0.7857142857142857,
+      "eval_f1": 0.7771340101366444,
+      "eval_loss": 0.30407363176345825,
+      "eval_runtime": 9.1656,
+      "eval_samples_per_second": 253.557,
+      "eval_steps_per_second": 31.749,
       "step": 978
     },
     {
       "epoch": 7.0,
-      "learning_rate": 6.5e-07,
-      "loss": 0.2585,
+      "learning_rate": 3e-07,
+      "loss": 0.2678,
       "step": 1141
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7908777969018933,
-      "eval_f1": 0.7821955847641968,
-      "eval_loss": 0.3001907467842102,
-      "eval_runtime": 8.8872,
-      "eval_samples_per_second": 261.499,
-      "eval_steps_per_second": 32.744,
+      "eval_accuracy": 0.7865748709122203,
+      "eval_f1": 0.7768983029852594,
+      "eval_loss": 0.3031200170516968,
+      "eval_runtime": 9.6529,
+      "eval_samples_per_second": 240.756,
+      "eval_steps_per_second": 30.146,
       "step": 1141
     },
     {
       "epoch": 8.0,
-      "learning_rate": 6e-07,
-      "loss": 0.2526,
+      "learning_rate": 2e-07,
+      "loss": 0.2641,
       "step": 1304
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.7943201376936316,
-      "eval_f1": 0.7876461988304093,
-      "eval_loss": 0.30516260862350464,
-      "eval_runtime": 9.0042,
-      "eval_samples_per_second": 258.103,
-      "eval_steps_per_second": 32.318,
+      "eval_accuracy": 0.7874354561101549,
+      "eval_f1": 0.7772050769924619,
+      "eval_loss": 0.3010457754135132,
+      "eval_runtime": 9.1465,
+      "eval_samples_per_second": 254.087,
+      "eval_steps_per_second": 31.816,
       "step": 1304
     },
     {
       "epoch": 9.0,
-      "learning_rate": 5.5e-07,
-      "loss": 0.2479,
+      "learning_rate": 1e-07,
+      "loss": 0.2631,
       "step": 1467
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7938898450946644,
-      "eval_f1": 0.784665589216992,
-      "eval_loss": 0.2996860444545746,
-      "eval_runtime": 9.1157,
-      "eval_samples_per_second": 254.944,
-      "eval_steps_per_second": 31.923,
+      "eval_accuracy": 0.7917383820998278,
+      "eval_f1": 0.7830140707046156,
+      "eval_loss": 0.29900965094566345,
+      "eval_runtime": 9.9577,
+      "eval_samples_per_second": 233.387,
+      "eval_steps_per_second": 29.224,
       "step": 1467
     },
     {
       "epoch": 10.0,
-      "learning_rate": 5e-07,
-      "loss": 0.2451,
+      "learning_rate": 0.0,
+      "loss": 0.2615,
       "step": 1630
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.7956110154905336,
-      "eval_f1": 0.787314593871832,
-      "eval_loss": 0.3013566732406616,
-      "eval_runtime": 9.3541,
-      "eval_samples_per_second": 248.446,
-      "eval_steps_per_second": 31.109,
+      "eval_accuracy": 0.7895869191049913,
+      "eval_f1": 0.7804941506557442,
+      "eval_loss": 0.3004284203052521,
+      "eval_runtime": 9.1779,
+      "eval_samples_per_second": 253.217,
+      "eval_steps_per_second": 31.707,
       "step": 1630
     }
   ],
-  "max_steps": 3260,
-  "num_train_epochs": 20,
-  "total_flos": 3.0441815576382144e+16,
+  "max_steps": 1630,
+  "num_train_epochs": 10,
+  "total_flos": 3.0430166394980544e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-1630/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50dde5eeea306f54118173d342686475ad9209b6c2cac103f7b114d5f582dc36
+oid sha256:cd48e93c542d5f8f840918341d20dc98e6dbd60ec7052cf9f5610075d1655eaf
 size 3119

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a424c985030a874a9e4bfc84349addc7b4ac2efb4d1b6b44956f1f85a1ff96e1
+oid sha256:a94b56e1da42f00703387b7c987adc79cef5d650801e77d14a802f4e92301a9c
 size 1340743917