Training in progress, epoch 7

checkpoint-1141/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:645b21030c95ae40ac3efa58aeb77c516129c1e21a7b6f96040b285dab281f95
-size 2681490814
+oid sha256:3d4397c0871aa8e40f1ccb350534d7ab81ba3fccb978f0d348c29c46fbebd3f4
+size 2681485310

checkpoint-1141/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a8a61834396ae4adc1129d107763a962535223b9f1f71f8fd0280f439a14cc6
+oid sha256:5a64b29d4c457a46664b7f17b3612c04de19d524459d4eb507a1824de9a48d8b
 size 1340743917

checkpoint-1141/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d8ba5811f005ed35376bd706a06398b37e12b1ccc81171222e0877b0309aae66
+oid sha256:127190e9a123a5ee5c7cae85536b7ea27da0902983a65fced8591323b53379bf
 size 14503

checkpoint-1141/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5cbabc59ca6629b30ec20ef4a37b261f5583c218af1ab376085065bb9a7ec5b
+oid sha256:9b6c415d3c794e7691e5da96fc4ee5df698b9430d4ddd779b2851bfafcde2856
 size 623

checkpoint-1141/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.2981628179550171,
-  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-978",
+  "best_metric": 0.3031200170516968,
+  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-1141",
   "epoch": 6.997323135755258,
   "global_step": 1141,
   "is_hyper_param_search": false,
@@ -9,120 +9,120 @@
   "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
     }
   ],
-  "max_steps": 3260,
-  "num_train_epochs": 20,
-  "total_flos": 2.131077947245901e+16,
+  "max_steps": 1630,
+  "num_train_epochs": 10,
+  "total_flos": 2.131107070199405e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-1141/training_args.bin

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

checkpoint-1304/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a995198957991a4778c9b0a4ff2acc3f852248799a63f63a112918961d52dec7
-size 2681490814
+oid sha256:8ef7936ae326c4eaaaf3a99ec813dac11f0e26791ed08a0d5bcb6ff80d2857ff
+size 2681485310

checkpoint-1304/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffff6f838e3747eee33fd07b6b5738e8124d6bf65b8019b93dc2b67e1e73aaec
+oid sha256:a424c985030a874a9e4bfc84349addc7b4ac2efb4d1b6b44956f1f85a1ff96e1
 size 1340743917

checkpoint-1304/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a7cdc490809685b8f3f1a14763799335b83c7009894dafbcb3f054ef356ff5d
+oid sha256:38a7aa61e17bfe1f7d2cd89233eb4f54a7784fec415bcbcb0b0ac32e158b1d30
 size 14503

checkpoint-1304/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0acd2889d07b3f459da1f99892f4b8cf79ba4808ac88cb08ca68b0256d9cd814
+oid sha256:75e4a97a3a66b98bf4e23a9d653c834f95839beaa653ab7298c03a6f2b4965f0
 size 623

checkpoint-1304/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.2981628179550171,
-  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-978",
+  "best_metric": 0.3010457754135132,
+  "best_model_checkpoint": "outputs/electra-nli-efl-tweeteval/checkpoint-1304",
   "epoch": 7.997323135755258,
   "global_step": 1304,
   "is_hyper_param_search": false,
@@ -9,136 +9,136 @@
   "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
     }
   ],
-  "max_steps": 3260,
-  "num_train_epochs": 20,
-  "total_flos": 2.435494355632512e+16,
+  "max_steps": 1630,
+  "num_train_epochs": 10,
+  "total_flos": 2.435546776948819e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-1304/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:e0655cfc868a392ac2b0850468debd2161b60753b0a1dc8c0e61845d86a8b31a
+oid sha256:a424c985030a874a9e4bfc84349addc7b4ac2efb4d1b6b44956f1f85a1ff96e1
 size 1340743917