Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd4d89ad8b9454f1d0b670fc7e447f6d0a2f1c8aa327f29fe93915db5e9510ab
+oid sha256:7a4da76d70a524eacec579338be444279371ca1212e89b1c3ff12452d480ce07
 size 442518124

run-1/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd4d89ad8b9454f1d0b670fc7e447f6d0a2f1c8aa327f29fe93915db5e9510ab
+oid sha256:7a4da76d70a524eacec579338be444279371ca1212e89b1c3ff12452d480ce07
 size 442518124

run-1/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7bfeb4cd61b7f89aef6c894a76e5cf6da2fa925d17a0256c9941bb027f4e1b22
+oid sha256:fcd679f38dd251c90164cd0590ce269851ff9023cbffab70a9465bfde2fc08c5
 size 885156090

run-1/checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8676fbd04f9752ca208326869aadf87a8e201d4c9c4328ccab03b20bbb933463
-size 14308
+oid sha256:9f1bda03e9ae8d21d65e8de146ec1797a54afdd7e04e8929ddf985180fe3791c
+size 14244

run-1/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b084160dc6107c5c64602fdf374da79d74a6e7e085133c00104fa7a15de63149
+oid sha256:e9793b73f5790244a8460edf3f1bad9bcbee7514c9949ab36a641baa1c5d874a
 size 1064

run-1/checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.8162388724528657,
+  "best_metric": 0.8284272507637513,
   "best_model_checkpoint": "test-klue/ynat/run-1/checkpoint-500",
-  "epoch": 0.43782837127845886,
+  "epoch": 3.4965034965034967,
   "eval_steps": 50,
   "global_step": 500,
   "is_hyper_param_search": true,
@@ -9,168 +9,168 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.04,
-      "learning_rate": 4.707876059953676e-06,
-      "loss": 1.9595,
+      "epoch": 0.35,
+      "learning_rate": 5.859444109925402e-05,
+      "loss": 1.497,
       "step": 50
     },
     {
-      "epoch": 0.04,
-      "eval_f1": 0.01705140497086806,
-      "eval_loss": 1.9648141860961914,
-      "eval_runtime": 13.2976,
-      "eval_samples_per_second": 684.86,
-      "eval_steps_per_second": 1.354,
+      "epoch": 0.35,
+      "eval_f1": 0.6734208042191661,
+      "eval_loss": 1.089546799659729,
+      "eval_runtime": 13.0037,
+      "eval_samples_per_second": 700.339,
+      "eval_steps_per_second": 1.384,
       "step": 50
     },
     {
-      "epoch": 0.09,
-      "learning_rate": 9.415752119907352e-06,
-      "loss": 1.9099,
+      "epoch": 0.7,
+      "learning_rate": 5.418884402412214e-05,
+      "loss": 0.5665,
       "step": 100
     },
     {
-      "epoch": 0.09,
-      "eval_f1": 0.07360850081330118,
-      "eval_loss": 2.032604217529297,
-      "eval_runtime": 13.3151,
-      "eval_samples_per_second": 683.961,
-      "eval_steps_per_second": 1.352,
+      "epoch": 0.7,
+      "eval_f1": 0.8012954608718272,
+      "eval_loss": 0.6758251190185547,
+      "eval_runtime": 13.0152,
+      "eval_samples_per_second": 699.72,
+      "eval_steps_per_second": 1.383,
       "step": 100
     },
     {
-      "epoch": 0.13,
-      "learning_rate": 1.4123628179861027e-05,
-      "loss": 1.6735,
+      "epoch": 1.05,
+      "learning_rate": 4.9783246948990256e-05,
+      "loss": 0.5083,
       "step": 150
     },
     {
-      "epoch": 0.13,
-      "eval_f1": 0.38393105234961034,
-      "eval_loss": 1.5790455341339111,
-      "eval_runtime": 12.7623,
-      "eval_samples_per_second": 713.584,
-      "eval_steps_per_second": 1.41,
+      "epoch": 1.05,
+      "eval_f1": 0.7895493614371508,
+      "eval_loss": 0.6960522532463074,
+      "eval_runtime": 12.9373,
+      "eval_samples_per_second": 703.936,
+      "eval_steps_per_second": 1.391,
       "step": 150
     },
     {
-      "epoch": 0.18,
-      "learning_rate": 1.8831504239814704e-05,
-      "loss": 0.9703,
+      "epoch": 1.4,
+      "learning_rate": 4.537764987385838e-05,
+      "loss": 0.3865,
       "step": 200
     },
     {
-      "epoch": 0.18,
-      "eval_f1": 0.7721326246350599,
-      "eval_loss": 0.9168348908424377,
-      "eval_runtime": 12.5558,
-      "eval_samples_per_second": 725.324,
-      "eval_steps_per_second": 1.434,
+      "epoch": 1.4,
+      "eval_f1": 0.8201909522421974,
+      "eval_loss": 0.6086650490760803,
+      "eval_runtime": 12.7238,
+      "eval_samples_per_second": 715.746,
+      "eval_steps_per_second": 1.415,
       "step": 200
     },
     {
-      "epoch": 0.22,
-      "learning_rate": 1.866061944634996e-05,
-      "loss": 0.7516,
+      "epoch": 1.75,
+      "learning_rate": 4.09720527987265e-05,
+      "loss": 0.385,
       "step": 250
     },
     {
-      "epoch": 0.22,
-      "eval_f1": 0.7687755468039507,
-      "eval_loss": 0.7031986713409424,
-      "eval_runtime": 12.6864,
-      "eval_samples_per_second": 717.854,
-      "eval_steps_per_second": 1.419,
+      "epoch": 1.75,
+      "eval_f1": 0.8359538639496097,
+      "eval_loss": 0.50257807970047,
+      "eval_runtime": 12.8233,
+      "eval_samples_per_second": 710.19,
+      "eval_steps_per_second": 1.404,
       "step": 250
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 1.8489734652885216e-05,
-      "loss": 0.6465,
+      "epoch": 2.1,
+      "learning_rate": 3.656645572359462e-05,
+      "loss": 0.357,
       "step": 300
     },
     {
-      "epoch": 0.26,
-      "eval_f1": 0.8144320202800481,
-      "eval_loss": 0.6549465656280518,
-      "eval_runtime": 12.8685,
-      "eval_samples_per_second": 707.697,
-      "eval_steps_per_second": 1.399,
+      "epoch": 2.1,
+      "eval_f1": 0.8294020659039051,
+      "eval_loss": 0.5500179529190063,
+      "eval_runtime": 12.8836,
+      "eval_samples_per_second": 706.867,
+      "eval_steps_per_second": 1.397,
       "step": 300
     },
     {
-      "epoch": 0.31,
-      "learning_rate": 1.8318849859420474e-05,
-      "loss": 0.6678,
+      "epoch": 2.45,
+      "learning_rate": 3.216085864846273e-05,
+      "loss": 0.2337,
       "step": 350
     },
     {
-      "epoch": 0.31,
-      "eval_f1": 0.793206481293857,
-      "eval_loss": 0.7163683176040649,
-      "eval_runtime": 12.9464,
-      "eval_samples_per_second": 703.437,
-      "eval_steps_per_second": 1.39,
+      "epoch": 2.45,
+      "eval_f1": 0.8324393645838855,
+      "eval_loss": 0.5851877927780151,
+      "eval_runtime": 12.9246,
+      "eval_samples_per_second": 704.625,
+      "eval_steps_per_second": 1.393,
       "step": 350
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 1.8147965065955732e-05,
-      "loss": 0.7099,
+      "epoch": 2.8,
+      "learning_rate": 2.775526157333085e-05,
+      "loss": 0.2136,
       "step": 400
     },
     {
-      "epoch": 0.35,
-      "eval_f1": 0.8066754005498158,
-      "eval_loss": 0.7112386226654053,
-      "eval_runtime": 12.8448,
-      "eval_samples_per_second": 709.006,
-      "eval_steps_per_second": 1.401,
+      "epoch": 2.8,
+      "eval_f1": 0.8449098012505971,
+      "eval_loss": 0.5285059809684753,
+      "eval_runtime": 12.7922,
+      "eval_samples_per_second": 711.919,
+      "eval_steps_per_second": 1.407,
       "step": 400
     },
     {
-      "epoch": 0.39,
-      "learning_rate": 1.7977080272490987e-05,
-      "loss": 0.5154,
+      "epoch": 3.15,
+      "learning_rate": 2.3349664498198968e-05,
+      "loss": 0.1926,
       "step": 450
     },
     {
-      "epoch": 0.39,
-      "eval_f1": 0.8128681854085146,
-      "eval_loss": 0.6707363724708557,
-      "eval_runtime": 12.7035,
-      "eval_samples_per_second": 716.892,
-      "eval_steps_per_second": 1.417,
+      "epoch": 3.15,
+      "eval_f1": 0.8313597336118761,
+      "eval_loss": 0.6143582463264465,
+      "eval_runtime": 12.8169,
+      "eval_samples_per_second": 710.548,
+      "eval_steps_per_second": 1.404,
       "step": 450
     },
     {
-      "epoch": 0.44,
-      "learning_rate": 1.7806195479026245e-05,
-      "loss": 0.6595,
+      "epoch": 3.5,
+      "learning_rate": 1.894406742306709e-05,
+      "loss": 0.1359,
       "step": 500
     },
     {
-      "epoch": 0.44,
-      "eval_f1": 0.8162388724528657,
-      "eval_loss": 0.6982755064964294,
-      "eval_runtime": 12.7345,
-      "eval_samples_per_second": 715.146,
-      "eval_steps_per_second": 1.413,
+      "epoch": 3.5,
+      "eval_f1": 0.8284272507637513,
+      "eval_loss": 0.7059531211853027,
+      "eval_runtime": 12.7876,
+      "eval_samples_per_second": 712.173,
+      "eval_steps_per_second": 1.408,
       "step": 500
     }
   ],
   "logging_steps": 50,
-  "max_steps": 5710,
+  "max_steps": 715,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 19068185282400.0,
-  "train_batch_size": 4,
+  "total_flos": 178689205892400.0,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 1.8831504239814704e-05,
+    "learning_rate": 5.859444109925402e-05,
     "num_train_epochs": 5,
-    "per_device_train_batch_size": 4,
-    "seed": 32
+    "per_device_train_batch_size": 32,
+    "seed": 11
   }
 }

run-1/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aff63987d55bdb29a46347243764a71254aa8765001a1e61e633d8ed4d1fb8a
+oid sha256:3bf57395bda7f210838beb0e7bc53917d611090e8578e2da74f8aa9db6d5cf24
 size 4728

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aff63987d55bdb29a46347243764a71254aa8765001a1e61e633d8ed4d1fb8a
+oid sha256:3bf57395bda7f210838beb0e7bc53917d611090e8578e2da74f8aa9db6d5cf24
 size 4728