End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 16.0,
-    "eval_accuracy": 0.8368167202572347,
-    "eval_f1": 0.8891074502089993,
-    "eval_loss": 0.6541090607643127,
-    "eval_precision": 0.8746976294146106,
-    "eval_recall": 0.904,
-    "eval_runtime": 3.562,
+    "epoch": 26.32,
+    "eval_accuracy": 0.8407211759301791,
+    "eval_f1": 0.9107187266849044,
+    "eval_loss": 0.8859658241271973,
+    "eval_precision": 0.9059871350816427,
+    "eval_recall": 0.9155,
+    "eval_runtime": 3.6172,
     "eval_samples": 54,
-    "eval_samples_per_second": 15.16,
-    "eval_steps_per_second": 1.123,
+    "eval_samples_per_second": 14.929,
+    "eval_steps_per_second": 1.106,
     "predict_accuracy": 0.8218373936014088,
     "predict_f1": 0.9038133181559477,
     "predict_loss": 0.6888472437858582,
@@ -17,9 +17,9 @@
     "predict_runtime": 2.0596,
     "predict_samples_per_second": 8.74,
     "predict_steps_per_second": 0.971,
-    "train_loss": 0.43604583740234376,
-    "train_runtime": 307.9087,
+    "train_loss": 0.23488603591918944,
+    "train_runtime": 622.7344,
     "train_samples": 150,
-    "train_samples_per_second": 7.795,
-    "train_steps_per_second": 1.299
+    "train_samples_per_second": 6.423,
+    "train_steps_per_second": 1.606
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 16.0,
-    "eval_accuracy": 0.8368167202572347,
-    "eval_f1": 0.8891074502089993,
-    "eval_loss": 0.6541090607643127,
-    "eval_precision": 0.8746976294146106,
-    "eval_recall": 0.904,
-    "eval_runtime": 3.562,
+    "epoch": 26.32,
+    "eval_accuracy": 0.8407211759301791,
+    "eval_f1": 0.9107187266849044,
+    "eval_loss": 0.8859658241271973,
+    "eval_precision": 0.9059871350816427,
+    "eval_recall": 0.9155,
+    "eval_runtime": 3.6172,
     "eval_samples": 54,
-    "eval_samples_per_second": 15.16,
-    "eval_steps_per_second": 1.123
+    "eval_samples_per_second": 14.929,
+    "eval_steps_per_second": 1.106
 }

runs/Aug24_19-33-30_bernini/events.out.tfevents.1724520821.bernini.27584.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ac730116a233cf6fc8abc046bc02614e8d3a9c01c1e8ca99c117dc08602b9ea
-size 9111
+oid sha256:a68deafa55d8eed676e7f062d3f1cc33b1f4abcda0c9306ce87574dac1f4ca5e
+size 9465

runs/Aug24_19-33-30_bernini/events.out.tfevents.1724521452.bernini.27584.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e230662cdece234002471f217f400afc9e5edf6815c5827b538e0860a40f05e9
+size 512

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 16.0,
-    "train_loss": 0.43604583740234376,
-    "train_runtime": 307.9087,
+    "epoch": 26.32,
+    "train_loss": 0.23488603591918944,
+    "train_runtime": 622.7344,
     "train_samples": 150,
-    "train_samples_per_second": 7.795,
-    "train_steps_per_second": 1.299
+    "train_samples_per_second": 6.423,
+    "train_steps_per_second": 1.606
 }

trainer_state.json

@@ -1,217 +1,157 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 16.0,
-  "global_step": 400,
+  "epoch": 26.31578947368421,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.6092099219108865,
-      "eval_f1": 0.43809958885335776,
-      "eval_loss": 1.2830967903137207,
-      "eval_precision": 0.40328006728343146,
-      "eval_recall": 0.4795,
-      "eval_runtime": 3.5046,
-      "eval_samples_per_second": 15.408,
-      "eval_steps_per_second": 1.141,
-      "step": 25
-    },
-    {
-      "epoch": 2.0,
-      "eval_accuracy": 0.7748047772163528,
-      "eval_f1": 0.7586042065009561,
-      "eval_loss": 0.8177912831306458,
-      "eval_precision": 0.7266483516483516,
-      "eval_recall": 0.7935,
-      "eval_runtime": 3.5776,
-      "eval_samples_per_second": 15.094,
-      "eval_steps_per_second": 1.118,
-      "step": 50
-    },
-    {
-      "epoch": 3.0,
-      "eval_accuracy": 0.7990353697749196,
-      "eval_f1": 0.814344962185899,
-      "eval_loss": 0.6842699646949768,
-      "eval_precision": 0.7951405431157694,
-      "eval_recall": 0.8345,
-      "eval_runtime": 3.5751,
-      "eval_samples_per_second": 15.104,
-      "eval_steps_per_second": 1.119,
-      "step": 75
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.8161460725769407,
-      "eval_f1": 0.8306801736613604,
-      "eval_loss": 0.6316555142402649,
-      "eval_precision": 0.8024231127679403,
-      "eval_recall": 0.861,
-      "eval_runtime": 3.6093,
-      "eval_samples_per_second": 14.962,
-      "eval_steps_per_second": 1.108,
+      "epoch": 2.63,
+      "eval_accuracy": 0.798690858980248,
+      "eval_f1": 0.8287769784172662,
+      "eval_loss": 0.6110699772834778,
+      "eval_precision": 0.7963133640552995,
+      "eval_recall": 0.864,
+      "eval_runtime": 3.5435,
+      "eval_samples_per_second": 15.239,
+      "eval_steps_per_second": 1.129,
       "step": 100
     },
     {
-      "epoch": 5.0,
-      "eval_accuracy": 0.8233807992650436,
-      "eval_f1": 0.8600191754554171,
-      "eval_loss": 0.5964206457138062,
-      "eval_precision": 0.8259668508287292,
-      "eval_recall": 0.897,
-      "eval_runtime": 3.5003,
-      "eval_samples_per_second": 15.427,
-      "eval_steps_per_second": 1.143,
-      "step": 125
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.8207395498392283,
-      "eval_f1": 0.8444552293132735,
-      "eval_loss": 0.6049793362617493,
-      "eval_precision": 0.8203677510608204,
-      "eval_recall": 0.87,
-      "eval_runtime": 3.5639,
-      "eval_samples_per_second": 15.152,
-      "eval_steps_per_second": 1.122,
-      "step": 150
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.8168350941662839,
-      "eval_f1": 0.8474740149867052,
-      "eval_loss": 0.628131091594696,
-      "eval_precision": 0.8203088441740758,
-      "eval_recall": 0.8765,
-      "eval_runtime": 3.5404,
-      "eval_samples_per_second": 15.252,
-      "eval_steps_per_second": 1.13,
-      "step": 175
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.8234956361966008,
-      "eval_f1": 0.8708504967288586,
-      "eval_loss": 0.6227801442146301,
-      "eval_precision": 0.844851904090268,
-      "eval_recall": 0.8985,
-      "eval_runtime": 3.5716,
-      "eval_samples_per_second": 15.119,
-      "eval_steps_per_second": 1.12,
-      "step": 200
-    },
-    {
-      "epoch": 9.0,
+      "epoch": 5.26,
       "eval_accuracy": 0.826596233348645,
-      "eval_f1": 0.8566561207106351,
-      "eval_loss": 0.621345579624176,
-      "eval_precision": 0.8345187292555714,
-      "eval_recall": 0.88,
-      "eval_runtime": 3.5239,
-      "eval_samples_per_second": 15.324,
-      "eval_steps_per_second": 1.135,
-      "step": 225
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.8356683509416628,
-      "eval_f1": 0.8702401164200824,
-      "eval_loss": 0.6172508597373962,
-      "eval_precision": 0.8450306170513424,
-      "eval_recall": 0.897,
-      "eval_runtime": 3.508,
-      "eval_samples_per_second": 15.394,
-      "eval_steps_per_second": 1.14,
-      "step": 250
-    },
-    {
-      "epoch": 11.0,
-      "eval_accuracy": 0.8299265043638034,
-      "eval_f1": 0.8633826741082262,
-      "eval_loss": 0.6476383209228516,
-      "eval_precision": 0.8387553041018387,
-      "eval_recall": 0.8895,
-      "eval_runtime": 3.4928,
-      "eval_samples_per_second": 15.46,
-      "eval_steps_per_second": 1.145,
-      "step": 275
+      "eval_f1": 0.8665358194308145,
+      "eval_loss": 0.5860626697540283,
+      "eval_precision": 0.850674373795761,
+      "eval_recall": 0.883,
+      "eval_runtime": 3.5697,
+      "eval_samples_per_second": 15.127,
+      "eval_steps_per_second": 1.121,
+      "step": 200
     },
     {
-      "epoch": 12.0,
-      "eval_accuracy": 0.838194763435921,
-      "eval_f1": 0.8761018609206659,
-      "eval_loss": 0.6358577609062195,
-      "eval_precision": 0.8584452975047985,
-      "eval_recall": 0.8945,
-      "eval_runtime": 3.4999,
-      "eval_samples_per_second": 15.429,
-      "eval_steps_per_second": 1.143,
+      "epoch": 7.89,
+      "eval_accuracy": 0.8425585668350941,
+      "eval_f1": 0.8826268071551091,
+      "eval_loss": 0.5855879783630371,
+      "eval_precision": 0.865449303219606,
+      "eval_recall": 0.9005,
+      "eval_runtime": 3.5797,
+      "eval_samples_per_second": 15.085,
+      "eval_steps_per_second": 1.117,
       "step": 300
     },
     {
-      "epoch": 13.0,
-      "eval_accuracy": 0.8394579696830501,
-      "eval_f1": 0.8911815278801277,
-      "eval_loss": 0.6469025611877441,
-      "eval_precision": 0.8759053597295993,
-      "eval_recall": 0.907,
-      "eval_runtime": 3.5529,
-      "eval_samples_per_second": 15.199,
-      "eval_steps_per_second": 1.126,
-      "step": 325
+      "epoch": 10.53,
+      "eval_accuracy": 0.8426734037666513,
+      "eval_f1": 0.8897131552917903,
+      "eval_loss": 0.6501617431640625,
+      "eval_precision": 0.8801369863013698,
+      "eval_recall": 0.8995,
+      "eval_runtime": 3.5894,
+      "eval_samples_per_second": 15.044,
+      "eval_steps_per_second": 1.114,
+      "step": 400
     },
     {
-      "epoch": 14.0,
-      "eval_accuracy": 0.8372760679834634,
-      "eval_f1": 0.8879606879606879,
-      "eval_loss": 0.651043176651001,
-      "eval_precision": 0.8729468599033816,
-      "eval_recall": 0.9035,
-      "eval_runtime": 3.5289,
-      "eval_samples_per_second": 15.302,
-      "eval_steps_per_second": 1.133,
-      "step": 350
+      "epoch": 13.16,
+      "learning_rate": 5e-06,
+      "loss": 0.4088,
+      "step": 500
     },
     {
-      "epoch": 15.0,
-      "eval_accuracy": 0.8354386770785485,
-      "eval_f1": 0.8834476003917728,
-      "eval_loss": 0.6554981470108032,
-      "eval_precision": 0.8656429942418427,
-      "eval_recall": 0.902,
-      "eval_runtime": 3.5307,
-      "eval_samples_per_second": 15.294,
-      "eval_steps_per_second": 1.133,
-      "step": 375
+      "epoch": 13.16,
+      "eval_accuracy": 0.8372760679834634,
+      "eval_f1": 0.8959365708622398,
+      "eval_loss": 0.7679007649421692,
+      "eval_precision": 0.888015717092338,
+      "eval_recall": 0.904,
+      "eval_runtime": 3.5844,
+      "eval_samples_per_second": 15.065,
+      "eval_steps_per_second": 1.116,
+      "step": 500
     },
     {
-      "epoch": 16.0,
-      "eval_accuracy": 0.8368167202572347,
-      "eval_f1": 0.8891074502089993,
-      "eval_loss": 0.6541090607643127,
-      "eval_precision": 0.8746976294146106,
+      "epoch": 15.79,
+      "eval_accuracy": 0.8332567753789619,
+      "eval_f1": 0.8928395061728396,
+      "eval_loss": 0.8370674848556519,
+      "eval_precision": 0.8819512195121951,
       "eval_recall": 0.904,
-      "eval_runtime": 3.5967,
-      "eval_samples_per_second": 15.014,
-      "eval_steps_per_second": 1.112,
-      "step": 400
+      "eval_runtime": 3.4921,
+      "eval_samples_per_second": 15.463,
+      "eval_steps_per_second": 1.145,
+      "step": 600
     },
     {
-      "epoch": 16.0,
-      "step": 400,
-      "total_flos": 632652823756800.0,
-      "train_loss": 0.43604583740234376,
-      "train_runtime": 307.9087,
-      "train_samples_per_second": 7.795,
-      "train_steps_per_second": 1.299
+      "epoch": 18.42,
+      "eval_accuracy": 0.8336012861736335,
+      "eval_f1": 0.9036561264822135,
+      "eval_loss": 0.8320176601409912,
+      "eval_precision": 0.89306640625,
+      "eval_recall": 0.9145,
+      "eval_runtime": 3.5735,
+      "eval_samples_per_second": 15.111,
+      "eval_steps_per_second": 1.119,
+      "step": 700
+    },
+    {
+      "epoch": 21.05,
+      "eval_accuracy": 0.8340606338998622,
+      "eval_f1": 0.9051275699777063,
+      "eval_loss": 0.8494298458099365,
+      "eval_precision": 0.8969072164948454,
+      "eval_recall": 0.9135,
+      "eval_runtime": 3.6561,
+      "eval_samples_per_second": 14.77,
+      "eval_steps_per_second": 1.094,
+      "step": 800
+    },
+    {
+      "epoch": 23.68,
+      "eval_accuracy": 0.8385392742305926,
+      "eval_f1": 0.907196029776675,
+      "eval_loss": 0.8700249195098877,
+      "eval_precision": 0.9004926108374385,
+      "eval_recall": 0.914,
+      "eval_runtime": 3.5188,
+      "eval_samples_per_second": 15.346,
+      "eval_steps_per_second": 1.137,
+      "step": 900
+    },
+    {
+      "epoch": 26.32,
+      "learning_rate": 0.0,
+      "loss": 0.061,
+      "step": 1000
+    },
+    {
+      "epoch": 26.32,
+      "eval_accuracy": 0.8407211759301791,
+      "eval_f1": 0.9107187266849044,
+      "eval_loss": 0.8859658241271973,
+      "eval_precision": 0.9059871350816427,
+      "eval_recall": 0.9155,
+      "eval_runtime": 3.6186,
+      "eval_samples_per_second": 14.923,
+      "eval_steps_per_second": 1.105,
+      "step": 1000
+    },
+    {
+      "epoch": 26.32,
+      "step": 1000,
+      "total_flos": 1040713895079936.0,
+      "train_loss": 0.23488603591918944,
+      "train_runtime": 622.7344,
+      "train_samples_per_second": 6.423,
+      "train_steps_per_second": 1.606
     }
   ],
-  "max_steps": 400,
-  "num_train_epochs": 16,
-  "total_flos": 632652823756800.0,
+  "max_steps": 1000,
+  "num_train_epochs": 27,
+  "total_flos": 1040713895079936.0,
   "trial_name": null,
   "trial_params": null
 }