End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 20.0,
-    "eval_accuracy": 0.8434772622875517,
-    "eval_f1": 0.8826979472140762,
-    "eval_loss": 0.5468625426292419,
-    "eval_precision": 0.8632887189292543,
-    "eval_recall": 0.903,
-    "eval_runtime": 3.697,
+    "epoch": 10.53,
+    "eval_accuracy": 0.833371612310519,
+    "eval_f1": 0.8794946550048591,
+    "eval_loss": 0.5784164071083069,
+    "eval_precision": 0.8553875236294896,
+    "eval_recall": 0.905,
+    "eval_runtime": 3.53,
     "eval_samples": 54,
-    "eval_samples_per_second": 14.607,
-    "eval_steps_per_second": 1.623,
+    "eval_samples_per_second": 15.297,
+    "eval_steps_per_second": 1.133,
     "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": 0.4636675516764323,
-    "train_runtime": 318.7136,
+    "train_loss": 0.50620361328125,
+    "train_runtime": 250.4274,
     "train_samples": 150,
-    "train_samples_per_second": 9.413,
-    "train_steps_per_second": 0.941
+    "train_samples_per_second": 6.389,
+    "train_steps_per_second": 1.597
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 20.0,
-    "eval_accuracy": 0.8434772622875517,
-    "eval_f1": 0.8826979472140762,
-    "eval_loss": 0.5468625426292419,
-    "eval_precision": 0.8632887189292543,
-    "eval_recall": 0.903,
-    "eval_runtime": 3.697,
+    "epoch": 10.53,
+    "eval_accuracy": 0.833371612310519,
+    "eval_f1": 0.8794946550048591,
+    "eval_loss": 0.5784164071083069,
+    "eval_precision": 0.8553875236294896,
+    "eval_recall": 0.905,
+    "eval_runtime": 3.53,
     "eval_samples": 54,
-    "eval_samples_per_second": 14.607,
-    "eval_steps_per_second": 1.623
+    "eval_samples_per_second": 15.297,
+    "eval_steps_per_second": 1.133
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6676119aed64fb98a2529aadd0eb6d8a586dc32bcd9cdd67336343409ea44e02
+oid sha256:09c539b37ae0e9a660c60e3ec980a33949184bb678cbc4ce1034cc7f63cfd9c6
 size 501420883

runs/Aug24_19-17-02_bernini/1724519834.0999153/events.out.tfevents.1724519834.bernini.2498.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f2f29a3eb043f87b51037aeef3b2bba41943231865c71223524ff532d67e558
+size 4665

runs/Aug24_19-17-02_bernini/events.out.tfevents.1724519834.bernini.2498.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3aead96a37123fe126d5f594e9261d05cd21fc383115bd0d934c104e5cee4e7
+size 11945

runs/Aug24_19-17-02_bernini/events.out.tfevents.1724520093.bernini.2498.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a691246cd5d6024bc0eba5baa25a3c7326465903a0736478eda04b748fe72ac
+size 512

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.4636675516764323,
-    "train_runtime": 318.7136,
+    "epoch": 10.53,
+    "train_loss": 0.50620361328125,
+    "train_runtime": 250.4274,
     "train_samples": 150,
-    "train_samples_per_second": 9.413,
-    "train_steps_per_second": 0.941
+    "train_samples_per_second": 6.389,
+    "train_steps_per_second": 1.597
 }

trainer_state.json

@@ -1,169 +1,217 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
-  "global_step": 300,
+  "epoch": 10.526315789473685,
+  "global_step": 400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.67,
-      "eval_accuracy": 0.6553743683968765,
-      "eval_f1": 0.49752416882810657,
-      "eval_loss": 1.2105501890182495,
-      "eval_precision": 0.4707719767960732,
-      "eval_recall": 0.5275,
-      "eval_runtime": 3.73,
-      "eval_samples_per_second": 14.477,
-      "eval_steps_per_second": 1.609,
+      "epoch": 0.66,
+      "eval_accuracy": 0.5721175930179145,
+      "eval_f1": 0.34372003835091086,
+      "eval_loss": 1.351142406463623,
+      "eval_precision": 0.3301104972375691,
+      "eval_recall": 0.3585,
+      "eval_runtime": 3.6101,
+      "eval_samples_per_second": 14.958,
+      "eval_steps_per_second": 1.108,
       "step": 25
     },
     {
-      "epoch": 3.33,
-      "eval_accuracy": 0.7687184198438217,
-      "eval_f1": 0.777563793933558,
-      "eval_loss": 0.7854474186897278,
-      "eval_precision": 0.7497678737233054,
-      "eval_recall": 0.8075,
-      "eval_runtime": 3.7553,
-      "eval_samples_per_second": 14.38,
-      "eval_steps_per_second": 1.598,
+      "epoch": 1.32,
+      "eval_accuracy": 0.7614836931557188,
+      "eval_f1": 0.7229437229437228,
+      "eval_loss": 0.905920684337616,
+      "eval_precision": 0.6964782205746061,
+      "eval_recall": 0.7515,
+      "eval_runtime": 3.5063,
+      "eval_samples_per_second": 15.401,
+      "eval_steps_per_second": 1.141,
       "step": 50
     },
     {
-      "epoch": 5.0,
-      "eval_accuracy": 0.8141938447404685,
-      "eval_f1": 0.8184898354307841,
-      "eval_loss": 0.6001709699630737,
-      "eval_precision": 0.7931519699812383,
-      "eval_recall": 0.8455,
-      "eval_runtime": 3.7036,
-      "eval_samples_per_second": 14.58,
-      "eval_steps_per_second": 1.62,
+      "epoch": 1.97,
+      "eval_accuracy": 0.7796279283417548,
+      "eval_f1": 0.7946449916327994,
+      "eval_loss": 0.7163704633712769,
+      "eval_precision": 0.7613376087952359,
+      "eval_recall": 0.831,
+      "eval_runtime": 3.5828,
+      "eval_samples_per_second": 15.072,
+      "eval_steps_per_second": 1.116,
       "step": 75
     },
     {
-      "epoch": 6.67,
-      "eval_accuracy": 0.7781350482315113,
-      "eval_f1": 0.827977315689981,
-      "eval_loss": 0.6523196697235107,
-      "eval_precision": 0.7849462365591398,
-      "eval_recall": 0.876,
-      "eval_runtime": 3.7641,
-      "eval_samples_per_second": 14.346,
-      "eval_steps_per_second": 1.594,
+      "epoch": 2.63,
+      "eval_accuracy": 0.799265043638034,
+      "eval_f1": 0.8249158249158249,
+      "eval_loss": 0.6392571926116943,
+      "eval_precision": 0.7947173308619092,
+      "eval_recall": 0.8575,
+      "eval_runtime": 3.5968,
+      "eval_samples_per_second": 15.013,
+      "eval_steps_per_second": 1.112,
       "step": 100
     },
     {
-      "epoch": 8.33,
-      "eval_accuracy": 0.8354386770785485,
-      "eval_f1": 0.8442622950819673,
-      "eval_loss": 0.5189960598945618,
-      "eval_precision": 0.8151769087523277,
-      "eval_recall": 0.8755,
-      "eval_runtime": 3.8049,
-      "eval_samples_per_second": 14.192,
-      "eval_steps_per_second": 1.577,
+      "epoch": 3.29,
+      "eval_accuracy": 0.8104042259990814,
+      "eval_f1": 0.8409859835669407,
+      "eval_loss": 0.5755508542060852,
+      "eval_precision": 0.813844714686623,
+      "eval_recall": 0.87,
+      "eval_runtime": 3.5975,
+      "eval_samples_per_second": 15.011,
+      "eval_steps_per_second": 1.112,
       "step": 125
     },
     {
-      "epoch": 10.0,
-      "eval_accuracy": 0.8338309600367478,
-      "eval_f1": 0.8588007736943907,
-      "eval_loss": 0.5064252018928528,
-      "eval_precision": 0.8314606741573034,
-      "eval_recall": 0.888,
-      "eval_runtime": 3.7776,
-      "eval_samples_per_second": 14.295,
-      "eval_steps_per_second": 1.588,
+      "epoch": 3.95,
+      "eval_accuracy": 0.8323380799265043,
+      "eval_f1": 0.8506134231416886,
+      "eval_loss": 0.5508233308792114,
+      "eval_precision": 0.8196569309225776,
+      "eval_recall": 0.884,
+      "eval_runtime": 3.6045,
+      "eval_samples_per_second": 14.981,
+      "eval_steps_per_second": 1.11,
       "step": 150
     },
     {
-      "epoch": 11.67,
-      "eval_accuracy": 0.8344051446945338,
-      "eval_f1": 0.8693320331545589,
-      "eval_loss": 0.5342020392417908,
-      "eval_precision": 0.8482397716460514,
-      "eval_recall": 0.8915,
-      "eval_runtime": 3.753,
-      "eval_samples_per_second": 14.388,
-      "eval_steps_per_second": 1.599,
+      "epoch": 4.61,
+      "eval_accuracy": 0.8327974276527331,
+      "eval_f1": 0.8600435097897026,
+      "eval_loss": 0.5458412170410156,
+      "eval_precision": 0.8324754328497894,
+      "eval_recall": 0.8895,
+      "eval_runtime": 3.5309,
+      "eval_samples_per_second": 15.294,
+      "eval_steps_per_second": 1.133,
       "step": 175
     },
     {
-      "epoch": 13.33,
-      "eval_accuracy": 0.8200505282498851,
-      "eval_f1": 0.8703071672354948,
-      "eval_loss": 0.5538159012794495,
-      "eval_precision": 0.8491912464319695,
-      "eval_recall": 0.8925,
-      "eval_runtime": 3.8395,
-      "eval_samples_per_second": 14.064,
-      "eval_steps_per_second": 1.563,
+      "epoch": 5.26,
+      "eval_accuracy": 0.826596233348645,
+      "eval_f1": 0.8491160087188181,
+      "eval_loss": 0.5740342736244202,
+      "eval_precision": 0.8233912635039925,
+      "eval_recall": 0.8765,
+      "eval_runtime": 3.622,
+      "eval_samples_per_second": 14.909,
+      "eval_steps_per_second": 1.104,
       "step": 200
     },
     {
-      "epoch": 15.0,
-      "eval_accuracy": 0.8348644924207625,
-      "eval_f1": 0.8777398928397467,
-      "eval_loss": 0.5335590243339539,
-      "eval_precision": 0.855650522317189,
-      "eval_recall": 0.901,
-      "eval_runtime": 3.7278,
-      "eval_samples_per_second": 14.486,
-      "eval_steps_per_second": 1.61,
+      "epoch": 5.92,
+      "eval_accuracy": 0.8361276986678916,
+      "eval_f1": 0.8709914320685433,
+      "eval_loss": 0.5719187259674072,
+      "eval_precision": 0.8532374100719424,
+      "eval_recall": 0.8895,
+      "eval_runtime": 3.5659,
+      "eval_samples_per_second": 15.143,
+      "eval_steps_per_second": 1.122,
       "step": 225
     },
     {
-      "epoch": 16.67,
-      "eval_accuracy": 0.8385392742305926,
-      "eval_f1": 0.8764648437499999,
-      "eval_loss": 0.5464726686477661,
-      "eval_precision": 0.8563931297709924,
-      "eval_recall": 0.8975,
-      "eval_runtime": 3.7679,
-      "eval_samples_per_second": 14.331,
-      "eval_steps_per_second": 1.592,
+      "epoch": 6.58,
+      "eval_accuracy": 0.8263665594855305,
+      "eval_f1": 0.8736131210805596,
+      "eval_loss": 0.5435599684715271,
+      "eval_precision": 0.8438956197576887,
+      "eval_recall": 0.9055,
+      "eval_runtime": 3.5266,
+      "eval_samples_per_second": 15.312,
+      "eval_steps_per_second": 1.134,
       "step": 250
     },
     {
-      "epoch": 18.33,
-      "eval_accuracy": 0.8439366100137804,
-      "eval_f1": 0.8787509148572822,
-      "eval_loss": 0.5402917265892029,
-      "eval_precision": 0.8580276322058122,
-      "eval_recall": 0.9005,
-      "eval_runtime": 3.7992,
-      "eval_samples_per_second": 14.214,
-      "eval_steps_per_second": 1.579,
+      "epoch": 7.24,
+      "eval_accuracy": 0.8290078089113458,
+      "eval_f1": 0.8783914728682171,
+      "eval_loss": 0.5714461207389832,
+      "eval_precision": 0.8519736842105263,
+      "eval_recall": 0.9065,
+      "eval_runtime": 3.6124,
+      "eval_samples_per_second": 14.948,
+      "eval_steps_per_second": 1.107,
       "step": 275
     },
     {
-      "epoch": 20.0,
-      "eval_accuracy": 0.8434772622875517,
-      "eval_f1": 0.8826979472140762,
-      "eval_loss": 0.5468625426292419,
-      "eval_precision": 0.8632887189292543,
-      "eval_recall": 0.903,
-      "eval_runtime": 3.7775,
-      "eval_samples_per_second": 14.295,
-      "eval_steps_per_second": 1.588,
+      "epoch": 7.89,
+      "eval_accuracy": 0.8280891134588884,
+      "eval_f1": 0.8791048406713695,
+      "eval_loss": 0.5853330492973328,
+      "eval_precision": 0.8559924206537186,
+      "eval_recall": 0.9035,
+      "eval_runtime": 3.6008,
+      "eval_samples_per_second": 14.997,
+      "eval_steps_per_second": 1.111,
       "step": 300
     },
     {
-      "epoch": 20.0,
-      "step": 300,
-      "total_flos": 790816029696000.0,
-      "train_loss": 0.4636675516764323,
-      "train_runtime": 318.7136,
-      "train_samples_per_second": 9.413,
-      "train_steps_per_second": 0.941
+      "epoch": 8.55,
+      "eval_accuracy": 0.8389986219568213,
+      "eval_f1": 0.8807785888077859,
+      "eval_loss": 0.570177435874939,
+      "eval_precision": 0.8578199052132701,
+      "eval_recall": 0.905,
+      "eval_runtime": 3.5637,
+      "eval_samples_per_second": 15.153,
+      "eval_steps_per_second": 1.122,
+      "step": 325
+    },
+    {
+      "epoch": 9.21,
+      "eval_accuracy": 0.8418695452457511,
+      "eval_f1": 0.8775261748234722,
+      "eval_loss": 0.5666728019714355,
+      "eval_precision": 0.8552444233507357,
+      "eval_recall": 0.901,
+      "eval_runtime": 3.5809,
+      "eval_samples_per_second": 15.08,
+      "eval_steps_per_second": 1.117,
+      "step": 350
+    },
+    {
+      "epoch": 9.87,
+      "eval_accuracy": 0.8338309600367478,
+      "eval_f1": 0.8786773644541697,
+      "eval_loss": 0.5793057084083557,
+      "eval_precision": 0.8551822053951728,
+      "eval_recall": 0.9035,
+      "eval_runtime": 3.5472,
+      "eval_samples_per_second": 15.223,
+      "eval_steps_per_second": 1.128,
+      "step": 375
+    },
+    {
+      "epoch": 10.53,
+      "eval_accuracy": 0.833371612310519,
+      "eval_f1": 0.8794946550048591,
+      "eval_loss": 0.5784164071083069,
+      "eval_precision": 0.8553875236294896,
+      "eval_recall": 0.905,
+      "eval_runtime": 3.5919,
+      "eval_samples_per_second": 15.034,
+      "eval_steps_per_second": 1.114,
+      "step": 400
+    },
+    {
+      "epoch": 10.53,
+      "step": 400,
+      "total_flos": 416496442306560.0,
+      "train_loss": 0.50620361328125,
+      "train_runtime": 250.4274,
+      "train_samples_per_second": 6.389,
+      "train_steps_per_second": 1.597
     }
   ],
-  "max_steps": 300,
-  "num_train_epochs": 20,
-  "total_flos": 790816029696000.0,
+  "max_steps": 400,
+  "num_train_epochs": 11,
+  "total_flos": 416496442306560.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad5c6ae451d7fcc096ada600794b9529838382a07bad9f882b8f80b359b5b02b
+oid sha256:1b9738bcfd98ccbf71720f6b0ac66e4c20f1ded32caa28f7663edc931468381e
 size 2927