End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 13.16,
-    "eval_accuracy": 0.8535829122645843,
-    "eval_f1": 0.9075504610017443,
-    "eval_loss": 0.8330431580543518,
-    "eval_precision": 0.9046199701937406,
-    "eval_recall": 0.9105,
-    "eval_runtime": 2.2353,
+    "epoch": 3.16,
+    "eval_accuracy": 0.7967386311437759,
+    "eval_f1": 0.807738814993954,
+    "eval_loss": 0.7162159085273743,
+    "eval_precision": 0.7822014051522248,
+    "eval_recall": 0.835,
+    "eval_runtime": 2.1313,
     "eval_samples": 54,
-    "eval_samples_per_second": 24.158,
-    "eval_steps_per_second": 1.789,
+    "eval_samples_per_second": 25.337,
+    "eval_steps_per_second": 1.877,
     "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": 0.256378173828125,
-    "train_runtime": 215.2806,
+    "train_loss": 1.0857223510742187,
+    "train_runtime": 75.5805,
     "train_samples": 150,
-    "train_samples_per_second": 9.29,
-    "train_steps_per_second": 2.323
+    "train_samples_per_second": 6.351,
+    "train_steps_per_second": 1.588
 }

eval_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 13.16,
-    "eval_accuracy": 0.8535829122645843,
-    "eval_f1": 0.9075504610017443,
-    "eval_loss": 0.8330431580543518,
-    "eval_precision": 0.9046199701937406,
-    "eval_recall": 0.9105,
-    "eval_runtime": 2.2353,
+    "epoch": 3.16,
+    "eval_accuracy": 0.7967386311437759,
+    "eval_f1": 0.807738814993954,
+    "eval_loss": 0.7162159085273743,
+    "eval_precision": 0.7822014051522248,
+    "eval_recall": 0.835,
+    "eval_runtime": 2.1313,
     "eval_samples": 54,
-    "eval_samples_per_second": 24.158,
-    "eval_steps_per_second": 1.789
+    "eval_samples_per_second": 25.337,
+    "eval_steps_per_second": 1.877
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a199f09773c9ccd3c9ed282276252abdd4fb21595348d9a4616fad33649056d
+oid sha256:df965fc7c0d588b5ca471107fa5d2f73c5e1eb94867e73fb69a2ebc2e50f0eb5
 size 501420883

runs/Aug24_17-20-33_bernini/1724512848.4423606/events.out.tfevents.1724512848.bernini.20428.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b863ef46714d92d8daa45f559ca486782934213bc3c1c372dea5be596969e34d
+size 4666

runs/Aug24_17-20-33_bernini/events.out.tfevents.1724512848.bernini.20428.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccf830463e0de5db049161a702eff31e9c9b376dc57830e6e518f33d5d836831
+size 9989

runs/Aug24_17-20-33_bernini/events.out.tfevents.1724512932.bernini.20428.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4665d1c1952a985b8905dba3a611dd49c4554fdceec82388583af1485ece2ff8
+size 503

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 13.16,
-    "train_loss": 0.256378173828125,
-    "train_runtime": 215.2806,
+    "epoch": 3.16,
+    "train_loss": 1.0857223510742187,
+    "train_runtime": 75.5805,
     "train_samples": 150,
-    "train_samples_per_second": 9.29,
-    "train_steps_per_second": 2.323
+    "train_samples_per_second": 6.351,
+    "train_steps_per_second": 1.588
 }

trainer_state.json

@@ -1,151 +1,169 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 13.157894736842104,
-  "global_step": 500,
+  "epoch": 3.1578947368421053,
+  "global_step": 120,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.32,
-      "eval_accuracy": 0.8087965089572806,
-      "eval_f1": 0.8503675598766895,
-      "eval_loss": 0.616326630115509,
-      "eval_precision": 0.8087505638249888,
-      "eval_recall": 0.8965,
-      "eval_runtime": 2.4397,
-      "eval_samples_per_second": 22.134,
-      "eval_steps_per_second": 1.64,
-      "step": 50
+      "epoch": 0.26,
+      "eval_accuracy": 0.3226917776757005,
+      "eval_f1": 0.07917620137299772,
+      "eval_loss": 1.7343385219573975,
+      "eval_precision": 0.0729957805907173,
+      "eval_recall": 0.0865,
+      "eval_runtime": 2.8916,
+      "eval_samples_per_second": 18.675,
+      "eval_steps_per_second": 1.383,
+      "step": 10
     },
     {
-      "epoch": 2.63,
-      "eval_accuracy": 0.8133899862195683,
-      "eval_f1": 0.8346153846153846,
-      "eval_loss": 0.541599690914154,
-      "eval_precision": 0.8037037037037037,
-      "eval_recall": 0.868,
-      "eval_runtime": 2.2528,
-      "eval_samples_per_second": 23.97,
-      "eval_steps_per_second": 1.776,
-      "step": 100
+      "epoch": 0.53,
+      "eval_accuracy": 0.45372071658245294,
+      "eval_f1": 0.14903357070193285,
+      "eval_loss": 1.4896737337112427,
+      "eval_precision": 0.15165631469979296,
+      "eval_recall": 0.1465,
+      "eval_runtime": 2.1137,
+      "eval_samples_per_second": 25.547,
+      "eval_steps_per_second": 1.892,
+      "step": 20
+    },
+    {
+      "epoch": 0.79,
+      "eval_accuracy": 0.5962333486449242,
+      "eval_f1": 0.4183606557377049,
+      "eval_loss": 1.293913722038269,
+      "eval_precision": 0.3716504854368932,
+      "eval_recall": 0.4785,
+      "eval_runtime": 2.1315,
+      "eval_samples_per_second": 25.334,
+      "eval_steps_per_second": 1.877,
+      "step": 30
     },
     {
-      "epoch": 3.95,
-      "eval_accuracy": 0.8385392742305926,
-      "eval_f1": 0.8659844054580897,
-      "eval_loss": 0.5571682453155518,
-      "eval_precision": 0.8445817490494296,
-      "eval_recall": 0.8885,
-      "eval_runtime": 2.2592,
-      "eval_samples_per_second": 23.902,
-      "eval_steps_per_second": 1.771,
-      "step": 150
+      "epoch": 1.05,
+      "eval_accuracy": 0.6367707854846119,
+      "eval_f1": 0.5357978474496958,
+      "eval_loss": 1.1471035480499268,
+      "eval_precision": 0.5035180299032542,
+      "eval_recall": 0.5725,
+      "eval_runtime": 2.3459,
+      "eval_samples_per_second": 23.019,
+      "eval_steps_per_second": 1.705,
+      "step": 40
     },
     {
-      "epoch": 5.26,
-      "eval_accuracy": 0.8123564538355535,
-      "eval_f1": 0.850608998260005,
-      "eval_loss": 0.7317042350769043,
-      "eval_precision": 0.8457736035590707,
-      "eval_recall": 0.8555,
-      "eval_runtime": 2.2554,
-      "eval_samples_per_second": 23.943,
-      "eval_steps_per_second": 1.774,
-      "step": 200
+      "epoch": 1.32,
+      "eval_accuracy": 0.7382866329811667,
+      "eval_f1": 0.6415929203539823,
+      "eval_loss": 1.002245545387268,
+      "eval_precision": 0.6310444874274661,
+      "eval_recall": 0.6525,
+      "eval_runtime": 2.1375,
+      "eval_samples_per_second": 25.263,
+      "eval_steps_per_second": 1.871,
+      "step": 50
     },
     {
-      "epoch": 6.58,
-      "eval_accuracy": 0.8385392742305926,
-      "eval_f1": 0.8906055320209318,
-      "eval_loss": 0.7219748497009277,
-      "eval_precision": 0.8877297565822156,
-      "eval_recall": 0.8935,
-      "eval_runtime": 2.2283,
-      "eval_samples_per_second": 24.234,
-      "eval_steps_per_second": 1.795,
-      "step": 250
+      "epoch": 1.58,
+      "eval_accuracy": 0.7381717960496096,
+      "eval_f1": 0.6929057337220602,
+      "eval_loss": 0.9246326088905334,
+      "eval_precision": 0.6739130434782609,
+      "eval_recall": 0.713,
+      "eval_runtime": 2.1452,
+      "eval_samples_per_second": 25.172,
+      "eval_steps_per_second": 1.865,
+      "step": 60
     },
     {
-      "epoch": 7.89,
-      "eval_accuracy": 0.8435920992191088,
-      "eval_f1": 0.8914595126753629,
-      "eval_loss": 0.8069794774055481,
-      "eval_precision": 0.8778477944740669,
-      "eval_recall": 0.9055,
-      "eval_runtime": 2.2476,
-      "eval_samples_per_second": 24.025,
-      "eval_steps_per_second": 1.78,
-      "step": 300
+      "epoch": 1.84,
+      "eval_accuracy": 0.7543638033991732,
+      "eval_f1": 0.7477305303392261,
+      "eval_loss": 0.8454749584197998,
+      "eval_precision": 0.7159194876486734,
+      "eval_recall": 0.7825,
+      "eval_runtime": 2.1307,
+      "eval_samples_per_second": 25.344,
+      "eval_steps_per_second": 1.877,
+      "step": 70
     },
     {
-      "epoch": 9.21,
-      "eval_accuracy": 0.8477262287551677,
-      "eval_f1": 0.9048562933597621,
-      "eval_loss": 0.7895194888114929,
-      "eval_precision": 0.8968565815324165,
-      "eval_recall": 0.913,
-      "eval_runtime": 2.3704,
-      "eval_samples_per_second": 22.781,
-      "eval_steps_per_second": 1.687,
-      "step": 350
+      "epoch": 2.11,
+      "eval_accuracy": 0.7816949931097841,
+      "eval_f1": 0.777831094049904,
+      "eval_loss": 0.7844430208206177,
+      "eval_precision": 0.7476937269372693,
+      "eval_recall": 0.8105,
+      "eval_runtime": 2.1317,
+      "eval_samples_per_second": 25.332,
+      "eval_steps_per_second": 1.876,
+      "step": 80
     },
     {
-      "epoch": 10.53,
-      "eval_accuracy": 0.8411805236564079,
-      "eval_f1": 0.8912280701754386,
-      "eval_loss": 0.8168221712112427,
-      "eval_precision": 0.893467336683417,
-      "eval_recall": 0.889,
-      "eval_runtime": 2.4315,
-      "eval_samples_per_second": 22.208,
-      "eval_steps_per_second": 1.645,
-      "step": 400
+      "epoch": 2.37,
+      "eval_accuracy": 0.7456361966008268,
+      "eval_f1": 0.7816750178954902,
+      "eval_loss": 0.794705867767334,
+      "eval_precision": 0.7476038338658147,
+      "eval_recall": 0.819,
+      "eval_runtime": 2.153,
+      "eval_samples_per_second": 25.081,
+      "eval_steps_per_second": 1.858,
+      "step": 90
     },
     {
-      "epoch": 11.84,
-      "eval_accuracy": 0.8520900321543409,
-      "eval_f1": 0.9061264822134387,
-      "eval_loss": 0.8232607841491699,
-      "eval_precision": 0.8955078125,
-      "eval_recall": 0.917,
-      "eval_runtime": 2.2665,
-      "eval_samples_per_second": 23.825,
-      "eval_steps_per_second": 1.765,
-      "step": 450
+      "epoch": 2.63,
+      "eval_accuracy": 0.7959347726228755,
+      "eval_f1": 0.8073572120038721,
+      "eval_loss": 0.7345592975616455,
+      "eval_precision": 0.7823639774859287,
+      "eval_recall": 0.834,
+      "eval_runtime": 2.1241,
+      "eval_samples_per_second": 25.423,
+      "eval_steps_per_second": 1.883,
+      "step": 100
     },
     {
-      "epoch": 13.16,
-      "learning_rate": 0.0,
-      "loss": 0.2564,
-      "step": 500
+      "epoch": 2.89,
+      "eval_accuracy": 0.7953605879650896,
+      "eval_f1": 0.8112430336806398,
+      "eval_loss": 0.7199715375900269,
+      "eval_precision": 0.7870239774330042,
+      "eval_recall": 0.837,
+      "eval_runtime": 2.1537,
+      "eval_samples_per_second": 25.073,
+      "eval_steps_per_second": 1.857,
+      "step": 110
     },
     {
-      "epoch": 13.16,
-      "eval_accuracy": 0.8535829122645843,
-      "eval_f1": 0.9075504610017443,
-      "eval_loss": 0.8330431580543518,
-      "eval_precision": 0.9046199701937406,
-      "eval_recall": 0.9105,
-      "eval_runtime": 2.224,
-      "eval_samples_per_second": 24.281,
-      "eval_steps_per_second": 1.799,
-      "step": 500
+      "epoch": 3.16,
+      "eval_accuracy": 0.7967386311437759,
+      "eval_f1": 0.807738814993954,
+      "eval_loss": 0.7162159085273743,
+      "eval_precision": 0.7822014051522248,
+      "eval_recall": 0.835,
+      "eval_runtime": 2.1423,
+      "eval_samples_per_second": 25.207,
+      "eval_steps_per_second": 1.867,
+      "step": 120
     },
     {
-      "epoch": 13.16,
-      "step": 500,
-      "total_flos": 520356947539968.0,
-      "train_loss": 0.256378173828125,
-      "train_runtime": 215.2806,
-      "train_samples_per_second": 9.29,
-      "train_steps_per_second": 2.323
+      "epoch": 3.16,
+      "step": 120,
+      "total_flos": 124948932691968.0,
+      "train_loss": 1.0857223510742187,
+      "train_runtime": 75.5805,
+      "train_samples_per_second": 6.351,
+      "train_steps_per_second": 1.588
     }
   ],
-  "max_steps": 500,
-  "num_train_epochs": 14,
-  "total_flos": 520356947539968.0,
+  "max_steps": 120,
+  "num_train_epochs": 4,
+  "total_flos": 124948932691968.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ff27db448ecc4f0766af29a31b5c5ce5b719f6114c6d3e3be5126e0327583fa
-size 2991
+oid sha256:e0c211d7dfcd1074643be63d26126032307d1c996190f44cd24fa0cb3fbccfde
+size 2927