update model with newer version

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d8a1fc823d3ba0d45afc06b1105997300ad38030351b439d85610fa9fa38a75
+oid sha256:a184558aeae2acda43c0b7d8660f3a40620a1e111e2788bf17507bc208a251a9
 size 430935892

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eed95c6f0459164abd29b411040228bdd2d3ab9ef8949ba8423fab4ddfecad90
+oid sha256:753a905dc66bdc951ff46dbc455dcef5a9601f90d54b21507aed3abb36edf4a1
 size 861991482

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4679e8a1fa9d31242916d045260107779d3bf18ad04f68edc0e2a86739d10fbd
+oid sha256:712caa512e2bd3a13bbc3abf2f8c256aa41b420c1f92a596df76f55140a4898f
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73ec52a655b60bd9b36d5e598a010f1db1893ca2657401a77e3d62edf772ae6f
+oid sha256:49ca8cfba8e912830d0bbc9af499faa3ae8fc7cef2c871274f7a0984db09f49f
 size 1064

trainer_state.json

@@ -1,125 +1,121 @@
 {
-  "best_global_step": 1376,
-  "best_metric": 0.008029412478208542,
-  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/case_sensitive/model/checkpoint-1376",
-  "epoch": 5.0,
+  "best_global_step": 899,
+  "best_metric": 0.009147428907454014,
+  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/augment/checkpoint-899",
+  "epoch": 4.0,
   "eval_steps": 500,
-  "global_step": 3440,
+  "global_step": 3596,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0,
-      "eval_f1": 0.010973861448813302,
-      "eval_loss": 2.4063162803649902,
-      "eval_precision": 0.09108980806958722,
-      "eval_recall": 0.10473207318294019,
-      "eval_runtime": 23.7722,
-      "eval_samples_per_second": 139.407,
-      "eval_steps_per_second": 4.375,
+      "eval_f1": 0.012204053693834897,
+      "eval_loss": 2.3885538578033447,
+      "eval_precision": 0.08893940326175621,
+      "eval_recall": 0.08310981045257043,
+      "eval_runtime": 31.2912,
+      "eval_samples_per_second": 139.592,
+      "eval_steps_per_second": 4.378,
       "step": 0
     },
     {
-      "epoch": 0.7267441860465116,
-      "grad_norm": 0.2666139304637909,
-      "learning_rate": 4.996373546511628e-05,
-      "loss": 0.0518,
+      "epoch": 0.5561735261401557,
+      "grad_norm": 0.35977280139923096,
+      "learning_rate": 4.997224694104561e-05,
+      "loss": 0.0524,
       "step": 500
     },
     {
       "epoch": 1.0,
-      "eval_f1": 0.9198586422350306,
-      "eval_loss": 0.010010140016674995,
-      "eval_precision": 0.9136306910527949,
-      "eval_recall": 0.9281047546302535,
-      "eval_runtime": 24.8489,
-      "eval_samples_per_second": 133.366,
-      "eval_steps_per_second": 4.185,
-      "step": 688
+      "eval_f1": 0.9528093236906681,
+      "eval_loss": 0.009147428907454014,
+      "eval_precision": 0.9461252678130232,
+      "eval_recall": 0.9597947571160605,
+      "eval_runtime": 32.73,
+      "eval_samples_per_second": 133.456,
+      "eval_steps_per_second": 4.186,
+      "step": 899
     },
     {
-      "epoch": 1.4534883720930232,
-      "grad_norm": 0.11741874366998672,
-      "learning_rate": 4.992739825581396e-05,
-      "loss": 0.0036,
+      "epoch": 1.1123470522803114,
+      "grad_norm": 0.11930424720048904,
+      "learning_rate": 4.99444382647386e-05,
+      "loss": 0.0043,
       "step": 1000
     },
     {
-      "epoch": 2.0,
-      "eval_f1": 0.9300254346144702,
-      "eval_loss": 0.008029412478208542,
-      "eval_precision": 0.9130114164496201,
-      "eval_recall": 0.9489397415434179,
-      "eval_runtime": 25.1412,
-      "eval_samples_per_second": 131.815,
-      "eval_steps_per_second": 4.137,
-      "step": 1376
-    },
-    {
-      "epoch": 2.1802325581395348,
-      "grad_norm": 0.3377048373222351,
-      "learning_rate": 4.989106104651163e-05,
+      "epoch": 1.668520578420467,
+      "grad_norm": 0.12083794176578522,
+      "learning_rate": 4.991662958843159e-05,
       "loss": 0.0026,
       "step": 1500
     },
     {
-      "epoch": 2.9069767441860463,
-      "grad_norm": 0.2929118573665619,
-      "learning_rate": 4.985472383720931e-05,
-      "loss": 0.0011,
-      "step": 2000
+      "epoch": 2.0,
+      "eval_f1": 0.9591844285588763,
+      "eval_loss": 0.009312924928963184,
+      "eval_precision": 0.9530780078627331,
+      "eval_recall": 0.9657284498000394,
+      "eval_runtime": 32.7085,
+      "eval_samples_per_second": 133.543,
+      "eval_steps_per_second": 4.189,
+      "step": 1798
     },
     {
-      "epoch": 3.0,
-      "eval_f1": 0.9402795283929368,
-      "eval_loss": 0.01124291867017746,
-      "eval_precision": 0.9422754585534239,
-      "eval_recall": 0.9386206401984958,
-      "eval_runtime": 24.8815,
-      "eval_samples_per_second": 133.191,
-      "eval_steps_per_second": 4.18,
-      "step": 2064
+      "epoch": 2.2246941045606228,
+      "grad_norm": 0.012293193489313126,
+      "learning_rate": 4.988882091212459e-05,
+      "loss": 0.0015,
+      "step": 2000
     },
     {
-      "epoch": 3.633720930232558,
-      "grad_norm": 0.002858501160517335,
-      "learning_rate": 4.981838662790698e-05,
-      "loss": 0.0012,
+      "epoch": 2.7808676307007785,
+      "grad_norm": 0.005511277820914984,
+      "learning_rate": 4.986101223581758e-05,
+      "loss": 0.0014,
       "step": 2500
     },
     {
-      "epoch": 4.0,
-      "eval_f1": 0.9321076846329125,
-      "eval_loss": 0.011350538581609726,
-      "eval_precision": 0.9585579423470411,
-      "eval_recall": 0.9092514508781232,
-      "eval_runtime": 24.8844,
-      "eval_samples_per_second": 133.176,
-      "eval_steps_per_second": 4.179,
-      "step": 2752
+      "epoch": 3.0,
+      "eval_f1": 0.9473572974918704,
+      "eval_loss": 0.013561395928263664,
+      "eval_precision": 0.9406871215514148,
+      "eval_recall": 0.9544992839810097,
+      "eval_runtime": 32.7338,
+      "eval_samples_per_second": 133.44,
+      "eval_steps_per_second": 4.185,
+      "step": 2697
     },
     {
-      "epoch": 4.3604651162790695,
-      "grad_norm": 0.3254820704460144,
-      "learning_rate": 4.978204941860465e-05,
-      "loss": 0.001,
+      "epoch": 3.337041156840934,
+      "grad_norm": 0.03908713161945343,
+      "learning_rate": 4.983320355951057e-05,
+      "loss": 0.0014,
       "step": 3000
     },
     {
-      "epoch": 5.0,
-      "eval_f1": 0.9375959937426187,
-      "eval_loss": 0.012251886539161205,
-      "eval_precision": 0.9347887957568951,
-      "eval_recall": 0.9407576824824448,
-      "eval_runtime": 24.848,
-      "eval_samples_per_second": 133.371,
-      "eval_steps_per_second": 4.185,
-      "step": 3440
+      "epoch": 3.89321468298109,
+      "grad_norm": 0.009426549077033997,
+      "learning_rate": 4.980539488320356e-05,
+      "loss": 0.0016,
+      "step": 3500
+    },
+    {
+      "epoch": 4.0,
+      "eval_f1": 0.953040195996418,
+      "eval_loss": 0.010148942470550537,
+      "eval_precision": 0.9406877737118386,
+      "eval_recall": 0.9674011453268957,
+      "eval_runtime": 32.788,
+      "eval_samples_per_second": 133.219,
+      "eval_steps_per_second": 4.178,
+      "step": 3596
     }
   ],
   "logging_steps": 500,
-  "max_steps": 688000,
+  "max_steps": 899000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1000,
   "save_steps": 500,
@@ -144,7 +140,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.874236880509952e+16,
+  "total_flos": 3.006829642867507e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7f8df01d5a1682106eccdf01898b00019477ad3b989dbc3741b897bbcf18bbf
+oid sha256:3b027a28b72c517e30a4a077c26526840f9f8e189cb0dc1de6469ad0645aeeb7
 size 5304