Upload folder using huggingface_hub

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3db702d2e68876cab46976c836710ff38917f1da1c2ec63db4b90aea0f31bfad
+oid sha256:cce933ce5c8a4bde9a0eb0ee13022c4f225cff3988bc6ed55a85890f24088ced
 size 430935892

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c1c9ff4bb5264948e1e60bae8d14def3974dd4e7f386d9137099068800cfbc8
+oid sha256:c66e7ffa6c77b15fb9c4503794a9bc2e14918bef02ddca514bf8efd41f2d31e6
 size 861991482

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2bc70652e6ef170320bcca82afe9a9f4bdf2996075a63ec8a251ac321429afb
+oid sha256:7668a091b4969a9418e6ed28d276de5c8789d5a9b992e0f553d5ee350ad98059
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c7c30b78cec564dd03bf88837e697ae6ef4fcc7a8780b9e87c556a24173894f
+oid sha256:a080ecbbd0c0861ab0de44ac68d5a5f317cdd67d42046c4fbf2e239ffefe4619
 size 1064

trainer_state.json

@@ -1,93 +1,249 @@
 {
-  "best_global_step": 476,
-  "best_metric": 0.010661174543201923,
-  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/augment/checkpoint-476",
-  "epoch": 4.0,
+  "best_global_step": 6085,
+  "best_metric": 0.010878827422857285,
+  "best_model_checkpoint": "projects/PetBERT_annonymisation/data/augment/checkpoint-6085",
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 1904,
+  "global_step": 9736,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0,
-      "eval_f1": 0.035302169832600705,
-      "eval_loss": 2.312753200531006,
-      "eval_precision": 0.09090508818188361,
-      "eval_recall": 0.06752732472754598,
-      "eval_runtime": 12.3558,
-      "eval_samples_per_second": 134.107,
-      "eval_steps_per_second": 4.209,
+      "eval_f1": 0.007529013027610962,
+      "eval_loss": 2.555938720703125,
+      "eval_precision": 0.09265172512259717,
+      "eval_recall": 0.07914537413587709,
+      "eval_runtime": 31.6606,
+      "eval_samples_per_second": 137.963,
+      "eval_steps_per_second": 4.327,
       "step": 0
     },
     {
-      "epoch": 1.0,
-      "eval_f1": 0.8995538111348774,
-      "eval_loss": 0.010661174543201923,
-      "eval_precision": 0.8815916042183172,
-      "eval_recall": 0.9190286419243201,
-      "eval_runtime": 12.59,
-      "eval_samples_per_second": 131.612,
-      "eval_steps_per_second": 4.13,
-      "step": 476
-    },
-    {
-      "epoch": 1.050420168067227,
-      "grad_norm": 0.16209110617637634,
-      "learning_rate": 4.994758403361345e-05,
-      "loss": 0.0477,
+      "epoch": 0.4108463434675431,
+      "grad_norm": 0.5834885835647583,
+      "learning_rate": 4.997949876746097e-05,
+      "loss": 0.0454,
       "step": 500
     },
     {
-      "epoch": 2.0,
-      "eval_f1": 0.9230744441736807,
-      "eval_loss": 0.010812721215188503,
-      "eval_precision": 0.8937084772102274,
-      "eval_recall": 0.9571896846985819,
-      "eval_runtime": 12.587,
-      "eval_samples_per_second": 131.644,
-      "eval_steps_per_second": 4.131,
-      "step": 952
-    },
-    {
-      "epoch": 2.100840336134454,
-      "grad_norm": 0.012455416843295097,
-      "learning_rate": 4.989506302521009e-05,
-      "loss": 0.0056,
+      "epoch": 0.8216926869350862,
+      "grad_norm": 0.2591548264026642,
+      "learning_rate": 4.99589564502876e-05,
+      "loss": 0.0064,
       "step": 1000
     },
     {
-      "epoch": 3.0,
-      "eval_f1": 0.9348757835672437,
-      "eval_loss": 0.01099320687353611,
-      "eval_precision": 0.9214871357949422,
-      "eval_recall": 0.9494909444630193,
-      "eval_runtime": 12.579,
-      "eval_samples_per_second": 131.728,
-      "eval_steps_per_second": 4.134,
-      "step": 1428
-    },
-    {
-      "epoch": 3.1512605042016806,
-      "grad_norm": 0.015093757770955563,
-      "learning_rate": 4.9842542016806724e-05,
-      "loss": 0.0029,
+      "epoch": 1.0,
+      "eval_f1": 0.9464294969230879,
+      "eval_loss": 0.011466185562312603,
+      "eval_precision": 0.9361671319399428,
+      "eval_recall": 0.957253674779859,
+      "eval_runtime": 35.8703,
+      "eval_samples_per_second": 121.772,
+      "eval_steps_per_second": 3.819,
+      "step": 1217
+    },
+    {
+      "epoch": 1.2325390304026294,
+      "grad_norm": 0.15915486216545105,
+      "learning_rate": 4.993841413311422e-05,
+      "loss": 0.0034,
       "step": 1500
     },
+    {
+      "epoch": 1.6433853738701725,
+      "grad_norm": 0.014851076528429985,
+      "learning_rate": 4.991787181594084e-05,
+      "loss": 0.002,
+      "step": 2000
+    },
+    {
+      "epoch": 2.0,
+      "eval_f1": 0.9440709901648987,
+      "eval_loss": 0.012830332852900028,
+      "eval_precision": 0.9421940783229126,
+      "eval_recall": 0.9483736950091484,
+      "eval_runtime": 35.7192,
+      "eval_samples_per_second": 122.287,
+      "eval_steps_per_second": 3.835,
+      "step": 2434
+    },
+    {
+      "epoch": 2.0542317173377156,
+      "grad_norm": 0.0038151259068399668,
+      "learning_rate": 4.9897329498767464e-05,
+      "loss": 0.0018,
+      "step": 2500
+    },
+    {
+      "epoch": 2.4650780608052587,
+      "grad_norm": 0.02055566944181919,
+      "learning_rate": 4.987678718159409e-05,
+      "loss": 0.0017,
+      "step": 3000
+    },
+    {
+      "epoch": 2.875924404272802,
+      "grad_norm": 0.0036607636138796806,
+      "learning_rate": 4.985624486442071e-05,
+      "loss": 0.0015,
+      "step": 3500
+    },
+    {
+      "epoch": 3.0,
+      "eval_f1": 0.9531088085786098,
+      "eval_loss": 0.011053766123950481,
+      "eval_precision": 0.9507374837523845,
+      "eval_recall": 0.9564085301689659,
+      "eval_runtime": 35.6019,
+      "eval_samples_per_second": 122.69,
+      "eval_steps_per_second": 3.848,
+      "step": 3651
+    },
+    {
+      "epoch": 3.286770747740345,
+      "grad_norm": 1.1118292808532715,
+      "learning_rate": 4.983570254724733e-05,
+      "loss": 0.0016,
+      "step": 4000
+    },
+    {
+      "epoch": 3.6976170912078885,
+      "grad_norm": 0.0022786306217312813,
+      "learning_rate": 4.9815160230073956e-05,
+      "loss": 0.0013,
+      "step": 4500
+    },
     {
       "epoch": 4.0,
-      "eval_f1": 0.9105180723758415,
-      "eval_loss": 0.012333991006016731,
-      "eval_precision": 0.8738638072175106,
-      "eval_recall": 0.9524189376430204,
-      "eval_runtime": 12.6079,
-      "eval_samples_per_second": 131.426,
-      "eval_steps_per_second": 4.124,
-      "step": 1904
+      "eval_f1": 0.9505122264666478,
+      "eval_loss": 0.016675475984811783,
+      "eval_precision": 0.9546462471072065,
+      "eval_recall": 0.9498644595715678,
+      "eval_runtime": 36.2085,
+      "eval_samples_per_second": 120.635,
+      "eval_steps_per_second": 3.784,
+      "step": 4868
+    },
+    {
+      "epoch": 4.108463434675431,
+      "grad_norm": 0.006485091522336006,
+      "learning_rate": 4.979461791290058e-05,
+      "loss": 0.001,
+      "step": 5000
+    },
+    {
+      "epoch": 4.519309778142975,
+      "grad_norm": 0.013346013613045216,
+      "learning_rate": 4.9774075595727196e-05,
+      "loss": 0.001,
+      "step": 5500
+    },
+    {
+      "epoch": 4.930156121610517,
+      "grad_norm": 0.007624845020473003,
+      "learning_rate": 4.975353327855382e-05,
+      "loss": 0.001,
+      "step": 6000
+    },
+    {
+      "epoch": 5.0,
+      "eval_f1": 0.9625014075305048,
+      "eval_loss": 0.010878827422857285,
+      "eval_precision": 0.9639181826511514,
+      "eval_recall": 0.9620998228374574,
+      "eval_runtime": 36.0383,
+      "eval_samples_per_second": 121.204,
+      "eval_steps_per_second": 3.802,
+      "step": 6085
+    },
+    {
+      "epoch": 5.341002465078061,
+      "grad_norm": 0.11334934085607529,
+      "learning_rate": 4.973299096138044e-05,
+      "loss": 0.0009,
+      "step": 6500
+    },
+    {
+      "epoch": 5.751848808545604,
+      "grad_norm": 0.011706724762916565,
+      "learning_rate": 4.971244864420707e-05,
+      "loss": 0.0012,
+      "step": 7000
+    },
+    {
+      "epoch": 6.0,
+      "eval_f1": 0.9523607014366674,
+      "eval_loss": 0.015461287461221218,
+      "eval_precision": 0.9474535885425385,
+      "eval_recall": 0.9589480349926223,
+      "eval_runtime": 41.6283,
+      "eval_samples_per_second": 104.929,
+      "eval_steps_per_second": 3.291,
+      "step": 7302
+    },
+    {
+      "epoch": 6.162695152013147,
+      "grad_norm": 0.0012815777445212007,
+      "learning_rate": 4.9691906327033696e-05,
+      "loss": 0.001,
+      "step": 7500
+    },
+    {
+      "epoch": 6.57354149548069,
+      "grad_norm": 0.038588881492614746,
+      "learning_rate": 4.967136400986032e-05,
+      "loss": 0.0008,
+      "step": 8000
+    },
+    {
+      "epoch": 6.9843878389482335,
+      "grad_norm": 0.37063008546829224,
+      "learning_rate": 4.9650821692686935e-05,
+      "loss": 0.0009,
+      "step": 8500
+    },
+    {
+      "epoch": 7.0,
+      "eval_f1": 0.9536398690810983,
+      "eval_loss": 0.01187444943934679,
+      "eval_precision": 0.9546899807371699,
+      "eval_recall": 0.9538560273395369,
+      "eval_runtime": 41.4124,
+      "eval_samples_per_second": 105.476,
+      "eval_steps_per_second": 3.308,
+      "step": 8519
+    },
+    {
+      "epoch": 7.395234182415776,
+      "grad_norm": 0.25895071029663086,
+      "learning_rate": 4.963027937551356e-05,
+      "loss": 0.0005,
+      "step": 9000
+    },
+    {
+      "epoch": 7.80608052588332,
+      "grad_norm": 0.00598322506994009,
+      "learning_rate": 4.960973705834018e-05,
+      "loss": 0.0009,
+      "step": 9500
+    },
+    {
+      "epoch": 8.0,
+      "eval_f1": 0.9573662673101133,
+      "eval_loss": 0.01274662185460329,
+      "eval_precision": 0.9657258982469304,
+      "eval_recall": 0.9512914903353223,
+      "eval_runtime": 41.6882,
+      "eval_samples_per_second": 104.778,
+      "eval_steps_per_second": 3.286,
+      "step": 9736
     }
   ],
   "logging_steps": 500,
-  "max_steps": 476000,
+  "max_steps": 1217000,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1000,
   "save_steps": 500,
@@ -112,7 +268,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.5921584203628544e+16,
+  "total_flos": 8.138070245258035e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null