Upload folder using huggingface_hub

training_checkpoints/checkpoint-321/adapter_config.json

@@ -21,8 +21,8 @@
   "revision": null,
   "target_modules": [
     "key",
-    "dense",
     "value",
+    "dense",
     "query"
   ],
   "task_type": "SEQ_CLS",

training_checkpoints/checkpoint-321/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e8ce28b8ebc91e689ec80061c241c1abf5a7aa54a7da14a9704cc621a252c5d
+oid sha256:3b6f48a99b8a0776d96bcd18b5d06cd14ec02895eb419fb40b70a83b23757524
 size 104549068

training_checkpoints/checkpoint-321/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c3edaf2813048a74d52f1a069b2dbef9266a5e4febda56c974150e04f7c1c04
+oid sha256:e19c600f29a63e02a875059c8268a06ccdb5740fa8f9f04675aceff318f821a1
 size 21646778

training_checkpoints/checkpoint-321/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e51ba565919d238b384a7aae526690d95a1b78040917688c10a9dfc4fc409eb
+oid sha256:54663e34a28062edb9d6bcb6dd4fd35fda9d8a05a0a75cd6cf78f5e44f84d13e
 size 1064

training_checkpoints/checkpoint-321/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.39668888106795264,
+  "best_metric": 0.3969800602789428,
   "best_model_checkpoint": "legal-bert-lora/checkpoint-321",
   "epoch": 1.9968895800933126,
   "eval_steps": 500,
@@ -10,52 +10,52 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6297443841982959,
-      "eval_f1_macro": 0.274206359888179,
-      "eval_f1_micro": 0.6297443841982959,
-      "eval_f1_weighted": 0.5693924444438366,
-      "eval_loss": 1.3252344131469727,
-      "eval_macro_fpr": 0.04170202326759921,
-      "eval_macro_sensitivity": 0.31095312319919677,
-      "eval_macro_specificity": 0.9716859606487191,
-      "eval_precision": 0.5643117258906628,
-      "eval_precision_macro": 0.2864620870649182,
-      "eval_recall": 0.6297443841982959,
-      "eval_recall_macro": 0.31095312319919677,
-      "eval_runtime": 29.2455,
-      "eval_samples_per_second": 44.143,
-      "eval_steps_per_second": 5.539,
-      "eval_weighted_fpr": 0.0403035413153457,
-      "eval_weighted_sensitivity": 0.6297443841982959,
-      "eval_weighted_specificity": 0.945545025532491,
+      "eval_accuracy": 0.6374903175832688,
+      "eval_f1_macro": 0.3394539856434376,
+      "eval_f1_micro": 0.6374903175832688,
+      "eval_f1_weighted": 0.5765523433851403,
+      "eval_loss": 1.3127473592758179,
+      "eval_macro_fpr": 0.040137717078130394,
+      "eval_macro_sensitivity": 0.4017115530536667,
+      "eval_macro_specificity": 0.9725650393599338,
+      "eval_precision": 0.5466227876047401,
+      "eval_precision_macro": 0.31141306924471374,
+      "eval_recall": 0.6374903175832688,
+      "eval_recall_macro": 0.4017115530536667,
+      "eval_runtime": 44.4128,
+      "eval_samples_per_second": 29.068,
+      "eval_steps_per_second": 3.648,
+      "eval_weighted_fpr": 0.0390325271059216,
+      "eval_weighted_sensitivity": 0.6374903175832688,
+      "eval_weighted_specificity": 0.950985272815738,
       "step": 160
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7312161115414407,
-      "eval_f1_macro": 0.39668888106795264,
-      "eval_f1_micro": 0.7312161115414407,
-      "eval_f1_weighted": 0.6996238601563893,
-      "eval_loss": 0.8870222568511963,
-      "eval_macro_fpr": 0.025654492164758257,
-      "eval_macro_sensitivity": 0.4525442260434048,
-      "eval_macro_specificity": 0.9800113592544849,
-      "eval_precision": 0.687310022738013,
-      "eval_precision_macro": 0.37417463878901125,
-      "eval_recall": 0.7312161115414407,
-      "eval_recall_macro": 0.4525442260434048,
-      "eval_runtime": 31.4769,
-      "eval_samples_per_second": 41.014,
-      "eval_steps_per_second": 5.147,
-      "eval_weighted_fpr": 0.025584310255843103,
-      "eval_weighted_sensitivity": 0.7312161115414407,
-      "eval_weighted_specificity": 0.9689542772758329,
+      "eval_accuracy": 0.7273431448489543,
+      "eval_f1_macro": 0.3969800602789428,
+      "eval_f1_micro": 0.7273431448489543,
+      "eval_f1_weighted": 0.6960627383550935,
+      "eval_loss": 0.8160030841827393,
+      "eval_macro_fpr": 0.026417996388603746,
+      "eval_macro_sensitivity": 0.450135625445356,
+      "eval_macro_specificity": 0.9796396430402219,
+      "eval_precision": 0.6997929823446861,
+      "eval_precision_macro": 0.39453164470926966,
+      "eval_recall": 0.7273431448489543,
+      "eval_recall_macro": 0.450135625445356,
+      "eval_runtime": 34.3148,
+      "eval_samples_per_second": 37.622,
+      "eval_steps_per_second": 4.721,
+      "eval_weighted_fpr": 0.026077937472218107,
+      "eval_weighted_sensitivity": 0.7273431448489543,
+      "eval_weighted_specificity": 0.9672515007543734,
       "step": 321
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1600,
-  "num_train_epochs": 10,
+  "max_steps": 2400,
+  "num_train_epochs": 15,
   "save_steps": 500,
   "total_flos": 2788972434726912.0,
   "trial_name": null,

training_checkpoints/checkpoint-321/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1be05b6bb510587fe353991f6f2836b34a5909cbc5adc58389938bd163228a1f
+oid sha256:36a33fe5941ace0a96599f0a2229d1ce6d7002fe79c5372925862ab481915414
 size 4600