Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07e441f4986c665ba05c1436f3931b684f8a579a6f815d79e1e3ed27cb33e05a
+oid sha256:f456d0ed5f5b7415ebb134592fcc08a7b745e4e3ab9df5aab72bcf1fb2849cdc
 size 598898116

run-2/checkpoint-954/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47be2248f485aff0a35e94afcd646ca83b8673d6cb2ca7d25fd55d7d6e73c92a
+oid sha256:e43d911a8ecdd349e67fd8eadb50aab03d3ecdaf6b68ada27b35e9435ed9b8de
 size 598898116

run-2/checkpoint-954/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f586bbdb940858b5985f8b53c63cc841de46c96a4492542fc261557dd1121a2
+oid sha256:82cf219a11787f54352b5f528ff3e4b5280d8cd4d55302a0da0e3d869088130d
 size 1197886411

run-2/checkpoint-954/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3cb0104f24fd5b1c74b1de8bb979af0837a0c8451e509740f57bcbac9292de81
+oid sha256:0ca29ceaa1ac87639bc24d09680a228e1ce164b6f4aa0b5f1189196aad30d93f
 size 1383

run-2/checkpoint-954/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 954,
-  "best_metric": 0.943495053065245,
+  "best_metric": 0.9451672573240332,
   "best_model_checkpoint": "ModernBERT-base-distilled/run-2/checkpoint-954",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -11,56 +11,53 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 6.1552348136901855,
+      "grad_norm": 7.241408824920654,
       "learning_rate": 1.3354297693920338e-05,
-      "loss": 2.9367,
+      "loss": 3.1039,
       "step": 318
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9012903225806451,
-      "eval_f1": 0.8990202628613445,
-      "eval_loss": 1.5702729225158691,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.5772,
-      "eval_samples_per_second": 176.365,
-      "eval_steps_per_second": 3.698,
+      "eval_accuracy": 0.9058064516129032,
+      "eval_f1": 0.9039413673060775,
+      "eval_loss": 1.5589689016342163,
+      "eval_runtime": 16.8256,
+      "eval_samples_per_second": 184.243,
+      "eval_steps_per_second": 3.863,
       "step": 318
     },
     {
       "epoch": 2.0,
-      "grad_norm": 3.8856751918792725,
+      "grad_norm": 4.324426651000977,
       "learning_rate": 6.6876310272536696e-06,
-      "loss": 1.1765,
+      "loss": 1.1524,
       "step": 636
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9412903225806452,
-      "eval_f1": 0.9402146858730125,
-      "eval_loss": 1.036224126815796,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.3187,
-      "eval_samples_per_second": 178.997,
-      "eval_steps_per_second": 3.753,
+      "eval_accuracy": 0.94,
+      "eval_f1": 0.9386540826458253,
+      "eval_loss": 1.0111862421035767,
+      "eval_runtime": 16.5402,
+      "eval_samples_per_second": 187.422,
+      "eval_steps_per_second": 3.93,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "grad_norm": 2.4737651348114014,
+      "grad_norm": 2.4562952518463135,
       "learning_rate": 2.0964360587002098e-08,
-      "loss": 0.8183,
+      "loss": 0.7791,
       "step": 954
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.944516129032258,
-      "eval_f1": 0.943495053065245,
-      "eval_loss": 0.8997413516044617,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.4751,
-      "eval_samples_per_second": 177.395,
-      "eval_steps_per_second": 3.72,
+      "eval_accuracy": 0.9461290322580646,
+      "eval_f1": 0.9451672573240332,
+      "eval_loss": 0.8669379949569702,
+      "eval_runtime": 16.7212,
+      "eval_samples_per_second": 185.393,
+      "eval_steps_per_second": 3.887,
       "step": 954
     }
   ],
@@ -94,8 +91,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.4894773949849175,
+    "alpha": 0.5616085539831922,
     "num_train_epochs": 3,
-    "temperature": 20
+    "temperature": 14
   }
 }

run-2/checkpoint-954/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d7b3769acf391cea03ef15fd35cf3a1a8ad375b715605c557f3506a2acf1661
+oid sha256:ac0ba789da4d8d3fec25ab5f556d37cdc9662d151e6cf3587f40da34600391df
 size 5905

run-3/checkpoint-318/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcb271e93031ae8b58b911b87fcc3f4c4180a86b75d33df2fb9492973d3fbff4
+oid sha256:f456d0ed5f5b7415ebb134592fcc08a7b745e4e3ab9df5aab72bcf1fb2849cdc
 size 598898116

run-3/checkpoint-318/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5a8626969a07ac753378a4efe9517ee8a28bf0f6570521bd12c99844f5db425
+oid sha256:b932a20b2993303a88772f0eb02d764870588e7d0fb3d96228531f8522c0e74c
 size 1197886411

run-3/checkpoint-318/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60c346c7dfa879ca55dd907360767222bc10e5f61875eced577253aa33298d1a
+oid sha256:c6ab09cf38d9cc45768ad6a2e8380f2fba62e9a29d4d1df08af0eb463f18f7c7
 size 1383

run-3/checkpoint-318/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 318,
-  "best_metric": 0.9011780696610004,
+  "best_metric": 0.840301912678485,
   "best_model_checkpoint": "ModernBERT-base-distilled/run-3/checkpoint-318",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -11,20 +11,19 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 5.834390640258789,
+      "grad_norm": 30.5030517578125,
       "learning_rate": 1.3354297693920338e-05,
-      "loss": 2.6495,
+      "loss": 13.5958,
       "step": 318
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9035483870967742,
-      "eval_f1": 0.9011780696610004,
-      "eval_loss": 1.3007111549377441,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.5576,
-      "eval_samples_per_second": 176.561,
-      "eval_steps_per_second": 3.702,
+      "eval_accuracy": 0.8512903225806452,
+      "eval_f1": 0.840301912678485,
+      "eval_loss": 4.8386430740356445,
+      "eval_runtime": 16.7421,
+      "eval_samples_per_second": 185.162,
+      "eval_steps_per_second": 3.882,
       "step": 318
     }
   ],
@@ -58,8 +57,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6526114010549311,
+    "alpha": 0.009270166823335702,
     "num_train_epochs": 3,
-    "temperature": 19
+    "temperature": 5
   }
 }

run-3/checkpoint-318/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e1f35db374dcdc060b0bc938245b3236dd35b1439b812f5c776c2f2f93e0792
+oid sha256:bfb3c5fcdf213574e08d6db11e83b561a9f50bdee78fb3ae2e8aeae12175b619
 size 5905

runs/Oct11_19-55-54_cd07aeb3aeb4/events.out.tfevents.1760215466.cd07aeb3aeb4.2948.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57c2daa5b8175b293830d0200cf907aedd89aa33ed4e69310d8d97d7fc73923d
-size 14370
+oid sha256:cd3cf32b264bff3341e0c311d9240fb12a345f71e43c7dec01fd6a2ba78ac6bd
+size 15304

runs/Oct11_19-55-54_cd07aeb3aeb4/events.out.tfevents.1760216161.cd07aeb3aeb4.2948.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90b86ff7f94da9a7ca742b19ff9e85d89d0f876837882df45bf6bd4ce4757308
+size 13790

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ac0ba789da4d8d3fec25ab5f556d37cdc9662d151e6cf3587f40da34600391df
+oid sha256:bfb3c5fcdf213574e08d6db11e83b561a9f50bdee78fb3ae2e8aeae12175b619
 size 5905