Training in progress, epoch 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:292eddb02fac1edf33401ecc1e251e4d581d48a4091caf3f188aa9500f827259
+oid sha256:1015c872c3fcaf54d15403bac0176b038c57086bcc4726bcbdcd08696c9af344
 size 598898116

run-3/checkpoint-954/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45e7e5e1a277054d0833ad760f0da28f644dbe91e1d428c2bf8f0e201f5c73f5
+oid sha256:7f68f44e2aaee9c9a07da075f21def66d31e94f30bddacb953028441953ff10d
 size 598898116

run-3/checkpoint-954/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9203b64be6fa732de2ed35de7e72f9a4f5eb97c4fb6c9e067668bf401d993980
+oid sha256:b309f7a2253593034f73b870db5123c463a1aea5f5331be9cd85515719b77ee1
 size 1197886411

run-3/checkpoint-954/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be7c13a083190cc7a9ae1e7cb2d1ca5308021de7fcfedc07442691984dcbc458
+oid sha256:99d48937bb4120c41ac86c553a7ed356bd690aa081ea87a74dbae27d72ef1556
 size 1383

run-3/checkpoint-954/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 954,
-  "best_metric": 0.9473542257969847,
+  "best_metric": 0.9401923454508052,
   "best_model_checkpoint": "ModernBERT-base-distilled/run-3/checkpoint-954",
   "epoch": 3.0,
   "eval_steps": 500,
@@ -11,56 +11,53 @@
   "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
     },
     {
       "epoch": 2.0,
-      "grad_norm": 3.2908949851989746,
+      "grad_norm": 24.2105655670166,
       "learning_rate": 6.6876310272536696e-06,
-      "loss": 0.9455,
+      "loss": 3.3328,
       "step": 636
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9438709677419355,
-      "eval_f1": 0.9429587534002278,
-      "eval_loss": 0.8770028948783875,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.715,
-      "eval_samples_per_second": 174.993,
-      "eval_steps_per_second": 3.669,
+      "eval_accuracy": 0.9290322580645162,
+      "eval_f1": 0.9273620820413128,
+      "eval_loss": 2.3744025230407715,
+      "eval_runtime": 17.0116,
+      "eval_samples_per_second": 182.229,
+      "eval_steps_per_second": 3.821,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "grad_norm": 1.6844093799591064,
+      "grad_norm": 14.264106750488281,
       "learning_rate": 2.0964360587002098e-08,
-      "loss": 0.6573,
+      "loss": 1.9833,
       "step": 954
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9480645161290323,
-      "eval_f1": 0.9473542257969847,
-      "eval_loss": 0.7567955851554871,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 19.0534,
-      "eval_samples_per_second": 162.7,
-      "eval_steps_per_second": 3.411,
+      "eval_accuracy": 0.9409677419354838,
+      "eval_f1": 0.9401923454508052,
+      "eval_loss": 2.032017469406128,
+      "eval_runtime": 21.0404,
+      "eval_samples_per_second": 147.336,
+      "eval_steps_per_second": 3.089,
       "step": 954
     }
   ],
@@ -94,8 +91,8 @@
   "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-954/training_args.bin

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

run-4/checkpoint-318/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22c49322943f554baab091395632de4a3db37b22aa3fe9802bf784893f0d9105
+oid sha256:1015c872c3fcaf54d15403bac0176b038c57086bcc4726bcbdcd08696c9af344
 size 598898116

run-4/checkpoint-318/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fe7102f944c73bba137bc4dc0b6c154da823ad0d9e0b334e9adf30b9541fc2e
+oid sha256:0d79728571525d41b843ab8d51aa45e8e7f66feb36dbc6f43e845d3a171b4a97
 size 1197886411

run-4/checkpoint-318/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:02f8c2754fd832f08f5aa75c427c3b3b9fb4462e7f0ad6ca534f5243c52e54a3
+oid sha256:60c346c7dfa879ca55dd907360767222bc10e5f61875eced577253aa33298d1a
 size 1383

run-4/checkpoint-318/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca5b8515073837df3c125ab4451274b1cb9e490d869e06ba0dc919d861680999
+oid sha256:c75430e9735ac2401cc5dd787d1ea32f7e26f6e97fda72fd2cd23a50888689e7
 size 1465

run-4/checkpoint-318/trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 318,
-  "best_metric": 0.9028755903745567,
+  "best_metric": 0.9001187520006838,
   "best_model_checkpoint": "ModernBERT-base-distilled/run-4/checkpoint-318",
   "epoch": 1.0,
   "eval_steps": 500,
@@ -11,27 +11,26 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 7.233837604522705,
-      "learning_rate": 1.6012578616352204e-05,
-      "loss": 3.1627,
+      "grad_norm": 7.909060478210449,
+      "learning_rate": 1.3354297693920338e-05,
+      "loss": 4.7051,
       "step": 318
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9058064516129032,
-      "eval_f1": 0.9028755903745567,
-      "eval_loss": 1.3762449026107788,
-      "eval_model_preparation_time": 0.0029,
-      "eval_runtime": 17.6173,
-      "eval_samples_per_second": 175.963,
-      "eval_steps_per_second": 3.69,
+      "eval_accuracy": 0.9032258064516129,
+      "eval_f1": 0.9001187520006838,
+      "eval_loss": 2.157456159591675,
+      "eval_runtime": 16.7346,
+      "eval_samples_per_second": 185.245,
+      "eval_steps_per_second": 3.884,
       "step": 318
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 954,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 3,
   "save_steps": 500,
   "stateful_callbacks": {
     "EarlyStoppingCallback": {
@@ -58,8 +57,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7276994242754504,
-    "num_train_epochs": 5,
-    "temperature": 9
+    "alpha": 0.47088612493766213,
+    "num_train_epochs": 3,
+    "temperature": 8
   }
 }

run-4/checkpoint-318/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f29a88c2d12f252b8308fe6f8e285da5ba597901bbebe1ee1a84b2c3a45b18b
+oid sha256:defbd072261b8c72729f3cf595bfbc7ee5a2f3c7987f31fc86bb7d55172232da
 size 5905

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

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c811df1635823ab2c528000930f94961d74b3a9c143c44f7ec0a27488dc7a6e5
-size 14370
+oid sha256:36d117a0af09eaadee1a79a4e79ee61eaa5b25f5a6cbdccef718cd59a7ef5bad
+size 15304

runs/Oct11_19-55-54_cd07aeb3aeb4/events.out.tfevents.1760216895.cd07aeb3aeb4.2948.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fdf1edf87fe2f3b9cf12e3f3a1d3811fc10fef36fcf14191208d2f68a0fe803
+size 13790

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfb3c5fcdf213574e08d6db11e83b561a9f50bdee78fb3ae2e8aeae12175b619
+oid sha256:defbd072261b8c72729f3cf595bfbc7ee5a2f3c7987f31fc86bb7d55172232da
 size 5905