Training in progress, step 3000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8b669e1bd7ff55a9121797a8e4a756a52cfbba25a8136fcf2b7646dfe31f152
+oid sha256:eed8dbca2945508ed4ac89d4bd5cb30c0dfef2c41a89fb1da33d35d9f95d297f
 size 268290900

run-0/checkpoint-3000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70a66fd616aed35b39bc6bb24990cc3e1133adcba9f957399b18b8e0b7fe0ac1
+oid sha256:eed8dbca2945508ed4ac89d4bd5cb30c0dfef2c41a89fb1da33d35d9f95d297f
 size 268290900

run-0/checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:449551f59bff91ad71ee3cee3bc164d266215234ab3e197650a1bf6f4451ccbe
+oid sha256:051d15c448b0310509870729e149f9764897cb069c35beb10ac3e570e876e589
 size 536643898

run-0/checkpoint-3000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9548b29c3da5cefcad3a24e99f04bf8fb9b742e0c4db1bfc7dbdc86bbedf4f42
+oid sha256:9359efd64229560bd90e80fe6ffa6de38090f6b7f033862212653f3ccf150d93
 size 1064

run-0/checkpoint-3000/trainer_state.json

@@ -10,132 +10,132 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5938709677419355,
-      "eval_loss": 0.19997140765190125,
-      "eval_runtime": 5.5473,
-      "eval_samples_per_second": 558.834,
-      "eval_steps_per_second": 11.717,
+      "eval_accuracy": 0.6187096774193548,
+      "eval_loss": 0.2273363471031189,
+      "eval_runtime": 5.4212,
+      "eval_samples_per_second": 571.832,
+      "eval_steps_per_second": 11.99,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5329213738441467,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3229,
+      "grad_norm": 0.5702099204063416,
+      "learning_rate": 1.7141223556317898e-05,
+      "loss": 0.3672,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8238709677419355,
-      "eval_loss": 0.09559512138366699,
-      "eval_runtime": 5.7978,
-      "eval_samples_per_second": 534.683,
-      "eval_steps_per_second": 11.211,
+      "eval_accuracy": 0.8416129032258064,
+      "eval_loss": 0.10179296880960464,
+      "eval_runtime": 5.7619,
+      "eval_samples_per_second": 538.015,
+      "eval_steps_per_second": 11.281,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8812903225806452,
-      "eval_loss": 0.06311405450105667,
-      "eval_runtime": 5.4988,
-      "eval_samples_per_second": 563.759,
-      "eval_steps_per_second": 11.821,
+      "eval_accuracy": 0.8906451612903226,
+      "eval_loss": 0.06358441710472107,
+      "eval_runtime": 5.4704,
+      "eval_samples_per_second": 566.684,
+      "eval_steps_per_second": 11.882,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.44504690170288086,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1111,
+      "grad_norm": 0.48276859521865845,
+      "learning_rate": 1.4282447112635793e-05,
+      "loss": 0.1191,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9006451612903226,
-      "eval_loss": 0.048286207020282745,
-      "eval_runtime": 5.5577,
-      "eval_samples_per_second": 557.787,
-      "eval_steps_per_second": 11.696,
+      "eval_accuracy": 0.905483870967742,
+      "eval_loss": 0.04695529490709305,
+      "eval_runtime": 5.4589,
+      "eval_samples_per_second": 567.879,
+      "eval_steps_per_second": 11.907,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3021944761276245,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0699,
+      "grad_norm": 0.3197612166404724,
+      "learning_rate": 1.1423670668953687e-05,
+      "loss": 0.0712,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9103225806451613,
-      "eval_loss": 0.03919174522161484,
-      "eval_runtime": 5.6898,
-      "eval_samples_per_second": 544.834,
-      "eval_steps_per_second": 11.424,
+      "eval_accuracy": 0.9167741935483871,
+      "eval_loss": 0.03740057349205017,
+      "eval_runtime": 5.721,
+      "eval_samples_per_second": 541.86,
+      "eval_steps_per_second": 11.362,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9132258064516129,
-      "eval_loss": 0.03407873958349228,
-      "eval_runtime": 5.5185,
-      "eval_samples_per_second": 561.746,
-      "eval_steps_per_second": 11.779,
+      "eval_accuracy": 0.9241935483870968,
+      "eval_loss": 0.032310228794813156,
+      "eval_runtime": 5.4558,
+      "eval_samples_per_second": 568.208,
+      "eval_steps_per_second": 11.914,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.2565140724182129,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0542,
+      "grad_norm": 0.2629503309726715,
+      "learning_rate": 8.564894225271586e-06,
+      "loss": 0.0539,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9229032258064516,
-      "eval_loss": 0.03131110221147537,
-      "eval_runtime": 5.6225,
-      "eval_samples_per_second": 551.352,
-      "eval_steps_per_second": 11.561,
+      "eval_accuracy": 0.9283870967741935,
+      "eval_loss": 0.029569072648882866,
+      "eval_runtime": 5.5263,
+      "eval_samples_per_second": 560.95,
+      "eval_steps_per_second": 11.762,
       "step": 2226
     },
     {
       "epoch": 7.861635220125786,
-      "grad_norm": 0.27681249380111694,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0468,
+      "grad_norm": 0.29348886013031006,
+      "learning_rate": 5.706117781589479e-06,
+      "loss": 0.046,
       "step": 2500
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9270967741935484,
-      "eval_loss": 0.028976548463106155,
-      "eval_runtime": 5.7358,
-      "eval_samples_per_second": 540.464,
-      "eval_steps_per_second": 11.332,
+      "eval_accuracy": 0.9309677419354838,
+      "eval_loss": 0.02714126743376255,
+      "eval_runtime": 5.6894,
+      "eval_samples_per_second": 544.869,
+      "eval_steps_per_second": 11.425,
       "step": 2544
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9283870967741935,
-      "eval_loss": 0.027913924306631088,
-      "eval_runtime": 5.5425,
-      "eval_samples_per_second": 559.315,
-      "eval_steps_per_second": 11.728,
+      "eval_accuracy": 0.9335483870967742,
+      "eval_loss": 0.02584407664835453,
+      "eval_runtime": 5.4141,
+      "eval_samples_per_second": 572.575,
+      "eval_steps_per_second": 12.006,
       "step": 2862
     },
     {
       "epoch": 9.433962264150944,
-      "grad_norm": 0.2555118501186371,
-      "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.0434,
+      "grad_norm": 0.2581123113632202,
+      "learning_rate": 2.847341337907376e-06,
+      "loss": 0.042,
       "step": 3000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 3498,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 11,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -153,8 +153,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2755961419937909,
-    "num_train_epochs": 10,
-    "temperature": 10
+    "alpha": 0.5264187954467557,
+    "num_train_epochs": 11,
+    "temperature": 5
   }
 }

run-0/checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2da5d40335246ac682533d0ca6c52c386b58cb072bae3a30bf87a5e28562733
+oid sha256:0589ed65fdaae56362b9a8760696297fdd9596b2ddfb3d7051a76a6dfd129716
 size 5240

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728715792.b76c1be2ae55.1423.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:946866cbedfb96b502f27e9c02d9c9bbd82dccb02e40d3b77bbcf740c7017d98
-size 16081
+oid sha256:fa61ebe05e1bdad6a73f3fc5960eabaad1cdbde610420e3a274161635895904d
+size 16615