Training in progress, step 2500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:699ac59474fac90f89ae1da1a1ae436d64ca16090ac9c3281a874f01a99b911d
+oid sha256:c8b669e1bd7ff55a9121797a8e4a756a52cfbba25a8136fcf2b7646dfe31f152
 size 268290900

run-0/checkpoint-2500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9b57207f5f549168a640169527295f596ec83ccd2fe90c99253381447529a16
+oid sha256:c8b669e1bd7ff55a9121797a8e4a756a52cfbba25a8136fcf2b7646dfe31f152
 size 268290900

run-0/checkpoint-2500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e249282b0bf3b499af091012dfa62a3ba371d8bacf08f0e366a94d81ef206e53
+oid sha256:057baa1957fabfb204afc33491e4a2b0e5bcfbe64508bc110f0658f30baca802
 size 536643898

run-0/checkpoint-2500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b997ea9328158f28ff50e49c476931e703b821360bc8325e0c4d100e032c865
+oid sha256:147e6378d980b6b4eca617864452375deaabd499be58b60839723c4f4d63dc40
 size 1064

run-0/checkpoint-2500/trainer_state.json

@@ -10,107 +10,107 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5796774193548387,
-      "eval_loss": 0.19598601758480072,
-      "eval_runtime": 5.7949,
-      "eval_samples_per_second": 534.952,
-      "eval_steps_per_second": 11.217,
+      "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": 0.5235025882720947,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.3143,
+      "grad_norm": 0.5702099204063416,
+      "learning_rate": 1.7141223556317898e-05,
+      "loss": 0.3672,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8158064516129032,
-      "eval_loss": 0.096437968313694,
-      "eval_runtime": 5.8632,
-      "eval_samples_per_second": 528.722,
-      "eval_steps_per_second": 11.086,
+      "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": 0.8754838709677419,
-      "eval_loss": 0.06548532098531723,
-      "eval_runtime": 5.6772,
-      "eval_samples_per_second": 546.048,
-      "eval_steps_per_second": 11.449,
+      "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": 0.44506680965423584,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1116,
+      "grad_norm": 0.48276859521865845,
+      "learning_rate": 1.4282447112635793e-05,
+      "loss": 0.1191,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8961290322580645,
-      "eval_loss": 0.05135016515851021,
-      "eval_runtime": 5.7349,
-      "eval_samples_per_second": 540.547,
-      "eval_steps_per_second": 11.334,
+      "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": 0.3050009608268738,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0728,
+      "grad_norm": 0.3197612166404724,
+      "learning_rate": 1.1423670668953687e-05,
+      "loss": 0.0712,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9061290322580645,
-      "eval_loss": 0.042891617864370346,
-      "eval_runtime": 5.8649,
-      "eval_samples_per_second": 528.57,
-      "eval_steps_per_second": 11.083,
+      "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": 0.9074193548387097,
-      "eval_loss": 0.03820948302745819,
-      "eval_runtime": 5.6184,
-      "eval_samples_per_second": 551.755,
-      "eval_steps_per_second": 11.569,
+      "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": 0.27222588658332825,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0584,
+      "grad_norm": 0.2629503309726715,
+      "learning_rate": 8.564894225271586e-06,
+      "loss": 0.0539,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.912258064516129,
-      "eval_loss": 0.0359175018966198,
-      "eval_runtime": 5.767,
-      "eval_samples_per_second": 537.543,
-      "eval_steps_per_second": 11.271,
+      "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": 0.2702564001083374,
-      "learning_rate": 3.459119496855346e-07,
-      "loss": 0.0526,
+      "grad_norm": 0.29348886013031006,
+      "learning_rate": 5.706117781589479e-06,
+      "loss": 0.046,
       "step": 2500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 3498,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 11,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -128,8 +128,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2580849577581932,
-    "num_train_epochs": 8,
-    "temperature": 14
+    "alpha": 0.5264187954467557,
+    "num_train_epochs": 11,
+    "temperature": 5
   }
 }

run-0/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa272c0492b3901fc1a76322efd782d0609261ca66d80f909b4731c779116aae
+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 sha256:19ed2ea323dfee9713b7064a54547ff0e54775f521d972d9e445956d74936ff5
-size 15224
+oid sha256:946866cbedfb96b502f27e9c02d9c9bbd82dccb02e40d3b77bbcf740c7017d98
+size 16081