Training in progress, step 2000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00c7efa1eb43f0dbc4ee52d425f1c719b0ac50968b1ac6f289ce19795f21d6ef
+oid sha256:a5ead783c8cd394f3b7cff5bb9210de7d7720379c213b3f0ee696e1fc5a9d7e7
 size 268290900

run-2/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d498327d12764475eb71d0eea4de1b4284e2e17948e55b2007edfae199ad4d05
+oid sha256:a5ead783c8cd394f3b7cff5bb9210de7d7720379c213b3f0ee696e1fc5a9d7e7
 size 268290900

run-2/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a49fb5692275ec2d2f7d7a2e9865052c626e385ecf90c4f5732844fb82de1755
+oid sha256:7693bc288d70b7b54973569fea930806f04586d73f16be34aaf2af84e3c3b51b
 size 536643898

run-2/checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bf6d4c2c13eda2c2dfacdbfc8684055c5603a92b6c0c386d7c757b108195d12
+oid sha256:67d1a8a9215f18c672cf798b531337b9e0636922d8c4373e76080f8fec19ceb2
 size 1064

run-2/checkpoint-2000/trainer_state.json

@@ -10,91 +10,91 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5738709677419355,
-      "eval_loss": 0.1914977878332138,
-      "eval_runtime": 5.3865,
-      "eval_samples_per_second": 575.517,
-      "eval_steps_per_second": 12.067,
+      "eval_accuracy": 0.5948387096774194,
+      "eval_loss": 0.2074165940284729,
+      "eval_runtime": 5.1592,
+      "eval_samples_per_second": 600.867,
+      "eval_steps_per_second": 12.599,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5165097713470459,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.3072,
+      "grad_norm": 0.538582980632782,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3336,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8141935483870968,
-      "eval_loss": 0.09490247070789337,
-      "eval_runtime": 5.6092,
-      "eval_samples_per_second": 552.66,
-      "eval_steps_per_second": 11.588,
+      "eval_accuracy": 0.8258064516129032,
+      "eval_loss": 0.09868818521499634,
+      "eval_runtime": 5.3751,
+      "eval_samples_per_second": 576.732,
+      "eval_steps_per_second": 12.093,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8741935483870967,
-      "eval_loss": 0.06484736502170563,
-      "eval_runtime": 6.0767,
-      "eval_samples_per_second": 510.146,
-      "eval_steps_per_second": 10.697,
+      "eval_accuracy": 0.8829032258064516,
+      "eval_loss": 0.06500214338302612,
+      "eval_runtime": 5.0949,
+      "eval_samples_per_second": 608.447,
+      "eval_steps_per_second": 12.758,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.43419399857521057,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1097,
+      "grad_norm": 0.4604446291923523,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1147,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8958064516129032,
-      "eval_loss": 0.05110108479857445,
-      "eval_runtime": 5.4249,
-      "eval_samples_per_second": 571.437,
-      "eval_steps_per_second": 11.982,
+      "eval_accuracy": 0.8996774193548387,
+      "eval_loss": 0.049765754491090775,
+      "eval_runtime": 5.2676,
+      "eval_samples_per_second": 588.5,
+      "eval_steps_per_second": 12.34,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.2985864579677582,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.072,
+      "grad_norm": 0.31128990650177,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.0724,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9048387096774193,
-      "eval_loss": 0.04280891269445419,
-      "eval_runtime": 5.5512,
-      "eval_samples_per_second": 558.437,
-      "eval_steps_per_second": 11.709,
+      "eval_accuracy": 0.91,
+      "eval_loss": 0.04065703973174095,
+      "eval_runtime": 5.2744,
+      "eval_samples_per_second": 587.744,
+      "eval_steps_per_second": 12.324,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9074193548387097,
-      "eval_loss": 0.038190603256225586,
-      "eval_runtime": 5.4902,
-      "eval_samples_per_second": 564.646,
-      "eval_steps_per_second": 11.839,
+      "eval_accuracy": 0.912258064516129,
+      "eval_loss": 0.03559587895870209,
+      "eval_runtime": 5.1021,
+      "eval_samples_per_second": 607.594,
+      "eval_steps_per_second": 12.74,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.2691451609134674,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0579,
+      "grad_norm": 0.269597589969635,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0567,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -112,8 +112,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5962321311067933,
-    "num_train_epochs": 8,
-    "temperature": 20
+    "alpha": 0.186668286700706,
+    "num_train_epochs": 9,
+    "temperature": 8
   }
 }

run-2/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:605b5083623cb90e46bc6aa64e7e8f2cfc283cd343d0688245d2c254a0200467
+oid sha256:e09a3826ab4ffc76c9981d2ccde475f4c95a27c3ed7ffae8225ce76a3cd34eeb
 size 5240

runs/Oct11_14-59-09_984d9c1bc9e1/events.out.tfevents.1728659967.984d9c1bc9e1.1239.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b51fa2f271ce8791e755f646a6f0db7c3e8a918599585c11e3ef8a396ad4d6d
-size 14689
+oid sha256:9b17341e5a06c4aec47da48b8c6de627b5668536e6c1b500d5592dafe02385de
+size 15546