Training in progress, step 2000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c9a63d3004610f74819b6489ca4b580024510d89b7576222160d376a94edd03
+oid sha256:159a38a62caabed9e1210e59e61e37481121dd41cf4c689d9dd1ff3a76ab2258
 size 268290900

run-2/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c113d65ac4f0f8bc117da0c4dad65c330e9bd6b2d59095fdba5b4fe9f2bfdae1
+oid sha256:159a38a62caabed9e1210e59e61e37481121dd41cf4c689d9dd1ff3a76ab2258
 size 268290900

run-2/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b72581f3bd190b8a202a0adc2bfd59c8b1b300486a89ed58cfc32e6989affb86
+oid sha256:d104b09460b4032759fd9392322e2f598820f8336681511fe529d62c7badd4ca
 size 536643898

run-2/checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43d45918bfeb622ba899798234ec7646c7a90dd3f5771db086f5b7dee1d5a530
+oid sha256:f38866eaf1d2baeb52a55cb38ece6ee67f3213265b0830dd06c82a9148795bea
 size 1064

run-2/checkpoint-2000/trainer_state.json

@@ -10,91 +10,91 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5796774193548387,
-      "eval_loss": 0.1907225400209427,
-      "eval_runtime": 5.6367,
-      "eval_samples_per_second": 549.965,
-      "eval_steps_per_second": 11.532,
+      "eval_accuracy": 0.6445161290322581,
+      "eval_loss": 0.29249975085258484,
+      "eval_runtime": 5.4218,
+      "eval_samples_per_second": 571.762,
+      "eval_steps_per_second": 11.989,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5174282193183899,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3083,
+      "grad_norm": 0.6633772850036621,
+      "learning_rate": 1.550763701707098e-05,
+      "loss": 0.4626,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.817741935483871,
-      "eval_loss": 0.09267137199640274,
-      "eval_runtime": 5.8965,
-      "eval_samples_per_second": 525.732,
-      "eval_steps_per_second": 11.023,
+      "eval_accuracy": 0.8380645161290322,
+      "eval_loss": 0.12248263508081436,
+      "eval_runtime": 5.3956,
+      "eval_samples_per_second": 574.542,
+      "eval_steps_per_second": 12.047,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8780645161290322,
-      "eval_loss": 0.06213448941707611,
-      "eval_runtime": 6.1922,
-      "eval_samples_per_second": 500.633,
-      "eval_steps_per_second": 10.497,
+      "eval_accuracy": 0.8887096774193548,
+      "eval_loss": 0.07343784719705582,
+      "eval_runtime": 5.2977,
+      "eval_samples_per_second": 585.157,
+      "eval_steps_per_second": 12.269,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.42976459860801697,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1074,
+      "grad_norm": 0.59869384765625,
+      "learning_rate": 1.101527403414196e-05,
+      "loss": 0.1438,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9006451612903226,
-      "eval_loss": 0.047995250672101974,
-      "eval_runtime": 5.9722,
-      "eval_samples_per_second": 519.074,
-      "eval_steps_per_second": 10.884,
+      "eval_accuracy": 0.9045161290322581,
+      "eval_loss": 0.05444410815834999,
+      "eval_runtime": 5.5714,
+      "eval_samples_per_second": 556.417,
+      "eval_steps_per_second": 11.667,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.29368260502815247,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0685,
+      "grad_norm": 0.4075869023799896,
+      "learning_rate": 6.522911051212939e-06,
+      "loss": 0.0845,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9093548387096774,
-      "eval_loss": 0.03920818120241165,
-      "eval_runtime": 5.8584,
-      "eval_samples_per_second": 529.158,
-      "eval_steps_per_second": 11.095,
+      "eval_accuracy": 0.9141935483870968,
+      "eval_loss": 0.045045968145132065,
+      "eval_runtime": 5.3274,
+      "eval_samples_per_second": 581.894,
+      "eval_steps_per_second": 12.201,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9125806451612903,
-      "eval_loss": 0.034214429557323456,
-      "eval_runtime": 5.6445,
-      "eval_samples_per_second": 549.203,
-      "eval_steps_per_second": 11.516,
+      "eval_accuracy": 0.9190322580645162,
+      "eval_loss": 0.04080257937312126,
+      "eval_runtime": 5.3427,
+      "eval_samples_per_second": 580.23,
+      "eval_steps_per_second": 12.166,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.25137194991111755,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0535,
+      "grad_norm": 0.352282851934433,
+      "learning_rate": 2.0305480682839176e-06,
+      "loss": 0.0669,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 2226,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 7,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -112,8 +112,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.4529889909540463,
-    "num_train_epochs": 10,
-    "temperature": 17
+    "alpha": 0.444035914041536,
+    "num_train_epochs": 7,
+    "temperature": 3
   }
 }

run-2/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:721f120a8e258fc8c48320ec67ae2a387816a0061225dcba3c70a5da25825846
+oid sha256:22d5df12fd1120947d14fad0a0388bc240150e4fa0277e30f926f1ec12ddb8e9
 size 5240

runs/Oct12_05-09-37_f46fcd247c16/events.out.tfevents.1728711374.f46fcd247c16.522.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96655c74898d6436df313562795fa6a996fa4d35e878e19019d1b079d714fdff
-size 14689
+oid sha256:ef8bb7a2be93dd9fa2fde9f12064d70b4392395c55c41eb87ff2ed7036d9fce0
+size 15900