Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0492e7682ec1250a0c1b9ae09319ee939d2f6b518ac9ce9c7d535d268986e268
+oid sha256:50a2593d152e12d05013cb2362467fb567a0d93fc5a692e243e1e3888f3f1273
 size 268290900

run-1/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:176ef3c507fb509247ed678ad7820d7b95ef91bb124c7c80f8f6d76583facf5a
+oid sha256:50a2593d152e12d05013cb2362467fb567a0d93fc5a692e243e1e3888f3f1273
 size 268290900

run-1/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b1d434063a810ac5a8e9f21a272ba532b984a3780a2fe594e1d2489d014ad75
+oid sha256:cdf3e5833ffc00644c4e4c059b0c7628920f5711ad5f642215f2ddbb91c632ad
 size 536643898

run-1/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91d97f7c1be085852e2ffd2fe2f8a493a080e2b2858e2f3baf642fe214e035a0
+oid sha256:71453465aad25f4c5a0a948496c64b1f74df850abda497954afe3695c00756ee
 size 1064

run-1/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6690322580645162,
-      "eval_loss": 0.42948082089424133,
-      "eval_runtime": 5.393,
-      "eval_samples_per_second": 574.82,
-      "eval_steps_per_second": 12.053,
+      "eval_accuracy": 0.6529032258064517,
+      "eval_loss": 0.28555235266685486,
+      "eval_runtime": 5.4121,
+      "eval_samples_per_second": 572.786,
+      "eval_steps_per_second": 12.01,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.8646999001502991,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.6591,
+      "grad_norm": 0.6621501445770264,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.4579,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8229032258064516,
-      "eval_loss": 0.16900987923145294,
-      "eval_runtime": 5.2956,
-      "eval_samples_per_second": 585.388,
-      "eval_steps_per_second": 12.274,
+      "eval_accuracy": 0.8451612903225807,
+      "eval_loss": 0.1139635443687439,
+      "eval_runtime": 5.9245,
+      "eval_samples_per_second": 523.247,
+      "eval_steps_per_second": 10.971,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8729032258064516,
-      "eval_loss": 0.0963619202375412,
-      "eval_runtime": 5.3142,
-      "eval_samples_per_second": 583.338,
-      "eval_steps_per_second": 12.231,
+      "eval_accuracy": 0.8961290322580645,
+      "eval_loss": 0.06579811871051788,
+      "eval_runtime": 5.4507,
+      "eval_samples_per_second": 568.736,
+      "eval_steps_per_second": 11.925,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.7748064398765564,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1958,
+      "grad_norm": 0.5677153468132019,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.1346,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8977419354838709,
-      "eval_loss": 0.07352616637945175,
-      "eval_runtime": 5.3982,
-      "eval_samples_per_second": 574.264,
-      "eval_steps_per_second": 12.041,
+      "eval_accuracy": 0.9138709677419354,
+      "eval_loss": 0.04725706949830055,
+      "eval_runtime": 5.5473,
+      "eval_samples_per_second": 558.83,
+      "eval_steps_per_second": 11.717,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.542826771736145,
-      "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.113,
+      "grad_norm": 0.3639396131038666,
+      "learning_rate": 1.0566037735849058e-05,
+      "loss": 0.0751,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 3180,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.4868927644085208,
-    "num_train_epochs": 5,
-    "temperature": 2
+    "alpha": 0.31688267813599835,
+    "num_train_epochs": 10,
+    "temperature": 3
   }
 }

run-1/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f061e43c65f095a686a86278f34a31824db2598d218eedeec610b734a0e65f9
+oid sha256:6a074013ebca87018fe538250c2f45bfdeb260149a400a0644508c302966c2ef
 size 5240

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728716534.b76c1be2ae55.1423.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2635f27678ee5945de34e08fe81a8b40cbf90c8a1788fafa3994fb3c8638c7d3
-size 13833
+oid sha256:6a700c6fd9582e1f32c83887ea59a4c328759ee6ca4583ad86db333e64e23686
+size 14690