Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a085d735cd8cb688348e3d616bb12ef7bafe15d11b3ddb22f501564224348220
+oid sha256:176ef3c507fb509247ed678ad7820d7b95ef91bb124c7c80f8f6d76583facf5a
 size 268290900

run-1/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9b90474c183b5df91636679be91a2c4bf97301a8ca21ab1cefdaff8b2a19677
+oid sha256:176ef3c507fb509247ed678ad7820d7b95ef91bb124c7c80f8f6d76583facf5a
 size 268290900

run-1/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55cc95a171607f842497d316fa4808c9775326be24dd4ee178f156156d3e48e6
+oid sha256:6b1d434063a810ac5a8e9f21a272ba532b984a3780a2fe594e1d2489d014ad75
 size 536643898

run-1/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb7b4645e59a5c5cb66f63e8bcb5f006c535fd0ea63db9ae152dd586fd465b28
+oid sha256:91d97f7c1be085852e2ffd2fe2f8a493a080e2b2858e2f3baf642fe214e035a0
 size 1064

run-1/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5829032258064516,
-      "eval_loss": 0.1985393762588501,
-      "eval_runtime": 5.6203,
-      "eval_samples_per_second": 551.572,
-      "eval_steps_per_second": 11.565,
+      "eval_accuracy": 0.6690322580645162,
+      "eval_loss": 0.42948082089424133,
+      "eval_runtime": 5.393,
+      "eval_samples_per_second": 574.82,
+      "eval_steps_per_second": 12.053,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5288612246513367,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.3184,
+      "grad_norm": 0.8646999001502991,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.6591,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8174193548387096,
-      "eval_loss": 0.09726663678884506,
-      "eval_runtime": 6.0438,
-      "eval_samples_per_second": 512.92,
-      "eval_steps_per_second": 10.755,
+      "eval_accuracy": 0.8229032258064516,
+      "eval_loss": 0.16900987923145294,
+      "eval_runtime": 5.2956,
+      "eval_samples_per_second": 585.388,
+      "eval_steps_per_second": 12.274,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8764516129032258,
-      "eval_loss": 0.06577406078577042,
-      "eval_runtime": 5.6348,
-      "eval_samples_per_second": 550.155,
-      "eval_steps_per_second": 11.535,
+      "eval_accuracy": 0.8729032258064516,
+      "eval_loss": 0.0963619202375412,
+      "eval_runtime": 5.3142,
+      "eval_samples_per_second": 583.338,
+      "eval_steps_per_second": 12.231,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.44873112440109253,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1127,
+      "grad_norm": 0.7748064398765564,
+      "learning_rate": 7.421383647798742e-06,
+      "loss": 0.1958,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8964516129032258,
-      "eval_loss": 0.05149334296584129,
-      "eval_runtime": 5.7578,
-      "eval_samples_per_second": 538.401,
-      "eval_steps_per_second": 11.289,
+      "eval_accuracy": 0.8977419354838709,
+      "eval_loss": 0.07352616637945175,
+      "eval_runtime": 5.3982,
+      "eval_samples_per_second": 574.264,
+      "eval_steps_per_second": 12.041,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.30973368883132935,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0733,
+      "grad_norm": 0.542826771736145,
+      "learning_rate": 1.1320754716981133e-06,
+      "loss": 0.113,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 1590,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.819089504077056,
-    "num_train_epochs": 8,
-    "temperature": 12
+    "alpha": 0.4868927644085208,
+    "num_train_epochs": 5,
+    "temperature": 2
   }
 }

run-1/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ad3761ee6c049b5f54cddb21d4403152bfaafdbfc2a13b71993419e99660838
+oid sha256:2f061e43c65f095a686a86278f34a31824db2598d218eedeec610b734a0e65f9
 size 5240

runs/Oct12_05-09-37_f46fcd247c16/events.out.tfevents.1728710945.f46fcd247c16.522.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e64a06ab630c758344d723022bf7ac136984017d397bad62bdd38efa9621dcd
-size 13832
+oid sha256:a960ff1dd8991b02f8ba0927d7649777297c794780a86ae6ab7ff466474f0f84
+size 14366