Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ced52e0685875b8d675e97aaf647b1f9ccce4784eb0c6a6f8c8641ca10b9da5
+oid sha256:ebd8f296a86cd8022eec6263fe9e1abf4ee83487f9bbcae7b7ade8d89d337d1a
 size 268290900

run-4/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:866a5f62148435f94b3a1364b88a97b7dcf18eedf77829138812f80eb018ad2a
+oid sha256:60b838338a2308724453de7476d9b1c3bba2249c23a350f295167211e17d9424
 size 268290900

run-4/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:313fb3854975931f25c0c54840393d72028b1ef9815be13b182f9d4b40f2a329
+oid sha256:63d428f737fbba0a4c7235dc0b71a8b84d4fa763e538391547e0e9460ca8ca2d
 size 536643898

run-4/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f182501c34e4ea3ebc7617d27edab7e1367582b147e518cd90295ec7f2eaa0f
+oid sha256:60ef01273aaa599804d51a93c6c0c61874ebd50ab9e135a21aab5f7cd0e6487a
 size 1064

run-4/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6325806451612903,
-      "eval_loss": 0.3024460971355438,
-      "eval_runtime": 5.4182,
-      "eval_samples_per_second": 572.151,
-      "eval_steps_per_second": 11.997,
+      "eval_accuracy": 0.6406451612903226,
+      "eval_loss": 0.2967454195022583,
+      "eval_runtime": 5.4866,
+      "eval_samples_per_second": 565.017,
+      "eval_steps_per_second": 11.847,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6661854386329651,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.4696,
+      "grad_norm": 0.6655566096305847,
+      "learning_rate": 1.4758909853249476e-05,
+      "loss": 0.4655,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8274193548387097,
-      "eval_loss": 0.13584715127944946,
-      "eval_runtime": 5.4989,
-      "eval_samples_per_second": 563.751,
-      "eval_steps_per_second": 11.821,
+      "eval_accuracy": 0.8335483870967741,
+      "eval_loss": 0.12780845165252686,
+      "eval_runtime": 5.7922,
+      "eval_samples_per_second": 535.206,
+      "eval_steps_per_second": 11.222,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8712903225806452,
-      "eval_loss": 0.08720238506793976,
-      "eval_runtime": 5.4432,
-      "eval_samples_per_second": 569.518,
-      "eval_steps_per_second": 11.941,
+      "eval_accuracy": 0.8851612903225806,
+      "eval_loss": 0.07867012172937393,
+      "eval_runtime": 5.484,
+      "eval_samples_per_second": 565.28,
+      "eval_steps_per_second": 11.853,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.6423004865646362,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1584,
+      "grad_norm": 0.6149155497550964,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.1495,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 1908,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 6,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.41539799624272716,
-    "num_train_epochs": 5,
+    "alpha": 0.6003183502544485,
+    "num_train_epochs": 6,
     "temperature": 3
   }
 }

run-4/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42fcda7114a3cb0f210fff0e0a799f161a86990fe91444d8130cb28445b06005
+oid sha256:39e2230760e0d209a41a97071088ff5c736b18829fcf11d0286149c73a45dbed
 size 5240

run-4/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40578156a126d8bc2399e1d908fa81069a4992a0bc195381dd64b7d3e04f3de9
+oid sha256:ebd8f296a86cd8022eec6263fe9e1abf4ee83487f9bbcae7b7ade8d89d337d1a
 size 268290900

run-4/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ce27375ff8342917b293709988fb26d89a19c91bdf65ed569fea01e2ded88e9
+oid sha256:b91d84f1993b5b55c784280f6726bf11abeb203b23b2d8be5149fcac0dea9f4c
 size 536643898

run-4/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91d97f7c1be085852e2ffd2fe2f8a493a080e2b2858e2f3baf642fe214e035a0
+oid sha256:55c8d3ce0734337fc0c187ca5543b4c70ca45d996531f199209b3a0c2a798109
 size 1064

run-4/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6325806451612903,
-      "eval_loss": 0.3024460971355438,
-      "eval_runtime": 5.4182,
-      "eval_samples_per_second": 572.151,
-      "eval_steps_per_second": 11.997,
+      "eval_accuracy": 0.6406451612903226,
+      "eval_loss": 0.2967454195022583,
+      "eval_runtime": 5.4866,
+      "eval_samples_per_second": 565.017,
+      "eval_steps_per_second": 11.847,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6661854386329651,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.4696,
+      "grad_norm": 0.6655566096305847,
+      "learning_rate": 1.4758909853249476e-05,
+      "loss": 0.4655,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8274193548387097,
-      "eval_loss": 0.13584715127944946,
-      "eval_runtime": 5.4989,
-      "eval_samples_per_second": 563.751,
-      "eval_steps_per_second": 11.821,
+      "eval_accuracy": 0.8335483870967741,
+      "eval_loss": 0.12780845165252686,
+      "eval_runtime": 5.7922,
+      "eval_samples_per_second": 535.206,
+      "eval_steps_per_second": 11.222,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8712903225806452,
-      "eval_loss": 0.08720238506793976,
-      "eval_runtime": 5.4432,
-      "eval_samples_per_second": 569.518,
-      "eval_steps_per_second": 11.941,
+      "eval_accuracy": 0.8851612903225806,
+      "eval_loss": 0.07867012172937393,
+      "eval_runtime": 5.484,
+      "eval_samples_per_second": 565.28,
+      "eval_steps_per_second": 11.853,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.6423004865646362,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1584,
+      "grad_norm": 0.6149155497550964,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.1495,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8945161290322581,
-      "eval_loss": 0.06926823407411575,
-      "eval_runtime": 5.4202,
-      "eval_samples_per_second": 571.937,
-      "eval_steps_per_second": 11.992,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.05978764593601227,
+      "eval_runtime": 5.547,
+      "eval_samples_per_second": 558.865,
+      "eval_steps_per_second": 11.718,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.44296959042549133,
-      "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.1029,
+      "grad_norm": 0.43114015460014343,
+      "learning_rate": 4.276729559748428e-06,
+      "loss": 0.0911,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 1908,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 6,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.41539799624272716,
-    "num_train_epochs": 5,
+    "alpha": 0.6003183502544485,
+    "num_train_epochs": 6,
     "temperature": 3
   }
 }

run-4/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42fcda7114a3cb0f210fff0e0a799f161a86990fe91444d8130cb28445b06005
+oid sha256:39e2230760e0d209a41a97071088ff5c736b18829fcf11d0286149c73a45dbed
 size 5240

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728718423.b76c1be2ae55.1423.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db198a5f0e14ccda4309965ca889bf39e9b67c3eab3499a77df07f843e475e4b
-size 14155
+oid sha256:9f732651deb890d44c0204f14e20053d5c4cbb6fa7e12cb3d59061a197906d16
+size 15366