Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:483d35db5a9e0274d13ceb4de46f67de2ae9de87d1c21da4b62e40ba3bbc1f19
+oid sha256:07161090aa695b836b3109e56422c48659d0ff109ddde1739e80c7dbac69dcee
 size 268290900

run-0/checkpoint-3000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04e267073b83a3774be9767ceb552868f5ad6ba81e9c23d4300b1b085f4b41bd
+oid sha256:46b3eeae7a828221d48c3afdc388d80e26d0a581d2e64ad92b94d9ed549da0f3
 size 268290900

run-0/checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bec0da93d5c7afc4f75be2b83d1cdf51aa6b187b10bc1a76ee309f76f2606cfa
+oid sha256:77ff61fd99d0714fffc3df5ec06b987ba9ac00abbdb3de905fc7286b0bf7008a
 size 536643898

run-0/checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5d96ed9889a2a02e4c7dbb49c09c58f3c154dfd2cb3d0452d915d37d9ed5e34
+oid sha256:235bcd30eb5caaf6d85f48a7cbef42afd59119224ef62ab684da9f5c869126f8
 size 5368

run-0/checkpoint-3180/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72a8680587911942633689b6b4125cba31cd037c3cc19cfad4d8ef38e10fcab3
+oid sha256:49f4f654fad95fcb02fa6113e96bd7fca48a3a9b57102a1e9cbac9fd6a1db37f
 size 268290900

run-0/checkpoint-3180/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d049f940b3dc867a6e9b52ba894ed598afd5815cc9844758f57857b61d4a87e6
+oid sha256:4fc6947ee09ed474860810aeccd221c427a3d4b87f6c326fc1ad0ce52001ace3
 size 536643898

run-0/checkpoint-3180/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e5d96ed9889a2a02e4c7dbb49c09c58f3c154dfd2cb3d0452d915d37d9ed5e34
+oid sha256:235bcd30eb5caaf6d85f48a7cbef42afd59119224ef62ab684da9f5c869126f8
 size 5368

run-1/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6364516129032258,
-      "eval_loss": 0.2942647337913513,
-      "eval_runtime": 5.7495,
-      "eval_samples_per_second": 539.181,
-      "eval_steps_per_second": 11.305,
+      "eval_accuracy": 0.6106451612903225,
+      "eval_loss": 0.2180573046207428,
+      "eval_runtime": 5.4534,
+      "eval_samples_per_second": 568.452,
+      "eval_steps_per_second": 11.919,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6808456778526306,
-      "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4631,
+      "grad_norm": 0.5000836253166199,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.3508,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.837741935483871,
-      "eval_loss": 0.12298235297203064,
-      "eval_runtime": 5.982,
-      "eval_samples_per_second": 518.218,
-      "eval_steps_per_second": 10.866,
+      "eval_accuracy": 0.8367741935483871,
+      "eval_loss": 0.10006564110517502,
+      "eval_runtime": 5.8746,
+      "eval_samples_per_second": 527.697,
+      "eval_steps_per_second": 11.065,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.886774193548387,
-      "eval_loss": 0.07365494966506958,
-      "eval_runtime": 5.4854,
-      "eval_samples_per_second": 565.141,
-      "eval_steps_per_second": 11.85,
+      "eval_accuracy": 0.8880645161290323,
+      "eval_loss": 0.06387896835803986,
+      "eval_runtime": 5.5911,
+      "eval_samples_per_second": 554.452,
+      "eval_steps_per_second": 11.626,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.7211179137229919,
-      "learning_rate": 1.101527403414196e-05,
-      "loss": 0.144,
+      "grad_norm": 0.5663716197013855,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.1169,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2226,
+  "max_steps": 3180,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.27523519283703446,
-    "num_train_epochs": 7,
-    "temperature": 3
+    "alpha": 0.9382110737628452,
+    "num_train_epochs": 10,
+    "temperature": 6
   }
 }

run-1/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2cd6c1fa18c08f721120761bd2b8759e0e49a2bed40a99f2e3983ef95cc42964
+oid sha256:07161090aa695b836b3109e56422c48659d0ff109ddde1739e80c7dbac69dcee
 size 268290900

run-1/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:892905aa43fd219554759899ef36ed6c84a1859236e2551bde01b6a9964f1199
+oid sha256:026cd375591dc5083103d510969258e5b3bfdfc5a1ced96fd5734113ba364898
 size 536643898

run-1/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11e744277c61f87520794334442fae36c5f9ff6e10cb79d4bfee5176ca7eafe2
+oid sha256:04366f62f8f88f5a8265df59adb051b320463277845db80e7fa43f13110c18c9
 size 1064

run-1/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6364516129032258,
-      "eval_loss": 0.2942647337913513,
-      "eval_runtime": 5.7495,
-      "eval_samples_per_second": 539.181,
-      "eval_steps_per_second": 11.305,
+      "eval_accuracy": 0.6106451612903225,
+      "eval_loss": 0.2180573046207428,
+      "eval_runtime": 5.4534,
+      "eval_samples_per_second": 568.452,
+      "eval_steps_per_second": 11.919,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6808456778526306,
-      "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4631,
+      "grad_norm": 0.5000836253166199,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.3508,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2226,
+  "max_steps": 3180,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.27523519283703446,
-    "num_train_epochs": 7,
-    "temperature": 3
+    "alpha": 0.9382110737628452,
+    "num_train_epochs": 10,
+    "temperature": 6
   }
 }

run-1/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c50be35dabf57488910ecef52dcf0c3f00eb115989d551ba66b159364e8ee11
+oid sha256:235bcd30eb5caaf6d85f48a7cbef42afd59119224ef62ab684da9f5c869126f8
 size 5368

runs/Nov25_08-56-26_a78cb449300a/events.out.tfevents.1732527563.a78cb449300a.701.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06181206223d415c8f4cacccfdbc0686a5356fd2f9d84b91484ded39cd6be4fa
-size 31454
+oid sha256:f3ada0fbdf4700ec40a3d76ca520d4fccd14de22b457fddc0a8e9c3c1f9c1e3e
+size 32131

runs/Nov25_08-56-26_a78cb449300a/events.out.tfevents.1732528693.a78cb449300a.701.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93bec0e1d68ce0be5270f2972e875d4a42e30ff869a6d555bc000776fbd7d027
+size 13952