Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76cadb93027e5ec338e6724a0b24825a1fc09abf2eef793105f9414585309730
+oid sha256:3e3dfa7be2ce14e7f0ea5acdcda3076bbb825e51fff60e158046ec8811a22473
 size 268290900

run-3/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7a84ca12312755d63f3d6cb7c3f98591fb856de7edceb500316fb8b01dec93a
+oid sha256:d269238e659920d175ea8529919070513eb475cc8cc4981d046f748105493cb2
 size 268290900

run-3/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:274859b63376561cadd84391124a2f6182fd6726468ee5dda887a00a719a88f4
+oid sha256:87d8bcd0279b17a6b825fa8fcd97f96b19c502400a65143d1d153cf56a92b506
 size 536643898

run-3/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f182501c34e4ea3ebc7617d27edab7e1367582b147e518cd90295ec7f2eaa0f
+oid sha256:c794bc4c67ef18245dd516031ce405ab557e4d551d225d8dd1e1abc0f2be8e33
 size 1064

run-3/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5574193548387096,
-      "eval_loss": 0.19900532066822052,
-      "eval_runtime": 5.3984,
-      "eval_samples_per_second": 574.243,
-      "eval_steps_per_second": 12.041,
+      "eval_accuracy": 0.6129032258064516,
+      "eval_loss": 0.22857815027236938,
+      "eval_runtime": 5.1291,
+      "eval_samples_per_second": 604.398,
+      "eval_steps_per_second": 12.673,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5346085429191589,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.3123,
+      "grad_norm": 0.5715988874435425,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.368,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7993548387096774,
-      "eval_loss": 0.1044413223862648,
-      "eval_runtime": 5.6691,
-      "eval_samples_per_second": 546.825,
-      "eval_steps_per_second": 11.466,
+      "eval_accuracy": 0.8406451612903226,
+      "eval_loss": 0.10308429598808289,
+      "eval_runtime": 5.3537,
+      "eval_samples_per_second": 579.034,
+      "eval_steps_per_second": 12.141,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.853225806451613,
-      "eval_loss": 0.07598130404949188,
-      "eval_runtime": 5.3839,
-      "eval_samples_per_second": 575.791,
-      "eval_steps_per_second": 12.073,
+      "eval_accuracy": 0.8909677419354839,
+      "eval_loss": 0.06482071429491043,
+      "eval_runtime": 5.1876,
+      "eval_samples_per_second": 597.576,
+      "eval_steps_per_second": 12.53,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4651307165622711,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1202,
+      "grad_norm": 0.4861523509025574,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.1205,
       "step": 1000
     }
   ],
   "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": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2168496843217581,
-    "num_train_epochs": 5,
-    "temperature": 20
+    "alpha": 0.34170044466363136,
+    "num_train_epochs": 10,
+    "temperature": 5
   }
 }

run-3/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e06cae4b29f817ae50e2d0e86cf0bfd0dcec8273319eafd13e35e5529d6b14e2
+oid sha256:13f3c53667f2d0c994b1e0580a4240dd5f1920dd508edcd4cd8ca7ddb067f7f3
 size 5240

run-3/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15d568dbca52ab4915bb81ee9bf99f287c33ed2785910941e8a2d5b6054c7f7b
+oid sha256:3e3dfa7be2ce14e7f0ea5acdcda3076bbb825e51fff60e158046ec8811a22473
 size 268290900

run-3/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a63704a30df7ca6ce1af5a393053db96871cf5e2ce3e3aea3257a30bac82654
+oid sha256:578039ccc687432b658165aa96f3b0f2e843d55f7e0de1398b1ee0f672a39638
 size 536643898

run-3/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-3/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5574193548387096,
-      "eval_loss": 0.19900532066822052,
-      "eval_runtime": 5.3984,
-      "eval_samples_per_second": 574.243,
-      "eval_steps_per_second": 12.041,
+      "eval_accuracy": 0.6129032258064516,
+      "eval_loss": 0.22857815027236938,
+      "eval_runtime": 5.1291,
+      "eval_samples_per_second": 604.398,
+      "eval_steps_per_second": 12.673,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5346085429191589,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.3123,
+      "grad_norm": 0.5715988874435425,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.368,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7993548387096774,
-      "eval_loss": 0.1044413223862648,
-      "eval_runtime": 5.6691,
-      "eval_samples_per_second": 546.825,
-      "eval_steps_per_second": 11.466,
+      "eval_accuracy": 0.8406451612903226,
+      "eval_loss": 0.10308429598808289,
+      "eval_runtime": 5.3537,
+      "eval_samples_per_second": 579.034,
+      "eval_steps_per_second": 12.141,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.853225806451613,
-      "eval_loss": 0.07598130404949188,
-      "eval_runtime": 5.3839,
-      "eval_samples_per_second": 575.791,
-      "eval_steps_per_second": 12.073,
+      "eval_accuracy": 0.8909677419354839,
+      "eval_loss": 0.06482071429491043,
+      "eval_runtime": 5.1876,
+      "eval_samples_per_second": 597.576,
+      "eval_steps_per_second": 12.53,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4651307165622711,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1202,
+      "grad_norm": 0.4861523509025574,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.1205,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8751612903225806,
-      "eval_loss": 0.0643545612692833,
-      "eval_runtime": 5.4223,
-      "eval_samples_per_second": 571.715,
-      "eval_steps_per_second": 11.988,
+      "eval_accuracy": 0.9032258064516129,
+      "eval_loss": 0.04816382750868797,
+      "eval_runtime": 5.1502,
+      "eval_samples_per_second": 601.919,
+      "eval_steps_per_second": 12.621,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3457496464252472,
-      "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.0873,
+      "grad_norm": 0.3236384689807892,
+      "learning_rate": 1.0566037735849058e-05,
+      "loss": 0.0727,
       "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": 0.2168496843217581,
-    "num_train_epochs": 5,
-    "temperature": 20
+    "alpha": 0.34170044466363136,
+    "num_train_epochs": 10,
+    "temperature": 5
   }
 }

run-3/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e06cae4b29f817ae50e2d0e86cf0bfd0dcec8273319eafd13e35e5529d6b14e2
+oid sha256:13f3c53667f2d0c994b1e0580a4240dd5f1920dd508edcd4cd8ca7ddb067f7f3
 size 5240

run-3/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de522ad9da2147f730c9bef13d580578dd5291d37418f02b528d6bd5928be7ca
+oid sha256:1ca6d89d77b30dc89001d0a2e7380a4c048d3716c2663a0e1d21881704782ad6
 size 268290900

run-3/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:38652956e0755a366a7180402ae805cc004533fa33ad338d15dba9ccde09ef16
+oid sha256:44a088b20b0694a4c988c166bcf8fbc1124a32ad577ee6ba35273769aa011959
 size 536643898

run-3/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e1264523e958cf7990dc5f42d876cc12129475c4603804cf66868aaf25c2c24
+oid sha256:04366f62f8f88f5a8265df59adb051b320463277845db80e7fa43f13110c18c9
 size 1064

run-3/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5574193548387096,
-      "eval_loss": 0.19900532066822052,
-      "eval_runtime": 5.3984,
-      "eval_samples_per_second": 574.243,
-      "eval_steps_per_second": 12.041,
+      "eval_accuracy": 0.6129032258064516,
+      "eval_loss": 0.22857815027236938,
+      "eval_runtime": 5.1291,
+      "eval_samples_per_second": 604.398,
+      "eval_steps_per_second": 12.673,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5346085429191589,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.3123,
+      "grad_norm": 0.5715988874435425,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.368,
       "step": 500
     }
   ],
   "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": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2168496843217581,
-    "num_train_epochs": 5,
-    "temperature": 20
+    "alpha": 0.34170044466363136,
+    "num_train_epochs": 10,
+    "temperature": 5
   }
 }

run-3/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e06cae4b29f817ae50e2d0e86cf0bfd0dcec8273319eafd13e35e5529d6b14e2
+oid sha256:13f3c53667f2d0c994b1e0580a4240dd5f1920dd508edcd4cd8ca7ddb067f7f3
 size 5240

runs/Oct11_14-59-09_984d9c1bc9e1/events.out.tfevents.1728660671.984d9c1bc9e1.1239.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb8ec65ff4fb66f0f721c336cad45260b11c987cd5c668b7899b2a95142003a3
+size 15224

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e09a3826ab4ffc76c9981d2ccde475f4c95a27c3ed7ffae8225ce76a3cd34eeb
+oid sha256:13f3c53667f2d0c994b1e0580a4240dd5f1920dd508edcd4cd8ca7ddb067f7f3
 size 5240