Training in progress, step 1590

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:847b25893f634ae7ae19c7fa981b031b671da55fceaff107c9c6fd1535ba9bd9
+oid sha256:b33a2a62d28195b19d499e7488708c875b8b4a2f93577e25e2b236442f40b382
 size 268290900

run-3/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e77ed91a39a7cfbd38ba21e4c29b57be04d44cc975ec8f89e4c9d7fa45dbaaf6
+oid sha256:278c22619c47ca29f599e7780b2b974771cc416649b263f524b88bb8b1ed6ac4
 size 268290900

run-3/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:308eb28881ddbf364fc834b2998ed4a5228666e33d4a58332e3280f3506c9079
+oid sha256:da1f08b5f90a40897bed8c4857f5874a16eb498c2423dd4769c586a7c878c84f
 size 536643898

run-3/checkpoint-1500/scheduler.pt

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

run-3/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5803225806451613,
-      "eval_loss": 0.19005867838859558,
-      "eval_runtime": 5.674,
-      "eval_samples_per_second": 546.354,
-      "eval_steps_per_second": 11.456,
+      "eval_accuracy": 0.6083870967741936,
+      "eval_loss": 0.26111724972724915,
+      "eval_runtime": 5.392,
+      "eval_samples_per_second": 574.924,
+      "eval_steps_per_second": 12.055,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5149380564689636,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3072,
+      "grad_norm": 0.612690269947052,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.4067,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.817741935483871,
-      "eval_loss": 0.0924694836139679,
-      "eval_runtime": 6.0294,
-      "eval_samples_per_second": 514.15,
-      "eval_steps_per_second": 10.781,
+      "eval_accuracy": 0.8261290322580646,
+      "eval_loss": 0.1251312643289566,
+      "eval_runtime": 5.4108,
+      "eval_samples_per_second": 572.929,
+      "eval_steps_per_second": 12.013,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8770967741935484,
-      "eval_loss": 0.06209348514676094,
-      "eval_runtime": 5.7978,
-      "eval_samples_per_second": 534.685,
-      "eval_steps_per_second": 11.211,
+      "eval_accuracy": 0.8719354838709678,
+      "eval_loss": 0.08443494886159897,
+      "eval_runtime": 5.5128,
+      "eval_samples_per_second": 562.324,
+      "eval_steps_per_second": 11.791,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4281909167766571,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1072,
+      "grad_norm": 0.5761224031448364,
+      "learning_rate": 7.421383647798742e-06,
+      "loss": 0.1453,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9009677419354839,
-      "eval_loss": 0.048012129962444305,
-      "eval_runtime": 5.7482,
-      "eval_samples_per_second": 539.302,
-      "eval_steps_per_second": 11.308,
+      "eval_accuracy": 0.8880645161290323,
+      "eval_loss": 0.06863659620285034,
+      "eval_runtime": 5.4981,
+      "eval_samples_per_second": 563.835,
+      "eval_steps_per_second": 11.822,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.2933551073074341,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0684,
+      "grad_norm": 0.40807998180389404,
+      "learning_rate": 1.1320754716981133e-06,
+      "loss": 0.0989,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 1590,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7637668053146042,
-    "num_train_epochs": 10,
-    "temperature": 18
+    "alpha": 0.4777308452793029,
+    "num_train_epochs": 5,
+    "temperature": 4
   }
 }

run-3/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33ceb3fabcbf71cb66944e6d8ec61e55f12751f2e6805da39ec868f61cfabfda
+oid sha256:5ef5b259539127ef5ddc697d31b7e7180caca610f845f70ff2cba869a6149cf5
 size 5240

run-3/checkpoint-1590/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d37750f42dc9ce583a38a28bf0f1d530bd5579d1c806f22a7f77be1362be93a
+oid sha256:b33a2a62d28195b19d499e7488708c875b8b4a2f93577e25e2b236442f40b382
 size 268290900

run-3/checkpoint-1590/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c5d92ba1a3fb2db1509be6ab2d18a366e40e1932b2f029f2769c5b7e48a32d66
+oid sha256:abfebc946e9b45a711c2bbb8d8dae504cbe21728af7cc12465fce10025454fd5
 size 536643898

run-3/checkpoint-1590/trainer_state.json

@@ -10,59 +10,59 @@
   "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": 0.6083870967741936,
+      "eval_loss": 0.26111724972724915,
+      "eval_runtime": 5.392,
+      "eval_samples_per_second": 574.924,
+      "eval_steps_per_second": 12.055,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5346085429191589,
+      "grad_norm": 0.612690269947052,
       "learning_rate": 1.371069182389937e-05,
-      "loss": 0.3123,
+      "loss": 0.4067,
       "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": 0.8261290322580646,
+      "eval_loss": 0.1251312643289566,
+      "eval_runtime": 5.4108,
+      "eval_samples_per_second": 572.929,
+      "eval_steps_per_second": 12.013,
       "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": 0.8719354838709678,
+      "eval_loss": 0.08443494886159897,
+      "eval_runtime": 5.5128,
+      "eval_samples_per_second": 562.324,
+      "eval_steps_per_second": 11.791,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4651307165622711,
+      "grad_norm": 0.5761224031448364,
       "learning_rate": 7.421383647798742e-06,
-      "loss": 0.1202,
+      "loss": 0.1453,
       "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": 0.8880645161290323,
+      "eval_loss": 0.06863659620285034,
+      "eval_runtime": 5.4981,
+      "eval_samples_per_second": 563.835,
+      "eval_steps_per_second": 11.822,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3457496464252472,
+      "grad_norm": 0.40807998180389404,
       "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.0873,
+      "loss": 0.0989,
       "step": 1500
     }
   ],
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2168496843217581,
+    "alpha": 0.4777308452793029,
     "num_train_epochs": 5,
-    "temperature": 20
+    "temperature": 4
   }
 }

run-3/checkpoint-1590/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e06cae4b29f817ae50e2d0e86cf0bfd0dcec8273319eafd13e35e5529d6b14e2
+oid sha256:5ef5b259539127ef5ddc697d31b7e7180caca610f845f70ff2cba869a6149cf5
 size 5240

runs/Oct12_05-09-37_f46fcd247c16/events.out.tfevents.1728711874.f46fcd247c16.522.4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d802ac01f6c2352b30e80ad15674bc6de269c70f80ea0a9a406db7e8587b62b
-size 14366
+oid sha256:a1732b6ca00bb6ea5f12377d90a9ba5ce0ff1a6f714ea9c8907e32bfe4372e37
+size 15043