Training in progress, step 2500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f09ba02af072f6dab1b31924b201f644e12e95c27ead4dc9623741e011906d93
+oid sha256:1f9ff50c9e5f7dbdfafe4634702c271461312412ada297046b7621f2bf299e45
 size 268290900

run-2/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5ead783c8cd394f3b7cff5bb9210de7d7720379c213b3f0ee696e1fc5a9d7e7
+oid sha256:ed55eb11a186a1c7dabd0cbef9219723149ae2bbdf0865338e8cfd2698758e51
 size 268290900

run-2/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7693bc288d70b7b54973569fea930806f04586d73f16be34aaf2af84e3c3b51b
+oid sha256:8956025203d7e5690f41e8fdbcf1e9305247e6216d8b75dd834116ccd35de15b
 size 536643898

run-2/checkpoint-2000/trainer_state.json

@@ -10,84 +10,84 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5948387096774194,
-      "eval_loss": 0.2074165940284729,
-      "eval_runtime": 5.1592,
-      "eval_samples_per_second": 600.867,
-      "eval_steps_per_second": 12.599,
+      "eval_accuracy": 0.5887096774193549,
+      "eval_loss": 0.19886387884616852,
+      "eval_runtime": 5.1582,
+      "eval_samples_per_second": 600.982,
+      "eval_steps_per_second": 12.601,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.538582980632782,
+      "grad_norm": 0.5294517874717712,
       "learning_rate": 1.650593990216632e-05,
-      "loss": 0.3336,
+      "loss": 0.3203,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8258064516129032,
-      "eval_loss": 0.09868818521499634,
-      "eval_runtime": 5.3751,
-      "eval_samples_per_second": 576.732,
-      "eval_steps_per_second": 12.093,
+      "eval_accuracy": 0.8193548387096774,
+      "eval_loss": 0.09618715196847916,
+      "eval_runtime": 5.4445,
+      "eval_samples_per_second": 569.379,
+      "eval_steps_per_second": 11.939,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8829032258064516,
-      "eval_loss": 0.06500214338302612,
-      "eval_runtime": 5.0949,
-      "eval_samples_per_second": 608.447,
-      "eval_steps_per_second": 12.758,
+      "eval_accuracy": 0.8783870967741936,
+      "eval_loss": 0.06421981006860733,
+      "eval_runtime": 5.3341,
+      "eval_samples_per_second": 581.17,
+      "eval_steps_per_second": 12.186,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4604446291923523,
+      "grad_norm": 0.44660821557044983,
       "learning_rate": 1.3011879804332637e-05,
-      "loss": 0.1147,
+      "loss": 0.1116,
       "step": 1000
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.8996774193548387,
-      "eval_loss": 0.049765754491090775,
-      "eval_runtime": 5.2676,
-      "eval_samples_per_second": 588.5,
-      "eval_steps_per_second": 12.34,
+      "eval_loss": 0.04962550476193428,
+      "eval_runtime": 5.1177,
+      "eval_samples_per_second": 605.746,
+      "eval_steps_per_second": 12.701,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.31128990650177,
+      "grad_norm": 0.3033629357814789,
       "learning_rate": 9.517819706498952e-06,
-      "loss": 0.0724,
+      "loss": 0.0713,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.91,
-      "eval_loss": 0.04065703973174095,
-      "eval_runtime": 5.2744,
-      "eval_samples_per_second": 587.744,
-      "eval_steps_per_second": 12.324,
+      "eval_accuracy": 0.9080645161290323,
+      "eval_loss": 0.040737707167863846,
+      "eval_runtime": 5.2196,
+      "eval_samples_per_second": 593.916,
+      "eval_steps_per_second": 12.453,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.912258064516129,
-      "eval_loss": 0.03559587895870209,
-      "eval_runtime": 5.1021,
-      "eval_samples_per_second": 607.594,
-      "eval_steps_per_second": 12.74,
+      "eval_accuracy": 0.9116129032258065,
+      "eval_loss": 0.0357489213347435,
+      "eval_runtime": 5.2038,
+      "eval_samples_per_second": 595.717,
+      "eval_steps_per_second": 12.491,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.269597589969635,
+      "grad_norm": 0.26608094573020935,
       "learning_rate": 6.02375960866527e-06,
-      "loss": 0.0567,
+      "loss": 0.0561,
       "step": 2000
     }
   ],
@@ -112,8 +112,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.186668286700706,
+    "alpha": 0.38078945785669316,
     "num_train_epochs": 9,
-    "temperature": 8
+    "temperature": 11
   }
 }

run-2/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e09a3826ab4ffc76c9981d2ccde475f4c95a27c3ed7ffae8225ce76a3cd34eeb
+oid sha256:a815016dfca6056394449b7eec53142e869a6e9ff738e6f9557bb4382010c025
 size 5240

run-2/checkpoint-2500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6623ae75fa42bc651bb78d6a329587e2a40ed8c59df585f178a98ba6e2df18ff
+oid sha256:1f9ff50c9e5f7dbdfafe4634702c271461312412ada297046b7621f2bf299e45
 size 268290900

run-2/checkpoint-2500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3aeb58ef529b4a5188f21c262ab45c9233aa2fcbad4984461dde21f2136039cf
+oid sha256:548ed17ad3c86e3edcb062f433cdb96dbe045a413d377bd7914f5848ef7e9605
 size 536643898

run-2/checkpoint-2500/trainer_state.json

@@ -10,100 +10,100 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5948387096774194,
-      "eval_loss": 0.2074165940284729,
-      "eval_runtime": 5.1592,
-      "eval_samples_per_second": 600.867,
-      "eval_steps_per_second": 12.599,
+      "eval_accuracy": 0.5887096774193549,
+      "eval_loss": 0.19886387884616852,
+      "eval_runtime": 5.1582,
+      "eval_samples_per_second": 600.982,
+      "eval_steps_per_second": 12.601,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.538582980632782,
+      "grad_norm": 0.5294517874717712,
       "learning_rate": 1.650593990216632e-05,
-      "loss": 0.3336,
+      "loss": 0.3203,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8258064516129032,
-      "eval_loss": 0.09868818521499634,
-      "eval_runtime": 5.3751,
-      "eval_samples_per_second": 576.732,
-      "eval_steps_per_second": 12.093,
+      "eval_accuracy": 0.8193548387096774,
+      "eval_loss": 0.09618715196847916,
+      "eval_runtime": 5.4445,
+      "eval_samples_per_second": 569.379,
+      "eval_steps_per_second": 11.939,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8829032258064516,
-      "eval_loss": 0.06500214338302612,
-      "eval_runtime": 5.0949,
-      "eval_samples_per_second": 608.447,
-      "eval_steps_per_second": 12.758,
+      "eval_accuracy": 0.8783870967741936,
+      "eval_loss": 0.06421981006860733,
+      "eval_runtime": 5.3341,
+      "eval_samples_per_second": 581.17,
+      "eval_steps_per_second": 12.186,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4604446291923523,
+      "grad_norm": 0.44660821557044983,
       "learning_rate": 1.3011879804332637e-05,
-      "loss": 0.1147,
+      "loss": 0.1116,
       "step": 1000
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.8996774193548387,
-      "eval_loss": 0.049765754491090775,
-      "eval_runtime": 5.2676,
-      "eval_samples_per_second": 588.5,
-      "eval_steps_per_second": 12.34,
+      "eval_loss": 0.04962550476193428,
+      "eval_runtime": 5.1177,
+      "eval_samples_per_second": 605.746,
+      "eval_steps_per_second": 12.701,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.31128990650177,
+      "grad_norm": 0.3033629357814789,
       "learning_rate": 9.517819706498952e-06,
-      "loss": 0.0724,
+      "loss": 0.0713,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.91,
-      "eval_loss": 0.04065703973174095,
-      "eval_runtime": 5.2744,
-      "eval_samples_per_second": 587.744,
-      "eval_steps_per_second": 12.324,
+      "eval_accuracy": 0.9080645161290323,
+      "eval_loss": 0.040737707167863846,
+      "eval_runtime": 5.2196,
+      "eval_samples_per_second": 593.916,
+      "eval_steps_per_second": 12.453,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.912258064516129,
-      "eval_loss": 0.03559587895870209,
-      "eval_runtime": 5.1021,
-      "eval_samples_per_second": 607.594,
-      "eval_steps_per_second": 12.74,
+      "eval_accuracy": 0.9116129032258065,
+      "eval_loss": 0.0357489213347435,
+      "eval_runtime": 5.2038,
+      "eval_samples_per_second": 595.717,
+      "eval_steps_per_second": 12.491,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.269597589969635,
+      "grad_norm": 0.26608094573020935,
       "learning_rate": 6.02375960866527e-06,
-      "loss": 0.0567,
+      "loss": 0.0561,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9203225806451613,
-      "eval_loss": 0.032886430621147156,
-      "eval_runtime": 5.44,
-      "eval_samples_per_second": 569.85,
-      "eval_steps_per_second": 11.948,
+      "eval_accuracy": 0.9187096774193548,
+      "eval_loss": 0.03307666257023811,
+      "eval_runtime": 5.1324,
+      "eval_samples_per_second": 604.011,
+      "eval_steps_per_second": 12.665,
       "step": 2226
     },
     {
       "epoch": 7.861635220125786,
-      "grad_norm": 0.28186649084091187,
+      "grad_norm": 0.27350321412086487,
       "learning_rate": 2.5296995108315863e-06,
-      "loss": 0.0496,
+      "loss": 0.0492,
       "step": 2500
     }
   ],
@@ -128,8 +128,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.186668286700706,
+    "alpha": 0.38078945785669316,
     "num_train_epochs": 9,
-    "temperature": 8
+    "temperature": 11
   }
 }

run-2/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e09a3826ab4ffc76c9981d2ccde475f4c95a27c3ed7ffae8225ce76a3cd34eeb
+oid sha256:a815016dfca6056394449b7eec53142e869a6e9ff738e6f9557bb4382010c025
 size 5240

runs/Oct11_17-03-17_c179252869f4/events.out.tfevents.1728667869.c179252869f4.732.3

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1ef5556b5a378227de21603a8fff9c519c2252f5e86117c14ad4fb68edaf881
-size 15223
+oid sha256:9a1e336397b4ead093fca941eb5557f38735b9a597c301dd701161a7999021f1
+size 16757