Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee3a1bac76d6f8b0064125152c1195b2177bf2b148f77e087ed54f72938cf290
+oid sha256:744965fc651adbb182875f98c309e15905b940ddfa5ac317af007d18b2f3085e
 size 268290900

run-1/checkpoint-3180/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d58a82305e2a5db362a66997e1c7fa82fc7f03422c45a6bfdfd9d2b3a7db306
+oid sha256:ddc9e803591f0ee8cc00808757fed099581d60b91bf5905391335e90c3ebc61d
 size 268290900

run-1/checkpoint-3180/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3967238e7e100b5c9e5a3cc8e0554629dc1547a5668bb22928119b48779f657d
+oid sha256:4b770125e0b14f83491c1b9486975926119247e57ae8caf1e11f720ed8867bb9
 size 536643898

run-1/checkpoint-3180/trainer_state.json

@@ -10,125 +10,125 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5838709677419355,
-      "eval_loss": 0.19346614181995392,
-      "eval_runtime": 5.1609,
-      "eval_samples_per_second": 600.675,
-      "eval_steps_per_second": 12.595,
+      "eval_accuracy": 0.6529032258064517,
+      "eval_loss": 0.28555235266685486,
+      "eval_runtime": 5.4121,
+      "eval_samples_per_second": 572.786,
+      "eval_steps_per_second": 12.01,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5179193019866943,
+      "grad_norm": 0.6621501445770264,
       "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3125,
+      "loss": 0.4579,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.82,
-      "eval_loss": 0.09356740862131119,
-      "eval_runtime": 5.2878,
-      "eval_samples_per_second": 586.255,
-      "eval_steps_per_second": 12.292,
+      "eval_accuracy": 0.8451612903225807,
+      "eval_loss": 0.1139635443687439,
+      "eval_runtime": 5.9245,
+      "eval_samples_per_second": 523.247,
+      "eval_steps_per_second": 10.971,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8790322580645161,
-      "eval_loss": 0.06245172396302223,
-      "eval_runtime": 5.0672,
-      "eval_samples_per_second": 611.776,
-      "eval_steps_per_second": 12.828,
+      "eval_accuracy": 0.8961290322580645,
+      "eval_loss": 0.06579811871051788,
+      "eval_runtime": 5.4507,
+      "eval_samples_per_second": 568.736,
+      "eval_steps_per_second": 11.925,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4341842532157898,
+      "grad_norm": 0.5677153468132019,
       "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1085,
+      "loss": 0.1346,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9006451612903226,
-      "eval_loss": 0.048103515058755875,
-      "eval_runtime": 5.1362,
-      "eval_samples_per_second": 603.564,
-      "eval_steps_per_second": 12.655,
+      "eval_accuracy": 0.9138709677419354,
+      "eval_loss": 0.04725706949830055,
+      "eval_runtime": 5.5473,
+      "eval_samples_per_second": 558.83,
+      "eval_steps_per_second": 11.717,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.29487845301628113,
+      "grad_norm": 0.3639396131038666,
       "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.069,
+      "loss": 0.0751,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9093548387096774,
-      "eval_loss": 0.03922608867287636,
-      "eval_runtime": 5.2603,
-      "eval_samples_per_second": 589.32,
-      "eval_steps_per_second": 12.357,
+      "eval_accuracy": 0.92,
+      "eval_loss": 0.03804958611726761,
+      "eval_runtime": 5.7319,
+      "eval_samples_per_second": 540.831,
+      "eval_steps_per_second": 11.34,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.912258064516129,
-      "eval_loss": 0.03417789563536644,
-      "eval_runtime": 5.0804,
-      "eval_samples_per_second": 610.188,
-      "eval_steps_per_second": 12.794,
+      "eval_accuracy": 0.927741935483871,
+      "eval_loss": 0.03348824381828308,
+      "eval_runtime": 5.4253,
+      "eval_samples_per_second": 571.402,
+      "eval_steps_per_second": 11.981,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.2546312212944031,
+      "grad_norm": 0.29170453548431396,
       "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0537,
+      "loss": 0.0566,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9219354838709677,
-      "eval_loss": 0.03144192695617676,
-      "eval_runtime": 5.2857,
-      "eval_samples_per_second": 586.49,
-      "eval_steps_per_second": 12.297,
+      "eval_accuracy": 0.932258064516129,
+      "eval_loss": 0.0307790357619524,
+      "eval_runtime": 5.6345,
+      "eval_samples_per_second": 550.182,
+      "eval_steps_per_second": 11.536,
       "step": 2226
     },
     {
       "epoch": 7.861635220125786,
-      "grad_norm": 0.2692946791648865,
+      "grad_norm": 0.32524076104164124,
       "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0465,
+      "loss": 0.0488,
       "step": 2500
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9267741935483871,
-      "eval_loss": 0.029110092669725418,
-      "eval_runtime": 5.2525,
-      "eval_samples_per_second": 590.192,
-      "eval_steps_per_second": 12.375,
+      "eval_accuracy": 0.9319354838709677,
+      "eval_loss": 0.028849748894572258,
+      "eval_runtime": 5.7512,
+      "eval_samples_per_second": 539.022,
+      "eval_steps_per_second": 11.302,
       "step": 2544
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9283870967741935,
-      "eval_loss": 0.028060223907232285,
-      "eval_runtime": 5.2234,
-      "eval_samples_per_second": 593.48,
-      "eval_steps_per_second": 12.444,
+      "eval_accuracy": 0.9341935483870968,
+      "eval_loss": 0.027793901041150093,
+      "eval_runtime": 5.4474,
+      "eval_samples_per_second": 569.075,
+      "eval_steps_per_second": 11.932,
       "step": 2862
     },
     {
       "epoch": 9.433962264150944,
-      "grad_norm": 0.2502853572368622,
+      "grad_norm": 0.28644028306007385,
       "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.0431,
+      "loss": 0.0453,
       "step": 3000
     }
   ],
@@ -153,8 +153,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5365842802514643,
+    "alpha": 0.31688267813599835,
     "num_train_epochs": 10,
-    "temperature": 14
+    "temperature": 3
   }
 }

run-1/checkpoint-3180/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aff76e4271bd0bf9763ec56f904c0eab2f55fe172f25f099aa8c57c7f7749007
+oid sha256:6a074013ebca87018fe538250c2f45bfdeb260149a400a0644508c302966c2ef
 size 5240

run-2/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ee948a20c89755d7a96becd052e4426cb6cdccf9b8ce076c95e8904c6697ce5
+oid sha256:4ef3d7b0babe94fa5166a9ac6f9db248a8c3aed522dd161e1d5841de8689beec
 size 268290900

run-2/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ceae458e1139240eed9e7d904d0cad7cb3b3fe26a87e4014a66ee16b5ed87def
+oid sha256:3846e1108a24cbd21e60e75fb69347d910affcbcc9e8b69193ba6cb0161346d7
 size 536643898

run-2/checkpoint-1000/trainer_state.json

@@ -10,43 +10,43 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6445161290322581,
-      "eval_loss": 0.29249975085258484,
-      "eval_runtime": 5.4218,
-      "eval_samples_per_second": 571.762,
-      "eval_steps_per_second": 11.989,
+      "eval_accuracy": 0.603225806451613,
+      "eval_loss": 0.23406817018985748,
+      "eval_runtime": 5.4802,
+      "eval_samples_per_second": 565.673,
+      "eval_steps_per_second": 11.861,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6633772850036621,
+      "grad_norm": 0.57925945520401,
       "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4626,
+      "loss": 0.3716,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8380645161290322,
-      "eval_loss": 0.12248263508081436,
-      "eval_runtime": 5.3956,
-      "eval_samples_per_second": 574.542,
-      "eval_steps_per_second": 12.047,
+      "eval_accuracy": 0.8309677419354838,
+      "eval_loss": 0.1094982922077179,
+      "eval_runtime": 5.8681,
+      "eval_samples_per_second": 528.276,
+      "eval_steps_per_second": 11.077,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8887096774193548,
-      "eval_loss": 0.07343784719705582,
-      "eval_runtime": 5.2977,
-      "eval_samples_per_second": 585.157,
-      "eval_steps_per_second": 12.269,
+      "eval_accuracy": 0.8819354838709678,
+      "eval_loss": 0.07135984301567078,
+      "eval_runtime": 5.4437,
+      "eval_samples_per_second": 569.468,
+      "eval_steps_per_second": 11.94,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.59869384765625,
+      "grad_norm": 0.5091490149497986,
       "learning_rate": 1.101527403414196e-05,
-      "loss": 0.1438,
+      "loss": 0.1274,
       "step": 1000
     }
   ],
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.444035914041536,
+    "alpha": 0.49068870611003523,
     "num_train_epochs": 7,
-    "temperature": 3
+    "temperature": 5
   }
 }

run-2/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22d5df12fd1120947d14fad0a0388bc240150e4fa0277e30f926f1ec12ddb8e9
+oid sha256:8759fe239e3e09af377de9d8ac9ed677e8c3fe7b8d3bf99d5396a42501d32db2
 size 5240

run-2/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c9a63d3004610f74819b6489ca4b580024510d89b7576222160d376a94edd03
+oid sha256:744965fc651adbb182875f98c309e15905b940ddfa5ac317af007d18b2f3085e
 size 268290900

run-2/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90b0a266aa6d2c7cee70af38c41264e284f7c2475c5c6c36d6483a6cc0cab912
+oid sha256:e93e14489b8389f84f048b08257878674ed1df885fa0f27111119a629e822b54
 size 536643898

run-2/checkpoint-1500/trainer_state.json

@@ -10,59 +10,59 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6445161290322581,
-      "eval_loss": 0.29249975085258484,
-      "eval_runtime": 5.4218,
-      "eval_samples_per_second": 571.762,
-      "eval_steps_per_second": 11.989,
+      "eval_accuracy": 0.603225806451613,
+      "eval_loss": 0.23406817018985748,
+      "eval_runtime": 5.4802,
+      "eval_samples_per_second": 565.673,
+      "eval_steps_per_second": 11.861,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6633772850036621,
+      "grad_norm": 0.57925945520401,
       "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4626,
+      "loss": 0.3716,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8380645161290322,
-      "eval_loss": 0.12248263508081436,
-      "eval_runtime": 5.3956,
-      "eval_samples_per_second": 574.542,
-      "eval_steps_per_second": 12.047,
+      "eval_accuracy": 0.8309677419354838,
+      "eval_loss": 0.1094982922077179,
+      "eval_runtime": 5.8681,
+      "eval_samples_per_second": 528.276,
+      "eval_steps_per_second": 11.077,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8887096774193548,
-      "eval_loss": 0.07343784719705582,
-      "eval_runtime": 5.2977,
-      "eval_samples_per_second": 585.157,
-      "eval_steps_per_second": 12.269,
+      "eval_accuracy": 0.8819354838709678,
+      "eval_loss": 0.07135984301567078,
+      "eval_runtime": 5.4437,
+      "eval_samples_per_second": 569.468,
+      "eval_steps_per_second": 11.94,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.59869384765625,
+      "grad_norm": 0.5091490149497986,
       "learning_rate": 1.101527403414196e-05,
-      "loss": 0.1438,
+      "loss": 0.1274,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9045161290322581,
-      "eval_loss": 0.05444410815834999,
-      "eval_runtime": 5.5714,
-      "eval_samples_per_second": 556.417,
-      "eval_steps_per_second": 11.667,
+      "eval_accuracy": 0.8980645161290323,
+      "eval_loss": 0.055049341171979904,
+      "eval_runtime": 5.4797,
+      "eval_samples_per_second": 565.721,
+      "eval_steps_per_second": 11.862,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.4075869023799896,
+      "grad_norm": 0.3538859486579895,
       "learning_rate": 6.522911051212939e-06,
-      "loss": 0.0845,
+      "loss": 0.0809,
       "step": 1500
     }
   ],
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.444035914041536,
+    "alpha": 0.49068870611003523,
     "num_train_epochs": 7,
-    "temperature": 3
+    "temperature": 5
   }
 }

run-2/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22d5df12fd1120947d14fad0a0388bc240150e4fa0277e30f926f1ec12ddb8e9
+oid sha256:8759fe239e3e09af377de9d8ac9ed677e8c3fe7b8d3bf99d5396a42501d32db2
 size 5240

run-2/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a43fed3db2fa6cb21249810fc8b7e54c07d7631ccde23bd46df3a9ded2e58a2c
+oid sha256:898cdf11d05981397287e11c7ef9415fbedfe915715e9c683335b09571b423e2
 size 268290900

run-2/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:317dab0e8dcb99802f66c349baa9bb2f8c8244a04507efcbf7e60de697eb7179
+oid sha256:c4c8ea22cc9486b65469687d5ae65fb6a13ea70d68548174383fc8f1096b89b7
 size 536643898

run-2/checkpoint-500/trainer_state.json

@@ -10,18 +10,18 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6445161290322581,
-      "eval_loss": 0.29249975085258484,
-      "eval_runtime": 5.4218,
-      "eval_samples_per_second": 571.762,
-      "eval_steps_per_second": 11.989,
+      "eval_accuracy": 0.603225806451613,
+      "eval_loss": 0.23406817018985748,
+      "eval_runtime": 5.4802,
+      "eval_samples_per_second": 565.673,
+      "eval_steps_per_second": 11.861,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6633772850036621,
+      "grad_norm": 0.57925945520401,
       "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4626,
+      "loss": 0.3716,
       "step": 500
     }
   ],
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.444035914041536,
+    "alpha": 0.49068870611003523,
     "num_train_epochs": 7,
-    "temperature": 3
+    "temperature": 5
   }
 }

run-2/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:22d5df12fd1120947d14fad0a0388bc240150e4fa0277e30f926f1ec12ddb8e9
+oid sha256:8759fe239e3e09af377de9d8ac9ed677e8c3fe7b8d3bf99d5396a42501d32db2
 size 5240

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728717352.b76c1be2ae55.1423.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d40e1bffd20d44f6d7d196bd10584be1460641c398a0457ce17732b1f4af602
+size 15223

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a074013ebca87018fe538250c2f45bfdeb260149a400a0644508c302966c2ef
+oid sha256:8759fe239e3e09af377de9d8ac9ed677e8c3fe7b8d3bf99d5396a42501d32db2
 size 5240