Training in progress, step 3000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0737b1fa29f840b7030bb5c9b520c81b6908fc48c8a845707681321814b8ff0d
+oid sha256:ee3a1bac76d6f8b0064125152c1195b2177bf2b148f77e087ed54f72938cf290
 size 268290900

run-1/checkpoint-3000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70712497b01ecf087836f70dd4d39eef90310ecb253f9573ff8ff185297407a4
+oid sha256:ee3a1bac76d6f8b0064125152c1195b2177bf2b148f77e087ed54f72938cf290
 size 268290900

run-1/checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23c055647b871de2bbce71cbee3d6db56a038560434362edc576b310cc5ad48d
+oid sha256:28826e960082685e8ba5f648aeec3c010772b44bcf6982a6ba6ef63f287424ce
 size 536643898

run-1/checkpoint-3000/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-3000/training_args.bin

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

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728716534.b76c1be2ae55.1423.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a105d06a7edeaac035833951b00d7fe58e68fa10c1b0676f1153501aea079fd
-size 16081
+oid sha256:f138e1a3ffb424fe2f082b0e8f935be3712c462381acbdd1067eba81eab5b165
+size 17292