End of training

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.14942141332434558,
+    "eval_loss": 6.609232425689697,
+    "eval_runtime": 0.5708,
+    "eval_samples": 479,
+    "eval_samples_per_second": 839.189,
+    "eval_steps_per_second": 14.016,
+    "perplexity": 741.9133264941478,
+    "train_loss": 7.326953015673104,
+    "train_runtime": 5761.7187,
+    "train_samples": 228639,
+    "train_samples_per_second": 595.236,
+    "train_steps_per_second": 9.302
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 15.0,
+    "eval_accuracy": 0.14942141332434558,
+    "eval_loss": 6.609232425689697,
+    "eval_runtime": 0.5708,
+    "eval_samples": 479,
+    "eval_samples_per_second": 839.189,
+    "eval_steps_per_second": 14.016,
+    "perplexity": 741.9133264941478
+}

logs/events.out.tfevents.1675181784.serv-3333.204260.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c4e396b16cf55986819870ccfb69c18818a5144b9f54a2eff9dd818964d9bb1
+size 369

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 15.0,
+    "train_loss": 7.326953015673104,
+    "train_runtime": 5761.7187,
+    "train_samples": 228639,
+    "train_samples_per_second": 595.236,
+    "train_steps_per_second": 9.302
+}

trainer_state.json

@@ -0,0 +1,250 @@
+{
+  "best_metric": 6.556983947753906,
+  "best_model_checkpoint": "distilbert_add_pre-training-dim-96/checkpoint-50022",
+  "epoch": 15.0,
+  "global_step": 53595,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.6756706234227496e-05,
+      "loss": 14.685,
+      "step": 3573
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.12398869012625627,
+      "eval_loss": 9.392213821411133,
+      "eval_runtime": 0.565,
+      "eval_samples_per_second": 847.75,
+      "eval_steps_per_second": 14.159,
+      "step": 3573
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.341807645574353e-05,
+      "loss": 8.0255,
+      "step": 7146
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.1315243012683803,
+      "eval_loss": 7.151001453399658,
+      "eval_runtime": 0.5512,
+      "eval_samples_per_second": 869.004,
+      "eval_steps_per_second": 14.514,
+      "step": 7146
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.008038134405085e-05,
+      "loss": 7.0152,
+      "step": 10719
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.14824699404427463,
+      "eval_loss": 6.786097049713135,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 862.599,
+      "eval_steps_per_second": 14.407,
+      "step": 10719
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.674175156556688e-05,
+      "loss": 6.8127,
+      "step": 14292
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.1492738182022094,
+      "eval_loss": 6.705262660980225,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 862.591,
+      "eval_steps_per_second": 14.407,
+      "step": 14292
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.3404056453874196e-05,
+      "loss": 6.74,
+      "step": 17865
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.1474488649134637,
+      "eval_loss": 6.669492244720459,
+      "eval_runtime": 0.5614,
+      "eval_samples_per_second": 853.267,
+      "eval_steps_per_second": 14.251,
+      "step": 17865
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 3.0065426675390224e-05,
+      "loss": 6.7067,
+      "step": 21438
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.1490975747320925,
+      "eval_loss": 6.643059253692627,
+      "eval_runtime": 0.5554,
+      "eval_samples_per_second": 862.513,
+      "eval_steps_per_second": 14.405,
+      "step": 21438
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.6726796896906253e-05,
+      "loss": 6.6871,
+      "step": 25011
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.14833291057185535,
+      "eval_loss": 6.620449542999268,
+      "eval_runtime": 0.5532,
+      "eval_samples_per_second": 865.795,
+      "eval_steps_per_second": 14.46,
+      "step": 25011
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 2.338910178521357e-05,
+      "loss": 6.6748,
+      "step": 28584
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.1472751167407656,
+      "eval_loss": 6.625016212463379,
+      "eval_runtime": 0.5524,
+      "eval_samples_per_second": 867.158,
+      "eval_steps_per_second": 14.483,
+      "step": 28584
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 2.005140667352089e-05,
+      "loss": 6.6649,
+      "step": 32157
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.14858899133836267,
+      "eval_loss": 6.610751152038574,
+      "eval_runtime": 0.5557,
+      "eval_samples_per_second": 862.053,
+      "eval_steps_per_second": 14.398,
+      "step": 32157
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 1.671371156182821e-05,
+      "loss": 6.6596,
+      "step": 35730
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.14966520045947385,
+      "eval_loss": 6.613977909088135,
+      "eval_runtime": 0.5538,
+      "eval_samples_per_second": 864.918,
+      "eval_steps_per_second": 14.445,
+      "step": 35730
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 1.3375081783344236e-05,
+      "loss": 6.6536,
+      "step": 39303
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.14933370739234114,
+      "eval_loss": 6.60673189163208,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 862.613,
+      "eval_steps_per_second": 14.407,
+      "step": 39303
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 1.0037386671651557e-05,
+      "loss": 6.6483,
+      "step": 42876
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.14892078326657765,
+      "eval_loss": 6.613958835601807,
+      "eval_runtime": 0.5595,
+      "eval_samples_per_second": 856.133,
+      "eval_steps_per_second": 14.299,
+      "step": 42876
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 6.699691559958874e-06,
+      "loss": 6.6463,
+      "step": 46449
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.14841714254033375,
+      "eval_loss": 6.609643459320068,
+      "eval_runtime": 0.5507,
+      "eval_samples_per_second": 869.79,
+      "eval_steps_per_second": 14.527,
+      "step": 46449
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 3.3619964482661936e-06,
+      "loss": 6.6434,
+      "step": 50022
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.15260783760924726,
+      "eval_loss": 6.556983947753906,
+      "eval_runtime": 0.5531,
+      "eval_samples_per_second": 866.025,
+      "eval_steps_per_second": 14.464,
+      "step": 50022
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 2.336666978222264e-08,
+      "loss": 6.6414,
+      "step": 53595
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.15261607518586057,
+      "eval_loss": 6.583581924438477,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 862.526,
+      "eval_steps_per_second": 14.405,
+      "step": 53595
+    },
+    {
+      "epoch": 15.0,
+      "step": 53595,
+      "total_flos": 3.972393508798464e+16,
+      "train_loss": 7.326953015673104,
+      "train_runtime": 5761.7187,
+      "train_samples_per_second": 595.236,
+      "train_steps_per_second": 9.302
+    }
+  ],
+  "max_steps": 53595,
+  "num_train_epochs": 15,
+  "total_flos": 3.972393508798464e+16,
+  "trial_name": null,
+  "trial_params": null
+}