End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 1.9459083080291748,
+    "eval_runtime": 80.6558,
+    "eval_samples": 5193,
+    "eval_samples_per_second": 64.385,
+    "eval_steps_per_second": 8.059,
+    "perplexity": 6.999987112828894,
+    "train_loss": 2.0216803000285624,
+    "train_runtime": 46553.0134,
+    "train_samples": 98673,
+    "train_samples_per_second": 16.957,
+    "train_steps_per_second": 0.265
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 8.0,
+    "eval_loss": 1.9459083080291748,
+    "eval_runtime": 80.6558,
+    "eval_samples": 5193,
+    "eval_samples_per_second": 64.385,
+    "eval_steps_per_second": 8.059,
+    "perplexity": 6.999987112828894
+}

runs/Dec16_06-22-04_jbuhmann/events.out.tfevents.1639635743.jbuhmann.11147.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fac91bec09538c6b157c7bb7384398cdfe0cbf88dad75bca946afa7b381d9fee
-size 9094
+oid sha256:6eb64187ed0e41ae9a14b44fb1d5af5266afdbd11fb71d50a9717f6654dd4b86
+size 9448

runs/Dec16_06-22-04_jbuhmann/events.out.tfevents.1639682378.jbuhmann.11147.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49a4aef9cbb61e89405f7b5ae7b4547c6d35fb79f52544bf4488b42514e87582
+size 311

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "train_loss": 2.0216803000285624,
+    "train_runtime": 46553.0134,
+    "train_samples": 98673,
+    "train_samples_per_second": 16.957,
+    "train_steps_per_second": 0.265
+}

trainer_state.json

@@ -0,0 +1,233 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "global_step": 12336,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.32,
+      "learning_rate": 1e-05,
+      "loss": 2.3805,
+      "step": 500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 2e-05,
+      "loss": 2.2005,
+      "step": 1000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.911785462244178e-05,
+      "loss": 2.1951,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.028538465499878,
+      "eval_runtime": 81.1924,
+      "eval_samples_per_second": 63.959,
+      "eval_steps_per_second": 8.006,
+      "step": 1542
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.8235709244883558e-05,
+      "loss": 2.1375,
+      "step": 2000
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.7353563867325337e-05,
+      "loss": 2.1094,
+      "step": 2500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.6471418489767114e-05,
+      "loss": 2.0918,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 1.998868703842163,
+      "eval_runtime": 81.2459,
+      "eval_samples_per_second": 63.917,
+      "eval_steps_per_second": 8.0,
+      "step": 3084
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.5589273112208893e-05,
+      "loss": 2.0758,
+      "step": 3500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 1.4707127734650673e-05,
+      "loss": 2.0536,
+      "step": 4000
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 1.382498235709245e-05,
+      "loss": 2.0562,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.0162477493286133,
+      "eval_runtime": 81.3116,
+      "eval_samples_per_second": 63.865,
+      "eval_steps_per_second": 7.994,
+      "step": 4626
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 1.2942836979534227e-05,
+      "loss": 2.014,
+      "step": 5000
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 1.2060691601976007e-05,
+      "loss": 1.9984,
+      "step": 5500
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 1.1178546224417785e-05,
+      "loss": 2.0012,
+      "step": 6000
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 1.9330477714538574,
+      "eval_runtime": 81.304,
+      "eval_samples_per_second": 63.871,
+      "eval_steps_per_second": 7.995,
+      "step": 6168
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 1.0296400846859562e-05,
+      "loss": 1.9951,
+      "step": 6500
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 9.414255469301343e-06,
+      "loss": 1.9791,
+      "step": 7000
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 8.53211009174312e-06,
+      "loss": 1.9705,
+      "step": 7500
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 1.9150911569595337,
+      "eval_runtime": 81.3731,
+      "eval_samples_per_second": 63.817,
+      "eval_steps_per_second": 7.988,
+      "step": 7710
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 7.649964714184899e-06,
+      "loss": 1.9555,
+      "step": 8000
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 6.767819336626676e-06,
+      "loss": 1.9358,
+      "step": 8500
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 5.885673959068455e-06,
+      "loss": 1.9571,
+      "step": 9000
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 1.9419493675231934,
+      "eval_runtime": 81.4575,
+      "eval_samples_per_second": 63.751,
+      "eval_steps_per_second": 7.98,
+      "step": 9252
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 5.003528581510233e-06,
+      "loss": 1.9418,
+      "step": 9500
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 4.121383203952012e-06,
+      "loss": 1.9208,
+      "step": 10000
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 3.2392378263937903e-06,
+      "loss": 1.9113,
+      "step": 10500
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.9175336360931396,
+      "eval_runtime": 81.2139,
+      "eval_samples_per_second": 63.942,
+      "eval_steps_per_second": 8.004,
+      "step": 10794
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 2.3570924488355683e-06,
+      "loss": 1.9097,
+      "step": 11000
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.4749470712773468e-06,
+      "loss": 1.8942,
+      "step": 11500
+    },
+    {
+      "epoch": 7.78,
+      "learning_rate": 5.92801693719125e-07,
+      "loss": 1.8988,
+      "step": 12000
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 1.914337396621704,
+      "eval_runtime": 81.2673,
+      "eval_samples_per_second": 63.9,
+      "eval_steps_per_second": 7.998,
+      "step": 12336
+    },
+    {
+      "epoch": 8.0,
+      "step": 12336,
+      "total_flos": 2.0776857912786125e+17,
+      "train_loss": 2.0216803000285624,
+      "train_runtime": 46553.0134,
+      "train_samples_per_second": 16.957,
+      "train_steps_per_second": 0.265
+    }
+  ],
+  "max_steps": 12336,
+  "num_train_epochs": 8,
+  "total_flos": 2.0776857912786125e+17,
+  "trial_name": null,
+  "trial_params": null
+}