End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 12.0,
+    "eval_loss": 2.392148971557617,
+    "eval_runtime": 45.4129,
+    "eval_samples": 2933,
+    "eval_samples_per_second": 64.585,
+    "eval_steps_per_second": 8.081,
+    "perplexity": 10.936971948867901,
+    "train_loss": 2.415703173118034,
+    "train_runtime": 39608.6159,
+    "train_samples": 55736,
+    "train_samples_per_second": 16.886,
+    "train_steps_per_second": 0.264
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.0,
+    "eval_loss": 2.392148971557617,
+    "eval_runtime": 45.4129,
+    "eval_samples": 2933,
+    "eval_samples_per_second": 64.585,
+    "eval_steps_per_second": 8.081,
+    "perplexity": 10.936971948867901
+}

runs/Dec09_20-52-20_jbuhmann/events.out.tfevents.1639083162.jbuhmann.8182.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e2b13c8e2d6c6621cb9af03cbfbe95f27233c5e6f81478f193a86198239909f
-size 9569
+oid sha256:26dda307a1ca0b49575df0ee58cfc619b047534ed9e6b8c79dd29633c918f42c
+size 9923

runs/Dec09_20-52-20_jbuhmann/events.out.tfevents.1639122817.jbuhmann.8182.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78f4985fc30635510a1b9209b022d6ee0b873e81cc4985f4f225a4c452096f1e
+size 311

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 2.415703173118034,
+    "train_runtime": 39608.6159,
+    "train_samples": 55736,
+    "train_samples_per_second": 16.886,
+    "train_steps_per_second": 0.264
+}

trainer_state.json

@@ -0,0 +1,241 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 12.0,
+  "global_step": 10452,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002856486796785304,
+      "loss": 2.9349,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.9642183780670166,
+      "eval_runtime": 45.9478,
+      "eval_samples_per_second": 63.833,
+      "eval_steps_per_second": 7.987,
+      "step": 871
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0002712973593570608,
+      "loss": 3.0115,
+      "step": 1000
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00025694603903559125,
+      "loss": 2.9261,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.9243454933166504,
+      "eval_runtime": 45.8718,
+      "eval_samples_per_second": 63.939,
+      "eval_steps_per_second": 8.001,
+      "step": 1742
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.0002425947187141217,
+      "loss": 2.8685,
+      "step": 2000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.0002282433983926521,
+      "loss": 2.8409,
+      "step": 2500
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.8895106315612793,
+      "eval_runtime": 45.9241,
+      "eval_samples_per_second": 63.866,
+      "eval_steps_per_second": 7.991,
+      "step": 2613
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 0.00021389207807118253,
+      "loss": 2.7308,
+      "step": 3000
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.839418888092041,
+      "eval_runtime": 45.9075,
+      "eval_samples_per_second": 63.889,
+      "eval_steps_per_second": 7.994,
+      "step": 3484
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 0.00019954075774971297,
+      "loss": 2.698,
+      "step": 3500
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 0.00018518943742824339,
+      "loss": 2.6042,
+      "step": 4000
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.7702574729919434,
+      "eval_runtime": 45.8928,
+      "eval_samples_per_second": 63.91,
+      "eval_steps_per_second": 7.997,
+      "step": 4355
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 0.00017083811710677383,
+      "loss": 2.548,
+      "step": 4500
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 0.00015648679678530425,
+      "loss": 2.4671,
+      "step": 5000
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.7521896362304688,
+      "eval_runtime": 45.9244,
+      "eval_samples_per_second": 63.866,
+      "eval_steps_per_second": 7.991,
+      "step": 5226
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 0.00014213547646383466,
+      "loss": 2.4155,
+      "step": 5500
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 0.00012778415614236508,
+      "loss": 2.3481,
+      "step": 6000
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.633867025375366,
+      "eval_runtime": 45.9347,
+      "eval_samples_per_second": 63.852,
+      "eval_steps_per_second": 7.99,
+      "step": 6097
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 0.00011343283582089551,
+      "loss": 2.2493,
+      "step": 6500
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.6223971843719482,
+      "eval_runtime": 45.9444,
+      "eval_samples_per_second": 63.838,
+      "eval_steps_per_second": 7.988,
+      "step": 6968
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 9.908151549942594e-05,
+      "loss": 2.217,
+      "step": 7000
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 8.473019517795637e-05,
+      "loss": 2.1233,
+      "step": 7500
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.5636513233184814,
+      "eval_runtime": 45.9668,
+      "eval_samples_per_second": 63.807,
+      "eval_steps_per_second": 7.984,
+      "step": 7839
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 7.03788748564868e-05,
+      "loss": 2.0935,
+      "step": 8000
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 5.6027554535017214e-05,
+      "loss": 2.0194,
+      "step": 8500
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.4896304607391357,
+      "eval_runtime": 45.9773,
+      "eval_samples_per_second": 63.792,
+      "eval_steps_per_second": 7.982,
+      "step": 8710
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 4.1676234213547644e-05,
+      "loss": 1.9725,
+      "step": 9000
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 2.7324913892078067e-05,
+      "loss": 1.9178,
+      "step": 9500
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.4688611030578613,
+      "eval_runtime": 46.0218,
+      "eval_samples_per_second": 63.731,
+      "eval_steps_per_second": 7.974,
+      "step": 9581
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 1.2973593570608495e-05,
+      "loss": 1.8588,
+      "step": 10000
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.46632981300354,
+      "eval_runtime": 45.9791,
+      "eval_samples_per_second": 63.79,
+      "eval_steps_per_second": 7.982,
+      "step": 10452
+    },
+    {
+      "epoch": 12.0,
+      "step": 10452,
+      "total_flos": 1.760388788159447e+17,
+      "train_loss": 2.415703173118034,
+      "train_runtime": 39608.6159,
+      "train_samples_per_second": 16.886,
+      "train_steps_per_second": 0.264
+    }
+  ],
+  "max_steps": 10452,
+  "num_train_epochs": 12,
+  "total_flos": 1.760388788159447e+17,
+  "trial_name": null,
+  "trial_params": null
+}