End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 12.0,
+    "eval_loss": 1.9043388366699219,
+    "eval_runtime": 45.4507,
+    "eval_samples": 2933,
+    "eval_samples_per_second": 64.531,
+    "eval_steps_per_second": 8.075,
+    "perplexity": 6.714966470038764,
+    "train_loss": 1.9039230375877523,
+    "train_runtime": 39633.1701,
+    "train_samples": 55736,
+    "train_samples_per_second": 16.876,
+    "train_steps_per_second": 0.264
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.0,
+    "eval_loss": 1.9043388366699219,
+    "eval_runtime": 45.4507,
+    "eval_samples": 2933,
+    "eval_samples_per_second": 64.531,
+    "eval_steps_per_second": 8.075,
+    "perplexity": 6.714966470038764
+}

runs/Dec08_10-38-27_jbuhmann/events.out.tfevents.1638959926.jbuhmann.17680.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4533ccf8f17cc400a83c82f4388ef1c7b1c9e1ccf9c08171fb59221ebade01ff
-size 9568
+oid sha256:e0b3f6a25a9bb0d398076b38c1764fad41dd0f35a312721aca74b04c1f6ad37b
+size 9922

runs/Dec08_10-38-27_jbuhmann/events.out.tfevents.1638999606.jbuhmann.17680.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6c526055ecf7dd1cbf6d94a1d3f847574851949e1759921d8f1721a40eeccfe
+size 311

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.0,
+    "train_loss": 1.9039230375877523,
+    "train_runtime": 39633.1701,
+    "train_samples": 55736,
+    "train_samples_per_second": 16.876,
+    "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": 4.760811327975507e-05,
+      "loss": 2.2438,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.1121535301208496,
+      "eval_runtime": 46.0061,
+      "eval_samples_per_second": 63.752,
+      "eval_steps_per_second": 7.977,
+      "step": 871
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.521622655951014e-05,
+      "loss": 2.1935,
+      "step": 1000
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 4.282433983926522e-05,
+      "loss": 2.1235,
+      "step": 1500
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.0784153938293457,
+      "eval_runtime": 45.9581,
+      "eval_samples_per_second": 63.819,
+      "eval_steps_per_second": 7.986,
+      "step": 1742
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 4.043245311902029e-05,
+      "loss": 2.0744,
+      "step": 2000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 3.8040566398775356e-05,
+      "loss": 2.0712,
+      "step": 2500
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.0678515434265137,
+      "eval_runtime": 45.9902,
+      "eval_samples_per_second": 63.774,
+      "eval_steps_per_second": 7.98,
+      "step": 2613
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 3.5648679678530425e-05,
+      "loss": 2.0034,
+      "step": 3000
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.0545804500579834,
+      "eval_runtime": 45.9685,
+      "eval_samples_per_second": 63.805,
+      "eval_steps_per_second": 7.984,
+      "step": 3484
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 3.32567929582855e-05,
+      "loss": 1.9854,
+      "step": 3500
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.086490623804057e-05,
+      "loss": 1.9375,
+      "step": 4000
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.027700662612915,
+      "eval_runtime": 45.9612,
+      "eval_samples_per_second": 63.815,
+      "eval_steps_per_second": 7.985,
+      "step": 4355
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 2.847301951779564e-05,
+      "loss": 1.9208,
+      "step": 4500
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 2.608113279755071e-05,
+      "loss": 1.8911,
+      "step": 5000
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.0363667011260986,
+      "eval_runtime": 45.9893,
+      "eval_samples_per_second": 63.776,
+      "eval_steps_per_second": 7.98,
+      "step": 5226
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.368924607730578e-05,
+      "loss": 1.8687,
+      "step": 5500
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 2.1297359357060852e-05,
+      "loss": 1.8454,
+      "step": 6000
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 1.981174111366272,
+      "eval_runtime": 45.9582,
+      "eval_samples_per_second": 63.819,
+      "eval_steps_per_second": 7.986,
+      "step": 6097
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.890547263681592e-05,
+      "loss": 1.808,
+      "step": 6500
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.0175108909606934,
+      "eval_runtime": 45.9528,
+      "eval_samples_per_second": 63.826,
+      "eval_steps_per_second": 7.986,
+      "step": 6968
+    },
+    {
+      "epoch": 8.04,
+      "learning_rate": 1.651358591657099e-05,
+      "loss": 1.81,
+      "step": 7000
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 1.4121699196326064e-05,
+      "loss": 1.7716,
+      "step": 7500
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.028625965118408,
+      "eval_runtime": 45.9318,
+      "eval_samples_per_second": 63.856,
+      "eval_steps_per_second": 7.99,
+      "step": 7839
+    },
+    {
+      "epoch": 9.18,
+      "learning_rate": 1.1729812476081135e-05,
+      "loss": 1.7693,
+      "step": 8000
+    },
+    {
+      "epoch": 9.76,
+      "learning_rate": 9.337925755836204e-06,
+      "loss": 1.7519,
+      "step": 8500
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 1.9653006792068481,
+      "eval_runtime": 45.9563,
+      "eval_samples_per_second": 63.821,
+      "eval_steps_per_second": 7.986,
+      "step": 8710
+    },
+    {
+      "epoch": 10.33,
+      "learning_rate": 6.946039035591274e-06,
+      "loss": 1.7484,
+      "step": 9000
+    },
+    {
+      "epoch": 10.91,
+      "learning_rate": 4.554152315346345e-06,
+      "loss": 1.7358,
+      "step": 9500
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 1.9817312955856323,
+      "eval_runtime": 45.9672,
+      "eval_samples_per_second": 63.806,
+      "eval_steps_per_second": 7.984,
+      "step": 9581
+    },
+    {
+      "epoch": 11.48,
+      "learning_rate": 2.1622655951014164e-06,
+      "loss": 1.7084,
+      "step": 10000
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 1.9633114337921143,
+      "eval_runtime": 45.9914,
+      "eval_samples_per_second": 63.773,
+      "eval_steps_per_second": 7.98,
+      "step": 10452
+    },
+    {
+      "epoch": 12.0,
+      "step": 10452,
+      "total_flos": 1.760388788159447e+17,
+      "train_loss": 1.9039230375877523,
+      "train_runtime": 39633.1701,
+      "train_samples_per_second": 16.876,
+      "train_steps_per_second": 0.264
+    }
+  ],
+  "max_steps": 10452,
+  "num_train_epochs": 12,
+  "total_flos": 1.760388788159447e+17,
+  "trial_name": null,
+  "trial_params": null
+}