End of training

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 4.95,
+    "eval_accuracy": 0.9815789473684211,
+    "eval_loss": 0.16959501802921295,
+    "eval_runtime": 2.5149,
+    "eval_samples_per_second": 302.196,
+    "eval_steps_per_second": 9.543,
+    "train_loss": 0.34560299594447297,
+    "train_runtime": 321.4966,
+    "train_samples_per_second": 106.284,
+    "train_steps_per_second": 0.824
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.95,
+    "eval_accuracy": 0.9815789473684211,
+    "eval_loss": 0.16959501802921295,
+    "eval_runtime": 2.5149,
+    "eval_samples_per_second": 302.196,
+    "eval_steps_per_second": 9.543
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 4.95,
+    "train_loss": 0.34560299594447297,
+    "train_runtime": 321.4966,
+    "train_samples_per_second": 106.284,
+    "train_steps_per_second": 0.824
+}

trainer_state.json

@@ -0,0 +1,226 @@
+{
+  "best_metric": 0.9815789473684211,
+  "best_model_checkpoint": "neunit-ks-20230509V3/checkpoint-107",
+  "epoch": 4.953271028037383,
+  "global_step": 265,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.19,
+      "learning_rate": 1.111111111111111e-05,
+      "loss": 1.3681,
+      "step": 10
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.222222222222222e-05,
+      "loss": 1.2913,
+      "step": 20
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 2.9621848739495802e-05,
+      "loss": 1.1358,
+      "step": 30
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 2.8361344537815126e-05,
+      "loss": 0.8529,
+      "step": 40
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.7100840336134453e-05,
+      "loss": 0.5857,
+      "step": 50
+    },
+    {
+      "epoch": 0.99,
+      "eval_accuracy": 0.9328947368421052,
+      "eval_loss": 0.40735843777656555,
+      "eval_runtime": 2.4866,
+      "eval_samples_per_second": 305.64,
+      "eval_steps_per_second": 9.652,
+      "step": 53
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.5840336134453784e-05,
+      "loss": 0.4539,
+      "step": 60
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 2.4579831932773108e-05,
+      "loss": 0.3856,
+      "step": 70
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 2.331932773109244e-05,
+      "loss": 0.2886,
+      "step": 80
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2.2058823529411766e-05,
+      "loss": 0.2826,
+      "step": 90
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 2.0798319327731094e-05,
+      "loss": 0.2251,
+      "step": 100
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9815789473684211,
+      "eval_loss": 0.16959501802921295,
+      "eval_runtime": 2.5415,
+      "eval_samples_per_second": 299.033,
+      "eval_steps_per_second": 9.443,
+      "step": 107
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 1.953781512605042e-05,
+      "loss": 0.1937,
+      "step": 110
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.8277310924369748e-05,
+      "loss": 0.1776,
+      "step": 120
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.7016806722689076e-05,
+      "loss": 0.18,
+      "step": 130
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 1.5756302521008403e-05,
+      "loss": 0.1676,
+      "step": 140
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 1.4495798319327732e-05,
+      "loss": 0.1526,
+      "step": 150
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 1.323529411764706e-05,
+      "loss": 0.1465,
+      "step": 160
+    },
+    {
+      "epoch": 2.99,
+      "eval_accuracy": 0.975,
+      "eval_loss": 0.12837235629558563,
+      "eval_runtime": 2.5633,
+      "eval_samples_per_second": 296.489,
+      "eval_steps_per_second": 9.363,
+      "step": 160
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 1.1974789915966388e-05,
+      "loss": 0.1262,
+      "step": 170
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 1.0714285714285714e-05,
+      "loss": 0.1375,
+      "step": 180
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 9.453781512605041e-06,
+      "loss": 0.1449,
+      "step": 190
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 8.193277310924369e-06,
+      "loss": 0.1212,
+      "step": 200
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 6.932773109243698e-06,
+      "loss": 0.1169,
+      "step": 210
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9802631578947368,
+      "eval_loss": 0.09386323392391205,
+      "eval_runtime": 2.5788,
+      "eval_samples_per_second": 294.714,
+      "eval_steps_per_second": 9.307,
+      "step": 214
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 5.672268907563025e-06,
+      "loss": 0.1209,
+      "step": 220
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 4.411764705882353e-06,
+      "loss": 0.1219,
+      "step": 230
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.1512605042016808e-06,
+      "loss": 0.1207,
+      "step": 240
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 1.8907563025210083e-06,
+      "loss": 0.0989,
+      "step": 250
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 6.302521008403362e-07,
+      "loss": 0.1153,
+      "step": 260
+    },
+    {
+      "epoch": 4.95,
+      "eval_accuracy": 0.9789473684210527,
+      "eval_loss": 0.09296128153800964,
+      "eval_runtime": 2.5784,
+      "eval_samples_per_second": 294.755,
+      "eval_steps_per_second": 9.308,
+      "step": 265
+    },
+    {
+      "epoch": 4.95,
+      "step": 265,
+      "total_flos": 4.1714064555397216e+17,
+      "train_loss": 0.34560299594447297,
+      "train_runtime": 321.4966,
+      "train_samples_per_second": 106.284,
+      "train_steps_per_second": 0.824
+    }
+  ],
+  "max_steps": 265,
+  "num_train_epochs": 5,
+  "total_flos": 4.1714064555397216e+17,
+  "trial_name": null,
+  "trial_params": null
+}