End of training

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.7772225450319582,
+    "eval_loss": 0.7830244302749634,
+    "eval_runtime": 9.0522,
+    "eval_samples": 8605,
+    "eval_samples_per_second": 950.598,
+    "eval_steps_per_second": 118.866,
+    "total_flos": 4.52871045169152e+16,
+    "train_loss": 0.6343187750493727,
+    "train_runtime": 497.1667,
+    "train_samples": 34420,
+    "train_samples_per_second": 346.162,
+    "train_steps_per_second": 2.705
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.7772225450319582,
+    "eval_loss": 0.7830244302749634,
+    "eval_runtime": 9.0522,
+    "eval_samples": 8605,
+    "eval_samples_per_second": 950.598,
+    "eval_steps_per_second": 118.866
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "total_flos": 4.52871045169152e+16,
+    "train_loss": 0.6343187750493727,
+    "train_runtime": 497.1667,
+    "train_samples": 34420,
+    "train_samples_per_second": 346.162,
+    "train_steps_per_second": 2.705
+}

trainer_state.json

@@ -0,0 +1,178 @@
+{
+  "best_metric": 0.7830244302749634,
+  "best_model_checkpoint": "./bert-base-multilingual-cased/checkpoint-1345",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 1345,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.37174721189591076,
+      "grad_norm": 3.511324882507324,
+      "learning_rate": 9.25650557620818e-06,
+      "loss": 2.1825,
+      "step": 100
+    },
+    {
+      "epoch": 0.7434944237918215,
+      "grad_norm": 6.582064628601074,
+      "learning_rate": 8.513011152416357e-06,
+      "loss": 1.2704,
+      "step": 200
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7317838466008135,
+      "eval_loss": 0.9832090139389038,
+      "eval_runtime": 9.2351,
+      "eval_samples_per_second": 931.767,
+      "eval_steps_per_second": 116.511,
+      "step": 269
+    },
+    {
+      "epoch": 1.1152416356877324,
+      "grad_norm": 4.232427597045898,
+      "learning_rate": 7.769516728624536e-06,
+      "loss": 0.9943,
+      "step": 300
+    },
+    {
+      "epoch": 1.486988847583643,
+      "grad_norm": 4.7747673988342285,
+      "learning_rate": 7.0260223048327144e-06,
+      "loss": 0.9604,
+      "step": 400
+    },
+    {
+      "epoch": 1.858736059479554,
+      "grad_norm": 4.983441352844238,
+      "learning_rate": 6.282527881040893e-06,
+      "loss": 0.8977,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7608367228355607,
+      "eval_loss": 0.8561291098594666,
+      "eval_runtime": 9.0456,
+      "eval_samples_per_second": 951.291,
+      "eval_steps_per_second": 118.953,
+      "step": 538
+    },
+    {
+      "epoch": 2.2304832713754648,
+      "grad_norm": 4.7981953620910645,
+      "learning_rate": 5.539033457249071e-06,
+      "loss": 0.8473,
+      "step": 600
+    },
+    {
+      "epoch": 2.6022304832713754,
+      "grad_norm": 6.707701206207275,
+      "learning_rate": 4.795539033457249e-06,
+      "loss": 0.8154,
+      "step": 700
+    },
+    {
+      "epoch": 2.973977695167286,
+      "grad_norm": 5.124625205993652,
+      "learning_rate": 4.052044609665427e-06,
+      "loss": 0.7954,
+      "step": 800
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7712957582800697,
+      "eval_loss": 0.8103732466697693,
+      "eval_runtime": 9.0145,
+      "eval_samples_per_second": 954.573,
+      "eval_steps_per_second": 119.363,
+      "step": 807
+    },
+    {
+      "epoch": 3.345724907063197,
+      "grad_norm": 5.56990909576416,
+      "learning_rate": 3.3085501858736065e-06,
+      "loss": 0.7345,
+      "step": 900
+    },
+    {
+      "epoch": 3.717472118959108,
+      "grad_norm": 7.059451580047607,
+      "learning_rate": 2.5650557620817845e-06,
+      "loss": 0.7413,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7769901220220802,
+      "eval_loss": 0.7903115749359131,
+      "eval_runtime": 8.9773,
+      "eval_samples_per_second": 958.534,
+      "eval_steps_per_second": 119.858,
+      "step": 1076
+    },
+    {
+      "epoch": 4.089219330855019,
+      "grad_norm": 7.092191219329834,
+      "learning_rate": 1.821561338289963e-06,
+      "loss": 0.7269,
+      "step": 1100
+    },
+    {
+      "epoch": 4.4609665427509295,
+      "grad_norm": 6.146620750427246,
+      "learning_rate": 1.0780669144981413e-06,
+      "loss": 0.7001,
+      "step": 1200
+    },
+    {
+      "epoch": 4.83271375464684,
+      "grad_norm": 6.915247917175293,
+      "learning_rate": 3.345724907063197e-07,
+      "loss": 0.6915,
+      "step": 1300
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7772225450319582,
+      "eval_loss": 0.7830244302749634,
+      "eval_runtime": 9.0703,
+      "eval_samples_per_second": 948.705,
+      "eval_steps_per_second": 118.63,
+      "step": 1345
+    },
+    {
+      "epoch": 5.0,
+      "step": 1345,
+      "total_flos": 4.52871045169152e+16,
+      "train_loss": 0.6343187750493727,
+      "train_runtime": 497.1667,
+      "train_samples_per_second": 346.162,
+      "train_steps_per_second": 2.705
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 1345,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 5000.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.52871045169152e+16,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}