End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.4031584309730005,
+    "eval_loss": 1.0859721899032593,
+    "eval_runtime": 27.2148,
+    "eval_samples": 9815,
+    "eval_samples_per_second": 360.649,
+    "eval_steps_per_second": 11.281,
+    "train_loss": 1.0949462632001457,
+    "train_runtime": 12056.6812,
+    "train_samples": 392702,
+    "train_samples_per_second": 130.285,
+    "train_steps_per_second": 1.018
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.4031584309730005,
+    "eval_loss": 1.0859721899032593,
+    "eval_runtime": 27.2148,
+    "eval_samples": 9815,
+    "eval_samples_per_second": 360.649,
+    "eval_steps_per_second": 11.281
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 1.0949462632001457,
+    "train_runtime": 12056.6812,
+    "train_samples": 392702,
+    "train_samples_per_second": 130.285,
+    "train_steps_per_second": 1.018
+}

trainer_state.json

@@ -0,0 +1,205 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "global_step": 12272,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.808099739243807e-05,
+      "loss": 1.1464,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.604383963494133e-05,
+      "loss": 1.121,
+      "step": 1000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.400668187744459e-05,
+      "loss": 1.1114,
+      "step": 1500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.196952411994785e-05,
+      "loss": 1.1053,
+      "step": 2000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.9932366362451115e-05,
+      "loss": 1.1036,
+      "step": 2500
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 3.7899282920469364e-05,
+      "loss": 1.1021,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.3860417727967397,
+      "eval_loss": 1.0996018648147583,
+      "eval_runtime": 27.3661,
+      "eval_samples_per_second": 358.655,
+      "eval_steps_per_second": 11.218,
+      "step": 3068
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 3.586212516297262e-05,
+      "loss": 1.0988,
+      "step": 3500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.3824967405475883e-05,
+      "loss": 1.0957,
+      "step": 4000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.178780964797914e-05,
+      "loss": 1.0942,
+      "step": 4500
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 2.9754726205997396e-05,
+      "loss": 1.092,
+      "step": 5000
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 2.7717568448500652e-05,
+      "loss": 1.0908,
+      "step": 5500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 2.5680410691003915e-05,
+      "loss": 1.0902,
+      "step": 6000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.39470198675496687,
+      "eval_loss": 1.0913997888565063,
+      "eval_runtime": 27.3848,
+      "eval_samples_per_second": 358.41,
+      "eval_steps_per_second": 11.211,
+      "step": 6136
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.364325293350717e-05,
+      "loss": 1.088,
+      "step": 6500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.1610169491525427e-05,
+      "loss": 1.0875,
+      "step": 7000
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.9573011734028683e-05,
+      "loss": 1.0873,
+      "step": 7500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.7535853976531943e-05,
+      "loss": 1.0853,
+      "step": 8000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 1.5498696219035202e-05,
+      "loss": 1.0865,
+      "step": 8500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 1.3465612777053455e-05,
+      "loss": 1.0871,
+      "step": 9000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.3979623025980642,
+      "eval_loss": 1.0878193378448486,
+      "eval_runtime": 27.352,
+      "eval_samples_per_second": 358.84,
+      "eval_steps_per_second": 11.224,
+      "step": 9204
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 1.1428455019556714e-05,
+      "loss": 1.0866,
+      "step": 9500
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 9.391297262059974e-06,
+      "loss": 1.0847,
+      "step": 10000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 7.354139504563234e-06,
+      "loss": 1.0844,
+      "step": 10500
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 5.316981747066493e-06,
+      "loss": 1.0849,
+      "step": 11000
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 3.283898305084746e-06,
+      "loss": 1.0832,
+      "step": 11500
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 1.2467405475880053e-06,
+      "loss": 1.0871,
+      "step": 12000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4031584309730005,
+      "eval_loss": 1.0859721899032593,
+      "eval_runtime": 27.3545,
+      "eval_samples_per_second": 358.808,
+      "eval_steps_per_second": 11.223,
+      "step": 12272
+    },
+    {
+      "epoch": 4.0,
+      "step": 12272,
+      "total_flos": 5.465758560807813e+17,
+      "train_loss": 1.0949462632001457,
+      "train_runtime": 12056.6812,
+      "train_samples_per_second": 130.285,
+      "train_steps_per_second": 1.018
+    }
+  ],
+  "max_steps": 12272,
+  "num_train_epochs": 4,
+  "total_flos": 5.465758560807813e+17,
+  "trial_name": null,
+  "trial_params": null
+}