End of training

all_results.json

@@ -0,0 +1,21 @@
+{
+    "epoch": 5.0,
+    "epoch_mm": 5.0,
+    "eval_accuracy": 0.8678553234844626,
+    "eval_accuracy_mm": 0.8695077298616761,
+    "eval_loss": 0.3816947042942047,
+    "eval_loss_mm": 0.38041895627975464,
+    "eval_runtime": 19.3128,
+    "eval_runtime_mm": 19.5375,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 508.212,
+    "eval_samples_per_second_mm": 503.237,
+    "eval_steps_per_second": 63.533,
+    "eval_steps_per_second_mm": 62.905,
+    "train_loss": 0.3547627031438948,
+    "train_runtime": 9507.0849,
+    "train_samples": 392702,
+    "train_samples_per_second": 206.531,
+    "train_steps_per_second": 0.516
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 5.0,
+    "epoch_mm": 5.0,
+    "eval_accuracy": 0.8678553234844626,
+    "eval_accuracy_mm": 0.8695077298616761,
+    "eval_loss": 0.3816947042942047,
+    "eval_loss_mm": 0.38041895627975464,
+    "eval_runtime": 19.3128,
+    "eval_runtime_mm": 19.5375,
+    "eval_samples": 9815,
+    "eval_samples_mm": 9832,
+    "eval_samples_per_second": 508.212,
+    "eval_samples_per_second_mm": 503.237,
+    "eval_steps_per_second": 63.533,
+    "eval_steps_per_second_mm": 62.905
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.3547627031438948,
+    "train_runtime": 9507.0849,
+    "train_samples": 392702,
+    "train_samples_per_second": 206.531,
+    "train_steps_per_second": 0.516
+}

trainer_state.json

@@ -0,0 +1,79 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "global_step": 4910,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.51,
+      "learning_rate": 1.7963340122199594e-05,
+      "loss": 0.6083,
+      "step": 500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 1.5926680244399187e-05,
+      "loss": 0.4428,
+      "step": 1000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 1.3890020366598779e-05,
+      "loss": 0.3786,
+      "step": 1500
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 1.1853360488798371e-05,
+      "loss": 0.3679,
+      "step": 2000
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 9.816700610997964e-06,
+      "loss": 0.3219,
+      "step": 2500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 7.780040733197556e-06,
+      "loss": 0.3177,
+      "step": 3000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 5.74338085539715e-06,
+      "loss": 0.2876,
+      "step": 3500
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.7067209775967417e-06,
+      "loss": 0.2815,
+      "step": 4000
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 1.670061099796334e-06,
+      "loss": 0.2619,
+      "step": 4500
+    },
+    {
+      "epoch": 5.0,
+      "step": 4910,
+      "total_flos": 0.0,
+      "train_loss": 0.3547627031438948,
+      "train_runtime": 9507.0849,
+      "train_samples_per_second": 206.531,
+      "train_steps_per_second": 0.516
+    }
+  ],
+  "max_steps": 4910,
+  "num_train_epochs": 5,
+  "total_flos": 0.0,
+  "trial_name": null,
+  "trial_params": null
+}