End of training

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9323394495412844,
+    "eval_loss": 0.2478291541337967,
+    "eval_runtime": 3.9031,
+    "eval_samples": 872,
+    "eval_samples_per_second": 223.413,
+    "eval_steps_per_second": 27.927,
+    "train_loss": 0.1281920957829588,
+    "train_runtime": 2481.1506,
+    "train_samples": 67349,
+    "train_samples_per_second": 81.433,
+    "train_steps_per_second": 2.545
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9323394495412844,
+    "eval_loss": 0.2478291541337967,
+    "eval_runtime": 3.9031,
+    "eval_samples": 872,
+    "eval_samples_per_second": 223.413,
+    "eval_steps_per_second": 27.927
+}

runs/Jun21_14-48-15_mewtwo/events.out.tfevents.1655825695.mewtwo

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c10309ce8b4d2b861f73c5a43ff51497da794f395df02380e990a67353b2458f
+size 40

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.1281920957829588,
+    "train_runtime": 2481.1506,
+    "train_samples": 67349,
+    "train_samples_per_second": 81.433,
+    "train_steps_per_second": 2.545
+}

trainer_state.json

@@ -0,0 +1,124 @@
+{
+  "best_metric": 0.9323394495412844,
+  "best_model_checkpoint": "./fine-tune/bert-base-uncased/sst2/checkpoint-4210",
+  "epoch": 3.0,
+  "global_step": 6315,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8416468725257326e-05,
+      "loss": 0.2904,
+      "step": 500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 1.6832937450514647e-05,
+      "loss": 0.2097,
+      "step": 1000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 1.5249406175771972e-05,
+      "loss": 0.1752,
+      "step": 1500
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.3665874901029297e-05,
+      "loss": 0.1668,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9174311926605505,
+      "eval_loss": 0.2512936294078827,
+      "eval_runtime": 4.0394,
+      "eval_samples_per_second": 215.874,
+      "eval_steps_per_second": 26.984,
+      "step": 2105
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.208234362628662e-05,
+      "loss": 0.1199,
+      "step": 2500
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 1.0498812351543943e-05,
+      "loss": 0.1049,
+      "step": 3000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 8.915281076801267e-06,
+      "loss": 0.1075,
+      "step": 3500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 7.331749802058591e-06,
+      "loss": 0.1119,
+      "step": 4000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9323394495412844,
+      "eval_loss": 0.2478291541337967,
+      "eval_runtime": 4.1684,
+      "eval_samples_per_second": 209.192,
+      "eval_steps_per_second": 26.149,
+      "step": 4210
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 5.748218527315916e-06,
+      "loss": 0.0828,
+      "step": 4500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.164687252573238e-06,
+      "loss": 0.0667,
+      "step": 5000
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.581155977830562e-06,
+      "loss": 0.069,
+      "step": 5500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 9.976247030878861e-07,
+      "loss": 0.0699,
+      "step": 6000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.926605504587156,
+      "eval_loss": 0.2763631343841553,
+      "eval_runtime": 4.0199,
+      "eval_samples_per_second": 216.923,
+      "eval_steps_per_second": 27.115,
+      "step": 6315
+    },
+    {
+      "epoch": 3.0,
+      "step": 6315,
+      "total_flos": 1.329019985058048e+16,
+      "train_loss": 0.1281920957829588,
+      "train_runtime": 2481.1506,
+      "train_samples_per_second": 81.433,
+      "train_steps_per_second": 2.545
+    }
+  ],
+  "max_steps": 6315,
+  "num_train_epochs": 3,
+  "total_flos": 1.329019985058048e+16,
+  "trial_name": null,
+  "trial_params": null
+}