End of training

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 9.954264524103833,
+    "eval_accuracy": 0.9693806541405706,
+    "eval_loss": 0.10104648023843765,
+    "eval_runtime": 35.9112,
+    "eval_samples_per_second": 80.031,
+    "eval_steps_per_second": 2.506,
+    "total_flos": 2.631109878706471e+19,
+    "train_loss": 0.47059566762187693,
+    "train_runtime": 5303.842,
+    "train_samples_per_second": 48.768,
+    "train_steps_per_second": 0.381
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.954264524103833,
+    "eval_accuracy": 0.9693806541405706,
+    "eval_loss": 0.10104648023843765,
+    "eval_runtime": 35.9112,
+    "eval_samples_per_second": 80.031,
+    "eval_steps_per_second": 2.506
+}

runs/Dec17_06-28-09_8d5a9cee7ecd/events.out.tfevents.1734422246.8d5a9cee7ecd.1424.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c7d742c9036d9a7caa0b00b966a92eb2d2d649461059abc1e34481f56930198
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 9.954264524103833,
+    "total_flos": 2.631109878706471e+19,
+    "train_loss": 0.47059566762187693,
+    "train_runtime": 5303.842,
+    "train_samples_per_second": 48.768,
+    "train_steps_per_second": 0.381
+}

trainer_state.json

@@ -0,0 +1,202 @@
+{
+  "best_metric": 0.9693806541405706,
+  "best_model_checkpoint": "ky-finetuned-skindiseaseicthuawei25/checkpoint-2020",
+  "epoch": 9.954264524103833,
+  "eval_steps": 500,
+  "global_step": 2020,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 123.11931610107422,
+      "learning_rate": 4.997249724972497e-05,
+      "loss": 1.5505,
+      "step": 203
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7397355601948504,
+      "eval_loss": 0.8480764031410217,
+      "eval_runtime": 35.7495,
+      "eval_samples_per_second": 80.393,
+      "eval_steps_per_second": 2.518,
+      "step": 203
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 55.656707763671875,
+      "learning_rate": 4.438943894389439e-05,
+      "loss": 0.7314,
+      "step": 406
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8851774530271399,
+      "eval_loss": 0.39373043179512024,
+      "eval_runtime": 35.699,
+      "eval_samples_per_second": 80.507,
+      "eval_steps_per_second": 2.521,
+      "step": 406
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 58.35697937011719,
+      "learning_rate": 3.880638063806381e-05,
+      "loss": 0.5221,
+      "step": 609
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9057063326374392,
+      "eval_loss": 0.32093346118927,
+      "eval_runtime": 35.5732,
+      "eval_samples_per_second": 80.791,
+      "eval_steps_per_second": 2.53,
+      "step": 609
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 82.97339630126953,
+      "learning_rate": 3.322332233223323e-05,
+      "loss": 0.4238,
+      "step": 812
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9283228949199722,
+      "eval_loss": 0.23579993844032288,
+      "eval_runtime": 35.7701,
+      "eval_samples_per_second": 80.346,
+      "eval_steps_per_second": 2.516,
+      "step": 812
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 131.42694091796875,
+      "learning_rate": 2.764026402640264e-05,
+      "loss": 0.3587,
+      "step": 1015
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9495476687543494,
+      "eval_loss": 0.16918884217739105,
+      "eval_runtime": 35.8374,
+      "eval_samples_per_second": 80.196,
+      "eval_steps_per_second": 2.511,
+      "step": 1015
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 33.754241943359375,
+      "learning_rate": 2.2057205720572056e-05,
+      "loss": 0.2975,
+      "step": 1218
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9467640918580376,
+      "eval_loss": 0.17080058157444,
+      "eval_runtime": 35.8741,
+      "eval_samples_per_second": 80.114,
+      "eval_steps_per_second": 2.509,
+      "step": 1218
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 27.54159927368164,
+      "learning_rate": 1.6474147414741474e-05,
+      "loss": 0.2536,
+      "step": 1421
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.9592901878914405,
+      "eval_loss": 0.14195019006729126,
+      "eval_runtime": 35.7489,
+      "eval_samples_per_second": 80.394,
+      "eval_steps_per_second": 2.518,
+      "step": 1421
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 60.64569854736328,
+      "learning_rate": 1.0891089108910891e-05,
+      "loss": 0.2198,
+      "step": 1624
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9631176061238692,
+      "eval_loss": 0.12305579334497452,
+      "eval_runtime": 35.7591,
+      "eval_samples_per_second": 80.371,
+      "eval_steps_per_second": 2.517,
+      "step": 1624
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 35.08037185668945,
+      "learning_rate": 5.308030803080309e-06,
+      "loss": 0.1793,
+      "step": 1827
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.9686847599164927,
+      "eval_loss": 0.10935774445533752,
+      "eval_runtime": 35.7858,
+      "eval_samples_per_second": 80.311,
+      "eval_steps_per_second": 2.515,
+      "step": 1827
+    },
+    {
+      "epoch": 9.954264524103833,
+      "grad_norm": 26.716156005859375,
+      "learning_rate": 0.0,
+      "loss": 0.1538,
+      "step": 2020
+    },
+    {
+      "epoch": 9.954264524103833,
+      "eval_accuracy": 0.9693806541405706,
+      "eval_loss": 0.10104648023843765,
+      "eval_runtime": 35.9987,
+      "eval_samples_per_second": 79.836,
+      "eval_steps_per_second": 2.5,
+      "step": 2020
+    },
+    {
+      "epoch": 9.954264524103833,
+      "step": 2020,
+      "total_flos": 2.631109878706471e+19,
+      "train_loss": 0.47059566762187693,
+      "train_runtime": 5303.842,
+      "train_samples_per_second": 48.768,
+      "train_steps_per_second": 0.381
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2020,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.631109878706471e+19,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}