Training in progress, step 100

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 55.769230769230774,
+    "eval_average_metrics": 55.769230769230774,
+    "eval_loss": 0.26899516582489014,
+    "eval_runtime": 0.2241,
+    "eval_samples_per_second": 232.063,
+    "test_accuracy": 61.53846153846154,
+    "test_average_metrics": 61.53846153846154,
+    "test_loss": 0.2528702914714813,
+    "test_runtime": 0.2148,
+    "test_samples_per_second": 242.088,
+    "train_loss": 0.4271513197157118,
+    "train_runtime": 68.8963,
+    "train_samples": 554,
+    "train_samples_per_second": 160.821,
+    "train_steps_per_second": 5.225
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 55.769230769230774,
+    "eval_average_metrics": 55.769230769230774,
+    "eval_loss": 0.26899516582489014,
+    "eval_runtime": 0.2241,
+    "eval_samples_per_second": 232.063
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce617a82380388074f3274e6cecc2826898c891cec91bf5bd703d915b0a182cb
-size 1084131
+oid sha256:640fc4d9189b0b756a68b86353b92a29fbd757ec67f97722f98e2aa929516a3c
+size 7551621

runs/Feb01_00-26-48_node1/events.out.tfevents.1643646517.node1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ed9c54355e2241e333aa192326cddbbabe37b3040f1868ff84abccaf11cef38
-size 4345
+oid sha256:7752258f4ed65ef3b718c79e03a9e6aea8b0dc6b7e35647f4e7d0911a5bd84e0
+size 5021

runs/Feb01_00-26-48_node1/events.out.tfevents.1643646586.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:027e7833239f92c6d3d46180d26a806869a05bf1d5fe8369825ccc473237e51f
+size 684

runs/Feb01_07-17-34_node1/1643671165.268331/events.out.tfevents.1643671165.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:146c31c8346cd89ac0bb51e54c9319502835be6e2dc64d0fa0d5b8e6d0a892df
+size 5056

runs/Feb01_07-17-34_node1/events.out.tfevents.1643671165.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f240a9ae5872c1335ab696246379e5c4662862c102da7dfe8a99dde30aa9bae
+size 4345

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "test_accuracy": 61.53846153846154,
+    "test_average_metrics": 61.53846153846154,
+    "test_loss": 0.2528702914714813,
+    "test_runtime": 0.2148,
+    "test_samples_per_second": 242.088
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "train_loss": 0.4271513197157118,
+    "train_runtime": 68.8963,
+    "train_samples": 554,
+    "train_samples_per_second": 160.821,
+    "train_steps_per_second": 5.225
+}

trainer_state.json

@@ -0,0 +1,34 @@
+{
+  "best_metric": 55.769230769230774,
+  "best_model_checkpoint": "outputs/bitfit/t5-base/superglue-wsc.fixed/checkpoint-200",
+  "epoch": 20.0,
+  "global_step": 360,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 11.11,
+      "eval_accuracy": 55.769230769230774,
+      "eval_average_metrics": 55.769230769230774,
+      "eval_loss": 0.26899516582489014,
+      "eval_runtime": 0.4189,
+      "eval_samples_per_second": 124.126,
+      "step": 200
+    },
+    {
+      "epoch": 20.0,
+      "step": 360,
+      "total_flos": 1490349041823024.0,
+      "train_loss": 0.4271513197157118,
+      "train_runtime": 68.8963,
+      "train_samples_per_second": 160.821,
+      "train_steps_per_second": 5.225
+    }
+  ],
+  "max_steps": 360,
+  "num_train_epochs": 20,
+  "total_flos": 1490349041823024.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ceb1d8d8920be731c337aff46060f3138c9f038b6ec9a18c68f775d2cdeb074b
+oid sha256:7bd02b28bab24c1face8a1dc29c40509cef5a1f6da4574d8ba3a16d8da091037
 size 3247

training_config.json

@@ -1 +1 @@
-{"dataset_config_name": ["en"], "delta_type": "bitfit", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "superglue-wsc.fixed", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 256, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 20, "output_dir": "outputs/bitfit/t5-base/superglue-wsc.fixed", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "split_validation_test": true, "task_name": "superglue-wsc.fixed", "test_dataset_config_name": ["en"], "test_dataset_name": "superglue-wsc.fixed", "tokenizer_name": "../../../../plm_cache/t5-base", "warmup_steps": 0}
+{"bottleneck_dim": 24, "dataset_config_name": ["en"], "delta_type": "adapter", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "superglue-wsc.fixed", "eval_steps": 100, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 256, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 20, "output_dir": "outputs/bitfit/t5-base/superglue-wsc.fixed", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "predict_with_generate": true, "push_to_hub": true, "save_steps": 100, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "split_validation_test": true, "task_name": "superglue-wsc.fixed", "test_dataset_config_name": ["en"], "test_dataset_name": "superglue-wsc.fixed", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "warmup_steps": 0}