Training in progress, step 200

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 96.7,
+    "eval_average_metrics": 96.7,
+    "eval_loss": 0.044097092002630234,
+    "eval_runtime": 2.8834,
+    "eval_samples_per_second": 346.813,
+    "test_accuracy": 93.34862385321101,
+    "test_average_metrics": 93.34862385321101,
+    "test_loss": 0.0836009681224823,
+    "test_runtime": 2.7323,
+    "test_samples_per_second": 319.145,
+    "train_loss": 0.3596552710935294,
+    "train_runtime": 685.371,
+    "train_samples": 66349,
+    "train_samples_per_second": 290.422,
+    "train_steps_per_second": 2.906
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 96.7,
+    "eval_average_metrics": 96.7,
+    "eval_loss": 0.044097092002630234,
+    "eval_runtime": 2.8834,
+    "eval_samples_per_second": 346.813
+}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5563c2e2460673852d723470e1e9fff2701e0bb0ca193b576a137a71ba84639
-size 1084131
+oid sha256:590cd97136a35765e55df0d19fa76ffb8cf347d609b8f9397ccab3d1a1ef3763
+size 7551621

runs/Feb01_01-01-41_node2/1643648613.4983714/events.out.tfevents.1643648613.node2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f39a5805d71ae68bd0d4ee714488a6dc92b93feb765d41556796c580fe8f53c
+size 5011

runs/Feb01_01-01-41_node2/events.out.tfevents.1643648613.node2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e66ac4100d225647961efb2280d52f26d4ada2de6047ec9760acc7294f5a51d2
+size 4303

runs/Jan31_20-32-44_node1/events.out.tfevents.1643632472.node1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fbf2112f44dbed5c384181bd2c4b280e3f5cf9db679fbc76baef51a7b37c9b99
-size 5418
+oid sha256:4752c331e5e161f14467d9daf76e8092116afdf5e3b8278a358e786d715c743f
+size 6094

runs/Jan31_20-32-44_node1/events.out.tfevents.1643633161.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82e765331c53a019307ea9ecf67aae3deea7ef9d531c84bbaeef48df19964137
+size 684

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "test_accuracy": 93.34862385321101,
+    "test_average_metrics": 93.34862385321101,
+    "test_loss": 0.0836009681224823,
+    "test_runtime": 2.7323,
+    "test_samples_per_second": 319.145
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.3596552710935294,
+    "train_runtime": 685.371,
+    "train_samples": 66349,
+    "train_samples_per_second": 290.422,
+    "train_steps_per_second": 2.906
+}

trainer_state.json

@@ -0,0 +1,70 @@
+{
+  "best_metric": 96.7,
+  "best_model_checkpoint": "outputs/bitfit/t5-base/sst2/checkpoint-500",
+  "epoch": 3.0,
+  "global_step": 1992,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0003,
+      "loss": 1.2354,
+      "step": 500
+    },
+    {
+      "epoch": 0.75,
+      "eval_accuracy": 96.7,
+      "eval_average_metrics": 96.7,
+      "eval_loss": 0.044097092002630234,
+      "eval_runtime": 2.8987,
+      "eval_samples_per_second": 344.985,
+      "step": 500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00019946380697050936,
+      "loss": 0.0704,
+      "step": 1000
+    },
+    {
+      "epoch": 1.51,
+      "eval_accuracy": 96.5,
+      "eval_average_metrics": 96.5,
+      "eval_loss": 0.042973704636096954,
+      "eval_runtime": 2.8885,
+      "eval_samples_per_second": 346.198,
+      "step": 1000
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 9.892761394101876e-05,
+      "loss": 0.0651,
+      "step": 1500
+    },
+    {
+      "epoch": 2.26,
+      "eval_accuracy": 96.5,
+      "eval_average_metrics": 96.5,
+      "eval_loss": 0.04074199125170708,
+      "eval_runtime": 2.8768,
+      "eval_samples_per_second": 347.613,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "step": 1992,
+      "total_flos": 1.5337491316024032e+16,
+      "train_loss": 0.3596552710935294,
+      "train_runtime": 685.371,
+      "train_samples_per_second": 290.422,
+      "train_steps_per_second": 2.906
+    }
+  ],
+  "max_steps": 1992,
+  "num_train_epochs": 3,
+  "total_flos": 1.5337491316024032e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48007d03701fe90a1362e795c1d43af79cfb253600770dc54a0c54176605fc6e
+oid sha256:f4d232d4e577478fd4816e9232ce9e0a60b78f395e047fb89a0e972636c59ef4
 size 3183

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": "sst2", "eval_steps": 500, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 3, "output_dir": "outputs/bitfit/t5-base/sst2", "overwrite_output_dir": true, "per_device_eval_batch_size": 100, "per_device_train_batch_size": 100, "predict_with_generate": true, "push_to_hub": true, "save_steps": 500, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "split_validation_test": true, "task_name": "sst2", "test_dataset_config_name": ["en"], "test_dataset_name": "sst2", "tokenizer_name": "../../../../plm_cache/t5-base", "warmup_steps": 500}
+{"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": "sst2", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 3, "output_dir": "outputs/bitfit/t5-base/sst2", "overwrite_output_dir": true, "per_device_eval_batch_size": 100, "per_device_train_batch_size": 100, "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": "sst2", "test_dataset_config_name": ["en"], "test_dataset_name": "sst2", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "warmup_steps": 500}