Training in progress, step 200

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44193738456d1aa1026955d0a9f9e752e5b37d1130a182b5627652fdf8a3e621
-size 7551621
+oid sha256:2e6e806c5df9c46503a099f53e09783ce85f1a21b64f2b934557d06db90a9406
+size 879301

runs/Feb02_17-59-15_node4/1643796061.211064/events.out.tfevents.1643796061.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c80fd709f75dbddfc08a25b32abffc08ef25780bbfb95c00bc2f5b95a308494
+size 5005

runs/Feb02_17-59-15_node4/events.out.tfevents.1643796061.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3659e7994fc02fd0df7fbc88e27906a3fa7cf1e97b4e423edd16e96396a17a1
+size 4294

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fefbef8122e9411dd16f8ffeebe201a8890e42d3c2d5898454de12fc6489eb55
+oid sha256:62f92480d5f15a84117d30f2a6a088445639336c20d95f7392fa3071d4e4eb97
 size 3183

training_config.json

@@ -1 +1 @@
-{"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": "mrpc", "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": 20, "output_dir": "outputs/bitfit/t5-base/mrpc", "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": "mrpc", "test_dataset_config_name": ["en"], "test_dataset_name": "mrpc", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "warmup_steps": 0}
+{"dataset_config_name": ["en"], "delta_type": "compacter", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "mrpc", "eval_steps": 200, "evaluation_strategy": "steps", "factorized_phm": true, "factorized_phm_rule": false, "gradient_clip": false, "greater_is_better": true, "hypercomplex_adapters": true, "hypercomplex_division": 4, "hypercomplex_nonlinearity": "glorot-uniform", "learn_phm": 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", "normalize_phm_weight": false, "num_train_epochs": 20, "output_dir": "outputs/lora/t5-base/mrpc", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "phm_c_init": "normal", "phm_clamp": false, "phm_init_range": 0.0001, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "shared_phm_rule": false, "split_validation_test": true, "task_name": "mrpc", "test_dataset_config_name": ["en"], "test_dataset_name": "mrpc", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "use_bias_down_sampler": true, "use_bias_up_sampler": true, "warmup_steps": 0}