add model weights and configs

added_tokens.json

@@ -0,0 +1 @@
+{}

cartography_config_dev.json

@@ -0,0 +1,49 @@
+{
+    "adam_epsilon": 1e-08,
+    "cache_dir": "",
+    "config_name": "",
+    "data_dir": "/content/data/glue/MNLI",
+    "dev": null,
+    "device": "cuda",
+    "do_eval": true,
+    "do_lower_case": true,
+    "do_test": false,
+    "do_train": true,
+    "eval_all_checkpoints": false,
+    "eval_batch_size": 96,
+    "evaluate_during_training": true,
+    "evaluate_during_training_epoch": false,
+    "features_cache_dir": "/content/data/glue/MNLI/cache_36891",
+    "fp16": false,
+    "fp16_opt_level": "01",
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 1.0993071205018916e-05,
+    "local_rank": -1,
+    "logging_steps": 1000,
+    "max_grad_norm": 1.0,
+    "max_seq_length": 128,
+    "max_steps": -1,
+    "model_name_or_path": "roberta-base",
+    "model_type": "roberta",
+    "n_gpu": 1,
+    "no_cuda": false,
+    "num_train_epochs": 6,
+    "output_dir": "/content/results",
+    "output_mode": "classification",
+    "overwrite_cache": false,
+    "overwrite_output_dir": false,
+    "patience": 3,
+    "per_gpu_eval_batch_size": 96,
+    "per_gpu_train_batch_size": 64,
+    "save_steps": 0,
+    "seed": 36891,
+    "server_ip": "",
+    "server_port": "",
+    "task_name": "mnli",
+    "test": null,
+    "tokenizer_name": "",
+    "train": null,
+    "train_batch_size": 64,
+    "warmup_steps": 0,
+    "weight_decay": 0.0
+}

cartography_config_train.json

@@ -0,0 +1,47 @@
+{
+    "adam_epsilon": 1e-08,
+    "cache_dir": "",
+    "config_name": "",
+    "data_dir": "/content/data/glue/MNLI",
+    "dev": null,
+    "device": "cuda",
+    "do_eval": true,
+    "do_lower_case": true,
+    "do_test": false,
+    "do_train": true,
+    "eval_all_checkpoints": false,
+    "evaluate_during_training": true,
+    "evaluate_during_training_epoch": false,
+    "features_cache_dir": "/content/data/glue/MNLI/cache_36891",
+    "fp16": false,
+    "fp16_opt_level": "01",
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 1.0993071205018916e-05,
+    "local_rank": -1,
+    "logging_steps": 1000,
+    "max_grad_norm": 1.0,
+    "max_seq_length": 128,
+    "max_steps": -1,
+    "model_name_or_path": "roberta-base",
+    "model_type": "roberta",
+    "n_gpu": 1,
+    "no_cuda": false,
+    "num_train_epochs": 6,
+    "output_dir": "/content/results",
+    "output_mode": "classification",
+    "overwrite_cache": false,
+    "overwrite_output_dir": false,
+    "patience": 3,
+    "per_gpu_eval_batch_size": 96,
+    "per_gpu_train_batch_size": 64,
+    "save_steps": 0,
+    "seed": 36891,
+    "server_ip": "",
+    "server_port": "",
+    "task_name": "mnli",
+    "test": null,
+    "tokenizer_name": "",
+    "train": null,
+    "warmup_steps": 0,
+    "weight_decay": 0.0
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "eos_token_id": 2,
+  "finetuning_task": "mnli",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "num_labels": 3,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 1,
+  "pruned_heads": {},
+  "torchscript": false,
+  "type_vocab_size": 1,
+  "use_bfloat16": false,
+  "vocab_size": 50265
+}

eval_metrics_mnli_dev_.json

@@ -0,0 +1 @@
+{"acc": 0.8799796230259807}

eval_metrics_mnli_dev_in_training.json

@@ -0,0 +1,6 @@
+{"acc": 0.8644931227712684}
+{"acc": 0.8760061130922058}
+{"acc": 0.8799796230259807}
+{"acc": 0.8785532348446256}
+{"acc": 0.8790626591951095}
+{"acc": 0.8786551197147223}

eval_metrics_train.json

@@ -0,0 +1,6 @@
+{"epoch": 0, "train_acc": 0.8231992706938085, "best_dev_performance": 0.8644931227712684, "avg_batch_loss": 43.14311441895552, "learning_rate": 9.160892670849097e-06}
+{"epoch": 1, "train_acc": 0.8737134010012682, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 32.02692841726821, "learning_rate": 7.3287141366792775e-06}
+{"epoch": 2, "train_acc": 0.8975024318694582, "best_dev_performance": 0.8799796230259807, "avg_batch_loss": 26.392389136832207, "learning_rate": 5.496535602509458e-06}
+{"epoch": 3, "train_acc": 0.9159388034692973, "best_dev_performance": 0.8799796230259807, "avg_batch_loss": 22.091516123269685, "learning_rate": 3.6643570683396387e-06}
+{"epoch": 4, "train_acc": 0.9287704162443787, "best_dev_performance": 0.8799796230259807, "avg_batch_loss": 18.9487100453116, "learning_rate": 1.8321785341698194e-06}
+{"epoch": 5, "train_acc": 0.9374309272680047, "best_dev_performance": 0.8799796230259807, "avg_batch_loss": 16.80406616543769, "learning_rate": 0.0}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f41c0853e3dee9a1f875dcb762e95fd09a171a343ee9f10e32b467a654426246
+size 501004058

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "max_len": 512, "init_inputs": []}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abf0f7e83ce6c435cea17e0d4d8214689e7f25aaa0690eb2655129c8f9e1d108
+size 1393