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/pretrained/filtered/cartography_variability_0.33/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/pretrained/filtered/cartography_variability_0.33/MNLI/cache_36891",
+    "fp16": false,
+    "fp16_opt_level": "01",
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 3.66e-06,
+    "local_rank": -1,
+    "logging_steps": 1000,
+    "max_grad_norm": 1.0,
+    "max_seq_length": 128,
+    "max_steps": -1,
+    "model_name_or_path": "/content/pretrained/results/",
+    "model_type": "roberta",
+    "n_gpu": 1,
+    "no_cuda": false,
+    "num_train_epochs": 9,
+    "output_dir": "/content/results",
+    "output_mode": "classification",
+    "overwrite_cache": false,
+    "overwrite_output_dir": false,
+    "patience": 9,
+    "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/pretrained/filtered/cartography_variability_0.33/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/pretrained/filtered/cartography_variability_0.33/MNLI/cache_36891",
+    "fp16": false,
+    "fp16_opt_level": "01",
+    "gradient_accumulation_steps": 1,
+    "learning_rate": 3.66e-06,
+    "local_rank": -1,
+    "logging_steps": 1000,
+    "max_grad_norm": 1.0,
+    "max_seq_length": 128,
+    "max_steps": -1,
+    "model_name_or_path": "/content/pretrained/results/",
+    "model_type": "roberta",
+    "n_gpu": 1,
+    "no_cuda": false,
+    "num_train_epochs": 9,
+    "output_dir": "/content/results",
+    "output_mode": "classification",
+    "overwrite_cache": false,
+    "overwrite_output_dir": false,
+    "patience": 9,
+    "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": 0.8760061130922058}

eval_metrics_mnli_dev_in_training.json

@@ -0,0 +1,9 @@
+{"acc": 0.8746816097809476}
+{"acc": 0.8757004584819155}
+{"acc": 0.8760061130922058}
+{"acc": 0.8751910341314315}
+{"acc": 0.8749872643912379}
+{"acc": 0.8737646459500764}
+{"acc": 0.8743759551706571}
+{"acc": 0.8735608762098829}
+{"acc": 0.87396841569027}

eval_metrics_train.json

@@ -0,0 +1,9 @@
+{"epoch": 0, "train_acc": 0.9124507050898434, "best_dev_performance": 0.8746816097809476, "avg_batch_loss": 7.11664024181664, "learning_rate": 3.2533333333333332e-06}
+{"epoch": 1, "train_acc": 0.9262839693719407, "best_dev_performance": 0.8757004584819155, "avg_batch_loss": 6.131773860484827, "learning_rate": 2.846666666666667e-06}
+{"epoch": 2, "train_acc": 0.9340714143227609, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 5.4619441903196275, "learning_rate": 2.44e-06}
+{"epoch": 3, "train_acc": 0.941505930733938, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.922465293901041, "learning_rate": 2.0333333333333335e-06}
+{"epoch": 4, "train_acc": 0.9477282143350366, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.450159241823712, "learning_rate": 1.6266666666666666e-06}
+{"epoch": 5, "train_acc": 0.9508585370344795, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.147213218006073, "learning_rate": 1.22e-06}
+{"epoch": 6, "train_acc": 0.9543954947904678, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.8465021344600245, "learning_rate": 8.133333333333333e-07}
+{"epoch": 7, "train_acc": 0.9573723702987617, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.6861731809622142, "learning_rate": 4.0666666666666666e-07}
+{"epoch": 8, "train_acc": 0.9600193343460848, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.479075360490242, "learning_rate": 0.0}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f27aed8bb4d091c89bb3f421e9d00b213d261c009640392c972e6865d696996
+size 501004048

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, "bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>", "init_inputs": []}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f734b2cd1e2b3f5535905586f3a844a97bf75a5e881f583777c68e9751ad4f9d
+size 1487