Training in progress, step 11000, checkpoint

checkpoint-11000/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<mask>": 64000
+}

checkpoint-11000/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "demdecuong/vihealthbert-base-word",
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "word-level",
+  "gradient_checkpointing": false,
+  "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": 258,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_hidden_states": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "PhobertTokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 64001
+}

checkpoint-11000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa1780afb0b8f34e93e663fed72ad4b53f6c47dff462d8a72f1703bd038c3336
+size 540279452

checkpoint-11000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58e4405e6a17da957832f5bc5807ee04866ff3a3e1710acb465c2ff5bf23d446
+size 1080630853

checkpoint-11000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b53701262582263e4c2e06d4d575e8cac644b5d435803056409cfeba9a601003
+size 14575

checkpoint-11000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32ae43497a368469089f2e25199e81f939abc36d307af7c22a118982b235bd93
+size 627

checkpoint-11000/special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-11000/tokenizer_config.json

@@ -0,0 +1,54 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "64000": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "PhobertTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-11000/trainer_state.json

@@ -0,0 +1,204 @@
+{
+  "best_metric": 0.613650998068255,
+  "best_model_checkpoint": "/root/crosslingual-mining-for-domain-nli/output/pretraining/vihealthbert-w_dual-ViNLI/lr3e-5_wr0.1_wd0.0/checkpoint-8000",
+  "epoch": 171.875,
+  "eval_steps": 1000,
+  "global_step": 11000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.015625,
+      "grad_norm": 34.54078674316406,
+      "learning_rate": 1e-08,
+      "loss": 10.2871,
+      "step": 1
+    },
+    {
+      "epoch": 15.625,
+      "grad_norm": 5.461294174194336,
+      "learning_rate": 9.999999999999999e-06,
+      "loss": 5.8126,
+      "step": 1000
+    },
+    {
+      "epoch": 15.625,
+      "eval_accuracy": 0.44503311258278144,
+      "eval_loss": 3.546058177947998,
+      "eval_runtime": 0.7421,
+      "eval_samples_per_second": 312.643,
+      "eval_steps_per_second": 20.214,
+      "step": 1000
+    },
+    {
+      "epoch": 31.25,
+      "grad_norm": 4.604820728302002,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 2.605,
+      "step": 2000
+    },
+    {
+      "epoch": 31.25,
+      "eval_accuracy": 0.540436456996149,
+      "eval_loss": 2.7789385318756104,
+      "eval_runtime": 0.799,
+      "eval_samples_per_second": 290.377,
+      "eval_steps_per_second": 18.774,
+      "step": 2000
+    },
+    {
+      "epoch": 46.875,
+      "grad_norm": 4.253050327301025,
+      "learning_rate": 3e-05,
+      "loss": 1.5924,
+      "step": 3000
+    },
+    {
+      "epoch": 46.875,
+      "eval_accuracy": 0.5809284818067754,
+      "eval_loss": 2.5431649684906006,
+      "eval_runtime": 0.7348,
+      "eval_samples_per_second": 315.742,
+      "eval_steps_per_second": 20.414,
+      "step": 3000
+    },
+    {
+      "epoch": 62.5,
+      "grad_norm": 4.554752349853516,
+      "learning_rate": 2.8888888888888888e-05,
+      "loss": 1.2233,
+      "step": 4000
+    },
+    {
+      "epoch": 62.5,
+      "eval_accuracy": 0.5567077122488658,
+      "eval_loss": 2.6662397384643555,
+      "eval_runtime": 0.7259,
+      "eval_samples_per_second": 319.609,
+      "eval_steps_per_second": 20.664,
+      "step": 4000
+    },
+    {
+      "epoch": 78.125,
+      "grad_norm": 3.5844781398773193,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.9236,
+      "step": 5000
+    },
+    {
+      "epoch": 78.125,
+      "eval_accuracy": 0.5927342256214149,
+      "eval_loss": 2.4690873622894287,
+      "eval_runtime": 0.7491,
+      "eval_samples_per_second": 309.699,
+      "eval_steps_per_second": 20.024,
+      "step": 5000
+    },
+    {
+      "epoch": 93.75,
+      "grad_norm": 5.544617176055908,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.7193,
+      "step": 6000
+    },
+    {
+      "epoch": 93.75,
+      "eval_accuracy": 0.6027219701879456,
+      "eval_loss": 2.405272960662842,
+      "eval_runtime": 0.6998,
+      "eval_samples_per_second": 331.502,
+      "eval_steps_per_second": 21.433,
+      "step": 6000
+    },
+    {
+      "epoch": 109.375,
+      "grad_norm": 4.963156700134277,
+      "learning_rate": 2.5555555555555557e-05,
+      "loss": 0.6259,
+      "step": 7000
+    },
+    {
+      "epoch": 109.375,
+      "eval_accuracy": 0.5781748564135291,
+      "eval_loss": 2.5938053131103516,
+      "eval_runtime": 0.6966,
+      "eval_samples_per_second": 333.065,
+      "eval_steps_per_second": 21.534,
+      "step": 7000
+    },
+    {
+      "epoch": 125.0,
+      "grad_norm": 4.307164192199707,
+      "learning_rate": 2.4444444444444445e-05,
+      "loss": 0.5082,
+      "step": 8000
+    },
+    {
+      "epoch": 125.0,
+      "eval_accuracy": 0.613650998068255,
+      "eval_loss": 2.480868101119995,
+      "eval_runtime": 0.7149,
+      "eval_samples_per_second": 324.519,
+      "eval_steps_per_second": 20.982,
+      "step": 8000
+    },
+    {
+      "epoch": 140.625,
+      "grad_norm": 3.2876017093658447,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 0.4438,
+      "step": 9000
+    },
+    {
+      "epoch": 140.625,
+      "eval_accuracy": 0.5819174757281553,
+      "eval_loss": 2.705582618713379,
+      "eval_runtime": 0.845,
+      "eval_samples_per_second": 274.54,
+      "eval_steps_per_second": 17.75,
+      "step": 9000
+    },
+    {
+      "epoch": 156.25,
+      "grad_norm": 14.243521690368652,
+      "learning_rate": 2.222222222222222e-05,
+      "loss": 0.4075,
+      "step": 10000
+    },
+    {
+      "epoch": 156.25,
+      "eval_accuracy": 0.5945945945945946,
+      "eval_loss": 2.650148630142212,
+      "eval_runtime": 0.8487,
+      "eval_samples_per_second": 273.36,
+      "eval_steps_per_second": 17.674,
+      "step": 10000
+    },
+    {
+      "epoch": 171.875,
+      "grad_norm": 12.720209121704102,
+      "learning_rate": 2.111111111111111e-05,
+      "loss": 0.3571,
+      "step": 11000
+    },
+    {
+      "epoch": 171.875,
+      "eval_accuracy": 0.6082406801831263,
+      "eval_loss": 2.53369140625,
+      "eval_runtime": 0.7729,
+      "eval_samples_per_second": 300.18,
+      "eval_steps_per_second": 19.408,
+      "step": 11000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 30000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 469,
+  "save_steps": 1000,
+  "total_flos": 0.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-11000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9e64209731498e7565598d2dc690afbf0a6acfe64afd255aade8f0d8cbfc3c9
+size 4987