Upload trained model

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "nguyenvulebinh/vi-mrc-large",
+  "architectures": [
+    "RobertaForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.0",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4121b1fee3e80eb1e560fd8912d092f2e6056ad037a080a7558122e38e15ae2
+size 2235420048

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a18d54d8054789dff6f06542d4f16916220d7632585a78d83c268624dd14aece
+size 4471073855

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:133ea15aa50a6e6209338fc82b56e2c6ff4b4604152b4b075c74ee582942a61b
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0818c2db39a12247f3a5aeef5aec0b169c515576ed57912eac8323d6d8266a19
+size 1064

trainer_state.json

@@ -0,0 +1,185 @@
+{
+  "best_metric": 0.4843710611445973,
+  "best_model_checkpoint": "./reader_model/checkpoint-8730",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 8730,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1718213058419244,
+      "grad_norm": 75.05482482910156,
+      "learning_rate": 4.713631156930126e-05,
+      "loss": 2.9216,
+      "step": 500
+    },
+    {
+      "epoch": 0.3436426116838488,
+      "grad_norm": 21.536182403564453,
+      "learning_rate": 4.427262313860252e-05,
+      "loss": 2.3951,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5154639175257731,
+      "grad_norm": 185.8365020751953,
+      "learning_rate": 4.140893470790378e-05,
+      "loss": 2.1697,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6872852233676976,
+      "grad_norm": 48.49885559082031,
+      "learning_rate": 3.854524627720504e-05,
+      "loss": 2.2481,
+      "step": 2000
+    },
+    {
+      "epoch": 0.8591065292096219,
+      "grad_norm": 56.232383728027344,
+      "learning_rate": 3.5681557846506306e-05,
+      "loss": 1.9251,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0,
+      "eval_bertscore": 0.9379365905851414,
+      "eval_bleu1": 0.0,
+      "eval_f1": 0.44161785163619677,
+      "eval_loss": 1.82194983959198,
+      "eval_runtime": 87.956,
+      "eval_samples_per_second": 16.44,
+      "eval_steps_per_second": 4.116,
+      "step": 2910
+    },
+    {
+      "epoch": 1.0309278350515463,
+      "grad_norm": 71.09529876708984,
+      "learning_rate": 3.2817869415807564e-05,
+      "loss": 1.8373,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2027491408934707,
+      "grad_norm": 58.3028450012207,
+      "learning_rate": 2.9954180985108822e-05,
+      "loss": 1.5451,
+      "step": 3500
+    },
+    {
+      "epoch": 1.3745704467353952,
+      "grad_norm": 84.97217559814453,
+      "learning_rate": 2.709049255441008e-05,
+      "loss": 1.461,
+      "step": 4000
+    },
+    {
+      "epoch": 1.5463917525773194,
+      "grad_norm": 41.77628707885742,
+      "learning_rate": 2.422680412371134e-05,
+      "loss": 1.3964,
+      "step": 4500
+    },
+    {
+      "epoch": 1.718213058419244,
+      "grad_norm": 4.4470438957214355,
+      "learning_rate": 2.13631156930126e-05,
+      "loss": 1.3665,
+      "step": 5000
+    },
+    {
+      "epoch": 1.8900343642611683,
+      "grad_norm": 43.41935729980469,
+      "learning_rate": 1.849942726231386e-05,
+      "loss": 1.3783,
+      "step": 5500
+    },
+    {
+      "epoch": 2.0,
+      "eval_bertscore": 0.9420511256774596,
+      "eval_bleu1": 0.0,
+      "eval_f1": 0.47942977396358094,
+      "eval_loss": 1.7279274463653564,
+      "eval_runtime": 81.3708,
+      "eval_samples_per_second": 17.77,
+      "eval_steps_per_second": 4.449,
+      "step": 5820
+    },
+    {
+      "epoch": 2.0618556701030926,
+      "grad_norm": 29.33169937133789,
+      "learning_rate": 1.5635738831615122e-05,
+      "loss": 1.2364,
+      "step": 6000
+    },
+    {
+      "epoch": 2.2336769759450172,
+      "grad_norm": 101.00111389160156,
+      "learning_rate": 1.277205040091638e-05,
+      "loss": 0.952,
+      "step": 6500
+    },
+    {
+      "epoch": 2.4054982817869415,
+      "grad_norm": 163.68463134765625,
+      "learning_rate": 9.90836197021764e-06,
+      "loss": 0.9447,
+      "step": 7000
+    },
+    {
+      "epoch": 2.5773195876288657,
+      "grad_norm": 105.01953125,
+      "learning_rate": 7.0446735395189e-06,
+      "loss": 0.8667,
+      "step": 7500
+    },
+    {
+      "epoch": 2.7491408934707904,
+      "grad_norm": 41.25511169433594,
+      "learning_rate": 4.18098510882016e-06,
+      "loss": 0.8508,
+      "step": 8000
+    },
+    {
+      "epoch": 2.9209621993127146,
+      "grad_norm": 22.426847457885742,
+      "learning_rate": 1.3172966781214204e-06,
+      "loss": 0.8592,
+      "step": 8500
+    },
+    {
+      "epoch": 3.0,
+      "eval_bertscore": 0.9449925996868113,
+      "eval_bleu1": 0.0,
+      "eval_f1": 0.4843710611445973,
+      "eval_loss": 1.917205572128296,
+      "eval_runtime": 81.3736,
+      "eval_samples_per_second": 17.77,
+      "eval_steps_per_second": 4.449,
+      "step": 8730
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8730,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.242487592315699e+16,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70ae6d8f432b3bcbf3955b42f8070a6cfeb79eab9fdcf845793fb5670d923d62
+size 5368