Add michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2

Upload fine-tuned rerankers for BioASQ 14B

Co-authored-by: André Ribeiro <andrepedro2004@hotmail.com>
Co-authored-by: Rúben Garrido <rubengarrido@ua.pt>

michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2/config.json

@@ -0,0 +1,36 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": null,
+  "classifier_dropout": null,
+  "dtype": "bfloat16",
+  "eos_token_id": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "LABEL_0"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "is_decoder": false,
+  "label2id": {
+    "LABEL_0": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.2.0",
+  "type_vocab_size": 2,
+  "use_cache": false,
+  "vocab_size": 28895
+}

michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10a0538ea67a353d98251ec451891f8103cf27d1d5d5c61f491f2fa81120b4ee
+size 667000818

michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "is_local": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2/trainer_state.json

@@ -0,0 +1,350 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 2226,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.022461814914645103,
+      "grad_norm": 38.0,
+      "learning_rate": 1.977987421383648e-05,
+      "loss": 0.9769232177734375,
+      "step": 50
+    },
+    {
+      "epoch": 0.044923629829290206,
+      "grad_norm": 29.25,
+      "learning_rate": 1.955525606469003e-05,
+      "loss": 0.8111329650878907,
+      "step": 100
+    },
+    {
+      "epoch": 0.0673854447439353,
+      "grad_norm": 20.25,
+      "learning_rate": 1.9330637915543576e-05,
+      "loss": 0.5833827209472656,
+      "step": 150
+    },
+    {
+      "epoch": 0.08984725965858041,
+      "grad_norm": 31.75,
+      "learning_rate": 1.9106019766397127e-05,
+      "loss": 0.5423316955566406,
+      "step": 200
+    },
+    {
+      "epoch": 0.11230907457322552,
+      "grad_norm": 17.125,
+      "learning_rate": 1.8881401617250677e-05,
+      "loss": 0.4619390869140625,
+      "step": 250
+    },
+    {
+      "epoch": 0.1347708894878706,
+      "grad_norm": 87.5,
+      "learning_rate": 1.8656783468104224e-05,
+      "loss": 0.39820796966552735,
+      "step": 300
+    },
+    {
+      "epoch": 0.15723270440251572,
+      "grad_norm": 12.8125,
+      "learning_rate": 1.843216531895777e-05,
+      "loss": 0.4356438446044922,
+      "step": 350
+    },
+    {
+      "epoch": 0.17969451931716082,
+      "grad_norm": 21.5,
+      "learning_rate": 1.8207547169811322e-05,
+      "loss": 0.40065696716308596,
+      "step": 400
+    },
+    {
+      "epoch": 0.20215633423180593,
+      "grad_norm": 11.6875,
+      "learning_rate": 1.7982929020664873e-05,
+      "loss": 0.3666468048095703,
+      "step": 450
+    },
+    {
+      "epoch": 0.22461814914645103,
+      "grad_norm": 15.5625,
+      "learning_rate": 1.775831087151842e-05,
+      "loss": 0.41607128143310546,
+      "step": 500
+    },
+    {
+      "epoch": 0.24707996406109614,
+      "grad_norm": 16.25,
+      "learning_rate": 1.753369272237197e-05,
+      "loss": 0.3779059600830078,
+      "step": 550
+    },
+    {
+      "epoch": 0.2695417789757412,
+      "grad_norm": 15.5,
+      "learning_rate": 1.7309074573225517e-05,
+      "loss": 0.36492431640625,
+      "step": 600
+    },
+    {
+      "epoch": 0.2920035938903863,
+      "grad_norm": 9.75,
+      "learning_rate": 1.7084456424079068e-05,
+      "loss": 0.371168212890625,
+      "step": 650
+    },
+    {
+      "epoch": 0.31446540880503143,
+      "grad_norm": 11.375,
+      "learning_rate": 1.6859838274932615e-05,
+      "loss": 0.38794273376464844,
+      "step": 700
+    },
+    {
+      "epoch": 0.33692722371967654,
+      "grad_norm": 15.375,
+      "learning_rate": 1.6635220125786166e-05,
+      "loss": 0.3640760803222656,
+      "step": 750
+    },
+    {
+      "epoch": 0.35938903863432164,
+      "grad_norm": 13.625,
+      "learning_rate": 1.6410601976639713e-05,
+      "loss": 0.38307479858398436,
+      "step": 800
+    },
+    {
+      "epoch": 0.38185085354896675,
+      "grad_norm": 15.8125,
+      "learning_rate": 1.6185983827493263e-05,
+      "loss": 0.3891305160522461,
+      "step": 850
+    },
+    {
+      "epoch": 0.40431266846361186,
+      "grad_norm": 21.875,
+      "learning_rate": 1.596136567834681e-05,
+      "loss": 0.35688247680664065,
+      "step": 900
+    },
+    {
+      "epoch": 0.42677448337825696,
+      "grad_norm": 13.4375,
+      "learning_rate": 1.573674752920036e-05,
+      "loss": 0.35335597991943357,
+      "step": 950
+    },
+    {
+      "epoch": 0.44923629829290207,
+      "grad_norm": 19.5,
+      "learning_rate": 1.5512129380053908e-05,
+      "loss": 0.3869240951538086,
+      "step": 1000
+    },
+    {
+      "epoch": 0.4716981132075472,
+      "grad_norm": 17.125,
+      "learning_rate": 1.528751123090746e-05,
+      "loss": 0.34456672668457033,
+      "step": 1050
+    },
+    {
+      "epoch": 0.4941599281221923,
+      "grad_norm": 15.4375,
+      "learning_rate": 1.5062893081761007e-05,
+      "loss": 0.32508480072021484,
+      "step": 1100
+    },
+    {
+      "epoch": 0.5166217430368374,
+      "grad_norm": 15.1875,
+      "learning_rate": 1.4838274932614556e-05,
+      "loss": 0.3791059875488281,
+      "step": 1150
+    },
+    {
+      "epoch": 0.5390835579514824,
+      "grad_norm": 12.0,
+      "learning_rate": 1.4613656783468107e-05,
+      "loss": 0.3071991729736328,
+      "step": 1200
+    },
+    {
+      "epoch": 0.5615453728661276,
+      "grad_norm": 13.375,
+      "learning_rate": 1.4389038634321656e-05,
+      "loss": 0.32035968780517576,
+      "step": 1250
+    },
+    {
+      "epoch": 0.5840071877807727,
+      "grad_norm": 13.625,
+      "learning_rate": 1.4164420485175203e-05,
+      "loss": 0.33290882110595704,
+      "step": 1300
+    },
+    {
+      "epoch": 0.6064690026954178,
+      "grad_norm": 17.0,
+      "learning_rate": 1.3939802336028752e-05,
+      "loss": 0.3300562286376953,
+      "step": 1350
+    },
+    {
+      "epoch": 0.6289308176100629,
+      "grad_norm": 14.0625,
+      "learning_rate": 1.3715184186882302e-05,
+      "loss": 0.3090822982788086,
+      "step": 1400
+    },
+    {
+      "epoch": 0.651392632524708,
+      "grad_norm": 13.1875,
+      "learning_rate": 1.3490566037735851e-05,
+      "loss": 0.33966907501220706,
+      "step": 1450
+    },
+    {
+      "epoch": 0.6738544474393531,
+      "grad_norm": 9.0,
+      "learning_rate": 1.3265947888589398e-05,
+      "loss": 0.3335183334350586,
+      "step": 1500
+    },
+    {
+      "epoch": 0.6963162623539982,
+      "grad_norm": 13.4375,
+      "learning_rate": 1.3041329739442947e-05,
+      "loss": 0.3270130157470703,
+      "step": 1550
+    },
+    {
+      "epoch": 0.7187780772686433,
+      "grad_norm": 13.1875,
+      "learning_rate": 1.2816711590296497e-05,
+      "loss": 0.3418006896972656,
+      "step": 1600
+    },
+    {
+      "epoch": 0.7412398921832885,
+      "grad_norm": 16.125,
+      "learning_rate": 1.2592093441150046e-05,
+      "loss": 0.3601032638549805,
+      "step": 1650
+    },
+    {
+      "epoch": 0.7637017070979335,
+      "grad_norm": 12.4375,
+      "learning_rate": 1.2367475292003595e-05,
+      "loss": 0.35117397308349607,
+      "step": 1700
+    },
+    {
+      "epoch": 0.7861635220125787,
+      "grad_norm": 17.875,
+      "learning_rate": 1.2142857142857142e-05,
+      "loss": 0.3449953842163086,
+      "step": 1750
+    },
+    {
+      "epoch": 0.8086253369272237,
+      "grad_norm": 12.4375,
+      "learning_rate": 1.1918238993710693e-05,
+      "loss": 0.3240993499755859,
+      "step": 1800
+    },
+    {
+      "epoch": 0.8310871518418689,
+      "grad_norm": 8.875,
+      "learning_rate": 1.1693620844564242e-05,
+      "loss": 0.28546058654785156,
+      "step": 1850
+    },
+    {
+      "epoch": 0.8535489667565139,
+      "grad_norm": 9.75,
+      "learning_rate": 1.146900269541779e-05,
+      "loss": 0.30159576416015627,
+      "step": 1900
+    },
+    {
+      "epoch": 0.876010781671159,
+      "grad_norm": 15.875,
+      "learning_rate": 1.124438454627134e-05,
+      "loss": 0.3210825729370117,
+      "step": 1950
+    },
+    {
+      "epoch": 0.8984725965858041,
+      "grad_norm": 14.25,
+      "learning_rate": 1.101976639712489e-05,
+      "loss": 0.28267333984375,
+      "step": 2000
+    },
+    {
+      "epoch": 0.9209344115004492,
+      "grad_norm": 16.5,
+      "learning_rate": 1.0795148247978437e-05,
+      "loss": 0.31453651428222656,
+      "step": 2050
+    },
+    {
+      "epoch": 0.9433962264150944,
+      "grad_norm": 26.25,
+      "learning_rate": 1.0570530098831986e-05,
+      "loss": 0.3257890701293945,
+      "step": 2100
+    },
+    {
+      "epoch": 0.9658580413297394,
+      "grad_norm": 14.9375,
+      "learning_rate": 1.0345911949685536e-05,
+      "loss": 0.3485770416259766,
+      "step": 2150
+    },
+    {
+      "epoch": 0.9883198562443846,
+      "grad_norm": 12.375,
+      "learning_rate": 1.0121293800539085e-05,
+      "loss": 0.36537406921386717,
+      "step": 2200
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.37241339683532715,
+      "eval_runtime": 3.9519,
+      "eval_samples_per_second": 85.528,
+      "eval_steps_per_second": 2.783,
+      "step": 2226
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 4452,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

michiyasunaga-BioLinkBERT-large-E2-S1-Mpairwise-FullDataTrue--BasicV2/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:719c02497535d75c1323ef840f61d9cbcd27ca39e675340b1ebee5bddf2187f7
+size 5393