Add BAAI-bge-reranker-v2-m3-E5-Pairwise

Upload fine-tuned rerankers for BioASQ 14B

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

.gitattributes

@@ -38,3 +38,4 @@ BAAI-bge-reranker-v2-m3-E2-S1-Mpairwise-FullDataTrue/tokenizer.json filter=lfs d
 nvidia-llama-nemotron-rerank-1b-v2-E2-S4-Mmulti_neg_pairwise-Linfonce-FullData/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 nvidia-llama-nemotron-rerank-1b-v2_llama-E2-Pairwise/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 BAAI-bge-reranker-base-E5-Pairwise/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+BAAI-bge-reranker-v2-m3-E5-Pairwise/tokenizer.json filter=lfs diff=lfs merge=lfs -text

BAAI-bge-reranker-v2-m3-E5-Pairwise/config.json

@@ -0,0 +1,36 @@
+{
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "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-05,
+  "max_position_embeddings": 8194,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.2.0",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

BAAI-bge-reranker-v2-m3-E5-Pairwise/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cc0e77c439d36b5c30cf2639cf48193e42ba95e3ee04078a37dfb23a9413493
+size 1135560042

BAAI-bge-reranker-v2-m3-E5-Pairwise/ranx_results.json

@@ -0,0 +1,31 @@
+{
+    "model": "BAAI-bge-reranker-v2-m3",
+    "val_files": [
+        "../../data/val_data/13B1_golden.json",
+        "../../data/val_data/13B2_golden.json"
+    ],
+    "total": {
+        "ndcg@5": 0.7864378418154898,
+        "mrr": 0.8856398013750955,
+        "recall@10": 0.4997563617949055,
+        "recall@100": 0.9633423964098493,
+        "map@10": 0.44423862118303203,
+        "map-bioasq@10": 0.6824153948616483
+    },
+    "13B1_golden.json": {
+        "ndcg@5": 0.6997345762298963,
+        "mrr": 0.8443188184364655,
+        "recall@10": 0.46983498885013975,
+        "recall@100": 0.9653401214405154,
+        "map@10": 0.3988493809509552,
+        "map-bioasq@10": 0.5855540493975366
+    },
+    "13B2_golden.json": {
+        "ndcg@5": 0.8731411074010831,
+        "mrr": 0.9269607843137255,
+        "recall@10": 0.5296777347396715,
+        "recall@100": 0.9613446713791834,
+        "map@10": 0.4896278614151088,
+        "map-bioasq@10": 0.77927674032576
+    }
+}

BAAI-bge-reranker-v2-m3-E5-Pairwise/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ea38d007ec4d85d31c1ce3fdce5adfd91cf82244ba8afad500e154beb38c1ea
+size 16781581

BAAI-bge-reranker-v2-m3-E5-Pairwise/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "mask_token": "<mask>",
+  "model_max_length": 8192,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

BAAI-bge-reranker-v2-m3-E5-Pairwise/trainer_state.json

@@ -0,0 +1,963 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 6375,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0392156862745098,
+      "grad_norm": 10.8125,
+      "learning_rate": 1.9846274509803922e-05,
+      "loss": 0.4558255386352539,
+      "step": 50
+    },
+    {
+      "epoch": 0.0784313725490196,
+      "grad_norm": 9.5,
+      "learning_rate": 1.9689411764705883e-05,
+      "loss": 0.39794933319091796,
+      "step": 100
+    },
+    {
+      "epoch": 0.11764705882352941,
+      "grad_norm": 9.1875,
+      "learning_rate": 1.9532549019607844e-05,
+      "loss": 0.36890380859375,
+      "step": 150
+    },
+    {
+      "epoch": 0.1568627450980392,
+      "grad_norm": 10.875,
+      "learning_rate": 1.9375686274509805e-05,
+      "loss": 0.36853790283203125,
+      "step": 200
+    },
+    {
+      "epoch": 0.19607843137254902,
+      "grad_norm": 10.4375,
+      "learning_rate": 1.9218823529411766e-05,
+      "loss": 0.3469864273071289,
+      "step": 250
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 9.25,
+      "learning_rate": 1.9061960784313727e-05,
+      "loss": 0.33170494079589846,
+      "step": 300
+    },
+    {
+      "epoch": 0.27450980392156865,
+      "grad_norm": 7.53125,
+      "learning_rate": 1.8905098039215688e-05,
+      "loss": 0.34885086059570314,
+      "step": 350
+    },
+    {
+      "epoch": 0.3137254901960784,
+      "grad_norm": 10.3125,
+      "learning_rate": 1.874823529411765e-05,
+      "loss": 0.3499496078491211,
+      "step": 400
+    },
+    {
+      "epoch": 0.35294117647058826,
+      "grad_norm": 8.8125,
+      "learning_rate": 1.8591372549019607e-05,
+      "loss": 0.3427734375,
+      "step": 450
+    },
+    {
+      "epoch": 0.39215686274509803,
+      "grad_norm": 9.9375,
+      "learning_rate": 1.843450980392157e-05,
+      "loss": 0.307832145690918,
+      "step": 500
+    },
+    {
+      "epoch": 0.43137254901960786,
+      "grad_norm": 9.4375,
+      "learning_rate": 1.8277647058823532e-05,
+      "loss": 0.3249082183837891,
+      "step": 550
+    },
+    {
+      "epoch": 0.47058823529411764,
+      "grad_norm": 7.40625,
+      "learning_rate": 1.8120784313725493e-05,
+      "loss": 0.3580419921875,
+      "step": 600
+    },
+    {
+      "epoch": 0.5098039215686274,
+      "grad_norm": 8.8125,
+      "learning_rate": 1.796392156862745e-05,
+      "loss": 0.30408864974975586,
+      "step": 650
+    },
+    {
+      "epoch": 0.5490196078431373,
+      "grad_norm": 9.375,
+      "learning_rate": 1.7807058823529412e-05,
+      "loss": 0.2944978332519531,
+      "step": 700
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 13.0,
+      "learning_rate": 1.7650196078431373e-05,
+      "loss": 0.2830428886413574,
+      "step": 750
+    },
+    {
+      "epoch": 0.6274509803921569,
+      "grad_norm": 9.6875,
+      "learning_rate": 1.7493333333333334e-05,
+      "loss": 0.3323691940307617,
+      "step": 800
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 15.125,
+      "learning_rate": 1.7336470588235295e-05,
+      "loss": 0.32856346130371095,
+      "step": 850
+    },
+    {
+      "epoch": 0.7058823529411765,
+      "grad_norm": 8.8125,
+      "learning_rate": 1.7179607843137256e-05,
+      "loss": 0.3277847671508789,
+      "step": 900
+    },
+    {
+      "epoch": 0.7450980392156863,
+      "grad_norm": 9.9375,
+      "learning_rate": 1.7022745098039217e-05,
+      "loss": 0.2982676696777344,
+      "step": 950
+    },
+    {
+      "epoch": 0.7843137254901961,
+      "grad_norm": 9.0,
+      "learning_rate": 1.6865882352941178e-05,
+      "loss": 0.30295835494995116,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8235294117647058,
+      "grad_norm": 11.5,
+      "learning_rate": 1.670901960784314e-05,
+      "loss": 0.2956972122192383,
+      "step": 1050
+    },
+    {
+      "epoch": 0.8627450980392157,
+      "grad_norm": 8.8125,
+      "learning_rate": 1.65521568627451e-05,
+      "loss": 0.33095306396484375,
+      "step": 1100
+    },
+    {
+      "epoch": 0.9019607843137255,
+      "grad_norm": 10.125,
+      "learning_rate": 1.639529411764706e-05,
+      "loss": 0.2780212211608887,
+      "step": 1150
+    },
+    {
+      "epoch": 0.9411764705882353,
+      "grad_norm": 9.0,
+      "learning_rate": 1.6238431372549022e-05,
+      "loss": 0.29875099182128906,
+      "step": 1200
+    },
+    {
+      "epoch": 0.9803921568627451,
+      "grad_norm": 48.75,
+      "learning_rate": 1.608156862745098e-05,
+      "loss": 0.3062861251831055,
+      "step": 1250
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.2955200672149658,
+      "eval_runtime": 1.9237,
+      "eval_samples_per_second": 87.331,
+      "eval_steps_per_second": 1.559,
+      "step": 1275
+    },
+    {
+      "epoch": 1.0196078431372548,
+      "grad_norm": 12.125,
+      "learning_rate": 1.592470588235294e-05,
+      "loss": 0.27703052520751953,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0588235294117647,
+      "grad_norm": 11.3125,
+      "learning_rate": 1.5767843137254902e-05,
+      "loss": 0.29377376556396484,
+      "step": 1350
+    },
+    {
+      "epoch": 1.0980392156862746,
+      "grad_norm": 10.25,
+      "learning_rate": 1.5610980392156866e-05,
+      "loss": 0.26873968124389647,
+      "step": 1400
+    },
+    {
+      "epoch": 1.1372549019607843,
+      "grad_norm": 12.75,
+      "learning_rate": 1.5454117647058824e-05,
+      "loss": 0.27458084106445313,
+      "step": 1450
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 9.375,
+      "learning_rate": 1.5297254901960785e-05,
+      "loss": 0.3068842887878418,
+      "step": 1500
+    },
+    {
+      "epoch": 1.215686274509804,
+      "grad_norm": 11.1875,
+      "learning_rate": 1.5140392156862746e-05,
+      "loss": 0.29080896377563475,
+      "step": 1550
+    },
+    {
+      "epoch": 1.2549019607843137,
+      "grad_norm": 12.875,
+      "learning_rate": 1.4983529411764707e-05,
+      "loss": 0.2903484344482422,
+      "step": 1600
+    },
+    {
+      "epoch": 1.2941176470588236,
+      "grad_norm": 9.1875,
+      "learning_rate": 1.4826666666666666e-05,
+      "loss": 0.27252548217773437,
+      "step": 1650
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 12.3125,
+      "learning_rate": 1.4669803921568627e-05,
+      "loss": 0.29381614685058594,
+      "step": 1700
+    },
+    {
+      "epoch": 1.3725490196078431,
+      "grad_norm": 11.75,
+      "learning_rate": 1.451294117647059e-05,
+      "loss": 0.2906353950500488,
+      "step": 1750
+    },
+    {
+      "epoch": 1.4117647058823528,
+      "grad_norm": 8.6875,
+      "learning_rate": 1.4356078431372551e-05,
+      "loss": 0.2829168701171875,
+      "step": 1800
+    },
+    {
+      "epoch": 1.4509803921568627,
+      "grad_norm": 16.0,
+      "learning_rate": 1.419921568627451e-05,
+      "loss": 0.28158676147460937,
+      "step": 1850
+    },
+    {
+      "epoch": 1.4901960784313726,
+      "grad_norm": 10.625,
+      "learning_rate": 1.4042352941176472e-05,
+      "loss": 0.3125568962097168,
+      "step": 1900
+    },
+    {
+      "epoch": 1.5294117647058822,
+      "grad_norm": 10.125,
+      "learning_rate": 1.3885490196078433e-05,
+      "loss": 0.27885986328125,
+      "step": 1950
+    },
+    {
+      "epoch": 1.5686274509803921,
+      "grad_norm": 11.125,
+      "learning_rate": 1.3728627450980394e-05,
+      "loss": 0.30357994079589845,
+      "step": 2000
+    },
+    {
+      "epoch": 1.607843137254902,
+      "grad_norm": 10.25,
+      "learning_rate": 1.3571764705882353e-05,
+      "loss": 0.28456764221191405,
+      "step": 2050
+    },
+    {
+      "epoch": 1.6470588235294117,
+      "grad_norm": 6.6875,
+      "learning_rate": 1.3414901960784314e-05,
+      "loss": 0.2740924644470215,
+      "step": 2100
+    },
+    {
+      "epoch": 1.6862745098039216,
+      "grad_norm": 12.5625,
+      "learning_rate": 1.3258039215686275e-05,
+      "loss": 0.25640689849853515,
+      "step": 2150
+    },
+    {
+      "epoch": 1.7254901960784315,
+      "grad_norm": 10.375,
+      "learning_rate": 1.3101176470588238e-05,
+      "loss": 0.27075279235839844,
+      "step": 2200
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 11.4375,
+      "learning_rate": 1.2944313725490197e-05,
+      "loss": 0.2601768493652344,
+      "step": 2250
+    },
+    {
+      "epoch": 1.803921568627451,
+      "grad_norm": 12.25,
+      "learning_rate": 1.2787450980392158e-05,
+      "loss": 0.2924202537536621,
+      "step": 2300
+    },
+    {
+      "epoch": 1.843137254901961,
+      "grad_norm": 11.0,
+      "learning_rate": 1.2630588235294119e-05,
+      "loss": 0.2794902610778809,
+      "step": 2350
+    },
+    {
+      "epoch": 1.8823529411764706,
+      "grad_norm": 10.375,
+      "learning_rate": 1.247372549019608e-05,
+      "loss": 0.2599605369567871,
+      "step": 2400
+    },
+    {
+      "epoch": 1.9215686274509802,
+      "grad_norm": 11.625,
+      "learning_rate": 1.231686274509804e-05,
+      "loss": 0.2974419212341309,
+      "step": 2450
+    },
+    {
+      "epoch": 1.9607843137254903,
+      "grad_norm": 8.0625,
+      "learning_rate": 1.216e-05,
+      "loss": 0.2864244842529297,
+      "step": 2500
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.0,
+      "learning_rate": 1.2003137254901961e-05,
+      "loss": 0.2730862998962402,
+      "step": 2550
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.2867780327796936,
+      "eval_runtime": 1.9261,
+      "eval_samples_per_second": 87.222,
+      "eval_steps_per_second": 1.558,
+      "step": 2550
+    },
+    {
+      "epoch": 2.0392156862745097,
+      "grad_norm": 13.5,
+      "learning_rate": 1.1846274509803923e-05,
+      "loss": 0.26746143341064454,
+      "step": 2600
+    },
+    {
+      "epoch": 2.0784313725490198,
+      "grad_norm": 11.8125,
+      "learning_rate": 1.1689411764705882e-05,
+      "loss": 0.2880428886413574,
+      "step": 2650
+    },
+    {
+      "epoch": 2.1176470588235294,
+      "grad_norm": 9.25,
+      "learning_rate": 1.1532549019607845e-05,
+      "loss": 0.2801370620727539,
+      "step": 2700
+    },
+    {
+      "epoch": 2.156862745098039,
+      "grad_norm": 12.375,
+      "learning_rate": 1.1375686274509806e-05,
+      "loss": 0.2570189476013184,
+      "step": 2750
+    },
+    {
+      "epoch": 2.196078431372549,
+      "grad_norm": 11.9375,
+      "learning_rate": 1.1218823529411767e-05,
+      "loss": 0.29558477401733396,
+      "step": 2800
+    },
+    {
+      "epoch": 2.235294117647059,
+      "grad_norm": 10.4375,
+      "learning_rate": 1.1061960784313726e-05,
+      "loss": 0.28774349212646483,
+      "step": 2850
+    },
+    {
+      "epoch": 2.2745098039215685,
+      "grad_norm": 14.875,
+      "learning_rate": 1.0905098039215687e-05,
+      "loss": 0.30488248825073244,
+      "step": 2900
+    },
+    {
+      "epoch": 2.313725490196078,
+      "grad_norm": 15.0,
+      "learning_rate": 1.0748235294117648e-05,
+      "loss": 0.28303050994873047,
+      "step": 2950
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 11.875,
+      "learning_rate": 1.0591372549019609e-05,
+      "loss": 0.2963134384155273,
+      "step": 3000
+    },
+    {
+      "epoch": 2.392156862745098,
+      "grad_norm": 9.625,
+      "learning_rate": 1.0434509803921568e-05,
+      "loss": 0.2827070999145508,
+      "step": 3050
+    },
+    {
+      "epoch": 2.431372549019608,
+      "grad_norm": 11.1875,
+      "learning_rate": 1.027764705882353e-05,
+      "loss": 0.2745362281799316,
+      "step": 3100
+    },
+    {
+      "epoch": 2.4705882352941178,
+      "grad_norm": 12.25,
+      "learning_rate": 1.0120784313725492e-05,
+      "loss": 0.2844767189025879,
+      "step": 3150
+    },
+    {
+      "epoch": 2.5098039215686274,
+      "grad_norm": 11.6875,
+      "learning_rate": 9.963921568627451e-06,
+      "loss": 0.2794179344177246,
+      "step": 3200
+    },
+    {
+      "epoch": 2.549019607843137,
+      "grad_norm": 14.9375,
+      "learning_rate": 9.807058823529412e-06,
+      "loss": 0.29083608627319335,
+      "step": 3250
+    },
+    {
+      "epoch": 2.588235294117647,
+      "grad_norm": 13.75,
+      "learning_rate": 9.650196078431373e-06,
+      "loss": 0.2788432312011719,
+      "step": 3300
+    },
+    {
+      "epoch": 2.627450980392157,
+      "grad_norm": 11.6875,
+      "learning_rate": 9.493333333333334e-06,
+      "loss": 0.2642219543457031,
+      "step": 3350
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 8.875,
+      "learning_rate": 9.336470588235294e-06,
+      "loss": 0.2453985595703125,
+      "step": 3400
+    },
+    {
+      "epoch": 2.7058823529411766,
+      "grad_norm": 8.75,
+      "learning_rate": 9.179607843137257e-06,
+      "loss": 0.2759024429321289,
+      "step": 3450
+    },
+    {
+      "epoch": 2.7450980392156863,
+      "grad_norm": 8.0625,
+      "learning_rate": 9.022745098039216e-06,
+      "loss": 0.27131601333618166,
+      "step": 3500
+    },
+    {
+      "epoch": 2.784313725490196,
+      "grad_norm": 12.5,
+      "learning_rate": 8.865882352941177e-06,
+      "loss": 0.26699224472045896,
+      "step": 3550
+    },
+    {
+      "epoch": 2.8235294117647056,
+      "grad_norm": 10.5625,
+      "learning_rate": 8.709019607843138e-06,
+      "loss": 0.28045074462890623,
+      "step": 3600
+    },
+    {
+      "epoch": 2.8627450980392157,
+      "grad_norm": 11.3125,
+      "learning_rate": 8.552156862745099e-06,
+      "loss": 0.2871367835998535,
+      "step": 3650
+    },
+    {
+      "epoch": 2.9019607843137254,
+      "grad_norm": 12.75,
+      "learning_rate": 8.39529411764706e-06,
+      "loss": 0.25553524017333984,
+      "step": 3700
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 7.375,
+      "learning_rate": 8.238431372549021e-06,
+      "loss": 0.27515832901000975,
+      "step": 3750
+    },
+    {
+      "epoch": 2.980392156862745,
+      "grad_norm": 8.1875,
+      "learning_rate": 8.08156862745098e-06,
+      "loss": 0.26676090240478517,
+      "step": 3800
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.2837378978729248,
+      "eval_runtime": 1.92,
+      "eval_samples_per_second": 87.502,
+      "eval_steps_per_second": 1.563,
+      "step": 3825
+    },
+    {
+      "epoch": 3.019607843137255,
+      "grad_norm": 9.0,
+      "learning_rate": 7.924705882352941e-06,
+      "loss": 0.2817222213745117,
+      "step": 3850
+    },
+    {
+      "epoch": 3.0588235294117645,
+      "grad_norm": 14.0625,
+      "learning_rate": 7.767843137254902e-06,
+      "loss": 0.29212448120117185,
+      "step": 3900
+    },
+    {
+      "epoch": 3.0980392156862746,
+      "grad_norm": 11.5625,
+      "learning_rate": 7.610980392156863e-06,
+      "loss": 0.29557653427124025,
+      "step": 3950
+    },
+    {
+      "epoch": 3.1372549019607843,
+      "grad_norm": 11.375,
+      "learning_rate": 7.4541176470588236e-06,
+      "loss": 0.2853967475891113,
+      "step": 4000
+    },
+    {
+      "epoch": 3.176470588235294,
+      "grad_norm": 13.75,
+      "learning_rate": 7.2972549019607854e-06,
+      "loss": 0.292034912109375,
+      "step": 4050
+    },
+    {
+      "epoch": 3.215686274509804,
+      "grad_norm": 11.625,
+      "learning_rate": 7.140392156862746e-06,
+      "loss": 0.29273807525634765,
+      "step": 4100
+    },
+    {
+      "epoch": 3.2549019607843137,
+      "grad_norm": 16.625,
+      "learning_rate": 6.983529411764707e-06,
+      "loss": 0.3163567352294922,
+      "step": 4150
+    },
+    {
+      "epoch": 3.2941176470588234,
+      "grad_norm": 10.0,
+      "learning_rate": 6.826666666666667e-06,
+      "loss": 0.3079881858825684,
+      "step": 4200
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 14.3125,
+      "learning_rate": 6.669803921568629e-06,
+      "loss": 0.33801631927490233,
+      "step": 4250
+    },
+    {
+      "epoch": 3.372549019607843,
+      "grad_norm": 13.75,
+      "learning_rate": 6.512941176470589e-06,
+      "loss": 0.3027102661132812,
+      "step": 4300
+    },
+    {
+      "epoch": 3.411764705882353,
+      "grad_norm": 8.8125,
+      "learning_rate": 6.35607843137255e-06,
+      "loss": 0.28146699905395506,
+      "step": 4350
+    },
+    {
+      "epoch": 3.450980392156863,
+      "grad_norm": 11.5625,
+      "learning_rate": 6.19921568627451e-06,
+      "loss": 0.2953393173217773,
+      "step": 4400
+    },
+    {
+      "epoch": 3.4901960784313726,
+      "grad_norm": 17.75,
+      "learning_rate": 6.042352941176471e-06,
+      "loss": 0.30299001693725586,
+      "step": 4450
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 13.9375,
+      "learning_rate": 5.885490196078431e-06,
+      "loss": 0.3037626075744629,
+      "step": 4500
+    },
+    {
+      "epoch": 3.568627450980392,
+      "grad_norm": 11.625,
+      "learning_rate": 5.728627450980393e-06,
+      "loss": 0.3012955856323242,
+      "step": 4550
+    },
+    {
+      "epoch": 3.607843137254902,
+      "grad_norm": 39.25,
+      "learning_rate": 5.571764705882353e-06,
+      "loss": 0.2885744857788086,
+      "step": 4600
+    },
+    {
+      "epoch": 3.6470588235294117,
+      "grad_norm": 10.3125,
+      "learning_rate": 5.414901960784314e-06,
+      "loss": 0.2811390686035156,
+      "step": 4650
+    },
+    {
+      "epoch": 3.686274509803922,
+      "grad_norm": 11.5625,
+      "learning_rate": 5.2580392156862745e-06,
+      "loss": 0.27347116470336913,
+      "step": 4700
+    },
+    {
+      "epoch": 3.7254901960784315,
+      "grad_norm": 14.125,
+      "learning_rate": 5.101176470588236e-06,
+      "loss": 0.2942588424682617,
+      "step": 4750
+    },
+    {
+      "epoch": 3.764705882352941,
+      "grad_norm": 9.25,
+      "learning_rate": 4.9443137254901966e-06,
+      "loss": 0.3216423034667969,
+      "step": 4800
+    },
+    {
+      "epoch": 3.803921568627451,
+      "grad_norm": 11.625,
+      "learning_rate": 4.787450980392157e-06,
+      "loss": 0.2835306930541992,
+      "step": 4850
+    },
+    {
+      "epoch": 3.843137254901961,
+      "grad_norm": 13.4375,
+      "learning_rate": 4.630588235294118e-06,
+      "loss": 0.29528648376464844,
+      "step": 4900
+    },
+    {
+      "epoch": 3.8823529411764706,
+      "grad_norm": 14.5625,
+      "learning_rate": 4.473725490196079e-06,
+      "loss": 0.30034454345703127,
+      "step": 4950
+    },
+    {
+      "epoch": 3.9215686274509802,
+      "grad_norm": 11.75,
+      "learning_rate": 4.316862745098039e-06,
+      "loss": 0.2962734985351563,
+      "step": 5000
+    },
+    {
+      "epoch": 3.9607843137254903,
+      "grad_norm": 9.875,
+      "learning_rate": 4.16e-06,
+      "loss": 0.2890406036376953,
+      "step": 5050
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 26.5,
+      "learning_rate": 4.003137254901961e-06,
+      "loss": 0.30147428512573243,
+      "step": 5100
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.27950868010520935,
+      "eval_runtime": 1.8861,
+      "eval_samples_per_second": 89.073,
+      "eval_steps_per_second": 1.591,
+      "step": 5100
+    },
+    {
+      "epoch": 4.03921568627451,
+      "grad_norm": 9.8125,
+      "learning_rate": 3.846274509803922e-06,
+      "loss": 0.30072477340698245,
+      "step": 5150
+    },
+    {
+      "epoch": 4.078431372549019,
+      "grad_norm": 10.875,
+      "learning_rate": 3.6894117647058826e-06,
+      "loss": 0.3238717651367187,
+      "step": 5200
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 12.9375,
+      "learning_rate": 3.5325490196078432e-06,
+      "loss": 0.32802242279052735,
+      "step": 5250
+    },
+    {
+      "epoch": 4.1568627450980395,
+      "grad_norm": 11.875,
+      "learning_rate": 3.375686274509804e-06,
+      "loss": 0.30969699859619143,
+      "step": 5300
+    },
+    {
+      "epoch": 4.196078431372549,
+      "grad_norm": 10.9375,
+      "learning_rate": 3.218823529411765e-06,
+      "loss": 0.30397790908813477,
+      "step": 5350
+    },
+    {
+      "epoch": 4.235294117647059,
+      "grad_norm": 11.375,
+      "learning_rate": 3.0619607843137255e-06,
+      "loss": 0.31028812408447265,
+      "step": 5400
+    },
+    {
+      "epoch": 4.2745098039215685,
+      "grad_norm": 22.625,
+      "learning_rate": 2.9050980392156865e-06,
+      "loss": 0.29748680114746096,
+      "step": 5450
+    },
+    {
+      "epoch": 4.313725490196078,
+      "grad_norm": 13.5625,
+      "learning_rate": 2.748235294117647e-06,
+      "loss": 0.32922935485839844,
+      "step": 5500
+    },
+    {
+      "epoch": 4.352941176470588,
+      "grad_norm": 11.3125,
+      "learning_rate": 2.5913725490196077e-06,
+      "loss": 0.3094036102294922,
+      "step": 5550
+    },
+    {
+      "epoch": 4.392156862745098,
+      "grad_norm": 13.9375,
+      "learning_rate": 2.434509803921569e-06,
+      "loss": 0.29232891082763673,
+      "step": 5600
+    },
+    {
+      "epoch": 4.431372549019608,
+      "grad_norm": 10.75,
+      "learning_rate": 2.2776470588235297e-06,
+      "loss": 0.3246950149536133,
+      "step": 5650
+    },
+    {
+      "epoch": 4.470588235294118,
+      "grad_norm": 12.625,
+      "learning_rate": 2.1207843137254903e-06,
+      "loss": 0.2943936538696289,
+      "step": 5700
+    },
+    {
+      "epoch": 4.509803921568627,
+      "grad_norm": 9.4375,
+      "learning_rate": 1.9639215686274514e-06,
+      "loss": 0.3104681396484375,
+      "step": 5750
+    },
+    {
+      "epoch": 4.549019607843137,
+      "grad_norm": 10.9375,
+      "learning_rate": 1.807058823529412e-06,
+      "loss": 0.30682653427124024,
+      "step": 5800
+    },
+    {
+      "epoch": 4.588235294117647,
+      "grad_norm": 12.5,
+      "learning_rate": 1.6501960784313728e-06,
+      "loss": 0.32573684692382815,
+      "step": 5850
+    },
+    {
+      "epoch": 4.627450980392156,
+      "grad_norm": 12.875,
+      "learning_rate": 1.4933333333333336e-06,
+      "loss": 0.3465504455566406,
+      "step": 5900
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 11.8125,
+      "learning_rate": 1.3364705882352944e-06,
+      "loss": 0.3234005355834961,
+      "step": 5950
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 10.625,
+      "learning_rate": 1.179607843137255e-06,
+      "loss": 0.3344119644165039,
+      "step": 6000
+    },
+    {
+      "epoch": 4.745098039215686,
+      "grad_norm": 12.0,
+      "learning_rate": 1.0227450980392158e-06,
+      "loss": 0.3160061264038086,
+      "step": 6050
+    },
+    {
+      "epoch": 4.784313725490196,
+      "grad_norm": 10.4375,
+      "learning_rate": 8.658823529411765e-07,
+      "loss": 0.3229077911376953,
+      "step": 6100
+    },
+    {
+      "epoch": 4.823529411764706,
+      "grad_norm": 13.6875,
+      "learning_rate": 7.090196078431373e-07,
+      "loss": 0.3019586944580078,
+      "step": 6150
+    },
+    {
+      "epoch": 4.862745098039216,
+      "grad_norm": 12.625,
+      "learning_rate": 5.521568627450981e-07,
+      "loss": 0.30265892028808594,
+      "step": 6200
+    },
+    {
+      "epoch": 4.901960784313726,
+      "grad_norm": 10.875,
+      "learning_rate": 3.9529411764705884e-07,
+      "loss": 0.33167587280273436,
+      "step": 6250
+    },
+    {
+      "epoch": 4.9411764705882355,
+      "grad_norm": 12.875,
+      "learning_rate": 2.3843137254901965e-07,
+      "loss": 0.30589717864990235,
+      "step": 6300
+    },
+    {
+      "epoch": 4.980392156862745,
+      "grad_norm": 10.9375,
+      "learning_rate": 8.156862745098041e-08,
+      "loss": 0.31032962799072267,
+      "step": 6350
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.2776467204093933,
+      "eval_runtime": 1.9003,
+      "eval_samples_per_second": 88.406,
+      "eval_steps_per_second": 1.579,
+      "step": 6375
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 6375,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "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": 0.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

BAAI-bge-reranker-v2-m3-E5-Pairwise/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:010d7ce072836ce48451791f0d5fbc2a37b50650e45049afddbfa3c2078439ab
+size 5265