First commit

Results of seed 2 (best seed) when training microsoft/Multilingual-MiniLM-L12-H384 on rcds/swiss_citation_extraction/original.

.gitattributes

@@ -32,3 +32,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+test_predictions_detailed.csv filter=lfs diff=lfs merge=lfs -text
+test_predictions_detailed.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

all_results.json

@@ -0,0 +1,105 @@
+{
+    "de_predict/_CITATION_f1": 0.9793131792857794,
+    "de_predict/_CITATION_precision": 0.9852522282458881,
+    "de_predict/_CITATION_recall": 0.9734453018610985,
+    "de_predict/_LAW_f1": 0.9207842961099632,
+    "de_predict/_LAW_precision": 0.8598544432559407,
+    "de_predict/_LAW_recall": 0.9910077594333921,
+    "de_predict/_accuracy_normalized": 0.9880353522464387,
+    "de_predict/_loss": 0.035749662667512894,
+    "de_predict/_macro-f1": 0.9504272924171073,
+    "de_predict/_macro-precision": 0.9822265306472453,
+    "de_predict/_macro-recall": 0.9232171398568052,
+    "de_predict/_micro-f1": 0.9405898834091524,
+    "de_predict/_micro-precision": 0.9849051246865093,
+    "de_predict/_micro-recall": 0.9000908134288556,
+    "de_predict/_runtime": 265.8036,
+    "de_predict/_samples_per_second": 70.277,
+    "de_predict/_steps_per_second": 0.549,
+    "de_predict/_weighted-f1": 0.939658320951984,
+    "de_predict/_weighted-precision": 0.9854977355183103,
+    "de_predict/_weighted-recall": 0.9000908134288556,
+    "epoch": 23.0,
+    "fr_predict/_CITATION_f1": 0.9554686901203342,
+    "fr_predict/_CITATION_precision": 0.9684586699813549,
+    "fr_predict/_CITATION_recall": 0.9428225684465286,
+    "fr_predict/_LAW_f1": 0.910095519316377,
+    "fr_predict/_LAW_precision": 0.8366717393986756,
+    "fr_predict/_LAW_recall": 0.9976459048553212,
+    "fr_predict/_accuracy_normalized": 0.9830767480044869,
+    "fr_predict/_loss": 0.04828711599111557,
+    "fr_predict/_macro-f1": 0.9330080903677362,
+    "fr_predict/_macro-precision": 0.9702342366509249,
+    "fr_predict/_macro-recall": 0.9029739799827206,
+    "fr_predict/_micro-f1": 0.920617324580396,
+    "fr_predict/_micro-precision": 0.9842228065627199,
+    "fr_predict/_micro-recall": 0.8647338279317974,
+    "fr_predict/_runtime": 102.8006,
+    "fr_predict/_samples_per_second": 74.863,
+    "fr_predict/_steps_per_second": 0.593,
+    "fr_predict/_weighted-f1": 0.9198669665372888,
+    "fr_predict/_weighted-precision": 0.9861681830521788,
+    "fr_predict/_weighted-recall": 0.8647338279317974,
+    "it_predict/_CITATION_f1": 0.9703896103896105,
+    "it_predict/_CITATION_precision": 0.9769874476987448,
+    "it_predict/_CITATION_recall": 0.9638802889576883,
+    "it_predict/_LAW_f1": 0.9099276791584483,
+    "it_predict/_LAW_precision": 0.8422590068159689,
+    "it_predict/_LAW_recall": 0.9894195024306548,
+    "it_predict/_accuracy_normalized": 0.9892137683075134,
+    "it_predict/_loss": 0.031956374645233154,
+    "it_predict/_macro-f1": 0.9413484848298093,
+    "it_predict/_macro-precision": 0.9766498956941716,
+    "it_predict/_macro-recall": 0.9119834901073706,
+    "it_predict/_micro-f1": 0.9311429570080392,
+    "it_predict/_micro-precision": 0.9803127874885005,
+    "it_predict/_micro-recall": 0.8866699950074888,
+    "it_predict/_runtime": 26.6394,
+    "it_predict/_samples_per_second": 69.146,
+    "it_predict/_steps_per_second": 0.563,
+    "it_predict/_weighted-f1": 0.929971077318579,
+    "it_predict/_weighted-precision": 0.9813271971464931,
+    "it_predict/_weighted-recall": 0.8866699950074888,
+    "predict/_CITATION_f1": 0.973621340187501,
+    "predict/_CITATION_precision": 0.981138340970977,
+    "predict/_CITATION_recall": 0.9662186467837405,
+    "predict/_LAW_f1": 0.9168199439712499,
+    "predict/_LAW_precision": 0.8514980289093298,
+    "predict/_LAW_recall": 0.9929968125536349,
+    "predict/_accuracy_normalized": 0.986841752305624,
+    "predict/_loss": 0.03871648758649826,
+    "predict/_macro-f1": 0.9455976917351873,
+    "predict/_macro-precision": 0.9796077296686877,
+    "predict/_macro-recall": 0.9169959471957758,
+    "predict/_micro-f1": 0.934344809828224,
+    "predict/_micro-precision": 0.9844524443053164,
+    "predict/_micro-recall": 0.8890909776278342,
+    "predict/_runtime": 396.5316,
+    "predict/_samples_per_second": 71.162,
+    "predict/_steps_per_second": 0.557,
+    "predict/_weighted-f1": 0.9333974918752409,
+    "predict/_weighted-precision": 0.9854002360022739,
+    "predict/_weighted-recall": 0.8890909776278342,
+    "predict_samples": 28218,
+    "train_CITATION_f1": 0.9814787122684838,
+    "train_CITATION_precision": 0.9844650624611626,
+    "train_CITATION_recall": 0.978510425318936,
+    "train_LAW_f1": 0.9211885809936669,
+    "train_LAW_precision": 0.8550646934923544,
+    "train_LAW_recall": 0.9983966169255928,
+    "train_accuracy_normalized": 0.9905667906095162,
+    "train_loss": 0.026775669306516647,
+    "train_macro-f1": 0.9515974221562933,
+    "train_macro-precision": 0.9884535211222644,
+    "train_macro-recall": 0.9202279237061262,
+    "train_micro-f1": 0.9429566347321088,
+    "train_micro-precision": 0.9909403074465961,
+    "train_micro-recall": 0.8994053033460294,
+    "train_runtime": 1297.076,
+    "train_samples": 90101,
+    "train_samples_per_second": 69.465,
+    "train_steps_per_second": 0.543,
+    "train_weighted-f1": 0.9419101385291693,
+    "train_weighted-precision": 0.9916707876514618,
+    "train_weighted-recall": 0.8994053033460294
+}

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "microsoft/Multilingual-MiniLM-L12-H384",
+  "architectures": [
+    "BertForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "finetuning_task": "all_original",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "id2label": {
+    "0": "0",
+    "1": "B-CITATION",
+    "2": "I-CITATION",
+    "3": "B-LAW",
+    "4": "I-LAW"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "label2id": {
+    "LABEL_0": "O",
+    "LABEL_1": "B-CITATION",
+    "LABEL_2": "I-CITATION",
+    "LABEL_3": "B-LAW",
+    "LABEL_4": "I-LAW"
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "torch_dtype": "float16",
+  "transformers_version": "4.28.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 250037
+}

predict_results.json

@@ -0,0 +1,83 @@
+{
+    "de_predict/_CITATION_f1": 0.9793131792857794,
+    "de_predict/_CITATION_precision": 0.9852522282458881,
+    "de_predict/_CITATION_recall": 0.9734453018610985,
+    "de_predict/_LAW_f1": 0.9207842961099632,
+    "de_predict/_LAW_precision": 0.8598544432559407,
+    "de_predict/_LAW_recall": 0.9910077594333921,
+    "de_predict/_accuracy_normalized": 0.9880353522464387,
+    "de_predict/_loss": 0.035749662667512894,
+    "de_predict/_macro-f1": 0.9504272924171073,
+    "de_predict/_macro-precision": 0.9822265306472453,
+    "de_predict/_macro-recall": 0.9232171398568052,
+    "de_predict/_micro-f1": 0.9405898834091524,
+    "de_predict/_micro-precision": 0.9849051246865093,
+    "de_predict/_micro-recall": 0.9000908134288556,
+    "de_predict/_runtime": 265.8036,
+    "de_predict/_samples_per_second": 70.277,
+    "de_predict/_steps_per_second": 0.549,
+    "de_predict/_weighted-f1": 0.939658320951984,
+    "de_predict/_weighted-precision": 0.9854977355183103,
+    "de_predict/_weighted-recall": 0.9000908134288556,
+    "fr_predict/_CITATION_f1": 0.9554686901203342,
+    "fr_predict/_CITATION_precision": 0.9684586699813549,
+    "fr_predict/_CITATION_recall": 0.9428225684465286,
+    "fr_predict/_LAW_f1": 0.910095519316377,
+    "fr_predict/_LAW_precision": 0.8366717393986756,
+    "fr_predict/_LAW_recall": 0.9976459048553212,
+    "fr_predict/_accuracy_normalized": 0.9830767480044869,
+    "fr_predict/_loss": 0.04828711599111557,
+    "fr_predict/_macro-f1": 0.9330080903677362,
+    "fr_predict/_macro-precision": 0.9702342366509249,
+    "fr_predict/_macro-recall": 0.9029739799827206,
+    "fr_predict/_micro-f1": 0.920617324580396,
+    "fr_predict/_micro-precision": 0.9842228065627199,
+    "fr_predict/_micro-recall": 0.8647338279317974,
+    "fr_predict/_runtime": 102.8006,
+    "fr_predict/_samples_per_second": 74.863,
+    "fr_predict/_steps_per_second": 0.593,
+    "fr_predict/_weighted-f1": 0.9198669665372888,
+    "fr_predict/_weighted-precision": 0.9861681830521788,
+    "fr_predict/_weighted-recall": 0.8647338279317974,
+    "it_predict/_CITATION_f1": 0.9703896103896105,
+    "it_predict/_CITATION_precision": 0.9769874476987448,
+    "it_predict/_CITATION_recall": 0.9638802889576883,
+    "it_predict/_LAW_f1": 0.9099276791584483,
+    "it_predict/_LAW_precision": 0.8422590068159689,
+    "it_predict/_LAW_recall": 0.9894195024306548,
+    "it_predict/_accuracy_normalized": 0.9892137683075134,
+    "it_predict/_loss": 0.031956374645233154,
+    "it_predict/_macro-f1": 0.9413484848298093,
+    "it_predict/_macro-precision": 0.9766498956941716,
+    "it_predict/_macro-recall": 0.9119834901073706,
+    "it_predict/_micro-f1": 0.9311429570080392,
+    "it_predict/_micro-precision": 0.9803127874885005,
+    "it_predict/_micro-recall": 0.8866699950074888,
+    "it_predict/_runtime": 26.6394,
+    "it_predict/_samples_per_second": 69.146,
+    "it_predict/_steps_per_second": 0.563,
+    "it_predict/_weighted-f1": 0.929971077318579,
+    "it_predict/_weighted-precision": 0.9813271971464931,
+    "it_predict/_weighted-recall": 0.8866699950074888,
+    "predict/_CITATION_f1": 0.973621340187501,
+    "predict/_CITATION_precision": 0.981138340970977,
+    "predict/_CITATION_recall": 0.9662186467837405,
+    "predict/_LAW_f1": 0.9168199439712499,
+    "predict/_LAW_precision": 0.8514980289093298,
+    "predict/_LAW_recall": 0.9929968125536349,
+    "predict/_accuracy_normalized": 0.986841752305624,
+    "predict/_loss": 0.03871648758649826,
+    "predict/_macro-f1": 0.9455976917351873,
+    "predict/_macro-precision": 0.9796077296686877,
+    "predict/_macro-recall": 0.9169959471957758,
+    "predict/_micro-f1": 0.934344809828224,
+    "predict/_micro-precision": 0.9844524443053164,
+    "predict/_micro-recall": 0.8890909776278342,
+    "predict/_runtime": 396.5316,
+    "predict/_samples_per_second": 71.162,
+    "predict/_steps_per_second": 0.557,
+    "predict/_weighted-f1": 0.9333974918752409,
+    "predict/_weighted-precision": 0.9854002360022739,
+    "predict/_weighted-recall": 0.8890909776278342,
+    "predict_samples": 28218
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db557e55138656512eaf5ccb97ae3df00697588e8157f005996d982a963fe9db
+size 235087853

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

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>"
+}

test_predictions_detailed.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5189f91e0ed539a87b24b370cbe66c7f00336dc4d4a734cc66b2cca812580d1
+size 138154683

test_predictions_detailed.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d732e378ec11d5346171b8e9b1d5e9e266b77ec17192d7e366ad128f1f71a6
+size 960496838

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2bae0289b9bb66bfbf85d834f6313ebd8e40f484df2bd0351539e74a0e4ddeb4
+size 17098081

tokenizer_config.json

@@ -0,0 +1,20 @@
+{
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,24 @@
+{
+    "epoch": 23.0,
+    "train_CITATION_f1": 0.9814787122684838,
+    "train_CITATION_precision": 0.9844650624611626,
+    "train_CITATION_recall": 0.978510425318936,
+    "train_LAW_f1": 0.9211885809936669,
+    "train_LAW_precision": 0.8550646934923544,
+    "train_LAW_recall": 0.9983966169255928,
+    "train_accuracy_normalized": 0.9905667906095162,
+    "train_loss": 0.026775669306516647,
+    "train_macro-f1": 0.9515974221562933,
+    "train_macro-precision": 0.9884535211222644,
+    "train_macro-recall": 0.9202279237061262,
+    "train_micro-f1": 0.9429566347321088,
+    "train_micro-precision": 0.9909403074465961,
+    "train_micro-recall": 0.8994053033460294,
+    "train_runtime": 1297.076,
+    "train_samples": 90101,
+    "train_samples_per_second": 69.465,
+    "train_steps_per_second": 0.543,
+    "train_weighted-f1": 0.9419101385291693,
+    "train_weighted-precision": 0.9916707876514618,
+    "train_weighted-recall": 0.8994053033460294
+}

trainer_state.json

@@ -0,0 +1,739 @@
+{
+  "best_metric": 0.038164205849170685,
+  "best_model_checkpoint": "swiss_citation_extraction/original/microsoft/Multilingual-MiniLM-L12-H384/seed_2/checkpoint-12672",
+  "epoch": 23.0,
+  "global_step": 16192,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "learning_rate": 9.800000000000001e-06,
+      "loss": 0.4621,
+      "step": 704
+    },
+    {
+      "epoch": 1.0,
+      "eval_CITATION_f1": 0.9656679647142971,
+      "eval_CITATION_precision": 0.9683614918712146,
+      "eval_CITATION_recall": 0.9629893802504359,
+      "eval_LAW_f1": 0.8924282681945703,
+      "eval_LAW_precision": 0.8161431325170364,
+      "eval_LAW_recall": 0.9844445899921405,
+      "eval_accuracy_normalized": 0.9826788171520043,
+      "eval_loss": 0.22096700966358185,
+      "eval_macro-f1": 0.9328819776713801,
+      "eval_macro-precision": 0.9737169851212881,
+      "eval_macro-recall": 0.8988391632624153,
+      "eval_micro-f1": 0.9176521739130435,
+      "eval_micro-precision": 0.9781712008156833,
+      "eval_micro-recall": 0.8641853990091307,
+      "eval_runtime": 183.4312,
+      "eval_samples_per_second": 69.427,
+      "eval_steps_per_second": 0.545,
+      "eval_weighted-f1": 0.9165196950110328,
+      "eval_weighted-precision": 0.9789422067379471,
+      "eval_weighted-recall": 0.8641853990091307,
+      "step": 704
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 9.600000000000001e-06,
+      "loss": 0.1514,
+      "step": 1408
+    },
+    {
+      "epoch": 2.0,
+      "eval_CITATION_f1": 0.967408173823073,
+      "eval_CITATION_precision": 0.9715450403644793,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9143357699510664,
+      "eval_LAW_precision": 0.8484695593676421,
+      "eval_LAW_recall": 0.9912889703955986,
+      "eval_accuracy_normalized": 0.9859679337789155,
+      "eval_loss": 0.12093725800514221,
+      "eval_macro-f1": 0.9415014323169688,
+      "eval_macro-precision": 0.9772976790478547,
+      "eval_macro-recall": 0.9111432819443076,
+      "eval_micro-f1": 0.9296999978086035,
+      "eval_micro-precision": 0.9831070120962135,
+      "eval_micro-recall": 0.881796641170602,
+      "eval_runtime": 182.5229,
+      "eval_samples_per_second": 69.772,
+      "eval_steps_per_second": 0.548,
+      "eval_weighted-f1": 0.9288920564740497,
+      "eval_weighted-precision": 0.9840176399820014,
+      "eval_weighted-recall": 0.881796641170602,
+      "step": 1408
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 9.4e-06,
+      "loss": 0.0915,
+      "step": 2112
+    },
+    {
+      "epoch": 3.0,
+      "eval_CITATION_f1": 0.967600700525394,
+      "eval_CITATION_precision": 0.9719334719334719,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9151176825588413,
+      "eval_LAW_precision": 0.8484500895255148,
+      "eval_LAW_recall": 0.9931556195965417,
+      "eval_accuracy_normalized": 0.9861328466436042,
+      "eval_loss": 0.08684782683849335,
+      "eval_macro-f1": 0.941871559216277,
+      "eval_macro-precision": 0.9782310036483264,
+      "eval_macro-recall": 0.9110962962834105,
+      "eval_micro-f1": 0.9302254289061389,
+      "eval_micro-precision": 0.9844278630022709,
+      "eval_micro-recall": 0.8816802606728514,
+      "eval_runtime": 182.8851,
+      "eval_samples_per_second": 69.634,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9294201322268216,
+      "eval_weighted-precision": 0.985410666040948,
+      "eval_weighted-recall": 0.8816802606728514,
+      "step": 2112
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 9.200000000000002e-06,
+      "loss": 0.0666,
+      "step": 2816
+    },
+    {
+      "epoch": 4.0,
+      "eval_CITATION_f1": 0.9689891581632653,
+      "eval_CITATION_precision": 0.9747393744987971,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9185858127034597,
+      "eval_LAW_precision": 0.854262299698668,
+      "eval_LAW_recall": 0.9933848572177102,
+      "eval_accuracy_normalized": 0.986596354625882,
+      "eval_loss": 0.06908567994832993,
+      "eval_macro-f1": 0.9441561554169013,
+      "eval_macro-precision": 0.9783456224589105,
+      "eval_macro-recall": 0.9151899141069411,
+      "eval_micro-f1": 0.9328400807938877,
+      "eval_micro-precision": 0.984590072762664,
+      "eval_micro-recall": 0.8862583955613033,
+      "eval_runtime": 182.9721,
+      "eval_samples_per_second": 69.601,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9320652823900626,
+      "eval_weighted-precision": 0.9855669145232293,
+      "eval_weighted-recall": 0.8862583955613033,
+      "step": 2816
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 9e-06,
+      "loss": 0.0536,
+      "step": 3520
+    },
+    {
+      "epoch": 5.0,
+      "eval_CITATION_f1": 0.9693755482893373,
+      "eval_CITATION_precision": 0.9755216693418941,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9195075040511275,
+      "eval_LAW_precision": 0.8552753909001268,
+      "eval_LAW_recall": 0.9941708147760021,
+      "eval_accuracy_normalized": 0.9867220387952753,
+      "eval_loss": 0.059190474450588226,
+      "eval_macro-f1": 0.9446558923319595,
+      "eval_macro-precision": 0.9787386012380566,
+      "eval_macro-recall": 0.9157752849313606,
+      "eval_micro-f1": 0.9335309617918313,
+      "eval_micro-precision": 0.9851462205125828,
+      "eval_micro-recall": 0.887055043191587,
+      "eval_runtime": 182.5981,
+      "eval_samples_per_second": 69.743,
+      "eval_steps_per_second": 0.548,
+      "eval_weighted-f1": 0.932770084197937,
+      "eval_weighted-precision": 0.9861471971793212,
+      "eval_weighted-recall": 0.887055043191587,
+      "step": 3520
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 8.8e-06,
+      "loss": 0.0458,
+      "step": 4224
+    },
+    {
+      "epoch": 6.0,
+      "eval_CITATION_f1": 0.9689891581632653,
+      "eval_CITATION_precision": 0.9747393744987971,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9197133376767018,
+      "eval_LAW_precision": 0.8558014944311293,
+      "eval_LAW_recall": 0.9939415771548337,
+      "eval_accuracy_normalized": 0.9867346072122146,
+      "eval_loss": 0.05432054027915001,
+      "eval_macro-f1": 0.944648370774704,
+      "eval_macro-precision": 0.9786239824274723,
+      "eval_macro-recall": 0.915818979128866,
+      "eval_micro-f1": 0.9336130116300778,
+      "eval_micro-precision": 0.9849840107521899,
+      "eval_micro-recall": 0.8873348224537085,
+      "eval_runtime": 182.9328,
+      "eval_samples_per_second": 69.616,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.93286298930164,
+      "eval_weighted-precision": 0.9859705314716058,
+      "eval_weighted-recall": 0.8873348224537085,
+      "step": 4224
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 8.6e-06,
+      "loss": 0.0407,
+      "step": 4928
+    },
+    {
+      "epoch": 7.0,
+      "eval_CITATION_f1": 0.9689505360915142,
+      "eval_CITATION_precision": 0.9746612140165183,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9199606090447693,
+      "eval_LAW_precision": 0.8559869181538808,
+      "eval_LAW_recall": 0.9942690594707886,
+      "eval_accuracy_normalized": 0.9867467947680346,
+      "eval_loss": 0.04946836084127426,
+      "eval_macro-f1": 0.94497928086637,
+      "eval_macro-precision": 0.9787877235854497,
+      "eval_macro-recall": 0.9163274179869876,
+      "eval_micro-f1": 0.9339147052685917,
+      "eval_micro-precision": 0.9852157389813228,
+      "eval_micro-recall": 0.8876918258690887,
+      "eval_runtime": 183.49,
+      "eval_samples_per_second": 69.404,
+      "eval_steps_per_second": 0.545,
+      "eval_weighted-f1": 0.9331590580868986,
+      "eval_weighted-precision": 0.9862179243996534,
+      "eval_weighted-recall": 0.8876918258690887,
+      "step": 4928
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 8.400000000000001e-06,
+      "loss": 0.0371,
+      "step": 5632
+    },
+    {
+      "epoch": 8.0,
+      "eval_CITATION_f1": 0.9696075303126994,
+      "eval_CITATION_precision": 0.975991649269311,
+      "eval_CITATION_recall": 0.9633063877001109,
+      "eval_LAW_f1": 0.9197064830925183,
+      "eval_LAW_precision": 0.85474844624427,
+      "eval_LAW_recall": 0.9953497511134399,
+      "eval_accuracy_normalized": 0.9866988062669936,
+      "eval_loss": 0.045815542340278625,
+      "eval_macro-f1": 0.9452922817050826,
+      "eval_macro-precision": 0.9793280694067754,
+      "eval_macro-recall": 0.9165665634793729,
+      "eval_micro-f1": 0.934054837222576,
+      "eval_micro-precision": 0.9859804421374612,
+      "eval_micro-recall": 0.8873248248248248,
+      "eval_runtime": 182.9672,
+      "eval_samples_per_second": 69.603,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9332733879110633,
+      "eval_weighted-precision": 0.9870388712591779,
+      "eval_weighted-recall": 0.8873248248248248,
+      "step": 5632
+    },
+    {
+      "epoch": 9.0,
+      "learning_rate": 8.2e-06,
+      "loss": 0.0345,
+      "step": 6336
+    },
+    {
+      "epoch": 9.0,
+      "eval_CITATION_f1": 0.9796242602375184,
+      "eval_CITATION_precision": 0.9819253125248826,
+      "eval_CITATION_recall": 0.9773339673482326,
+      "eval_LAW_f1": 0.9201594688414254,
+      "eval_LAW_precision": 0.8565743798154263,
+      "eval_LAW_recall": 0.9939415771548337,
+      "eval_accuracy_normalized": 0.9868583870760111,
+      "eval_loss": 0.04324862360954285,
+      "eval_macro-f1": 0.9502039808194804,
+      "eval_macro-precision": 0.9856377722515332,
+      "eval_macro-recall": 0.9198232236455686,
+      "eval_micro-f1": 0.9369347616120818,
+      "eval_micro-precision": 0.9890856004078417,
+      "eval_micro-recall": 0.8900079235998165,
+      "eval_runtime": 182.6619,
+      "eval_samples_per_second": 69.719,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9360739408863803,
+      "eval_weighted-precision": 0.9895945375388725,
+      "eval_weighted-recall": 0.8900079235998165,
+      "step": 6336
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.0326,
+      "step": 7040
+    },
+    {
+      "epoch": 10.0,
+      "eval_CITATION_f1": 0.98077534159517,
+      "eval_CITATION_precision": 0.9831183309444179,
+      "eval_CITATION_recall": 0.9784434934220955,
+      "eval_LAW_f1": 0.9204929588139884,
+      "eval_LAW_precision": 0.8568848248807609,
+      "eval_LAW_recall": 0.9943018077023841,
+      "eval_accuracy_normalized": 0.9868957114657098,
+      "eval_loss": 0.04360693693161011,
+      "eval_macro-f1": 0.9508766633650136,
+      "eval_macro-precision": 0.9863726505622398,
+      "eval_macro-recall": 0.9204543086360064,
+      "eval_micro-f1": 0.9373909417148626,
+      "eval_micro-precision": 0.9896649209806738,
+      "eval_micro-recall": 0.8903621239602226,
+      "eval_runtime": 182.4045,
+      "eval_samples_per_second": 69.817,
+      "eval_steps_per_second": 0.548,
+      "eval_weighted-f1": 0.9365143149914062,
+      "eval_weighted-precision": 0.9901520055290943,
+      "eval_weighted-recall": 0.8903621239602226,
+      "step": 7040
+    },
+    {
+      "epoch": 11.0,
+      "learning_rate": 7.800000000000002e-06,
+      "loss": 0.0312,
+      "step": 7744
+    },
+    {
+      "epoch": 11.0,
+      "eval_CITATION_f1": 0.9814785373608903,
+      "eval_CITATION_precision": 0.9844522404720141,
+      "eval_CITATION_recall": 0.9785227452845142,
+      "eval_LAW_f1": 0.920169278140974,
+      "eval_LAW_precision": 0.8570540532903845,
+      "eval_LAW_recall": 0.9933193607545192,
+      "eval_accuracy_normalized": 0.9868713363540699,
+      "eval_loss": 0.041282109916210175,
+      "eval_macro-f1": 0.9510066434146158,
+      "eval_macro-precision": 0.9859210530195167,
+      "eval_macro-recall": 0.9210918842900306,
+      "eval_micro-f1": 0.937299600298678,
+      "eval_micro-precision": 0.9889929091161885,
+      "eval_micro-recall": 0.8907417456275828,
+      "eval_runtime": 183.3833,
+      "eval_samples_per_second": 69.445,
+      "eval_steps_per_second": 0.545,
+      "eval_weighted-f1": 0.936407960886005,
+      "eval_weighted-precision": 0.9894474242690886,
+      "eval_weighted-recall": 0.8907417456275828,
+      "step": 7744
+    },
+    {
+      "epoch": 12.0,
+      "learning_rate": 7.600000000000001e-06,
+      "loss": 0.0302,
+      "step": 8448
+    },
+    {
+      "epoch": 12.0,
+      "eval_CITATION_f1": 0.9821840451761711,
+      "eval_CITATION_precision": 0.9857120051085568,
+      "eval_CITATION_recall": 0.9786812490093517,
+      "eval_LAW_f1": 0.9202834903231668,
+      "eval_LAW_precision": 0.8559637162657051,
+      "eval_LAW_recall": 0.9950550170290804,
+      "eval_accuracy_normalized": 0.9868378205755649,
+      "eval_loss": 0.0409509539604187,
+      "eval_macro-f1": 0.9515210413529415,
+      "eval_macro-precision": 0.986868133019216,
+      "eval_macro-recall": 0.9213572734980617,
+      "eval_micro-f1": 0.937704342483488,
+      "eval_micro-precision": 0.9902674143764193,
+      "eval_micro-recall": 0.8904400733455576,
+      "eval_runtime": 182.6717,
+      "eval_samples_per_second": 69.715,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9367869766694694,
+      "eval_weighted-precision": 0.9907821159064123,
+      "eval_weighted-recall": 0.8904400733455576,
+      "step": 8448
+    },
+    {
+      "epoch": 13.0,
+      "learning_rate": 7.4e-06,
+      "loss": 0.0294,
+      "step": 9152
+    },
+    {
+      "epoch": 13.0,
+      "eval_CITATION_f1": 0.9822635806887775,
+      "eval_CITATION_precision": 0.9857918263090677,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9207781394292636,
+      "eval_LAW_precision": 0.8561896449787437,
+      "eval_LAW_recall": 0.9959064710505633,
+      "eval_accuracy_normalized": 0.986947889439064,
+      "eval_loss": 0.039835721254348755,
+      "eval_macro-f1": 0.9517621380923427,
+      "eval_macro-precision": 0.9873334859611669,
+      "eval_macro-recall": 0.9214192450961509,
+      "eval_micro-f1": 0.9380497970823737,
+      "eval_micro-precision": 0.9908930805950781,
+      "eval_micro-recall": 0.8905573142285905,
+      "eval_runtime": 183.2667,
+      "eval_samples_per_second": 69.489,
+      "eval_steps_per_second": 0.546,
+      "eval_weighted-f1": 0.937137235367138,
+      "eval_weighted-precision": 0.9914335782136015,
+      "eval_weighted-recall": 0.8905573142285905,
+      "step": 9152
+    },
+    {
+      "epoch": 14.0,
+      "learning_rate": 7.2000000000000005e-06,
+      "loss": 0.0288,
+      "step": 9856
+    },
+    {
+      "epoch": 14.0,
+      "eval_CITATION_f1": 0.9823417117403754,
+      "eval_CITATION_precision": 0.9859492256107297,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9209932279909707,
+      "eval_LAW_precision": 0.8569253756589891,
+      "eval_LAW_recall": 0.9954152475766309,
+      "eval_accuracy_normalized": 0.986973407134062,
+      "eval_loss": 0.03937051072716713,
+      "eval_macro-f1": 0.9518572703321915,
+      "eval_macro-precision": 0.9870878742242006,
+      "eval_macro-recall": 0.9217881303971995,
+      "eval_micro-f1": 0.9381220440903755,
+      "eval_micro-precision": 0.9905454882513788,
+      "eval_micro-recall": 0.8909685891156179,
+      "eval_runtime": 183.2841,
+      "eval_samples_per_second": 69.482,
+      "eval_steps_per_second": 0.546,
+      "eval_weighted-f1": 0.937211000059121,
+      "eval_weighted-precision": 0.9910683576452416,
+      "eval_weighted-recall": 0.8909685891156179,
+      "step": 9856
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 7.000284090909092e-06,
+      "loss": 0.0282,
+      "step": 10560
+    },
+    {
+      "epoch": 15.0,
+      "eval_CITATION_f1": 0.9819121447028423,
+      "eval_CITATION_precision": 0.9850841509132966,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9207796338083628,
+      "eval_LAW_precision": 0.8564586561487534,
+      "eval_LAW_recall": 0.9955462405030129,
+      "eval_accuracy_normalized": 0.9870015908568956,
+      "eval_loss": 0.039993565529584885,
+      "eval_macro-f1": 0.9517671412400261,
+      "eval_macro-precision": 0.9871533706873916,
+      "eval_macro-recall": 0.9215724095097284,
+      "eval_micro-f1": 0.9380245532040943,
+      "eval_micro-precision": 0.9906381795430319,
+      "eval_micro-recall": 0.8907177831024065,
+      "eval_runtime": 183.1632,
+      "eval_samples_per_second": 69.528,
+      "eval_steps_per_second": 0.546,
+      "eval_weighted-f1": 0.937111701553557,
+      "eval_weighted-precision": 0.9911661050067804,
+      "eval_weighted-recall": 0.8907177831024065,
+      "step": 10560
+    },
+    {
+      "epoch": 16.0,
+      "learning_rate": 6.800284090909092e-06,
+      "loss": 0.0278,
+      "step": 11264
+    },
+    {
+      "epoch": 16.0,
+      "eval_CITATION_f1": 0.9828889773179467,
+      "eval_CITATION_precision": 0.9870524296675192,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9207239434911781,
+      "eval_LAW_precision": 0.8570662603002597,
+      "eval_LAW_recall": 0.9945965417867435,
+      "eval_accuracy_normalized": 0.9870153018571931,
+      "eval_loss": 0.03851037099957466,
+      "eval_macro-f1": 0.9520466388916144,
+      "eval_macro-precision": 0.986678521329257,
+      "eval_macro-recall": 0.9224923590174634,
+      "eval_micro-f1": 0.9381704786270355,
+      "eval_micro-precision": 0.9899661676785466,
+      "eval_micro-recall": 0.8915252822471254,
+      "eval_runtime": 183.6471,
+      "eval_samples_per_second": 69.345,
+      "eval_steps_per_second": 0.545,
+      "eval_weighted-f1": 0.937250674098129,
+      "eval_weighted-precision": 0.990462627560541,
+      "eval_weighted-recall": 0.8915252822471254,
+      "step": 11264
+    },
+    {
+      "epoch": 17.0,
+      "learning_rate": 6.6002840909090915e-06,
+      "loss": 0.0275,
+      "step": 11968
+    },
+    {
+      "epoch": 17.0,
+      "eval_CITATION_f1": 0.9827716547964827,
+      "eval_CITATION_precision": 0.9868158210147823,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9203515684194575,
+      "eval_LAW_precision": 0.8565183054098268,
+      "eval_LAW_recall": 0.9944655488603615,
+      "eval_accuracy_normalized": 0.9869817860786881,
+      "eval_loss": 0.03849661350250244,
+      "eval_macro-f1": 0.9518505107642844,
+      "eval_macro-precision": 0.9866130248660661,
+      "eval_macro-recall": 0.9221843554847858,
+      "eval_micro-f1": 0.9379384544446518,
+      "eval_micro-precision": 0.9898734763868935,
+      "eval_micro-recall": 0.8911814407610623,
+      "eval_runtime": 182.5915,
+      "eval_samples_per_second": 69.746,
+      "eval_steps_per_second": 0.548,
+      "eval_weighted-f1": 0.9370160495313582,
+      "eval_weighted-precision": 0.990366044125977,
+      "eval_weighted-recall": 0.8911814407610623,
+      "step": 11968
+    },
+    {
+      "epoch": 18.0,
+      "learning_rate": 6.400568181818182e-06,
+      "loss": 0.0272,
+      "step": 12672
+    },
+    {
+      "epoch": 18.0,
+      "eval_CITATION_f1": 0.9828889773179467,
+      "eval_CITATION_precision": 0.9870524296675192,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9205657006147603,
+      "eval_LAW_precision": 0.8561369909311103,
+      "eval_LAW_recall": 0.9954807440398219,
+      "eval_accuracy_normalized": 0.9869932119122694,
+      "eval_loss": 0.038164205849170685,
+      "eval_macro-f1": 0.9519616843549739,
+      "eval_macro-precision": 0.9871206224557962,
+      "eval_macro-recall": 0.9220114522283747,
+      "eval_micro-f1": 0.9380527089596453,
+      "eval_micro-precision": 0.9905918338972054,
+      "eval_micro-recall": 0.8908060348420438,
+      "eval_runtime": 182.7469,
+      "eval_samples_per_second": 69.687,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9371218313775519,
+      "eval_weighted-precision": 0.9911219409633794,
+      "eval_weighted-recall": 0.8908060348420438,
+      "step": 12672
+    },
+    {
+      "epoch": 19.0,
+      "learning_rate": 6.200568181818183e-06,
+      "loss": 0.027,
+      "step": 13376
+    },
+    {
+      "epoch": 19.0,
+      "eval_CITATION_f1": 0.982810759191469,
+      "eval_CITATION_precision": 0.9868946779606841,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9201540514313441,
+      "eval_LAW_precision": 0.8567596566523605,
+      "eval_LAW_recall": 0.9936795913020697,
+      "eval_accuracy_normalized": 0.986947889439064,
+      "eval_loss": 0.03846826031804085,
+      "eval_macro-f1": 0.9517588185840318,
+      "eval_macro-precision": 0.9862200460869202,
+      "eval_macro-recall": 0.922317690725456,
+      "eval_micro-f1": 0.9378130217028381,
+      "eval_micro-precision": 0.9893173286369745,
+      "eval_micro-recall": 0.891406021631102,
+      "eval_runtime": 184.1158,
+      "eval_samples_per_second": 69.168,
+      "eval_steps_per_second": 0.543,
+      "eval_weighted-f1": 0.9368937680506739,
+      "eval_weighted-precision": 0.9897820747141067,
+      "eval_weighted-recall": 0.891406021631102,
+      "step": 13376
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 6.0005681818181824e-06,
+      "loss": 0.0266,
+      "step": 14080
+    },
+    {
+      "epoch": 20.0,
+      "eval_CITATION_f1": 0.982928091050181,
+      "eval_CITATION_precision": 0.9871313244344977,
+      "eval_CITATION_recall": 0.9787605008717705,
+      "eval_LAW_f1": 0.9199981821211617,
+      "eval_LAW_precision": 0.8559548978153629,
+      "eval_LAW_recall": 0.9944000523971706,
+      "eval_accuracy_normalized": 0.9869730262729426,
+      "eval_loss": 0.03889286145567894,
+      "eval_macro-f1": 0.9518160523849399,
+      "eval_macro-precision": 0.9865802766344705,
+      "eval_macro-recall": 0.9221656032832765,
+      "eval_micro-f1": 0.9378327643177853,
+      "eval_micro-precision": 0.9898271307410669,
+      "eval_micro-recall": 0.8910281816475104,
+      "eval_runtime": 183.0494,
+      "eval_samples_per_second": 69.571,
+      "eval_steps_per_second": 0.546,
+      "eval_weighted-f1": 0.936903094166417,
+      "eval_weighted-precision": 0.9903191339376443,
+      "eval_weighted-recall": 0.8910281816475104,
+      "step": 14080
+    },
+    {
+      "epoch": 21.0,
+      "learning_rate": 5.800568181818183e-06,
+      "loss": 0.0264,
+      "step": 14784
+    },
+    {
+      "epoch": 21.0,
+      "eval_CITATION_f1": 0.9821030862233535,
+      "eval_CITATION_precision": 0.9857097237745489,
+      "eval_CITATION_recall": 0.9785227452845142,
+      "eval_LAW_f1": 0.9203370820576556,
+      "eval_LAW_precision": 0.856639015856893,
+      "eval_LAW_recall": 0.9942690594707886,
+      "eval_accuracy_normalized": 0.9869707411062264,
+      "eval_loss": 0.038498904556035995,
+      "eval_macro-f1": 0.9515075473026433,
+      "eval_macro-precision": 0.9863959023776514,
+      "eval_macro-recall": 0.9216946659139149,
+      "eval_micro-f1": 0.9377305462849113,
+      "eval_micro-precision": 0.9896649209806738,
+      "eval_micro-recall": 0.890975090749781,
+      "eval_runtime": 182.7777,
+      "eval_samples_per_second": 69.675,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9368225906951636,
+      "eval_weighted-precision": 0.9901555837282984,
+      "eval_weighted-recall": 0.890975090749781,
+      "step": 14784
+    },
+    {
+      "epoch": 22.0,
+      "learning_rate": 5.600568181818182e-06,
+      "loss": 0.0262,
+      "step": 15488
+    },
+    {
+      "epoch": 22.0,
+      "eval_CITATION_f1": 0.9823403070559223,
+      "eval_CITATION_precision": 0.9860268284893006,
+      "eval_CITATION_recall": 0.9786812490093517,
+      "eval_LAW_f1": 0.9201017657570342,
+      "eval_LAW_precision": 0.8557946926587413,
+      "eval_LAW_recall": 0.9948585276395074,
+      "eval_accuracy_normalized": 0.986917420549514,
+      "eval_loss": 0.040023379027843475,
+      "eval_macro-f1": 0.951632458392357,
+      "eval_macro-precision": 0.9867698883244296,
+      "eval_macro-recall": 0.9216507907951708,
+      "eval_micro-f1": 0.9377784605930252,
+      "eval_micro-precision": 0.9901283774389397,
+      "eval_micro-recall": 0.8906862336362878,
+      "eval_runtime": 182.7513,
+      "eval_samples_per_second": 69.685,
+      "eval_steps_per_second": 0.547,
+      "eval_weighted-f1": 0.9368585268788033,
+      "eval_weighted-precision": 0.9906368284989479,
+      "eval_weighted-recall": 0.8906862336362878,
+      "step": 15488
+    },
+    {
+      "epoch": 23.0,
+      "learning_rate": 5.400568181818182e-06,
+      "loss": 0.0259,
+      "step": 16192
+    },
+    {
+      "epoch": 23.0,
+      "eval_CITATION_f1": 0.9820626019170345,
+      "eval_CITATION_precision": 0.9857085828343314,
+      "eval_CITATION_recall": 0.9784434934220955,
+      "eval_LAW_f1": 0.9201865140642503,
+      "eval_LAW_precision": 0.8556506560054057,
+      "eval_LAW_recall": 0.9952515064186533,
+      "eval_accuracy_normalized": 0.9869360827443634,
+      "eval_loss": 0.03981756791472435,
+      "eval_macro-f1": 0.9515610106395767,
+      "eval_macro-precision": 0.9868474999203745,
+      "eval_macro-recall": 0.9214737971120037,
+      "eval_micro-f1": 0.9377598823877911,
+      "eval_micro-precision": 0.9903369328451592,
+      "eval_micro-recall": 0.8904840289208843,
+      "eval_runtime": 182.5734,
+      "eval_samples_per_second": 69.753,
+      "eval_steps_per_second": 0.548,
+      "eval_weighted-f1": 0.9368400401851952,
+      "eval_weighted-precision": 0.9908674773583548,
+      "eval_weighted-recall": 0.8904840289208843,
+      "step": 16192
+    },
+    {
+      "epoch": 23.0,
+      "step": 16192,
+      "total_flos": 1.3557575111977882e+17,
+      "train_loss": 0.05992582664188189,
+      "train_runtime": 10758.5934,
+      "train_samples_per_second": 418.74,
+      "train_steps_per_second": 3.272
+    },
+    {
+      "epoch": 23.0,
+      "eval_CITATION_f1": 0.9814787122684838,
+      "eval_CITATION_precision": 0.9844650624611626,
+      "eval_CITATION_recall": 0.978510425318936,
+      "eval_LAW_f1": 0.9211885809936669,
+      "eval_LAW_precision": 0.8550646934923544,
+      "eval_LAW_recall": 0.9983966169255928,
+      "eval_accuracy_normalized": 0.9905667906095162,
+      "eval_loss": 0.026775669306516647,
+      "eval_macro-f1": 0.9515974221562933,
+      "eval_macro-precision": 0.9884535211222644,
+      "eval_macro-recall": 0.9202279237061262,
+      "eval_micro-f1": 0.9429566347321088,
+      "eval_micro-precision": 0.9909403074465961,
+      "eval_micro-recall": 0.8994053033460294,
+      "eval_runtime": 1297.076,
+      "eval_samples_per_second": 69.465,
+      "eval_steps_per_second": 0.543,
+      "eval_weighted-f1": 0.9419101385291693,
+      "eval_weighted-precision": 0.9916707876514618,
+      "eval_weighted-recall": 0.8994053033460294,
+      "step": 16192
+    }
+  ],
+  "max_steps": 35200,
+  "num_train_epochs": 50,
+  "total_flos": 1.3557575111977882e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f786e36e587f57aa16e36e866bdbaa5961b59563fddb5e05c500499e1282f912
+size 3707