Training in progress, step 471, checkpoint

checkpoint-471/1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 1536,
+  "pooling_mode_cls_token": false,
+  "pooling_mode_mean_tokens": true,
+  "pooling_mode_max_tokens": false,
+  "pooling_mode_mean_sqrt_len_tokens": false,
+  "pooling_mode_weightedmean_tokens": false,
+  "pooling_mode_lasttoken": false,
+  "include_prompt": true
+}

checkpoint-471/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "[MASK]": 128000
+}

checkpoint-471/config.json

@@ -0,0 +1,38 @@
+{
+  "_name_or_path": "microsoft/deberta-v2-xlarge",
+  "architectures": [
+    "DebertaV2Model"
+  ],
+  "attention_head_size": 64,
+  "attention_probs_dropout_prob": 0.1,
+  "conv_act": "gelu",
+  "conv_kernel_size": 3,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1536,
+  "initializer_range": 0.02,
+  "intermediate_size": 6144,
+  "layer_norm_eps": 1e-07,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta-v2",
+  "norm_rel_ebd": "layer_norm",
+  "num_attention_heads": 24,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 1536,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.4",
+  "type_vocab_size": 0,
+  "vocab_size": 128100
+}

checkpoint-471/config_sentence_transformers.json

@@ -0,0 +1,10 @@
+{
+  "__version__": {
+    "sentence_transformers": "3.0.1",
+    "transformers": "4.42.4",
+    "pytorch": "2.4.0+cu121"
+  },
+  "prompts": {},
+  "default_prompt_name": null,
+  "similarity_fn_name": null
+}

checkpoint-471/modules.json

@@ -0,0 +1,14 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  }
+]

checkpoint-471/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cef56177e678b7edd0d2390d91447e0cce8bc03cc73505794b2edfcdbc6d09b
+size 7077084396

checkpoint-471/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e76c76c4521b1d29132ae9dbef9dd3e44618f6f6905afdd96bd1df39582a37e9
+size 3538506546

checkpoint-471/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd48142d105fcb3b2e718e7f9a42160e502204c22765699cb3e3d4408b6dd128
+size 14244

checkpoint-471/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb8c18e498bcfd8328aa5e44335d38f5f09ef6f04cac30adbe287f46813164b3
+size 1064

checkpoint-471/sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+  "max_seq_length": 512,
+  "do_lower_case": false
+}

checkpoint-471/special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "[CLS]",
+  "cls_token": "[CLS]",
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-471/spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5598d5e96f339a8d980c15f9afd405a2e5e1be7db41de3ed13b0f03fac1e8c17
+size 2447305

checkpoint-471/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128000": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "[CLS]",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": false,
+  "eos_token": "[SEP]",
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "sp_model_kwargs": {},
+  "split_by_punct": false,
+  "tokenizer_class": "DebertaV2Tokenizer",
+  "unk_token": "[UNK]",
+  "vocab_type": "spm"
+}

checkpoint-471/trainer_state.json

@@ -0,0 +1,815 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.30115089514066495,
+  "eval_steps": 157,
+  "global_step": 471,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.010230179028132993,
+      "grad_norm": 200.60667419433594,
+      "learning_rate": 2.129925452609159e-07,
+      "loss": 7.1882,
+      "step": 16
+    },
+    {
+      "epoch": 0.020460358056265986,
+      "grad_norm": 174.4742889404297,
+      "learning_rate": 5.324813631522898e-07,
+      "loss": 9.1489,
+      "step": 32
+    },
+    {
+      "epoch": 0.030690537084398978,
+      "grad_norm": 193.9903106689453,
+      "learning_rate": 8.732694355697552e-07,
+      "loss": 8.805,
+      "step": 48
+    },
+    {
+      "epoch": 0.04092071611253197,
+      "grad_norm": 3077.94775390625,
+      "learning_rate": 1.192758253461129e-06,
+      "loss": 5.7489,
+      "step": 64
+    },
+    {
+      "epoch": 0.05115089514066496,
+      "grad_norm": 80.12879943847656,
+      "learning_rate": 1.5335463258785943e-06,
+      "loss": 4.7163,
+      "step": 80
+    },
+    {
+      "epoch": 0.061381074168797956,
+      "grad_norm": 99.18599700927734,
+      "learning_rate": 1.8743343982960599e-06,
+      "loss": 3.176,
+      "step": 96
+    },
+    {
+      "epoch": 0.07161125319693094,
+      "grad_norm": 23.476715087890625,
+      "learning_rate": 2.1938232161874336e-06,
+      "loss": 2.034,
+      "step": 112
+    },
+    {
+      "epoch": 0.08184143222506395,
+      "grad_norm": 39.820777893066406,
+      "learning_rate": 2.534611288604899e-06,
+      "loss": 1.1278,
+      "step": 128
+    },
+    {
+      "epoch": 0.09207161125319693,
+      "grad_norm": 12.513270378112793,
+      "learning_rate": 2.8753993610223648e-06,
+      "loss": 0.7996,
+      "step": 144
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_Qnli-dev_cosine_accuracy": 0.662109375,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8656402826309204,
+      "eval_Qnli-dev_cosine_ap": 0.6654869266843996,
+      "eval_Qnli-dev_cosine_f1": 0.6504559270516718,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.6888570785522461,
+      "eval_Qnli-dev_cosine_precision": 0.5071090047393365,
+      "eval_Qnli-dev_cosine_recall": 0.9067796610169492,
+      "eval_Qnli-dev_dot_accuracy": 0.654296875,
+      "eval_Qnli-dev_dot_accuracy_threshold": 905.306640625,
+      "eval_Qnli-dev_dot_ap": 0.6262830221036416,
+      "eval_Qnli-dev_dot_f1": 0.65625,
+      "eval_Qnli-dev_dot_f1_threshold": 810.4493408203125,
+      "eval_Qnli-dev_dot_precision": 0.5558823529411765,
+      "eval_Qnli-dev_dot_recall": 0.8008474576271186,
+      "eval_Qnli-dev_euclidean_accuracy": 0.662109375,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 17.05492401123047,
+      "eval_Qnli-dev_euclidean_ap": 0.6671756125841228,
+      "eval_Qnli-dev_euclidean_f1": 0.6495176848874598,
+      "eval_Qnli-dev_euclidean_f1_threshold": 23.73143768310547,
+      "eval_Qnli-dev_euclidean_precision": 0.5233160621761658,
+      "eval_Qnli-dev_euclidean_recall": 0.8559322033898306,
+      "eval_Qnli-dev_manhattan_accuracy": 0.65625,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 526.1962890625,
+      "eval_Qnli-dev_manhattan_ap": 0.6662986170174698,
+      "eval_Qnli-dev_manhattan_f1": 0.6487341772151899,
+      "eval_Qnli-dev_manhattan_f1_threshold": 748.320068359375,
+      "eval_Qnli-dev_manhattan_precision": 0.5176767676767676,
+      "eval_Qnli-dev_manhattan_recall": 0.8686440677966102,
+      "eval_Qnli-dev_max_accuracy": 0.662109375,
+      "eval_Qnli-dev_max_accuracy_threshold": 905.306640625,
+      "eval_Qnli-dev_max_ap": 0.6671756125841228,
+      "eval_Qnli-dev_max_f1": 0.65625,
+      "eval_Qnli-dev_max_f1_threshold": 810.4493408203125,
+      "eval_Qnli-dev_max_precision": 0.5558823529411765,
+      "eval_Qnli-dev_max_recall": 0.9067796610169492,
+      "eval_allNLI-dev_cosine_accuracy": 0.6875,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.9388645887374878,
+      "eval_allNLI-dev_cosine_ap": 0.5213973850248246,
+      "eval_allNLI-dev_cosine_f1": 0.5590361445783132,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.7874077558517456,
+      "eval_allNLI-dev_cosine_precision": 0.4793388429752066,
+      "eval_allNLI-dev_cosine_recall": 0.6705202312138728,
+      "eval_allNLI-dev_dot_accuracy": 0.6953125,
+      "eval_allNLI-dev_dot_accuracy_threshold": 942.6085205078125,
+      "eval_allNLI-dev_dot_ap": 0.520777376545363,
+      "eval_allNLI-dev_dot_f1": 0.5538461538461539,
+      "eval_allNLI-dev_dot_f1_threshold": 817.1332397460938,
+      "eval_allNLI-dev_dot_precision": 0.44680851063829785,
+      "eval_allNLI-dev_dot_recall": 0.7283236994219653,
+      "eval_allNLI-dev_euclidean_accuracy": 0.6875,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 12.643478393554688,
+      "eval_allNLI-dev_euclidean_ap": 0.518083458930424,
+      "eval_allNLI-dev_euclidean_f1": 0.5605381165919282,
+      "eval_allNLI-dev_euclidean_f1_threshold": 22.352760314941406,
+      "eval_allNLI-dev_euclidean_precision": 0.45787545787545786,
+      "eval_allNLI-dev_euclidean_recall": 0.7225433526011561,
+      "eval_allNLI-dev_manhattan_accuracy": 0.6875,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 392.8375549316406,
+      "eval_allNLI-dev_manhattan_ap": 0.5229585168795372,
+      "eval_allNLI-dev_manhattan_f1": 0.5590909090909091,
+      "eval_allNLI-dev_manhattan_f1_threshold": 690.5735473632812,
+      "eval_allNLI-dev_manhattan_precision": 0.4606741573033708,
+      "eval_allNLI-dev_manhattan_recall": 0.7109826589595376,
+      "eval_allNLI-dev_max_accuracy": 0.6953125,
+      "eval_allNLI-dev_max_accuracy_threshold": 942.6085205078125,
+      "eval_allNLI-dev_max_ap": 0.5229585168795372,
+      "eval_allNLI-dev_max_f1": 0.5605381165919282,
+      "eval_allNLI-dev_max_f1_threshold": 817.1332397460938,
+      "eval_allNLI-dev_max_precision": 0.4793388429752066,
+      "eval_allNLI-dev_max_recall": 0.7283236994219653,
+      "eval_sequential_score": 0.6671756125841228,
+      "eval_sts-test_pearson_cosine": 0.8718645025760672,
+      "eval_sts-test_pearson_dot": 0.8442027854618407,
+      "eval_sts-test_pearson_euclidean": 0.8902978035400368,
+      "eval_sts-test_pearson_manhattan": 0.8909390707647828,
+      "eval_sts-test_pearson_max": 0.8909390707647828,
+      "eval_sts-test_spearman_cosine": 0.8775905391512535,
+      "eval_sts-test_spearman_dot": 0.8282565695883292,
+      "eval_sts-test_spearman_euclidean": 0.8785767674497972,
+      "eval_sts-test_spearman_manhattan": 0.8796654313856657,
+      "eval_sts-test_spearman_max": 0.8796654313856657,
+      "eval_vitaminc-pairs_loss": 3.82141375541687,
+      "eval_vitaminc-pairs_runtime": 4.6321,
+      "eval_vitaminc-pairs_samples_per_second": 27.633,
+      "eval_vitaminc-pairs_steps_per_second": 0.216,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_negation-triplets_loss": 1.2526816129684448,
+      "eval_negation-triplets_runtime": 3.4008,
+      "eval_negation-triplets_samples_per_second": 37.638,
+      "eval_negation-triplets_steps_per_second": 0.294,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_scitail-pairs-pos_loss": 0.07802588492631912,
+      "eval_scitail-pairs-pos_runtime": 2.7191,
+      "eval_scitail-pairs-pos_samples_per_second": 47.075,
+      "eval_scitail-pairs-pos_steps_per_second": 0.368,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_scitail-pairs-qa_loss": 0.11051780730485916,
+      "eval_scitail-pairs-qa_runtime": 2.1249,
+      "eval_scitail-pairs-qa_samples_per_second": 60.237,
+      "eval_scitail-pairs-qa_steps_per_second": 0.471,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_xsum-pairs_loss": 0.1444225311279297,
+      "eval_xsum-pairs_runtime": 3.075,
+      "eval_xsum-pairs_samples_per_second": 41.626,
+      "eval_xsum-pairs_steps_per_second": 0.325,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_sciq_pairs_loss": 0.10900198668241501,
+      "eval_sciq_pairs_runtime": 4.6201,
+      "eval_sciq_pairs_samples_per_second": 27.705,
+      "eval_sciq_pairs_steps_per_second": 0.216,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_qasc_pairs_loss": 0.8441975116729736,
+      "eval_qasc_pairs_runtime": 2.0926,
+      "eval_qasc_pairs_samples_per_second": 61.168,
+      "eval_qasc_pairs_steps_per_second": 0.478,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_openbookqa_pairs_loss": 0.7852775454521179,
+      "eval_openbookqa_pairs_runtime": 2.2378,
+      "eval_openbookqa_pairs_samples_per_second": 57.198,
+      "eval_openbookqa_pairs_steps_per_second": 0.447,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_msmarco_pairs_loss": 0.7973570823669434,
+      "eval_msmarco_pairs_runtime": 2.2477,
+      "eval_msmarco_pairs_samples_per_second": 56.948,
+      "eval_msmarco_pairs_steps_per_second": 0.445,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_nq_pairs_loss": 1.149829387664795,
+      "eval_nq_pairs_runtime": 3.1982,
+      "eval_nq_pairs_samples_per_second": 40.023,
+      "eval_nq_pairs_steps_per_second": 0.313,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_trivia_pairs_loss": 0.46642398834228516,
+      "eval_trivia_pairs_runtime": 3.2778,
+      "eval_trivia_pairs_samples_per_second": 39.05,
+      "eval_trivia_pairs_steps_per_second": 0.305,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_gooaq_pairs_loss": 0.7757962346076965,
+      "eval_gooaq_pairs_runtime": 2.1385,
+      "eval_gooaq_pairs_samples_per_second": 59.856,
+      "eval_gooaq_pairs_steps_per_second": 0.468,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_paws-pos_loss": 0.03750575706362724,
+      "eval_paws-pos_runtime": 2.2521,
+      "eval_paws-pos_samples_per_second": 56.836,
+      "eval_paws-pos_steps_per_second": 0.444,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_global_dataset_loss": 0.48674729466438293,
+      "eval_global_dataset_runtime": 8.6002,
+      "eval_global_dataset_samples_per_second": 37.79,
+      "eval_global_dataset_steps_per_second": 0.349,
+      "step": 157
+    },
+    {
+      "epoch": 0.10230179028132992,
+      "grad_norm": 45.195247650146484,
+      "learning_rate": 3.2161874334398298e-06,
+      "loss": 0.54,
+      "step": 160
+    },
+    {
+      "epoch": 0.11253196930946291,
+      "grad_norm": 64.59953308105469,
+      "learning_rate": 3.556975505857295e-06,
+      "loss": 0.6267,
+      "step": 176
+    },
+    {
+      "epoch": 0.12276214833759591,
+      "grad_norm": 108.30097961425781,
+      "learning_rate": 3.8977635782747605e-06,
+      "loss": 0.401,
+      "step": 192
+    },
+    {
+      "epoch": 0.1329923273657289,
+      "grad_norm": 10.606786727905273,
+      "learning_rate": 4.238551650692226e-06,
+      "loss": 0.455,
+      "step": 208
+    },
+    {
+      "epoch": 0.1432225063938619,
+      "grad_norm": 10.502503395080566,
+      "learning_rate": 4.579339723109691e-06,
+      "loss": 0.308,
+      "step": 224
+    },
+    {
+      "epoch": 0.1534526854219949,
+      "grad_norm": 10.550800323486328,
+      "learning_rate": 4.920127795527157e-06,
+      "loss": 0.2808,
+      "step": 240
+    },
+    {
+      "epoch": 0.1636828644501279,
+      "grad_norm": 0.047671813517808914,
+      "learning_rate": 5.260915867944623e-06,
+      "loss": 0.319,
+      "step": 256
+    },
+    {
+      "epoch": 0.17391304347826086,
+      "grad_norm": 4.752868175506592,
+      "learning_rate": 5.6017039403620875e-06,
+      "loss": 0.3241,
+      "step": 272
+    },
+    {
+      "epoch": 0.18414322250639387,
+      "grad_norm": 3.744504690170288,
+      "learning_rate": 5.942492012779553e-06,
+      "loss": 0.2181,
+      "step": 288
+    },
+    {
+      "epoch": 0.19437340153452684,
+      "grad_norm": 0.10304329544305801,
+      "learning_rate": 6.283280085197019e-06,
+      "loss": 0.3247,
+      "step": 304
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_Qnli-dev_cosine_accuracy": 0.68359375,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.728395938873291,
+      "eval_Qnli-dev_cosine_ap": 0.7208629559865216,
+      "eval_Qnli-dev_cosine_f1": 0.6887661141804787,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.671984076499939,
+      "eval_Qnli-dev_cosine_precision": 0.6091205211726385,
+      "eval_Qnli-dev_cosine_recall": 0.7923728813559322,
+      "eval_Qnli-dev_dot_accuracy": 0.681640625,
+      "eval_Qnli-dev_dot_accuracy_threshold": 713.8800048828125,
+      "eval_Qnli-dev_dot_ap": 0.6851873907589782,
+      "eval_Qnli-dev_dot_f1": 0.6788685524126457,
+      "eval_Qnli-dev_dot_f1_threshold": 571.084228515625,
+      "eval_Qnli-dev_dot_precision": 0.5589041095890411,
+      "eval_Qnli-dev_dot_recall": 0.864406779661017,
+      "eval_Qnli-dev_euclidean_accuracy": 0.689453125,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 23.01095962524414,
+      "eval_Qnli-dev_euclidean_ap": 0.7234217762808675,
+      "eval_Qnli-dev_euclidean_f1": 0.6831858407079647,
+      "eval_Qnli-dev_euclidean_f1_threshold": 25.087703704833984,
+      "eval_Qnli-dev_euclidean_precision": 0.5866261398176292,
+      "eval_Qnli-dev_euclidean_recall": 0.8177966101694916,
+      "eval_Qnli-dev_manhattan_accuracy": 0.6875,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 715.2478637695312,
+      "eval_Qnli-dev_manhattan_ap": 0.7243486016703601,
+      "eval_Qnli-dev_manhattan_f1": 0.6844919786096257,
+      "eval_Qnli-dev_manhattan_f1_threshold": 780.095947265625,
+      "eval_Qnli-dev_manhattan_precision": 0.5907692307692308,
+      "eval_Qnli-dev_manhattan_recall": 0.8135593220338984,
+      "eval_Qnli-dev_max_accuracy": 0.689453125,
+      "eval_Qnli-dev_max_accuracy_threshold": 715.2478637695312,
+      "eval_Qnli-dev_max_ap": 0.7243486016703601,
+      "eval_Qnli-dev_max_f1": 0.6887661141804787,
+      "eval_Qnli-dev_max_f1_threshold": 780.095947265625,
+      "eval_Qnli-dev_max_precision": 0.6091205211726385,
+      "eval_Qnli-dev_max_recall": 0.864406779661017,
+      "eval_allNLI-dev_cosine_accuracy": 0.720703125,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8353008031845093,
+      "eval_allNLI-dev_cosine_ap": 0.5917629293541244,
+      "eval_allNLI-dev_cosine_f1": 0.6000000000000001,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.7087839245796204,
+      "eval_allNLI-dev_cosine_precision": 0.5019455252918288,
+      "eval_allNLI-dev_cosine_recall": 0.7456647398843931,
+      "eval_allNLI-dev_dot_accuracy": 0.71484375,
+      "eval_allNLI-dev_dot_accuracy_threshold": 817.4105224609375,
+      "eval_allNLI-dev_dot_ap": 0.5858245181656638,
+      "eval_allNLI-dev_dot_f1": 0.5980392156862746,
+      "eval_allNLI-dev_dot_f1_threshold": 674.387939453125,
+      "eval_allNLI-dev_dot_precision": 0.5191489361702127,
+      "eval_allNLI-dev_dot_recall": 0.7052023121387283,
+      "eval_allNLI-dev_euclidean_accuracy": 0.716796875,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 17.109304428100586,
+      "eval_allNLI-dev_euclidean_ap": 0.5912447191211185,
+      "eval_allNLI-dev_euclidean_f1": 0.6013986013986015,
+      "eval_allNLI-dev_euclidean_f1_threshold": 23.201446533203125,
+      "eval_allNLI-dev_euclidean_precision": 0.50390625,
+      "eval_allNLI-dev_euclidean_recall": 0.7456647398843931,
+      "eval_allNLI-dev_manhattan_accuracy": 0.71875,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 552.2703857421875,
+      "eval_allNLI-dev_manhattan_ap": 0.5910112781514775,
+      "eval_allNLI-dev_manhattan_f1": 0.5981735159817351,
+      "eval_allNLI-dev_manhattan_f1_threshold": 731.6765747070312,
+      "eval_allNLI-dev_manhattan_precision": 0.49433962264150944,
+      "eval_allNLI-dev_manhattan_recall": 0.7572254335260116,
+      "eval_allNLI-dev_max_accuracy": 0.720703125,
+      "eval_allNLI-dev_max_accuracy_threshold": 817.4105224609375,
+      "eval_allNLI-dev_max_ap": 0.5917629293541244,
+      "eval_allNLI-dev_max_f1": 0.6013986013986015,
+      "eval_allNLI-dev_max_f1_threshold": 731.6765747070312,
+      "eval_allNLI-dev_max_precision": 0.5191489361702127,
+      "eval_allNLI-dev_max_recall": 0.7572254335260116,
+      "eval_sequential_score": 0.7243486016703601,
+      "eval_sts-test_pearson_cosine": 0.91107907465776,
+      "eval_sts-test_pearson_dot": 0.8802333333831622,
+      "eval_sts-test_pearson_euclidean": 0.92175721802385,
+      "eval_sts-test_pearson_manhattan": 0.9207411620779591,
+      "eval_sts-test_pearson_max": 0.92175721802385,
+      "eval_sts-test_spearman_cosine": 0.9170956996722621,
+      "eval_sts-test_spearman_dot": 0.8733774316475602,
+      "eval_sts-test_spearman_euclidean": 0.9165704988685727,
+      "eval_sts-test_spearman_manhattan": 0.9153493264477239,
+      "eval_sts-test_spearman_max": 0.9170956996722621,
+      "eval_vitaminc-pairs_loss": 3.3627231121063232,
+      "eval_vitaminc-pairs_runtime": 4.4746,
+      "eval_vitaminc-pairs_samples_per_second": 28.606,
+      "eval_vitaminc-pairs_steps_per_second": 0.223,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_negation-triplets_loss": 0.7009554505348206,
+      "eval_negation-triplets_runtime": 3.3424,
+      "eval_negation-triplets_samples_per_second": 38.296,
+      "eval_negation-triplets_steps_per_second": 0.299,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_scitail-pairs-pos_loss": 0.027083460241556168,
+      "eval_scitail-pairs-pos_runtime": 2.6253,
+      "eval_scitail-pairs-pos_samples_per_second": 48.757,
+      "eval_scitail-pairs-pos_steps_per_second": 0.381,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_scitail-pairs-qa_loss": 0.0013978304341435432,
+      "eval_scitail-pairs-qa_runtime": 2.0929,
+      "eval_scitail-pairs-qa_samples_per_second": 61.159,
+      "eval_scitail-pairs-qa_steps_per_second": 0.478,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_xsum-pairs_loss": 0.03819597512483597,
+      "eval_xsum-pairs_runtime": 3.0764,
+      "eval_xsum-pairs_samples_per_second": 41.607,
+      "eval_xsum-pairs_steps_per_second": 0.325,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_sciq_pairs_loss": 0.04831065982580185,
+      "eval_sciq_pairs_runtime": 4.6239,
+      "eval_sciq_pairs_samples_per_second": 27.682,
+      "eval_sciq_pairs_steps_per_second": 0.216,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_qasc_pairs_loss": 0.11294831335544586,
+      "eval_qasc_pairs_runtime": 2.1013,
+      "eval_qasc_pairs_samples_per_second": 60.915,
+      "eval_qasc_pairs_steps_per_second": 0.476,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_openbookqa_pairs_loss": 0.4045051634311676,
+      "eval_openbookqa_pairs_runtime": 2.2313,
+      "eval_openbookqa_pairs_samples_per_second": 57.367,
+      "eval_openbookqa_pairs_steps_per_second": 0.448,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_msmarco_pairs_loss": 0.17428961396217346,
+      "eval_msmarco_pairs_runtime": 2.2379,
+      "eval_msmarco_pairs_samples_per_second": 57.197,
+      "eval_msmarco_pairs_steps_per_second": 0.447,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_nq_pairs_loss": 0.18139179050922394,
+      "eval_nq_pairs_runtime": 3.1825,
+      "eval_nq_pairs_samples_per_second": 40.22,
+      "eval_nq_pairs_steps_per_second": 0.314,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_trivia_pairs_loss": 0.10668718814849854,
+      "eval_trivia_pairs_runtime": 3.2891,
+      "eval_trivia_pairs_samples_per_second": 38.917,
+      "eval_trivia_pairs_steps_per_second": 0.304,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_gooaq_pairs_loss": 0.2038198560476303,
+      "eval_gooaq_pairs_runtime": 2.209,
+      "eval_gooaq_pairs_samples_per_second": 57.945,
+      "eval_gooaq_pairs_steps_per_second": 0.453,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_paws-pos_loss": 0.02337946929037571,
+      "eval_paws-pos_runtime": 2.289,
+      "eval_paws-pos_samples_per_second": 55.92,
+      "eval_paws-pos_steps_per_second": 0.437,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_global_dataset_loss": 0.23699811100959778,
+      "eval_global_dataset_runtime": 8.6273,
+      "eval_global_dataset_samples_per_second": 37.671,
+      "eval_global_dataset_steps_per_second": 0.348,
+      "step": 314
+    },
+    {
+      "epoch": 0.20460358056265984,
+      "grad_norm": 7.919135570526123,
+      "learning_rate": 6.624068157614484e-06,
+      "loss": 0.233,
+      "step": 320
+    },
+    {
+      "epoch": 0.21483375959079284,
+      "grad_norm": 8.169248580932617,
+      "learning_rate": 6.96485623003195e-06,
+      "loss": 0.1946,
+      "step": 336
+    },
+    {
+      "epoch": 0.22506393861892582,
+      "grad_norm": 4.551661968231201,
+      "learning_rate": 7.305644302449415e-06,
+      "loss": 0.2393,
+      "step": 352
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 49.21135330200195,
+      "learning_rate": 7.64643237486688e-06,
+      "loss": 0.1476,
+      "step": 368
+    },
+    {
+      "epoch": 0.24552429667519182,
+      "grad_norm": 12.119938850402832,
+      "learning_rate": 7.987220447284347e-06,
+      "loss": 0.1976,
+      "step": 384
+    },
+    {
+      "epoch": 0.2557544757033248,
+      "grad_norm": 6.922146797180176,
+      "learning_rate": 8.32800851970181e-06,
+      "loss": 0.1139,
+      "step": 400
+    },
+    {
+      "epoch": 0.2659846547314578,
+      "grad_norm": 1.659225583076477,
+      "learning_rate": 8.668796592119276e-06,
+      "loss": 0.1986,
+      "step": 416
+    },
+    {
+      "epoch": 0.27621483375959077,
+      "grad_norm": 11.489442825317383,
+      "learning_rate": 9.009584664536743e-06,
+      "loss": 0.2405,
+      "step": 432
+    },
+    {
+      "epoch": 0.2864450127877238,
+      "grad_norm": 6.8070149421691895,
+      "learning_rate": 9.350372736954207e-06,
+      "loss": 0.1519,
+      "step": 448
+    },
+    {
+      "epoch": 0.2966751918158568,
+      "grad_norm": 8.569122314453125,
+      "learning_rate": 9.691160809371672e-06,
+      "loss": 0.1508,
+      "step": 464
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_Qnli-dev_cosine_accuracy": 0.689453125,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.7045497894287109,
+      "eval_Qnli-dev_cosine_ap": 0.7262372250492908,
+      "eval_Qnli-dev_cosine_f1": 0.693103448275862,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.6277430057525635,
+      "eval_Qnli-dev_cosine_precision": 0.5843023255813954,
+      "eval_Qnli-dev_cosine_recall": 0.8516949152542372,
+      "eval_Qnli-dev_dot_accuracy": 0.671875,
+      "eval_Qnli-dev_dot_accuracy_threshold": 652.669677734375,
+      "eval_Qnli-dev_dot_ap": 0.6928244257428712,
+      "eval_Qnli-dev_dot_f1": 0.6923076923076922,
+      "eval_Qnli-dev_dot_f1_threshold": 571.573486328125,
+      "eval_Qnli-dev_dot_precision": 0.5892857142857143,
+      "eval_Qnli-dev_dot_recall": 0.8389830508474576,
+      "eval_Qnli-dev_euclidean_accuracy": 0.689453125,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 23.253864288330078,
+      "eval_Qnli-dev_euclidean_ap": 0.7293266479878985,
+      "eval_Qnli-dev_euclidean_f1": 0.6928571428571428,
+      "eval_Qnli-dev_euclidean_f1_threshold": 25.396923065185547,
+      "eval_Qnli-dev_euclidean_precision": 0.5987654320987654,
+      "eval_Qnli-dev_euclidean_recall": 0.8220338983050848,
+      "eval_Qnli-dev_manhattan_accuracy": 0.6875,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 704.4737548828125,
+      "eval_Qnli-dev_manhattan_ap": 0.7283792835739776,
+      "eval_Qnli-dev_manhattan_f1": 0.6925675675675677,
+      "eval_Qnli-dev_manhattan_f1_threshold": 820.7425537109375,
+      "eval_Qnli-dev_manhattan_precision": 0.5758426966292135,
+      "eval_Qnli-dev_manhattan_recall": 0.8686440677966102,
+      "eval_Qnli-dev_max_accuracy": 0.689453125,
+      "eval_Qnli-dev_max_accuracy_threshold": 704.4737548828125,
+      "eval_Qnli-dev_max_ap": 0.7293266479878985,
+      "eval_Qnli-dev_max_f1": 0.693103448275862,
+      "eval_Qnli-dev_max_f1_threshold": 820.7425537109375,
+      "eval_Qnli-dev_max_precision": 0.5987654320987654,
+      "eval_Qnli-dev_max_recall": 0.8686440677966102,
+      "eval_allNLI-dev_cosine_accuracy": 0.716796875,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.805169939994812,
+      "eval_allNLI-dev_cosine_ap": 0.5991158471614924,
+      "eval_allNLI-dev_cosine_f1": 0.6095617529880478,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.6292704343795776,
+      "eval_allNLI-dev_cosine_precision": 0.46504559270516715,
+      "eval_allNLI-dev_cosine_recall": 0.884393063583815,
+      "eval_allNLI-dev_dot_accuracy": 0.7265625,
+      "eval_allNLI-dev_dot_accuracy_threshold": 685.67724609375,
+      "eval_allNLI-dev_dot_ap": 0.6022852154957883,
+      "eval_allNLI-dev_dot_f1": 0.603112840466926,
+      "eval_allNLI-dev_dot_f1_threshold": 566.2506103515625,
+      "eval_allNLI-dev_dot_precision": 0.45454545454545453,
+      "eval_allNLI-dev_dot_recall": 0.8959537572254336,
+      "eval_allNLI-dev_euclidean_accuracy": 0.71484375,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 20.33218765258789,
+      "eval_allNLI-dev_euclidean_ap": 0.5957908065503393,
+      "eval_allNLI-dev_euclidean_f1": 0.6140724946695097,
+      "eval_allNLI-dev_euclidean_f1_threshold": 25.308191299438477,
+      "eval_allNLI-dev_euclidean_precision": 0.4864864864864865,
+      "eval_allNLI-dev_euclidean_recall": 0.8323699421965318,
+      "eval_allNLI-dev_manhattan_accuracy": 0.716796875,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 638.1854248046875,
+      "eval_allNLI-dev_manhattan_ap": 0.5968486499581244,
+      "eval_allNLI-dev_manhattan_f1": 0.616052060737527,
+      "eval_allNLI-dev_manhattan_f1_threshold": 783.5068359375,
+      "eval_allNLI-dev_manhattan_precision": 0.4930555555555556,
+      "eval_allNLI-dev_manhattan_recall": 0.8208092485549133,
+      "eval_allNLI-dev_max_accuracy": 0.7265625,
+      "eval_allNLI-dev_max_accuracy_threshold": 685.67724609375,
+      "eval_allNLI-dev_max_ap": 0.6022852154957883,
+      "eval_allNLI-dev_max_f1": 0.616052060737527,
+      "eval_allNLI-dev_max_f1_threshold": 783.5068359375,
+      "eval_allNLI-dev_max_precision": 0.4930555555555556,
+      "eval_allNLI-dev_max_recall": 0.8959537572254336,
+      "eval_sequential_score": 0.7293266479878985,
+      "eval_sts-test_pearson_cosine": 0.9163990296196276,
+      "eval_sts-test_pearson_dot": 0.9067925652682547,
+      "eval_sts-test_pearson_euclidean": 0.9258813754113262,
+      "eval_sts-test_pearson_manhattan": 0.9250575950562571,
+      "eval_sts-test_pearson_max": 0.9258813754113262,
+      "eval_sts-test_spearman_cosine": 0.9200439127727275,
+      "eval_sts-test_spearman_dot": 0.9030931642366772,
+      "eval_sts-test_spearman_euclidean": 0.920441080217576,
+      "eval_sts-test_spearman_manhattan": 0.9196469690841858,
+      "eval_sts-test_spearman_max": 0.920441080217576,
+      "eval_vitaminc-pairs_loss": 2.4166860580444336,
+      "eval_vitaminc-pairs_runtime": 4.558,
+      "eval_vitaminc-pairs_samples_per_second": 28.082,
+      "eval_vitaminc-pairs_steps_per_second": 0.219,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_negation-triplets_loss": 0.6827869415283203,
+      "eval_negation-triplets_runtime": 3.2935,
+      "eval_negation-triplets_samples_per_second": 38.864,
+      "eval_negation-triplets_steps_per_second": 0.304,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_scitail-pairs-pos_loss": 0.0062057748436927795,
+      "eval_scitail-pairs-pos_runtime": 2.638,
+      "eval_scitail-pairs-pos_samples_per_second": 48.521,
+      "eval_scitail-pairs-pos_steps_per_second": 0.379,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_scitail-pairs-qa_loss": 0.0004358906007837504,
+      "eval_scitail-pairs-qa_runtime": 2.0989,
+      "eval_scitail-pairs-qa_samples_per_second": 60.985,
+      "eval_scitail-pairs-qa_steps_per_second": 0.476,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_xsum-pairs_loss": 0.03286559134721756,
+      "eval_xsum-pairs_runtime": 3.0772,
+      "eval_xsum-pairs_samples_per_second": 41.596,
+      "eval_xsum-pairs_steps_per_second": 0.325,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_sciq_pairs_loss": 0.04141838476061821,
+      "eval_sciq_pairs_runtime": 4.69,
+      "eval_sciq_pairs_samples_per_second": 27.292,
+      "eval_sciq_pairs_steps_per_second": 0.213,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_qasc_pairs_loss": 0.03906266763806343,
+      "eval_qasc_pairs_runtime": 2.104,
+      "eval_qasc_pairs_samples_per_second": 60.836,
+      "eval_qasc_pairs_steps_per_second": 0.475,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_openbookqa_pairs_loss": 0.38474294543266296,
+      "eval_openbookqa_pairs_runtime": 2.2386,
+      "eval_openbookqa_pairs_samples_per_second": 57.18,
+      "eval_openbookqa_pairs_steps_per_second": 0.447,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_msmarco_pairs_loss": 0.08647920191287994,
+      "eval_msmarco_pairs_runtime": 2.2858,
+      "eval_msmarco_pairs_samples_per_second": 55.999,
+      "eval_msmarco_pairs_steps_per_second": 0.437,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_nq_pairs_loss": 0.10326902568340302,
+      "eval_nq_pairs_runtime": 3.1905,
+      "eval_nq_pairs_samples_per_second": 40.12,
+      "eval_nq_pairs_steps_per_second": 0.313,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_trivia_pairs_loss": 0.09608602523803711,
+      "eval_trivia_pairs_runtime": 3.256,
+      "eval_trivia_pairs_samples_per_second": 39.312,
+      "eval_trivia_pairs_steps_per_second": 0.307,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_gooaq_pairs_loss": 0.12804344296455383,
+      "eval_gooaq_pairs_runtime": 2.1753,
+      "eval_gooaq_pairs_samples_per_second": 58.841,
+      "eval_gooaq_pairs_steps_per_second": 0.46,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_paws-pos_loss": 0.025196708738803864,
+      "eval_paws-pos_runtime": 2.3068,
+      "eval_paws-pos_samples_per_second": 55.488,
+      "eval_paws-pos_steps_per_second": 0.434,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_global_dataset_loss": 0.17351965606212616,
+      "eval_global_dataset_runtime": 8.6124,
+      "eval_global_dataset_samples_per_second": 37.736,
+      "eval_global_dataset_steps_per_second": 0.348,
+      "step": 471
+    }
+  ],
+  "logging_steps": 16,
+  "max_steps": 3128,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 157,
+  "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": 64,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-471/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae5d47b05b44fe0990966324c4a681d83a347e47ed07c3ca3b6e4479e0e161a9
+size 5560