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 sha256:5015000af1a0d50ae3b2d6ea3e180524a71081a03b911d9e89b6314a18efdb69
+size 7077084396

checkpoint-471/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4568c0e4a883b874a34cc3e2fc23f0e1f1977ddc8415131c390aa2ea2581c7e
+size 3538506546

checkpoint-471/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a14718398d1c1fc7d548cc02e1dcb77914d09f9d022a9105b822a78bfbbee2a2
+size 14244

checkpoint-471/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff41f33b90b042dbcfabbc23038d5f6b557b4133dcaa940601ab0fb15b2f228
+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": 127.07197570800781,
+      "learning_rate": 2.129925452609159e-07,
+      "loss": 7.0465,
+      "step": 16
+    },
+    {
+      "epoch": 0.020460358056265986,
+      "grad_norm": 193.34689331054688,
+      "learning_rate": 5.537806176783813e-07,
+      "loss": 9.0583,
+      "step": 32
+    },
+    {
+      "epoch": 0.030690537084398978,
+      "grad_norm": 183.065185546875,
+      "learning_rate": 8.732694355697552e-07,
+      "loss": 8.6271,
+      "step": 48
+    },
+    {
+      "epoch": 0.04092071611253197,
+      "grad_norm": 150.517333984375,
+      "learning_rate": 1.2140575079872206e-06,
+      "loss": 5.6026,
+      "step": 64
+    },
+    {
+      "epoch": 0.05115089514066496,
+      "grad_norm": 78.38551330566406,
+      "learning_rate": 1.5335463258785943e-06,
+      "loss": 4.6908,
+      "step": 80
+    },
+    {
+      "epoch": 0.061381074168797956,
+      "grad_norm": 106.0997314453125,
+      "learning_rate": 1.8743343982960599e-06,
+      "loss": 3.1954,
+      "step": 96
+    },
+    {
+      "epoch": 0.07161125319693094,
+      "grad_norm": 47.50294876098633,
+      "learning_rate": 2.215122470713525e-06,
+      "loss": 2.1179,
+      "step": 112
+    },
+    {
+      "epoch": 0.08184143222506395,
+      "grad_norm": 24.581972122192383,
+      "learning_rate": 2.5559105431309904e-06,
+      "loss": 1.2636,
+      "step": 128
+    },
+    {
+      "epoch": 0.09207161125319693,
+      "grad_norm": 219.01101684570312,
+      "learning_rate": 2.8966986155484563e-06,
+      "loss": 0.8325,
+      "step": 144
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_Qnli-dev_cosine_accuracy": 0.65234375,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8284608125686646,
+      "eval_Qnli-dev_cosine_ap": 0.6663718886737695,
+      "eval_Qnli-dev_cosine_f1": 0.6469673405909797,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.702437698841095,
+      "eval_Qnli-dev_cosine_precision": 0.5110565110565111,
+      "eval_Qnli-dev_cosine_recall": 0.8813559322033898,
+      "eval_Qnli-dev_dot_accuracy": 0.654296875,
+      "eval_Qnli-dev_dot_accuracy_threshold": 899.9195556640625,
+      "eval_Qnli-dev_dot_ap": 0.6265727343636489,
+      "eval_Qnli-dev_dot_f1": 0.6462480857580397,
+      "eval_Qnli-dev_dot_f1_threshold": 708.03076171875,
+      "eval_Qnli-dev_dot_precision": 0.5059952038369304,
+      "eval_Qnli-dev_dot_recall": 0.8940677966101694,
+      "eval_Qnli-dev_euclidean_accuracy": 0.654296875,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 17.465450286865234,
+      "eval_Qnli-dev_euclidean_ap": 0.6678841105609593,
+      "eval_Qnli-dev_euclidean_f1": 0.6485568760611206,
+      "eval_Qnli-dev_euclidean_f1_threshold": 22.50777244567871,
+      "eval_Qnli-dev_euclidean_precision": 0.5410764872521246,
+      "eval_Qnli-dev_euclidean_recall": 0.809322033898305,
+      "eval_Qnli-dev_manhattan_accuracy": 0.654296875,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 541.6007080078125,
+      "eval_Qnli-dev_manhattan_ap": 0.6671352135215549,
+      "eval_Qnli-dev_manhattan_f1": 0.6472602739726028,
+      "eval_Qnli-dev_manhattan_f1_threshold": 693.766845703125,
+      "eval_Qnli-dev_manhattan_precision": 0.5431034482758621,
+      "eval_Qnli-dev_manhattan_recall": 0.8008474576271186,
+      "eval_Qnli-dev_max_accuracy": 0.654296875,
+      "eval_Qnli-dev_max_accuracy_threshold": 899.9195556640625,
+      "eval_Qnli-dev_max_ap": 0.6678841105609593,
+      "eval_Qnli-dev_max_f1": 0.6485568760611206,
+      "eval_Qnli-dev_max_f1_threshold": 708.03076171875,
+      "eval_Qnli-dev_max_precision": 0.5431034482758621,
+      "eval_Qnli-dev_max_recall": 0.8940677966101694,
+      "eval_allNLI-dev_cosine_accuracy": 0.69140625,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8886671662330627,
+      "eval_allNLI-dev_cosine_ap": 0.5184375875165067,
+      "eval_allNLI-dev_cosine_f1": 0.5584415584415584,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.7475587725639343,
+      "eval_allNLI-dev_cosine_precision": 0.4463667820069204,
+      "eval_allNLI-dev_cosine_recall": 0.7456647398843931,
+      "eval_allNLI-dev_dot_accuracy": 0.689453125,
+      "eval_allNLI-dev_dot_accuracy_threshold": 935.1204223632812,
+      "eval_allNLI-dev_dot_ap": 0.4904119502439996,
+      "eval_allNLI-dev_dot_f1": 0.5509259259259259,
+      "eval_allNLI-dev_dot_f1_threshold": 803.4815673828125,
+      "eval_allNLI-dev_dot_precision": 0.4594594594594595,
+      "eval_allNLI-dev_dot_recall": 0.6878612716763006,
+      "eval_allNLI-dev_euclidean_accuracy": 0.6953125,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 15.452061653137207,
+      "eval_allNLI-dev_euclidean_ap": 0.5162188247814465,
+      "eval_allNLI-dev_euclidean_f1": 0.5557986870897156,
+      "eval_allNLI-dev_euclidean_f1_threshold": 22.695791244506836,
+      "eval_allNLI-dev_euclidean_precision": 0.4471830985915493,
+      "eval_allNLI-dev_euclidean_recall": 0.7341040462427746,
+      "eval_allNLI-dev_manhattan_accuracy": 0.693359375,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 484.3291015625,
+      "eval_allNLI-dev_manhattan_ap": 0.5171097761848361,
+      "eval_allNLI-dev_manhattan_f1": 0.5550660792951542,
+      "eval_allNLI-dev_manhattan_f1_threshold": 707.2572631835938,
+      "eval_allNLI-dev_manhattan_precision": 0.4483985765124555,
+      "eval_allNLI-dev_manhattan_recall": 0.7283236994219653,
+      "eval_allNLI-dev_max_accuracy": 0.6953125,
+      "eval_allNLI-dev_max_accuracy_threshold": 935.1204223632812,
+      "eval_allNLI-dev_max_ap": 0.5184375875165067,
+      "eval_allNLI-dev_max_f1": 0.5584415584415584,
+      "eval_allNLI-dev_max_f1_threshold": 803.4815673828125,
+      "eval_allNLI-dev_max_precision": 0.4594594594594595,
+      "eval_allNLI-dev_max_recall": 0.7456647398843931,
+      "eval_sequential_score": 0.6678841105609593,
+      "eval_sts-test_pearson_cosine": 0.8691399425075377,
+      "eval_sts-test_pearson_dot": 0.84633945645717,
+      "eval_sts-test_pearson_euclidean": 0.8863973274037125,
+      "eval_sts-test_pearson_manhattan": 0.887205994467606,
+      "eval_sts-test_pearson_max": 0.887205994467606,
+      "eval_sts-test_spearman_cosine": 0.8753474716765325,
+      "eval_sts-test_spearman_dot": 0.8345210301629428,
+      "eval_sts-test_spearman_euclidean": 0.8755160049269324,
+      "eval_sts-test_spearman_manhattan": 0.8763851191744148,
+      "eval_sts-test_spearman_max": 0.8763851191744148,
+      "eval_vitaminc-pairs_loss": 3.764756917953491,
+      "eval_vitaminc-pairs_runtime": 4.7024,
+      "eval_vitaminc-pairs_samples_per_second": 27.22,
+      "eval_vitaminc-pairs_steps_per_second": 0.213,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_negation-triplets_loss": 1.4374932050704956,
+      "eval_negation-triplets_runtime": 3.2218,
+      "eval_negation-triplets_samples_per_second": 39.73,
+      "eval_negation-triplets_steps_per_second": 0.31,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_scitail-pairs-pos_loss": 0.07158112525939941,
+      "eval_scitail-pairs-pos_runtime": 2.7614,
+      "eval_scitail-pairs-pos_samples_per_second": 46.353,
+      "eval_scitail-pairs-pos_steps_per_second": 0.362,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_scitail-pairs-qa_loss": 0.11121156811714172,
+      "eval_scitail-pairs-qa_runtime": 2.2983,
+      "eval_scitail-pairs-qa_samples_per_second": 55.694,
+      "eval_scitail-pairs-qa_steps_per_second": 0.435,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_xsum-pairs_loss": 0.28410428762435913,
+      "eval_xsum-pairs_runtime": 3.1437,
+      "eval_xsum-pairs_samples_per_second": 40.717,
+      "eval_xsum-pairs_steps_per_second": 0.318,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_sciq_pairs_loss": 0.15557295083999634,
+      "eval_sciq_pairs_runtime": 4.4086,
+      "eval_sciq_pairs_samples_per_second": 29.034,
+      "eval_sciq_pairs_steps_per_second": 0.227,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_qasc_pairs_loss": 0.9191333651542664,
+      "eval_qasc_pairs_runtime": 2.1323,
+      "eval_qasc_pairs_samples_per_second": 60.03,
+      "eval_qasc_pairs_steps_per_second": 0.469,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_openbookqa_pairs_loss": 0.8187942504882812,
+      "eval_openbookqa_pairs_runtime": 2.2771,
+      "eval_openbookqa_pairs_samples_per_second": 56.213,
+      "eval_openbookqa_pairs_steps_per_second": 0.439,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_msmarco_pairs_loss": 0.6015267372131348,
+      "eval_msmarco_pairs_runtime": 2.2879,
+      "eval_msmarco_pairs_samples_per_second": 55.946,
+      "eval_msmarco_pairs_steps_per_second": 0.437,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_nq_pairs_loss": 0.9822638630867004,
+      "eval_nq_pairs_runtime": 3.3732,
+      "eval_nq_pairs_samples_per_second": 37.946,
+      "eval_nq_pairs_steps_per_second": 0.296,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_trivia_pairs_loss": 0.3045535683631897,
+      "eval_trivia_pairs_runtime": 3.3462,
+      "eval_trivia_pairs_samples_per_second": 38.253,
+      "eval_trivia_pairs_steps_per_second": 0.299,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_gooaq_pairs_loss": 0.4534245729446411,
+      "eval_gooaq_pairs_runtime": 2.1775,
+      "eval_gooaq_pairs_samples_per_second": 58.783,
+      "eval_gooaq_pairs_steps_per_second": 0.459,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_paws-pos_loss": 0.03816518932580948,
+      "eval_paws-pos_runtime": 2.3035,
+      "eval_paws-pos_samples_per_second": 55.567,
+      "eval_paws-pos_steps_per_second": 0.434,
+      "step": 157
+    },
+    {
+      "epoch": 0.100383631713555,
+      "eval_global_dataset_loss": 0.4413764476776123,
+      "eval_global_dataset_runtime": 8.8968,
+      "eval_global_dataset_samples_per_second": 36.53,
+      "eval_global_dataset_steps_per_second": 0.337,
+      "step": 157
+    },
+    {
+      "epoch": 0.10230179028132992,
+      "grad_norm": 4.65382719039917,
+      "learning_rate": 3.2374866879659217e-06,
+      "loss": 0.5674,
+      "step": 160
+    },
+    {
+      "epoch": 0.11253196930946291,
+      "grad_norm": 16.067514419555664,
+      "learning_rate": 3.5782747603833866e-06,
+      "loss": 0.6077,
+      "step": 176
+    },
+    {
+      "epoch": 0.12276214833759591,
+      "grad_norm": 13.119791984558105,
+      "learning_rate": 3.919062832800852e-06,
+      "loss": 0.4102,
+      "step": 192
+    },
+    {
+      "epoch": 0.1329923273657289,
+      "grad_norm": 27.51828384399414,
+      "learning_rate": 4.259850905218318e-06,
+      "loss": 0.4442,
+      "step": 208
+    },
+    {
+      "epoch": 0.1432225063938619,
+      "grad_norm": 76.78152465820312,
+      "learning_rate": 4.600638977635783e-06,
+      "loss": 0.3306,
+      "step": 224
+    },
+    {
+      "epoch": 0.1534526854219949,
+      "grad_norm": 9.617141723632812,
+      "learning_rate": 4.941427050053248e-06,
+      "loss": 0.3002,
+      "step": 240
+    },
+    {
+      "epoch": 0.1636828644501279,
+      "grad_norm": 0.007532110437750816,
+      "learning_rate": 5.2822151224707144e-06,
+      "loss": 0.2485,
+      "step": 256
+    },
+    {
+      "epoch": 0.17391304347826086,
+      "grad_norm": 9.366889953613281,
+      "learning_rate": 5.623003194888179e-06,
+      "loss": 0.274,
+      "step": 272
+    },
+    {
+      "epoch": 0.18414322250639387,
+      "grad_norm": 2.958397626876831,
+      "learning_rate": 5.963791267305645e-06,
+      "loss": 0.2093,
+      "step": 288
+    },
+    {
+      "epoch": 0.19437340153452684,
+      "grad_norm": 0.2244662344455719,
+      "learning_rate": 6.304579339723111e-06,
+      "loss": 0.3521,
+      "step": 304
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_Qnli-dev_cosine_accuracy": 0.6796875,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.750099241733551,
+      "eval_Qnli-dev_cosine_ap": 0.7156359777017246,
+      "eval_Qnli-dev_cosine_f1": 0.6803418803418803,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.6485489010810852,
+      "eval_Qnli-dev_cosine_precision": 0.5702005730659025,
+      "eval_Qnli-dev_cosine_recall": 0.8432203389830508,
+      "eval_Qnli-dev_dot_accuracy": 0.681640625,
+      "eval_Qnli-dev_dot_accuracy_threshold": 724.4273681640625,
+      "eval_Qnli-dev_dot_ap": 0.7008881822413993,
+      "eval_Qnli-dev_dot_f1": 0.6768189509306259,
+      "eval_Qnli-dev_dot_f1_threshold": 601.5821533203125,
+      "eval_Qnli-dev_dot_precision": 0.5633802816901409,
+      "eval_Qnli-dev_dot_recall": 0.847457627118644,
+      "eval_Qnli-dev_euclidean_accuracy": 0.6796875,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 21.8489990234375,
+      "eval_Qnli-dev_euclidean_ap": 0.7160572797142014,
+      "eval_Qnli-dev_euclidean_f1": 0.6843033509700176,
+      "eval_Qnli-dev_euclidean_f1_threshold": 25.212352752685547,
+      "eval_Qnli-dev_euclidean_precision": 0.5861027190332326,
+      "eval_Qnli-dev_euclidean_recall": 0.8220338983050848,
+      "eval_Qnli-dev_manhattan_accuracy": 0.6796875,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 667.1034545898438,
+      "eval_Qnli-dev_manhattan_ap": 0.7161884702166013,
+      "eval_Qnli-dev_manhattan_f1": 0.6877192982456141,
+      "eval_Qnli-dev_manhattan_f1_threshold": 791.084716796875,
+      "eval_Qnli-dev_manhattan_precision": 0.5868263473053892,
+      "eval_Qnli-dev_manhattan_recall": 0.8305084745762712,
+      "eval_Qnli-dev_max_accuracy": 0.681640625,
+      "eval_Qnli-dev_max_accuracy_threshold": 724.4273681640625,
+      "eval_Qnli-dev_max_ap": 0.7161884702166013,
+      "eval_Qnli-dev_max_f1": 0.6877192982456141,
+      "eval_Qnli-dev_max_f1_threshold": 791.084716796875,
+      "eval_Qnli-dev_max_precision": 0.5868263473053892,
+      "eval_Qnli-dev_max_recall": 0.847457627118644,
+      "eval_allNLI-dev_cosine_accuracy": 0.71875,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.834626317024231,
+      "eval_allNLI-dev_cosine_ap": 0.5835584554560046,
+      "eval_allNLI-dev_cosine_f1": 0.6094808126410836,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.6867948174476624,
+      "eval_allNLI-dev_cosine_precision": 0.5,
+      "eval_allNLI-dev_cosine_recall": 0.7803468208092486,
+      "eval_allNLI-dev_dot_accuracy": 0.716796875,
+      "eval_allNLI-dev_dot_accuracy_threshold": 792.005615234375,
+      "eval_allNLI-dev_dot_ap": 0.5699574122566177,
+      "eval_allNLI-dev_dot_f1": 0.6130434782608695,
+      "eval_allNLI-dev_dot_f1_threshold": 633.9556274414062,
+      "eval_allNLI-dev_dot_precision": 0.4912891986062718,
+      "eval_allNLI-dev_dot_recall": 0.815028901734104,
+      "eval_allNLI-dev_euclidean_accuracy": 0.716796875,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 17.933330535888672,
+      "eval_allNLI-dev_euclidean_ap": 0.583705987979273,
+      "eval_allNLI-dev_euclidean_f1": 0.602510460251046,
+      "eval_allNLI-dev_euclidean_f1_threshold": 25.284334182739258,
+      "eval_allNLI-dev_euclidean_precision": 0.4721311475409836,
+      "eval_allNLI-dev_euclidean_recall": 0.8323699421965318,
+      "eval_allNLI-dev_manhattan_accuracy": 0.720703125,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 559.5651245117188,
+      "eval_allNLI-dev_manhattan_ap": 0.5843640673658059,
+      "eval_allNLI-dev_manhattan_f1": 0.6072186836518046,
+      "eval_allNLI-dev_manhattan_f1_threshold": 784.5065307617188,
+      "eval_allNLI-dev_manhattan_precision": 0.4798657718120805,
+      "eval_allNLI-dev_manhattan_recall": 0.8265895953757225,
+      "eval_allNLI-dev_max_accuracy": 0.720703125,
+      "eval_allNLI-dev_max_accuracy_threshold": 792.005615234375,
+      "eval_allNLI-dev_max_ap": 0.5843640673658059,
+      "eval_allNLI-dev_max_f1": 0.6130434782608695,
+      "eval_allNLI-dev_max_f1_threshold": 784.5065307617188,
+      "eval_allNLI-dev_max_precision": 0.5,
+      "eval_allNLI-dev_max_recall": 0.8323699421965318,
+      "eval_sequential_score": 0.7161884702166013,
+      "eval_sts-test_pearson_cosine": 0.9083914036604525,
+      "eval_sts-test_pearson_dot": 0.8981723163151856,
+      "eval_sts-test_pearson_euclidean": 0.9173432574357514,
+      "eval_sts-test_pearson_manhattan": 0.9168751297887556,
+      "eval_sts-test_pearson_max": 0.9173432574357514,
+      "eval_sts-test_spearman_cosine": 0.9121196279131069,
+      "eval_sts-test_spearman_dot": 0.894677286770672,
+      "eval_sts-test_spearman_euclidean": 0.9124939274633844,
+      "eval_sts-test_spearman_manhattan": 0.9125480317384066,
+      "eval_sts-test_spearman_max": 0.9125480317384066,
+      "eval_vitaminc-pairs_loss": 3.233806610107422,
+      "eval_vitaminc-pairs_runtime": 4.507,
+      "eval_vitaminc-pairs_samples_per_second": 28.401,
+      "eval_vitaminc-pairs_steps_per_second": 0.222,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_negation-triplets_loss": 0.7851067781448364,
+      "eval_negation-triplets_runtime": 3.158,
+      "eval_negation-triplets_samples_per_second": 40.532,
+      "eval_negation-triplets_steps_per_second": 0.317,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_scitail-pairs-pos_loss": 0.021762849763035774,
+      "eval_scitail-pairs-pos_runtime": 2.6751,
+      "eval_scitail-pairs-pos_samples_per_second": 47.849,
+      "eval_scitail-pairs-pos_steps_per_second": 0.374,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_scitail-pairs-qa_loss": 0.010233570821583271,
+      "eval_scitail-pairs-qa_runtime": 2.2385,
+      "eval_scitail-pairs-qa_samples_per_second": 57.181,
+      "eval_scitail-pairs-qa_steps_per_second": 0.447,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_xsum-pairs_loss": 0.07829513400793076,
+      "eval_xsum-pairs_runtime": 3.1055,
+      "eval_xsum-pairs_samples_per_second": 41.217,
+      "eval_xsum-pairs_steps_per_second": 0.322,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_sciq_pairs_loss": 0.06304117292165756,
+      "eval_sciq_pairs_runtime": 4.3422,
+      "eval_sciq_pairs_samples_per_second": 29.478,
+      "eval_sciq_pairs_steps_per_second": 0.23,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_qasc_pairs_loss": 0.11734917759895325,
+      "eval_qasc_pairs_runtime": 2.1055,
+      "eval_qasc_pairs_samples_per_second": 60.793,
+      "eval_qasc_pairs_steps_per_second": 0.475,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_openbookqa_pairs_loss": 0.45362916588783264,
+      "eval_openbookqa_pairs_runtime": 2.2519,
+      "eval_openbookqa_pairs_samples_per_second": 56.84,
+      "eval_openbookqa_pairs_steps_per_second": 0.444,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_msmarco_pairs_loss": 0.1522417515516281,
+      "eval_msmarco_pairs_runtime": 2.2506,
+      "eval_msmarco_pairs_samples_per_second": 56.874,
+      "eval_msmarco_pairs_steps_per_second": 0.444,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_nq_pairs_loss": 0.18318293988704681,
+      "eval_nq_pairs_runtime": 3.3369,
+      "eval_nq_pairs_samples_per_second": 38.359,
+      "eval_nq_pairs_steps_per_second": 0.3,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_trivia_pairs_loss": 0.13312742114067078,
+      "eval_trivia_pairs_runtime": 3.3622,
+      "eval_trivia_pairs_samples_per_second": 38.07,
+      "eval_trivia_pairs_steps_per_second": 0.297,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_gooaq_pairs_loss": 0.13270226120948792,
+      "eval_gooaq_pairs_runtime": 2.2422,
+      "eval_gooaq_pairs_samples_per_second": 57.086,
+      "eval_gooaq_pairs_steps_per_second": 0.446,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_paws-pos_loss": 0.02471838891506195,
+      "eval_paws-pos_runtime": 2.3061,
+      "eval_paws-pos_samples_per_second": 55.506,
+      "eval_paws-pos_steps_per_second": 0.434,
+      "step": 314
+    },
+    {
+      "epoch": 0.20076726342711,
+      "eval_global_dataset_loss": 0.23317644000053406,
+      "eval_global_dataset_runtime": 8.8835,
+      "eval_global_dataset_samples_per_second": 36.585,
+      "eval_global_dataset_steps_per_second": 0.338,
+      "step": 314
+    },
+    {
+      "epoch": 0.20460358056265984,
+      "grad_norm": 14.963007926940918,
+      "learning_rate": 6.645367412140575e-06,
+      "loss": 0.2628,
+      "step": 320
+    },
+    {
+      "epoch": 0.21483375959079284,
+      "grad_norm": 33.163265228271484,
+      "learning_rate": 6.986155484558041e-06,
+      "loss": 0.2167,
+      "step": 336
+    },
+    {
+      "epoch": 0.22506393861892582,
+      "grad_norm": 239.8547821044922,
+      "learning_rate": 7.326943556975507e-06,
+      "loss": 0.1816,
+      "step": 352
+    },
+    {
+      "epoch": 0.23529411764705882,
+      "grad_norm": 6.623231410980225,
+      "learning_rate": 7.667731629392972e-06,
+      "loss": 0.1327,
+      "step": 368
+    },
+    {
+      "epoch": 0.24552429667519182,
+      "grad_norm": 9.30941390991211,
+      "learning_rate": 8.008519701810438e-06,
+      "loss": 0.2178,
+      "step": 384
+    },
+    {
+      "epoch": 0.2557544757033248,
+      "grad_norm": 4.750399112701416,
+      "learning_rate": 8.32800851970181e-06,
+      "loss": 0.0993,
+      "step": 400
+    },
+    {
+      "epoch": 0.2659846547314578,
+      "grad_norm": 6.0740461349487305,
+      "learning_rate": 8.668796592119276e-06,
+      "loss": 0.2293,
+      "step": 416
+    },
+    {
+      "epoch": 0.27621483375959077,
+      "grad_norm": 39.3314323425293,
+      "learning_rate": 9.009584664536743e-06,
+      "loss": 0.4346,
+      "step": 432
+    },
+    {
+      "epoch": 0.2864450127877238,
+      "grad_norm": 1288.867431640625,
+      "learning_rate": 9.329073482428116e-06,
+      "loss": 1.0288,
+      "step": 448
+    },
+    {
+      "epoch": 0.2966751918158568,
+      "grad_norm": 9.365520477294922,
+      "learning_rate": 9.64856230031949e-06,
+      "loss": 0.4893,
+      "step": 464
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_Qnli-dev_cosine_accuracy": 0.69921875,
+      "eval_Qnli-dev_cosine_accuracy_threshold": 0.7410688400268555,
+      "eval_Qnli-dev_cosine_ap": 0.7331344916543854,
+      "eval_Qnli-dev_cosine_f1": 0.6888111888111887,
+      "eval_Qnli-dev_cosine_f1_threshold": 0.6412021517753601,
+      "eval_Qnli-dev_cosine_precision": 0.5863095238095238,
+      "eval_Qnli-dev_cosine_recall": 0.8347457627118644,
+      "eval_Qnli-dev_dot_accuracy": 0.69140625,
+      "eval_Qnli-dev_dot_accuracy_threshold": 673.3492431640625,
+      "eval_Qnli-dev_dot_ap": 0.702049578471496,
+      "eval_Qnli-dev_dot_f1": 0.6897810218978102,
+      "eval_Qnli-dev_dot_f1_threshold": 611.4298706054688,
+      "eval_Qnli-dev_dot_precision": 0.6057692307692307,
+      "eval_Qnli-dev_dot_recall": 0.8008474576271186,
+      "eval_Qnli-dev_euclidean_accuracy": 0.697265625,
+      "eval_Qnli-dev_euclidean_accuracy_threshold": 22.84914207458496,
+      "eval_Qnli-dev_euclidean_ap": 0.7357836494194735,
+      "eval_Qnli-dev_euclidean_f1": 0.6843177189409368,
+      "eval_Qnli-dev_euclidean_f1_threshold": 23.908557891845703,
+      "eval_Qnli-dev_euclidean_precision": 0.6588235294117647,
+      "eval_Qnli-dev_euclidean_recall": 0.711864406779661,
+      "eval_Qnli-dev_manhattan_accuracy": 0.69921875,
+      "eval_Qnli-dev_manhattan_accuracy_threshold": 712.7897338867188,
+      "eval_Qnli-dev_manhattan_ap": 0.736884863348749,
+      "eval_Qnli-dev_manhattan_f1": 0.6865079365079366,
+      "eval_Qnli-dev_manhattan_f1_threshold": 753.1321411132812,
+      "eval_Qnli-dev_manhattan_precision": 0.6455223880597015,
+      "eval_Qnli-dev_manhattan_recall": 0.7330508474576272,
+      "eval_Qnli-dev_max_accuracy": 0.69921875,
+      "eval_Qnli-dev_max_accuracy_threshold": 712.7897338867188,
+      "eval_Qnli-dev_max_ap": 0.736884863348749,
+      "eval_Qnli-dev_max_f1": 0.6897810218978102,
+      "eval_Qnli-dev_max_f1_threshold": 753.1321411132812,
+      "eval_Qnli-dev_max_precision": 0.6588235294117647,
+      "eval_Qnli-dev_max_recall": 0.8347457627118644,
+      "eval_allNLI-dev_cosine_accuracy": 0.716796875,
+      "eval_allNLI-dev_cosine_accuracy_threshold": 0.835934042930603,
+      "eval_allNLI-dev_cosine_ap": 0.5816321594253542,
+      "eval_allNLI-dev_cosine_f1": 0.6029106029106028,
+      "eval_allNLI-dev_cosine_f1_threshold": 0.6776033043861389,
+      "eval_allNLI-dev_cosine_precision": 0.4707792207792208,
+      "eval_allNLI-dev_cosine_recall": 0.838150289017341,
+      "eval_allNLI-dev_dot_accuracy": 0.703125,
+      "eval_allNLI-dev_dot_accuracy_threshold": 810.56591796875,
+      "eval_allNLI-dev_dot_ap": 0.5677886575854101,
+      "eval_allNLI-dev_dot_f1": 0.6004228329809725,
+      "eval_allNLI-dev_dot_f1_threshold": 647.8101806640625,
+      "eval_allNLI-dev_dot_precision": 0.47333333333333333,
+      "eval_allNLI-dev_dot_recall": 0.8208092485549133,
+      "eval_allNLI-dev_euclidean_accuracy": 0.71875,
+      "eval_allNLI-dev_euclidean_accuracy_threshold": 17.530292510986328,
+      "eval_allNLI-dev_euclidean_ap": 0.5790276724859877,
+      "eval_allNLI-dev_euclidean_f1": 0.6070686070686071,
+      "eval_allNLI-dev_euclidean_f1_threshold": 24.67755126953125,
+      "eval_allNLI-dev_euclidean_precision": 0.474025974025974,
+      "eval_allNLI-dev_euclidean_recall": 0.8439306358381503,
+      "eval_allNLI-dev_manhattan_accuracy": 0.71875,
+      "eval_allNLI-dev_manhattan_accuracy_threshold": 547.3074951171875,
+      "eval_allNLI-dev_manhattan_ap": 0.5803704124451106,
+      "eval_allNLI-dev_manhattan_f1": 0.6008583690987125,
+      "eval_allNLI-dev_manhattan_f1_threshold": 760.1802368164062,
+      "eval_allNLI-dev_manhattan_precision": 0.4778156996587031,
+      "eval_allNLI-dev_manhattan_recall": 0.8092485549132948,
+      "eval_allNLI-dev_max_accuracy": 0.71875,
+      "eval_allNLI-dev_max_accuracy_threshold": 810.56591796875,
+      "eval_allNLI-dev_max_ap": 0.5816321594253542,
+      "eval_allNLI-dev_max_f1": 0.6070686070686071,
+      "eval_allNLI-dev_max_f1_threshold": 760.1802368164062,
+      "eval_allNLI-dev_max_precision": 0.4778156996587031,
+      "eval_allNLI-dev_max_recall": 0.8439306358381503,
+      "eval_sequential_score": 0.736884863348749,
+      "eval_sts-test_pearson_cosine": 0.9094229035285761,
+      "eval_sts-test_pearson_dot": 0.8982590057670237,
+      "eval_sts-test_pearson_euclidean": 0.9199260651959468,
+      "eval_sts-test_pearson_manhattan": 0.9194219727619668,
+      "eval_sts-test_pearson_max": 0.9199260651959468,
+      "eval_sts-test_spearman_cosine": 0.9158958736030971,
+      "eval_sts-test_spearman_dot": 0.8963403778977608,
+      "eval_sts-test_spearman_euclidean": 0.9158618178931783,
+      "eval_sts-test_spearman_manhattan": 0.9156180577723795,
+      "eval_sts-test_spearman_max": 0.9158958736030971,
+      "eval_vitaminc-pairs_loss": 3.671621084213257,
+      "eval_vitaminc-pairs_runtime": 4.6059,
+      "eval_vitaminc-pairs_samples_per_second": 27.791,
+      "eval_vitaminc-pairs_steps_per_second": 0.217,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_negation-triplets_loss": 0.6801220774650574,
+      "eval_negation-triplets_runtime": 3.2119,
+      "eval_negation-triplets_samples_per_second": 39.852,
+      "eval_negation-triplets_steps_per_second": 0.311,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_scitail-pairs-pos_loss": 0.01225399598479271,
+      "eval_scitail-pairs-pos_runtime": 2.7794,
+      "eval_scitail-pairs-pos_samples_per_second": 46.052,
+      "eval_scitail-pairs-pos_steps_per_second": 0.36,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_scitail-pairs-qa_loss": 0.00501797953620553,
+      "eval_scitail-pairs-qa_runtime": 2.3318,
+      "eval_scitail-pairs-qa_samples_per_second": 54.893,
+      "eval_scitail-pairs-qa_steps_per_second": 0.429,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_xsum-pairs_loss": 0.05804339051246643,
+      "eval_xsum-pairs_runtime": 3.1527,
+      "eval_xsum-pairs_samples_per_second": 40.6,
+      "eval_xsum-pairs_steps_per_second": 0.317,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_sciq_pairs_loss": 0.04724589362740517,
+      "eval_sciq_pairs_runtime": 4.3481,
+      "eval_sciq_pairs_samples_per_second": 29.438,
+      "eval_sciq_pairs_steps_per_second": 0.23,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_qasc_pairs_loss": 0.04424997419118881,
+      "eval_qasc_pairs_runtime": 2.1476,
+      "eval_qasc_pairs_samples_per_second": 59.6,
+      "eval_qasc_pairs_steps_per_second": 0.466,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_openbookqa_pairs_loss": 0.46001699566841125,
+      "eval_openbookqa_pairs_runtime": 2.3534,
+      "eval_openbookqa_pairs_samples_per_second": 54.389,
+      "eval_openbookqa_pairs_steps_per_second": 0.425,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_msmarco_pairs_loss": 0.22556719183921814,
+      "eval_msmarco_pairs_runtime": 2.2946,
+      "eval_msmarco_pairs_samples_per_second": 55.782,
+      "eval_msmarco_pairs_steps_per_second": 0.436,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_nq_pairs_loss": 0.13123252987861633,
+      "eval_nq_pairs_runtime": 3.3642,
+      "eval_nq_pairs_samples_per_second": 38.048,
+      "eval_nq_pairs_steps_per_second": 0.297,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_trivia_pairs_loss": 0.11183707416057587,
+      "eval_trivia_pairs_runtime": 3.3501,
+      "eval_trivia_pairs_samples_per_second": 38.208,
+      "eval_trivia_pairs_steps_per_second": 0.299,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_gooaq_pairs_loss": 0.17444811761379242,
+      "eval_gooaq_pairs_runtime": 2.2434,
+      "eval_gooaq_pairs_samples_per_second": 57.056,
+      "eval_gooaq_pairs_steps_per_second": 0.446,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_paws-pos_loss": 0.02457336150109768,
+      "eval_paws-pos_runtime": 2.3035,
+      "eval_paws-pos_samples_per_second": 55.567,
+      "eval_paws-pos_steps_per_second": 0.434,
+      "step": 471
+    },
+    {
+      "epoch": 0.30115089514066495,
+      "eval_global_dataset_loss": 0.259395956993103,
+      "eval_global_dataset_runtime": 8.8545,
+      "eval_global_dataset_samples_per_second": 36.704,
+      "eval_global_dataset_steps_per_second": 0.339,
+      "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:d523e4a0d91ff1db0041a8e72663ad04f154bb6b99fc03ec758da489cf961449
+size 5560