Training in progress, step 214, checkpoint

checkpoint-214/1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+  "word_embedding_dimension": 768,
+  "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-214/added_tokens.json

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

checkpoint-214/config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "microsoft/deberta-v3-small",
+  "architectures": [
+    "DebertaV2Model"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "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": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "p2c",
+    "c2p"
+  ],
+  "position_biased_input": false,
+  "position_buckets": 256,
+  "relative_attention": true,
+  "share_att_key": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.42.3",
+  "type_vocab_size": 0,
+  "vocab_size": 128100
+}

checkpoint-214/config_sentence_transformers.json

@@ -0,0 +1,10 @@
+{
+  "__version__": {
+    "sentence_transformers": "3.0.1",
+    "transformers": "4.42.3",
+    "pytorch": "2.1.2"
+  },
+  "prompts": {},
+  "default_prompt_name": null,
+  "similarity_fn_name": null
+}

checkpoint-214/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-214/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52c905c776e5ae71066e2801101078aabbec9b5878ff9fd90c19897056191b45
+size 1130520122

checkpoint-214/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddba6327b1a377cfdcef6125292be85fa16e40cb025172fd1e124afb94b69984
+size 565251810

checkpoint-214/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c970bca00c34ca2adef33f13d94cd397d493fc8e23507c794e76d9ad3a07652
+size 14244

checkpoint-214/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7306392a8e5051864ce8e84fe594d9ecbfcfe132886d49eb79d53d8d1b602099
+size 1064

checkpoint-214/sentence_bert_config.json

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

checkpoint-214/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-214/spm.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
+size 2464616

checkpoint-214/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-214/trainer_state.json

@@ -0,0 +1,1559 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.2081863091037404,
+  "eval_steps": 27,
+  "global_step": 214,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.016937191249117856,
+      "grad_norm": 34.22002029418945,
+      "learning_rate": 6.818181818181818e-07,
+      "loss": 7.2372,
+      "step": 3
+    },
+    {
+      "epoch": 0.03387438249823571,
+      "grad_norm": 21.76839828491211,
+      "learning_rate": 1.3636363636363636e-06,
+      "loss": 6.855,
+      "step": 6
+    },
+    {
+      "epoch": 0.05081157374735357,
+      "grad_norm": 21.260774612426758,
+      "learning_rate": 2.0454545454545457e-06,
+      "loss": 7.4707,
+      "step": 9
+    },
+    {
+      "epoch": 0.06774876499647142,
+      "grad_norm": 16.885921478271484,
+      "learning_rate": 2.7272727272727272e-06,
+      "loss": 7.0187,
+      "step": 12
+    },
+    {
+      "epoch": 0.08468595624558928,
+      "grad_norm": 19.509899139404297,
+      "learning_rate": 3.409090909090909e-06,
+      "loss": 6.6756,
+      "step": 15
+    },
+    {
+      "epoch": 0.10162314749470713,
+      "grad_norm": 7.9427289962768555,
+      "learning_rate": 4.0909090909090915e-06,
+      "loss": 6.0155,
+      "step": 18
+    },
+    {
+      "epoch": 0.11856033874382499,
+      "grad_norm": 7.325345039367676,
+      "learning_rate": 4.772727272727273e-06,
+      "loss": 6.1644,
+      "step": 21
+    },
+    {
+      "epoch": 0.13549752999294284,
+      "grad_norm": 7.544689655303955,
+      "learning_rate": 5.4545454545454545e-06,
+      "loss": 6.2158,
+      "step": 24
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "grad_norm": 5.141758918762207,
+      "learning_rate": 6.136363636363637e-06,
+      "loss": 6.1369,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.109375,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.5546875,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.9544724822044373,
+      "eval_VitaminC_cosine_ap": 0.5356492030729136,
+      "eval_VitaminC_cosine_f1": 0.6542553191489362,
+      "eval_VitaminC_cosine_f1_threshold": 0.7148199081420898,
+      "eval_VitaminC_cosine_precision": 0.48616600790513836,
+      "eval_VitaminC_cosine_recall": 1.0,
+      "eval_VitaminC_dot_accuracy": 0.55078125,
+      "eval_VitaminC_dot_accuracy_threshold": 414.4264831542969,
+      "eval_VitaminC_dot_ap": 0.5108219546857565,
+      "eval_VitaminC_dot_f1": 0.6507936507936508,
+      "eval_VitaminC_dot_f1_threshold": 271.6522521972656,
+      "eval_VitaminC_dot_precision": 0.4823529411764706,
+      "eval_VitaminC_dot_recall": 1.0,
+      "eval_VitaminC_euclidean_accuracy": 0.55078125,
+      "eval_VitaminC_euclidean_accuracy_threshold": 6.519885063171387,
+      "eval_VitaminC_euclidean_ap": 0.5226419655984281,
+      "eval_VitaminC_euclidean_f1": 0.6505376344086021,
+      "eval_VitaminC_euclidean_f1_threshold": 15.194067001342773,
+      "eval_VitaminC_euclidean_precision": 0.4859437751004016,
+      "eval_VitaminC_euclidean_recall": 0.983739837398374,
+      "eval_VitaminC_manhattan_accuracy": 0.546875,
+      "eval_VitaminC_manhattan_accuracy_threshold": 149.20114135742188,
+      "eval_VitaminC_manhattan_ap": 0.5237451656134715,
+      "eval_VitaminC_manhattan_f1": 0.6542553191489362,
+      "eval_VitaminC_manhattan_f1_threshold": 259.007080078125,
+      "eval_VitaminC_manhattan_precision": 0.48616600790513836,
+      "eval_VitaminC_manhattan_recall": 1.0,
+      "eval_VitaminC_max_accuracy": 0.5546875,
+      "eval_VitaminC_max_accuracy_threshold": 414.4264831542969,
+      "eval_VitaminC_max_ap": 0.5356492030729136,
+      "eval_VitaminC_max_f1": 0.6542553191489362,
+      "eval_VitaminC_max_f1_threshold": 271.6522521972656,
+      "eval_VitaminC_max_precision": 0.48616600790513836,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.5356492030729136,
+      "eval_sts-test_pearson_cosine": 0.056062031998983373,
+      "eval_sts-test_pearson_dot": 0.2979259445723872,
+      "eval_sts-test_pearson_euclidean": 0.0498319208592713,
+      "eval_sts-test_pearson_manhattan": 0.07381429239121526,
+      "eval_sts-test_pearson_max": 0.2979259445723872,
+      "eval_sts-test_spearman_cosine": 0.1066788491614481,
+      "eval_sts-test_spearman_dot": 0.315952670306405,
+      "eval_sts-test_spearman_euclidean": 0.07303394554435191,
+      "eval_sts-test_spearman_manhattan": 0.09039525717692232,
+      "eval_sts-test_spearman_max": 0.315952670306405,
+      "eval_vitaminc-pairs_loss": 2.698580741882324,
+      "eval_vitaminc-pairs_runtime": 1.4747,
+      "eval_vitaminc-pairs_samples_per_second": 73.236,
+      "eval_vitaminc-pairs_steps_per_second": 1.356,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_negation-triplets_loss": 5.142906665802002,
+      "eval_negation-triplets_runtime": 0.2993,
+      "eval_negation-triplets_samples_per_second": 213.865,
+      "eval_negation-triplets_steps_per_second": 3.342,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_scitail-pairs-pos_loss": 1.9216958284378052,
+      "eval_scitail-pairs-pos_runtime": 0.3834,
+      "eval_scitail-pairs-pos_samples_per_second": 140.842,
+      "eval_scitail-pairs-pos_steps_per_second": 2.608,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_xsum-pairs_loss": 6.073049545288086,
+      "eval_xsum-pairs_runtime": 3.1587,
+      "eval_xsum-pairs_samples_per_second": 40.523,
+      "eval_xsum-pairs_steps_per_second": 0.633,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_sciq_pairs_loss": 0.3449864387512207,
+      "eval_sciq_pairs_runtime": 3.3747,
+      "eval_sciq_pairs_samples_per_second": 37.93,
+      "eval_sciq_pairs_steps_per_second": 0.593,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_qasc_pairs_loss": 3.2267842292785645,
+      "eval_qasc_pairs_runtime": 0.6576,
+      "eval_qasc_pairs_samples_per_second": 194.646,
+      "eval_qasc_pairs_steps_per_second": 3.041,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_openbookqa_pairs_loss": 4.405983924865723,
+      "eval_openbookqa_pairs_runtime": 0.6107,
+      "eval_openbookqa_pairs_samples_per_second": 209.594,
+      "eval_openbookqa_pairs_steps_per_second": 3.275,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_msmarco_pairs_loss": 6.937691688537598,
+      "eval_msmarco_pairs_runtime": 1.3091,
+      "eval_msmarco_pairs_samples_per_second": 97.779,
+      "eval_msmarco_pairs_steps_per_second": 1.528,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_nq_pairs_loss": 6.794108867645264,
+      "eval_nq_pairs_runtime": 2.3968,
+      "eval_nq_pairs_samples_per_second": 53.404,
+      "eval_nq_pairs_steps_per_second": 0.834,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_trivia_pairs_loss": 6.3355631828308105,
+      "eval_trivia_pairs_runtime": 4.4974,
+      "eval_trivia_pairs_samples_per_second": 28.461,
+      "eval_trivia_pairs_steps_per_second": 0.445,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_gooaq_pairs_loss": 6.405998706817627,
+      "eval_gooaq_pairs_runtime": 0.8745,
+      "eval_gooaq_pairs_samples_per_second": 146.37,
+      "eval_gooaq_pairs_steps_per_second": 2.287,
+      "step": 27
+    },
+    {
+      "epoch": 0.1524347212420607,
+      "eval_paws-pos_loss": 2.2308223247528076,
+      "eval_paws-pos_runtime": 0.6998,
+      "eval_paws-pos_samples_per_second": 182.908,
+      "eval_paws-pos_steps_per_second": 2.858,
+      "step": 27
+    },
+    {
+      "epoch": 0.16937191249117856,
+      "grad_norm": 5.885251522064209,
+      "learning_rate": 6.818181818181818e-06,
+      "loss": 5.7653,
+      "step": 30
+    },
+    {
+      "epoch": 0.1863091037402964,
+      "grad_norm": 7.357480049133301,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 6.1259,
+      "step": 33
+    },
+    {
+      "epoch": 0.20324629498941427,
+      "grad_norm": 7.321795463562012,
+      "learning_rate": 8.181818181818183e-06,
+      "loss": 5.7539,
+      "step": 36
+    },
+    {
+      "epoch": 0.22018348623853212,
+      "grad_norm": 4.239792346954346,
+      "learning_rate": 8.863636363636365e-06,
+      "loss": 6.0131,
+      "step": 39
+    },
+    {
+      "epoch": 0.23712067748764998,
+      "grad_norm": 3.9554407596588135,
+      "learning_rate": 9.545454545454547e-06,
+      "loss": 6.0074,
+      "step": 42
+    },
+    {
+      "epoch": 0.25405786873676783,
+      "grad_norm": 4.406026840209961,
+      "learning_rate": 1.0227272727272729e-05,
+      "loss": 5.7125,
+      "step": 45
+    },
+    {
+      "epoch": 0.2709950599858857,
+      "grad_norm": 7.235893249511719,
+      "learning_rate": 1.0909090909090909e-05,
+      "loss": 5.5634,
+      "step": 48
+    },
+    {
+      "epoch": 0.28793225123500354,
+      "grad_norm": 5.330288410186768,
+      "learning_rate": 1.1590909090909093e-05,
+      "loss": 5.2924,
+      "step": 51
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "grad_norm": 7.216403961181641,
+      "learning_rate": 1.2272727272727274e-05,
+      "loss": 5.2286,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.046875,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.54296875,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.9328227043151855,
+      "eval_VitaminC_cosine_ap": 0.5212059026196154,
+      "eval_VitaminC_cosine_f1": 0.6576819407008085,
+      "eval_VitaminC_cosine_f1_threshold": 0.7373804450035095,
+      "eval_VitaminC_cosine_precision": 0.49193548387096775,
+      "eval_VitaminC_cosine_recall": 0.991869918699187,
+      "eval_VitaminC_dot_accuracy": 0.55078125,
+      "eval_VitaminC_dot_accuracy_threshold": 418.2774658203125,
+      "eval_VitaminC_dot_ap": 0.5160594099493883,
+      "eval_VitaminC_dot_f1": 0.6521739130434782,
+      "eval_VitaminC_dot_f1_threshold": 291.5081481933594,
+      "eval_VitaminC_dot_precision": 0.4897959183673469,
+      "eval_VitaminC_dot_recall": 0.975609756097561,
+      "eval_VitaminC_euclidean_accuracy": 0.5390625,
+      "eval_VitaminC_euclidean_accuracy_threshold": 8.120429039001465,
+      "eval_VitaminC_euclidean_ap": 0.5224837623095228,
+      "eval_VitaminC_euclidean_f1": 0.6576819407008085,
+      "eval_VitaminC_euclidean_f1_threshold": 14.879999160766602,
+      "eval_VitaminC_euclidean_precision": 0.49193548387096775,
+      "eval_VitaminC_euclidean_recall": 0.991869918699187,
+      "eval_VitaminC_manhattan_accuracy": 0.53515625,
+      "eval_VitaminC_manhattan_accuracy_threshold": 137.40658569335938,
+      "eval_VitaminC_manhattan_ap": 0.5186382518671783,
+      "eval_VitaminC_manhattan_f1": 0.6576086956521738,
+      "eval_VitaminC_manhattan_f1_threshold": 263.32452392578125,
+      "eval_VitaminC_manhattan_precision": 0.49387755102040815,
+      "eval_VitaminC_manhattan_recall": 0.983739837398374,
+      "eval_VitaminC_max_accuracy": 0.55078125,
+      "eval_VitaminC_max_accuracy_threshold": 418.2774658203125,
+      "eval_VitaminC_max_ap": 0.5224837623095228,
+      "eval_VitaminC_max_f1": 0.6576819407008085,
+      "eval_VitaminC_max_f1_threshold": 291.5081481933594,
+      "eval_VitaminC_max_precision": 0.49387755102040815,
+      "eval_VitaminC_max_recall": 0.991869918699187,
+      "eval_sequential_score": 0.5224837623095228,
+      "eval_sts-test_pearson_cosine": 0.14377091128453176,
+      "eval_sts-test_pearson_dot": 0.24728387094758872,
+      "eval_sts-test_pearson_euclidean": 0.14604155960515372,
+      "eval_sts-test_pearson_manhattan": 0.1446467532231986,
+      "eval_sts-test_pearson_max": 0.24728387094758872,
+      "eval_sts-test_spearman_cosine": 0.1968510434344728,
+      "eval_sts-test_spearman_dot": 0.29467218283745694,
+      "eval_sts-test_spearman_euclidean": 0.17218164683969664,
+      "eval_sts-test_spearman_manhattan": 0.17741843340856742,
+      "eval_sts-test_spearman_max": 0.29467218283745694,
+      "eval_vitaminc-pairs_loss": 2.664700746536255,
+      "eval_vitaminc-pairs_runtime": 1.4487,
+      "eval_vitaminc-pairs_samples_per_second": 74.551,
+      "eval_vitaminc-pairs_steps_per_second": 1.381,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_negation-triplets_loss": 4.6218037605285645,
+      "eval_negation-triplets_runtime": 0.2971,
+      "eval_negation-triplets_samples_per_second": 215.438,
+      "eval_negation-triplets_steps_per_second": 3.366,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_scitail-pairs-pos_loss": 1.2413936853408813,
+      "eval_scitail-pairs-pos_runtime": 0.372,
+      "eval_scitail-pairs-pos_samples_per_second": 145.175,
+      "eval_scitail-pairs-pos_steps_per_second": 2.688,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_xsum-pairs_loss": 5.249766826629639,
+      "eval_xsum-pairs_runtime": 3.1506,
+      "eval_xsum-pairs_samples_per_second": 40.627,
+      "eval_xsum-pairs_steps_per_second": 0.635,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_sciq_pairs_loss": 0.2961578667163849,
+      "eval_sciq_pairs_runtime": 3.2909,
+      "eval_sciq_pairs_samples_per_second": 38.895,
+      "eval_sciq_pairs_steps_per_second": 0.608,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_qasc_pairs_loss": 2.530872344970703,
+      "eval_qasc_pairs_runtime": 0.6255,
+      "eval_qasc_pairs_samples_per_second": 204.63,
+      "eval_qasc_pairs_steps_per_second": 3.197,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_openbookqa_pairs_loss": 3.8855104446411133,
+      "eval_openbookqa_pairs_runtime": 0.5742,
+      "eval_openbookqa_pairs_samples_per_second": 222.914,
+      "eval_openbookqa_pairs_steps_per_second": 3.483,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_msmarco_pairs_loss": 5.246406555175781,
+      "eval_msmarco_pairs_runtime": 1.2872,
+      "eval_msmarco_pairs_samples_per_second": 99.442,
+      "eval_msmarco_pairs_steps_per_second": 1.554,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_nq_pairs_loss": 5.332630157470703,
+      "eval_nq_pairs_runtime": 2.3739,
+      "eval_nq_pairs_samples_per_second": 53.92,
+      "eval_nq_pairs_steps_per_second": 0.843,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_trivia_pairs_loss": 5.647429943084717,
+      "eval_trivia_pairs_runtime": 4.4729,
+      "eval_trivia_pairs_samples_per_second": 28.617,
+      "eval_trivia_pairs_steps_per_second": 0.447,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_gooaq_pairs_loss": 5.225871562957764,
+      "eval_gooaq_pairs_runtime": 0.8715,
+      "eval_gooaq_pairs_samples_per_second": 146.868,
+      "eval_gooaq_pairs_steps_per_second": 2.295,
+      "step": 54
+    },
+    {
+      "epoch": 0.3048694424841214,
+      "eval_paws-pos_loss": 0.8335962891578674,
+      "eval_paws-pos_runtime": 0.6844,
+      "eval_paws-pos_samples_per_second": 187.036,
+      "eval_paws-pos_steps_per_second": 2.922,
+      "step": 54
+    },
+    {
+      "epoch": 0.32180663373323926,
+      "grad_norm": 6.847682952880859,
+      "learning_rate": 1.2954545454545455e-05,
+      "loss": 4.4811,
+      "step": 57
+    },
+    {
+      "epoch": 0.3387438249823571,
+      "grad_norm": 8.383002281188965,
+      "learning_rate": 1.3636363636363637e-05,
+      "loss": 4.4239,
+      "step": 60
+    },
+    {
+      "epoch": 0.35568101623147497,
+      "grad_norm": 7.014843463897705,
+      "learning_rate": 1.431818181818182e-05,
+      "loss": 4.0273,
+      "step": 63
+    },
+    {
+      "epoch": 0.3726182074805928,
+      "grad_norm": 5.9739885330200195,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 3.4508,
+      "step": 66
+    },
+    {
+      "epoch": 0.3895553987297107,
+      "grad_norm": 11.202752113342285,
+      "learning_rate": 1.5681818181818182e-05,
+      "loss": 3.9702,
+      "step": 69
+    },
+    {
+      "epoch": 0.40649258997882853,
+      "grad_norm": 7.064818859100342,
+      "learning_rate": 1.6363636363636366e-05,
+      "loss": 3.5295,
+      "step": 72
+    },
+    {
+      "epoch": 0.4234297812279464,
+      "grad_norm": 5.912719249725342,
+      "learning_rate": 1.7045454545454546e-05,
+      "loss": 3.6395,
+      "step": 75
+    },
+    {
+      "epoch": 0.44036697247706424,
+      "grad_norm": 5.033207893371582,
+      "learning_rate": 1.772727272727273e-05,
+      "loss": 3.2398,
+      "step": 78
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "grad_norm": 5.218384265899658,
+      "learning_rate": 1.840909090909091e-05,
+      "loss": 3.116,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.0,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.5546875,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.9041332006454468,
+      "eval_VitaminC_cosine_ap": 0.5292859731465609,
+      "eval_VitaminC_cosine_f1": 0.6542553191489362,
+      "eval_VitaminC_cosine_f1_threshold": 0.452939510345459,
+      "eval_VitaminC_cosine_precision": 0.48616600790513836,
+      "eval_VitaminC_cosine_recall": 1.0,
+      "eval_VitaminC_dot_accuracy": 0.5546875,
+      "eval_VitaminC_dot_accuracy_threshold": 414.42559814453125,
+      "eval_VitaminC_dot_ap": 0.5222732504955002,
+      "eval_VitaminC_dot_f1": 0.6542553191489362,
+      "eval_VitaminC_dot_f1_threshold": 212.6934814453125,
+      "eval_VitaminC_dot_precision": 0.48616600790513836,
+      "eval_VitaminC_dot_recall": 1.0,
+      "eval_VitaminC_euclidean_accuracy": 0.5546875,
+      "eval_VitaminC_euclidean_accuracy_threshold": 9.18377685546875,
+      "eval_VitaminC_euclidean_ap": 0.5291787221346742,
+      "eval_VitaminC_euclidean_f1": 0.6542553191489362,
+      "eval_VitaminC_euclidean_f1_threshold": 22.683509826660156,
+      "eval_VitaminC_euclidean_precision": 0.48616600790513836,
+      "eval_VitaminC_euclidean_recall": 1.0,
+      "eval_VitaminC_manhattan_accuracy": 0.55859375,
+      "eval_VitaminC_manhattan_accuracy_threshold": 173.8212127685547,
+      "eval_VitaminC_manhattan_ap": 0.5305698453165033,
+      "eval_VitaminC_manhattan_f1": 0.6542553191489362,
+      "eval_VitaminC_manhattan_f1_threshold": 415.5366516113281,
+      "eval_VitaminC_manhattan_precision": 0.48616600790513836,
+      "eval_VitaminC_manhattan_recall": 1.0,
+      "eval_VitaminC_max_accuracy": 0.55859375,
+      "eval_VitaminC_max_accuracy_threshold": 414.42559814453125,
+      "eval_VitaminC_max_ap": 0.5305698453165033,
+      "eval_VitaminC_max_f1": 0.6542553191489362,
+      "eval_VitaminC_max_f1_threshold": 415.5366516113281,
+      "eval_VitaminC_max_precision": 0.48616600790513836,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.5305698453165033,
+      "eval_sts-test_pearson_cosine": 0.45494716382349193,
+      "eval_sts-test_pearson_dot": 0.44837123659858896,
+      "eval_sts-test_pearson_euclidean": 0.4480861256491879,
+      "eval_sts-test_pearson_manhattan": 0.4417008219313264,
+      "eval_sts-test_pearson_max": 0.45494716382349193,
+      "eval_sts-test_spearman_cosine": 0.48921418507251446,
+      "eval_sts-test_spearman_dot": 0.46707725062744593,
+      "eval_sts-test_spearman_euclidean": 0.4610824798409968,
+      "eval_sts-test_spearman_manhattan": 0.46068648052845956,
+      "eval_sts-test_spearman_max": 0.48921418507251446,
+      "eval_vitaminc-pairs_loss": 2.5043575763702393,
+      "eval_vitaminc-pairs_runtime": 1.4778,
+      "eval_vitaminc-pairs_samples_per_second": 73.079,
+      "eval_vitaminc-pairs_steps_per_second": 1.353,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_negation-triplets_loss": 3.4229447841644287,
+      "eval_negation-triplets_runtime": 0.2991,
+      "eval_negation-triplets_samples_per_second": 213.954,
+      "eval_negation-triplets_steps_per_second": 3.343,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_scitail-pairs-pos_loss": 0.2784869372844696,
+      "eval_scitail-pairs-pos_runtime": 0.3633,
+      "eval_scitail-pairs-pos_samples_per_second": 148.649,
+      "eval_scitail-pairs-pos_steps_per_second": 2.753,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_xsum-pairs_loss": 2.428964614868164,
+      "eval_xsum-pairs_runtime": 3.1548,
+      "eval_xsum-pairs_samples_per_second": 40.573,
+      "eval_xsum-pairs_steps_per_second": 0.634,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_sciq_pairs_loss": 0.15256048738956451,
+      "eval_sciq_pairs_runtime": 3.2432,
+      "eval_sciq_pairs_samples_per_second": 39.467,
+      "eval_sciq_pairs_steps_per_second": 0.617,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_qasc_pairs_loss": 1.2902077436447144,
+      "eval_qasc_pairs_runtime": 0.6211,
+      "eval_qasc_pairs_samples_per_second": 206.085,
+      "eval_qasc_pairs_steps_per_second": 3.22,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_openbookqa_pairs_loss": 2.4784862995147705,
+      "eval_openbookqa_pairs_runtime": 0.5758,
+      "eval_openbookqa_pairs_samples_per_second": 222.308,
+      "eval_openbookqa_pairs_steps_per_second": 3.474,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_msmarco_pairs_loss": 2.967724084854126,
+      "eval_msmarco_pairs_runtime": 1.2944,
+      "eval_msmarco_pairs_samples_per_second": 98.885,
+      "eval_msmarco_pairs_steps_per_second": 1.545,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_nq_pairs_loss": 3.358661413192749,
+      "eval_nq_pairs_runtime": 2.3827,
+      "eval_nq_pairs_samples_per_second": 53.722,
+      "eval_nq_pairs_steps_per_second": 0.839,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_trivia_pairs_loss": 3.1391680240631104,
+      "eval_trivia_pairs_runtime": 4.4155,
+      "eval_trivia_pairs_samples_per_second": 28.989,
+      "eval_trivia_pairs_steps_per_second": 0.453,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_gooaq_pairs_loss": 2.8774912357330322,
+      "eval_gooaq_pairs_runtime": 0.8746,
+      "eval_gooaq_pairs_samples_per_second": 146.346,
+      "eval_gooaq_pairs_steps_per_second": 2.287,
+      "step": 81
+    },
+    {
+      "epoch": 0.4573041637261821,
+      "eval_paws-pos_loss": 0.19754411280155182,
+      "eval_paws-pos_runtime": 0.684,
+      "eval_paws-pos_samples_per_second": 187.141,
+      "eval_paws-pos_steps_per_second": 2.924,
+      "step": 81
+    },
+    {
+      "epoch": 0.47424135497529996,
+      "grad_norm": 5.149569988250732,
+      "learning_rate": 1.9090909090909094e-05,
+      "loss": 2.6049,
+      "step": 84
+    },
+    {
+      "epoch": 0.4911785462244178,
+      "grad_norm": 5.012928009033203,
+      "learning_rate": 1.9772727272727274e-05,
+      "loss": 2.7738,
+      "step": 87
+    },
+    {
+      "epoch": 0.5081157374735357,
+      "grad_norm": 4.880725383758545,
+      "learning_rate": 2.0454545454545457e-05,
+      "loss": 2.5416,
+      "step": 90
+    },
+    {
+      "epoch": 0.5250529287226535,
+      "grad_norm": 5.618528366088867,
+      "learning_rate": 2.113636363636364e-05,
+      "loss": 2.3913,
+      "step": 93
+    },
+    {
+      "epoch": 0.5419901199717714,
+      "grad_norm": 5.020515441894531,
+      "learning_rate": 2.1818181818181818e-05,
+      "loss": 2.3144,
+      "step": 96
+    },
+    {
+      "epoch": 0.5589273112208892,
+      "grad_norm": 4.818451404571533,
+      "learning_rate": 2.25e-05,
+      "loss": 2.1857,
+      "step": 99
+    },
+    {
+      "epoch": 0.5758645024700071,
+      "grad_norm": 5.094771385192871,
+      "learning_rate": 2.3181818181818185e-05,
+      "loss": 1.8881,
+      "step": 102
+    },
+    {
+      "epoch": 0.592801693719125,
+      "grad_norm": 3.795962333679199,
+      "learning_rate": 2.3863636363636365e-05,
+      "loss": 2.2699,
+      "step": 105
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "grad_norm": 4.46245813369751,
+      "learning_rate": 2.454545454545455e-05,
+      "loss": 2.1425,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.0,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.5546875,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.8830112218856812,
+      "eval_VitaminC_cosine_ap": 0.5302172957740995,
+      "eval_VitaminC_cosine_f1": 0.6558265582655827,
+      "eval_VitaminC_cosine_f1_threshold": 0.5253933668136597,
+      "eval_VitaminC_cosine_precision": 0.491869918699187,
+      "eval_VitaminC_cosine_recall": 0.983739837398374,
+      "eval_VitaminC_dot_accuracy": 0.5390625,
+      "eval_VitaminC_dot_accuracy_threshold": 427.5576171875,
+      "eval_VitaminC_dot_ap": 0.517120157327104,
+      "eval_VitaminC_dot_f1": 0.6542553191489362,
+      "eval_VitaminC_dot_f1_threshold": 175.80963134765625,
+      "eval_VitaminC_dot_precision": 0.48616600790513836,
+      "eval_VitaminC_dot_recall": 1.0,
+      "eval_VitaminC_euclidean_accuracy": 0.5625,
+      "eval_VitaminC_euclidean_accuracy_threshold": 10.817148208618164,
+      "eval_VitaminC_euclidean_ap": 0.532255112376416,
+      "eval_VitaminC_euclidean_f1": 0.6558265582655827,
+      "eval_VitaminC_euclidean_f1_threshold": 21.10729217529297,
+      "eval_VitaminC_euclidean_precision": 0.491869918699187,
+      "eval_VitaminC_euclidean_recall": 0.983739837398374,
+      "eval_VitaminC_manhattan_accuracy": 0.5546875,
+      "eval_VitaminC_manhattan_accuracy_threshold": 224.70416259765625,
+      "eval_VitaminC_manhattan_ap": 0.5298930718604624,
+      "eval_VitaminC_manhattan_f1": 0.6558265582655827,
+      "eval_VitaminC_manhattan_f1_threshold": 415.3311767578125,
+      "eval_VitaminC_manhattan_precision": 0.491869918699187,
+      "eval_VitaminC_manhattan_recall": 0.983739837398374,
+      "eval_VitaminC_max_accuracy": 0.5625,
+      "eval_VitaminC_max_accuracy_threshold": 427.5576171875,
+      "eval_VitaminC_max_ap": 0.532255112376416,
+      "eval_VitaminC_max_f1": 0.6558265582655827,
+      "eval_VitaminC_max_f1_threshold": 415.3311767578125,
+      "eval_VitaminC_max_precision": 0.491869918699187,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.532255112376416,
+      "eval_sts-test_pearson_cosine": 0.755963151531783,
+      "eval_sts-test_pearson_dot": 0.7384823091540473,
+      "eval_sts-test_pearson_euclidean": 0.764089555623164,
+      "eval_sts-test_pearson_manhattan": 0.7670467479701304,
+      "eval_sts-test_pearson_max": 0.7670467479701304,
+      "eval_sts-test_spearman_cosine": 0.7806331583677342,
+      "eval_sts-test_spearman_dot": 0.7442842883778696,
+      "eval_sts-test_spearman_euclidean": 0.7674205303105437,
+      "eval_sts-test_spearman_manhattan": 0.7664974867050092,
+      "eval_sts-test_spearman_max": 0.7806331583677342,
+      "eval_vitaminc-pairs_loss": 2.721674919128418,
+      "eval_vitaminc-pairs_runtime": 1.4468,
+      "eval_vitaminc-pairs_samples_per_second": 74.65,
+      "eval_vitaminc-pairs_steps_per_second": 1.382,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_negation-triplets_loss": 2.338909387588501,
+      "eval_negation-triplets_runtime": 0.3017,
+      "eval_negation-triplets_samples_per_second": 212.101,
+      "eval_negation-triplets_steps_per_second": 3.314,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_scitail-pairs-pos_loss": 0.23291125893592834,
+      "eval_scitail-pairs-pos_runtime": 0.3664,
+      "eval_scitail-pairs-pos_samples_per_second": 147.385,
+      "eval_scitail-pairs-pos_steps_per_second": 2.729,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_xsum-pairs_loss": 1.2065516710281372,
+      "eval_xsum-pairs_runtime": 3.1488,
+      "eval_xsum-pairs_samples_per_second": 40.65,
+      "eval_xsum-pairs_steps_per_second": 0.635,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_sciq_pairs_loss": 0.09487833082675934,
+      "eval_sciq_pairs_runtime": 3.2618,
+      "eval_sciq_pairs_samples_per_second": 39.242,
+      "eval_sciq_pairs_steps_per_second": 0.613,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_qasc_pairs_loss": 0.8461999297142029,
+      "eval_qasc_pairs_runtime": 0.6246,
+      "eval_qasc_pairs_samples_per_second": 204.93,
+      "eval_qasc_pairs_steps_per_second": 3.202,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_openbookqa_pairs_loss": 1.5739191770553589,
+      "eval_openbookqa_pairs_runtime": 0.5751,
+      "eval_openbookqa_pairs_samples_per_second": 222.568,
+      "eval_openbookqa_pairs_steps_per_second": 3.478,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_msmarco_pairs_loss": 1.6446179151535034,
+      "eval_msmarco_pairs_runtime": 1.2828,
+      "eval_msmarco_pairs_samples_per_second": 99.784,
+      "eval_msmarco_pairs_steps_per_second": 1.559,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_nq_pairs_loss": 2.364896535873413,
+      "eval_nq_pairs_runtime": 2.3802,
+      "eval_nq_pairs_samples_per_second": 53.777,
+      "eval_nq_pairs_steps_per_second": 0.84,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_trivia_pairs_loss": 1.7080069780349731,
+      "eval_trivia_pairs_runtime": 4.4372,
+      "eval_trivia_pairs_samples_per_second": 28.847,
+      "eval_trivia_pairs_steps_per_second": 0.451,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_gooaq_pairs_loss": 1.7924479246139526,
+      "eval_gooaq_pairs_runtime": 0.8761,
+      "eval_gooaq_pairs_samples_per_second": 146.094,
+      "eval_gooaq_pairs_steps_per_second": 2.283,
+      "step": 108
+    },
+    {
+      "epoch": 0.6097388849682428,
+      "eval_paws-pos_loss": 0.08000019192695618,
+      "eval_paws-pos_runtime": 0.6839,
+      "eval_paws-pos_samples_per_second": 187.168,
+      "eval_paws-pos_steps_per_second": 2.924,
+      "step": 108
+    },
+    {
+      "epoch": 0.6266760762173607,
+      "grad_norm": 4.418070316314697,
+      "learning_rate": 2.5227272727272732e-05,
+      "loss": 2.1276,
+      "step": 111
+    },
+    {
+      "epoch": 0.6436132674664785,
+      "grad_norm": 4.3495259284973145,
+      "learning_rate": 2.590909090909091e-05,
+      "loss": 1.7531,
+      "step": 114
+    },
+    {
+      "epoch": 0.6605504587155964,
+      "grad_norm": 4.294332027435303,
+      "learning_rate": 2.6590909090909093e-05,
+      "loss": 2.0179,
+      "step": 117
+    },
+    {
+      "epoch": 0.6774876499647142,
+      "grad_norm": 3.4215610027313232,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 1.5305,
+      "step": 120
+    },
+    {
+      "epoch": 0.6944248412138321,
+      "grad_norm": 4.37844181060791,
+      "learning_rate": 2.7954545454545457e-05,
+      "loss": 1.6925,
+      "step": 123
+    },
+    {
+      "epoch": 0.7113620324629499,
+      "grad_norm": 4.019878387451172,
+      "learning_rate": 2.863636363636364e-05,
+      "loss": 1.5248,
+      "step": 126
+    },
+    {
+      "epoch": 0.7282992237120678,
+      "grad_norm": 4.662445068359375,
+      "learning_rate": 2.931818181818182e-05,
+      "loss": 1.523,
+      "step": 129
+    },
+    {
+      "epoch": 0.7452364149611856,
+      "grad_norm": 4.6323161125183105,
+      "learning_rate": 3.0000000000000004e-05,
+      "loss": 1.5474,
+      "step": 132
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "grad_norm": 4.586575984954834,
+      "learning_rate": 3.068181818181819e-05,
+      "loss": 1.7221,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.0,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.56640625,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.8478574156761169,
+      "eval_VitaminC_cosine_ap": 0.5325579595957614,
+      "eval_VitaminC_cosine_f1": 0.6559999999999999,
+      "eval_VitaminC_cosine_f1_threshold": 0.35839784145355225,
+      "eval_VitaminC_cosine_precision": 0.4880952380952381,
+      "eval_VitaminC_cosine_recall": 1.0,
+      "eval_VitaminC_dot_accuracy": 0.5625,
+      "eval_VitaminC_dot_accuracy_threshold": 366.9839172363281,
+      "eval_VitaminC_dot_ap": 0.5326813797607027,
+      "eval_VitaminC_dot_f1": 0.6559999999999999,
+      "eval_VitaminC_dot_f1_threshold": 157.35829162597656,
+      "eval_VitaminC_dot_precision": 0.4880952380952381,
+      "eval_VitaminC_dot_recall": 1.0,
+      "eval_VitaminC_euclidean_accuracy": 0.5625,
+      "eval_VitaminC_euclidean_accuracy_threshold": 12.044445037841797,
+      "eval_VitaminC_euclidean_ap": 0.5304103559932005,
+      "eval_VitaminC_euclidean_f1": 0.6542553191489362,
+      "eval_VitaminC_euclidean_f1_threshold": 24.461441040039062,
+      "eval_VitaminC_euclidean_precision": 0.48616600790513836,
+      "eval_VitaminC_euclidean_recall": 1.0,
+      "eval_VitaminC_manhattan_accuracy": 0.5625,
+      "eval_VitaminC_manhattan_accuracy_threshold": 239.24815368652344,
+      "eval_VitaminC_manhattan_ap": 0.5314780667834758,
+      "eval_VitaminC_manhattan_f1": 0.6575342465753424,
+      "eval_VitaminC_manhattan_f1_threshold": 400.6834716796875,
+      "eval_VitaminC_manhattan_precision": 0.49586776859504134,
+      "eval_VitaminC_manhattan_recall": 0.975609756097561,
+      "eval_VitaminC_max_accuracy": 0.56640625,
+      "eval_VitaminC_max_accuracy_threshold": 366.9839172363281,
+      "eval_VitaminC_max_ap": 0.5326813797607027,
+      "eval_VitaminC_max_f1": 0.6575342465753424,
+      "eval_VitaminC_max_f1_threshold": 400.6834716796875,
+      "eval_VitaminC_max_precision": 0.49586776859504134,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.5326813797607027,
+      "eval_sts-test_pearson_cosine": 0.7919597804368175,
+      "eval_sts-test_pearson_dot": 0.7994867531185785,
+      "eval_sts-test_pearson_euclidean": 0.8117960113303863,
+      "eval_sts-test_pearson_manhattan": 0.8144714466358016,
+      "eval_sts-test_pearson_max": 0.8144714466358016,
+      "eval_sts-test_spearman_cosine": 0.831478610786181,
+      "eval_sts-test_spearman_dot": 0.8192534746855707,
+      "eval_sts-test_spearman_euclidean": 0.8185577905406703,
+      "eval_sts-test_spearman_manhattan": 0.8154771593606782,
+      "eval_sts-test_spearman_max": 0.831478610786181,
+      "eval_vitaminc-pairs_loss": 2.852091073989868,
+      "eval_vitaminc-pairs_runtime": 1.4427,
+      "eval_vitaminc-pairs_samples_per_second": 74.858,
+      "eval_vitaminc-pairs_steps_per_second": 1.386,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_negation-triplets_loss": 2.074247121810913,
+      "eval_negation-triplets_runtime": 0.3,
+      "eval_negation-triplets_samples_per_second": 213.353,
+      "eval_negation-triplets_steps_per_second": 3.334,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_scitail-pairs-pos_loss": 0.2149849385023117,
+      "eval_scitail-pairs-pos_runtime": 0.3744,
+      "eval_scitail-pairs-pos_samples_per_second": 144.219,
+      "eval_scitail-pairs-pos_steps_per_second": 2.671,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_xsum-pairs_loss": 0.7706837058067322,
+      "eval_xsum-pairs_runtime": 3.1609,
+      "eval_xsum-pairs_samples_per_second": 40.495,
+      "eval_xsum-pairs_steps_per_second": 0.633,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_sciq_pairs_loss": 0.07513368874788284,
+      "eval_sciq_pairs_runtime": 3.2949,
+      "eval_sciq_pairs_samples_per_second": 38.848,
+      "eval_sciq_pairs_steps_per_second": 0.607,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_qasc_pairs_loss": 0.6355602741241455,
+      "eval_qasc_pairs_runtime": 0.6392,
+      "eval_qasc_pairs_samples_per_second": 200.246,
+      "eval_qasc_pairs_steps_per_second": 3.129,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_openbookqa_pairs_loss": 1.4014525413513184,
+      "eval_openbookqa_pairs_runtime": 0.622,
+      "eval_openbookqa_pairs_samples_per_second": 205.786,
+      "eval_openbookqa_pairs_steps_per_second": 3.215,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_msmarco_pairs_loss": 1.1524099111557007,
+      "eval_msmarco_pairs_runtime": 1.31,
+      "eval_msmarco_pairs_samples_per_second": 97.709,
+      "eval_msmarco_pairs_steps_per_second": 1.527,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_nq_pairs_loss": 1.7768574953079224,
+      "eval_nq_pairs_runtime": 2.3979,
+      "eval_nq_pairs_samples_per_second": 53.379,
+      "eval_nq_pairs_steps_per_second": 0.834,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_trivia_pairs_loss": 1.4495295286178589,
+      "eval_trivia_pairs_runtime": 4.4194,
+      "eval_trivia_pairs_samples_per_second": 28.964,
+      "eval_trivia_pairs_steps_per_second": 0.453,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_gooaq_pairs_loss": 1.3955378532409668,
+      "eval_gooaq_pairs_runtime": 0.8788,
+      "eval_gooaq_pairs_samples_per_second": 145.649,
+      "eval_gooaq_pairs_steps_per_second": 2.276,
+      "step": 135
+    },
+    {
+      "epoch": 0.7621736062103035,
+      "eval_paws-pos_loss": 0.06006813049316406,
+      "eval_paws-pos_runtime": 0.6896,
+      "eval_paws-pos_samples_per_second": 185.603,
+      "eval_paws-pos_steps_per_second": 2.9,
+      "step": 135
+    },
+    {
+      "epoch": 0.7791107974594214,
+      "grad_norm": 3.864208936691284,
+      "learning_rate": 3.1363636363636365e-05,
+      "loss": 1.5366,
+      "step": 138
+    },
+    {
+      "epoch": 0.7960479887085392,
+      "grad_norm": 3.837550640106201,
+      "learning_rate": 3.204545454545455e-05,
+      "loss": 1.3045,
+      "step": 141
+    },
+    {
+      "epoch": 0.8129851799576571,
+      "grad_norm": 3.5258102416992188,
+      "learning_rate": 3.272727272727273e-05,
+      "loss": 1.1999,
+      "step": 144
+    },
+    {
+      "epoch": 0.8299223712067749,
+      "grad_norm": 3.4431183338165283,
+      "learning_rate": 3.340909090909091e-05,
+      "loss": 1.3483,
+      "step": 147
+    },
+    {
+      "epoch": 0.8468595624558928,
+      "grad_norm": 3.6455864906311035,
+      "learning_rate": 3.409090909090909e-05,
+      "loss": 1.2009,
+      "step": 150
+    },
+    {
+      "epoch": 0.8637967537050106,
+      "grad_norm": 4.508525371551514,
+      "learning_rate": 3.4772727272727276e-05,
+      "loss": 1.4495,
+      "step": 153
+    },
+    {
+      "epoch": 0.8807339449541285,
+      "grad_norm": 3.0432400703430176,
+      "learning_rate": 3.545454545454546e-05,
+      "loss": 1.2329,
+      "step": 156
+    },
+    {
+      "epoch": 0.8976711362032463,
+      "grad_norm": 3.0190365314483643,
+      "learning_rate": 3.613636363636364e-05,
+      "loss": 1.1905,
+      "step": 159
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "grad_norm": 3.74668288230896,
+      "learning_rate": 3.681818181818182e-05,
+      "loss": 1.277,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.0,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.57421875,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.8101799488067627,
+      "eval_VitaminC_cosine_ap": 0.5298515171639175,
+      "eval_VitaminC_cosine_f1": 0.6542553191489362,
+      "eval_VitaminC_cosine_f1_threshold": 0.345889687538147,
+      "eval_VitaminC_cosine_precision": 0.48616600790513836,
+      "eval_VitaminC_cosine_recall": 1.0,
+      "eval_VitaminC_dot_accuracy": 0.55078125,
+      "eval_VitaminC_dot_accuracy_threshold": 373.5804443359375,
+      "eval_VitaminC_dot_ap": 0.5310954683437364,
+      "eval_VitaminC_dot_f1": 0.6542553191489362,
+      "eval_VitaminC_dot_f1_threshold": 155.41326904296875,
+      "eval_VitaminC_dot_precision": 0.48616600790513836,
+      "eval_VitaminC_dot_recall": 1.0,
+      "eval_VitaminC_euclidean_accuracy": 0.57421875,
+      "eval_VitaminC_euclidean_accuracy_threshold": 13.60124683380127,
+      "eval_VitaminC_euclidean_ap": 0.5286057955992807,
+      "eval_VitaminC_euclidean_f1": 0.6577540106951871,
+      "eval_VitaminC_euclidean_f1_threshold": 22.904512405395508,
+      "eval_VitaminC_euclidean_precision": 0.4900398406374502,
+      "eval_VitaminC_euclidean_recall": 1.0,
+      "eval_VitaminC_manhattan_accuracy": 0.57421875,
+      "eval_VitaminC_manhattan_accuracy_threshold": 262.37322998046875,
+      "eval_VitaminC_manhattan_ap": 0.5253560845853567,
+      "eval_VitaminC_manhattan_f1": 0.6559999999999999,
+      "eval_VitaminC_manhattan_f1_threshold": 465.94549560546875,
+      "eval_VitaminC_manhattan_precision": 0.4880952380952381,
+      "eval_VitaminC_manhattan_recall": 1.0,
+      "eval_VitaminC_max_accuracy": 0.57421875,
+      "eval_VitaminC_max_accuracy_threshold": 373.5804443359375,
+      "eval_VitaminC_max_ap": 0.5310954683437364,
+      "eval_VitaminC_max_f1": 0.6577540106951871,
+      "eval_VitaminC_max_f1_threshold": 465.94549560546875,
+      "eval_VitaminC_max_precision": 0.4900398406374502,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.5310954683437364,
+      "eval_sts-test_pearson_cosine": 0.8067612938723231,
+      "eval_sts-test_pearson_dot": 0.8217874837658639,
+      "eval_sts-test_pearson_euclidean": 0.827948115812785,
+      "eval_sts-test_pearson_manhattan": 0.8261527694953693,
+      "eval_sts-test_pearson_max": 0.827948115812785,
+      "eval_sts-test_spearman_cosine": 0.8547777638284432,
+      "eval_sts-test_spearman_dot": 0.8498786150097738,
+      "eval_sts-test_spearman_euclidean": 0.8373845860667446,
+      "eval_sts-test_spearman_manhattan": 0.8324507067477893,
+      "eval_sts-test_spearman_max": 0.8547777638284432,
+      "eval_vitaminc-pairs_loss": 2.776399612426758,
+      "eval_vitaminc-pairs_runtime": 1.4503,
+      "eval_vitaminc-pairs_samples_per_second": 74.467,
+      "eval_vitaminc-pairs_steps_per_second": 1.379,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_negation-triplets_loss": 2.005451202392578,
+      "eval_negation-triplets_runtime": 0.2981,
+      "eval_negation-triplets_samples_per_second": 214.709,
+      "eval_negation-triplets_steps_per_second": 3.355,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_scitail-pairs-pos_loss": 0.19877880811691284,
+      "eval_scitail-pairs-pos_runtime": 0.3623,
+      "eval_scitail-pairs-pos_samples_per_second": 149.043,
+      "eval_scitail-pairs-pos_steps_per_second": 2.76,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_xsum-pairs_loss": 0.5586928725242615,
+      "eval_xsum-pairs_runtime": 3.1466,
+      "eval_xsum-pairs_samples_per_second": 40.679,
+      "eval_xsum-pairs_steps_per_second": 0.636,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_sciq_pairs_loss": 0.06038254499435425,
+      "eval_sciq_pairs_runtime": 3.4092,
+      "eval_sciq_pairs_samples_per_second": 37.545,
+      "eval_sciq_pairs_steps_per_second": 0.587,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_qasc_pairs_loss": 0.49434012174606323,
+      "eval_qasc_pairs_runtime": 0.6342,
+      "eval_qasc_pairs_samples_per_second": 201.832,
+      "eval_qasc_pairs_steps_per_second": 3.154,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_openbookqa_pairs_loss": 1.1903400421142578,
+      "eval_openbookqa_pairs_runtime": 0.5754,
+      "eval_openbookqa_pairs_samples_per_second": 222.449,
+      "eval_openbookqa_pairs_steps_per_second": 3.476,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_msmarco_pairs_loss": 0.8656420707702637,
+      "eval_msmarco_pairs_runtime": 1.2858,
+      "eval_msmarco_pairs_samples_per_second": 99.547,
+      "eval_msmarco_pairs_steps_per_second": 1.555,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_nq_pairs_loss": 1.1553651094436646,
+      "eval_nq_pairs_runtime": 2.3754,
+      "eval_nq_pairs_samples_per_second": 53.885,
+      "eval_nq_pairs_steps_per_second": 0.842,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_trivia_pairs_loss": 1.2928619384765625,
+      "eval_trivia_pairs_runtime": 4.4084,
+      "eval_trivia_pairs_samples_per_second": 29.035,
+      "eval_trivia_pairs_steps_per_second": 0.454,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_gooaq_pairs_loss": 1.1580811738967896,
+      "eval_gooaq_pairs_runtime": 0.8731,
+      "eval_gooaq_pairs_samples_per_second": 146.607,
+      "eval_gooaq_pairs_steps_per_second": 2.291,
+      "step": 162
+    },
+    {
+      "epoch": 0.9146083274523642,
+      "eval_paws-pos_loss": 0.052534349262714386,
+      "eval_paws-pos_runtime": 0.6835,
+      "eval_paws-pos_samples_per_second": 187.258,
+      "eval_paws-pos_steps_per_second": 2.926,
+      "step": 162
+    },
+    {
+      "epoch": 0.9315455187014821,
+      "grad_norm": 4.7817864418029785,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 1.339,
+      "step": 165
+    },
+    {
+      "epoch": 0.9484827099505999,
+      "grad_norm": 4.000570774078369,
+      "learning_rate": 3.818181818181819e-05,
+      "loss": 1.1535,
+      "step": 168
+    },
+    {
+      "epoch": 0.9654199011997178,
+      "grad_norm": 3.5971670150756836,
+      "learning_rate": 3.8863636363636364e-05,
+      "loss": 1.1643,
+      "step": 171
+    },
+    {
+      "epoch": 0.9823570924488356,
+      "grad_norm": 3.6582131385803223,
+      "learning_rate": 3.954545454545455e-05,
+      "loss": 1.2221,
+      "step": 174
+    },
+    {
+      "epoch": 0.9992942836979535,
+      "grad_norm": 4.0953898429870605,
+      "learning_rate": 3.9999477905707075e-05,
+      "loss": 1.0974,
+      "step": 177
+    },
+    {
+      "epoch": 1.0162314749470713,
+      "grad_norm": 4.092026233673096,
+      "learning_rate": 3.999164730903481e-05,
+      "loss": 1.0984,
+      "step": 180
+    },
+    {
+      "epoch": 1.0331686661961892,
+      "grad_norm": 3.6480906009674072,
+      "learning_rate": 3.997442539262898e-05,
+      "loss": 1.0543,
+      "step": 183
+    },
+    {
+      "epoch": 1.050105857445307,
+      "grad_norm": 3.433056592941284,
+      "learning_rate": 3.99478242943326e-05,
+      "loss": 1.0994,
+      "step": 186
+    },
+    {
+      "epoch": 1.067043048694425,
+      "grad_norm": 3.507981777191162,
+      "learning_rate": 3.991186276234698e-05,
+      "loss": 1.0621,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_NLI-v2_cosine_accuracy": 1.0,
+      "eval_NLI-v2_dot_accuracy": 0.0,
+      "eval_NLI-v2_euclidean_accuracy": 1.0,
+      "eval_NLI-v2_manhattan_accuracy": 1.0,
+      "eval_NLI-v2_max_accuracy": 1.0,
+      "eval_VitaminC_cosine_accuracy": 0.578125,
+      "eval_VitaminC_cosine_accuracy_threshold": 0.7840081453323364,
+      "eval_VitaminC_cosine_ap": 0.5400770399437144,
+      "eval_VitaminC_cosine_f1": 0.6577540106951871,
+      "eval_VitaminC_cosine_f1_threshold": 0.39448243379592896,
+      "eval_VitaminC_cosine_precision": 0.4900398406374502,
+      "eval_VitaminC_cosine_recall": 1.0,
+      "eval_VitaminC_dot_accuracy": 0.5625,
+      "eval_VitaminC_dot_accuracy_threshold": 323.20281982421875,
+      "eval_VitaminC_dot_ap": 0.5420016101916201,
+      "eval_VitaminC_dot_f1": 0.6575342465753424,
+      "eval_VitaminC_dot_f1_threshold": 198.04354858398438,
+      "eval_VitaminC_dot_precision": 0.49586776859504134,
+      "eval_VitaminC_dot_recall": 0.975609756097561,
+      "eval_VitaminC_euclidean_accuracy": 0.5859375,
+      "eval_VitaminC_euclidean_accuracy_threshold": 13.84214973449707,
+      "eval_VitaminC_euclidean_ap": 0.5392157650683609,
+      "eval_VitaminC_euclidean_f1": 0.6577540106951871,
+      "eval_VitaminC_euclidean_f1_threshold": 22.595678329467773,
+      "eval_VitaminC_euclidean_precision": 0.4900398406374502,
+      "eval_VitaminC_euclidean_recall": 1.0,
+      "eval_VitaminC_manhattan_accuracy": 0.5703125,
+      "eval_VitaminC_manhattan_accuracy_threshold": 275.1253356933594,
+      "eval_VitaminC_manhattan_ap": 0.5341380380767263,
+      "eval_VitaminC_manhattan_f1": 0.6576819407008085,
+      "eval_VitaminC_manhattan_f1_threshold": 457.04986572265625,
+      "eval_VitaminC_manhattan_precision": 0.49193548387096775,
+      "eval_VitaminC_manhattan_recall": 0.991869918699187,
+      "eval_VitaminC_max_accuracy": 0.5859375,
+      "eval_VitaminC_max_accuracy_threshold": 323.20281982421875,
+      "eval_VitaminC_max_ap": 0.5420016101916201,
+      "eval_VitaminC_max_f1": 0.6577540106951871,
+      "eval_VitaminC_max_f1_threshold": 457.04986572265625,
+      "eval_VitaminC_max_precision": 0.49586776859504134,
+      "eval_VitaminC_max_recall": 1.0,
+      "eval_sequential_score": 0.5420016101916201,
+      "eval_sts-test_pearson_cosine": 0.8193410747427454,
+      "eval_sts-test_pearson_dot": 0.8275444476338831,
+      "eval_sts-test_pearson_euclidean": 0.8464528142983967,
+      "eval_sts-test_pearson_manhattan": 0.8440476980962159,
+      "eval_sts-test_pearson_max": 0.8464528142983967,
+      "eval_sts-test_spearman_cosine": 0.8680272706642878,
+      "eval_sts-test_spearman_dot": 0.8555529342729671,
+      "eval_sts-test_spearman_euclidean": 0.8542457068859202,
+      "eval_sts-test_spearman_manhattan": 0.8510265117511795,
+      "eval_sts-test_spearman_max": 0.8680272706642878,
+      "eval_vitaminc-pairs_loss": 2.6755428314208984,
+      "eval_vitaminc-pairs_runtime": 1.4509,
+      "eval_vitaminc-pairs_samples_per_second": 74.437,
+      "eval_vitaminc-pairs_steps_per_second": 1.378,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_negation-triplets_loss": 1.9071491956710815,
+      "eval_negation-triplets_runtime": 0.3051,
+      "eval_negation-triplets_samples_per_second": 209.756,
+      "eval_negation-triplets_steps_per_second": 3.277,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_scitail-pairs-pos_loss": 0.18539850413799286,
+      "eval_scitail-pairs-pos_runtime": 0.4199,
+      "eval_scitail-pairs-pos_samples_per_second": 128.604,
+      "eval_scitail-pairs-pos_steps_per_second": 2.382,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_xsum-pairs_loss": 0.38365328311920166,
+      "eval_xsum-pairs_runtime": 3.1907,
+      "eval_xsum-pairs_samples_per_second": 40.116,
+      "eval_xsum-pairs_steps_per_second": 0.627,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_sciq_pairs_loss": 0.05558515340089798,
+      "eval_sciq_pairs_runtime": 3.2891,
+      "eval_sciq_pairs_samples_per_second": 38.917,
+      "eval_sciq_pairs_steps_per_second": 0.608,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_qasc_pairs_loss": 0.40469691157341003,
+      "eval_qasc_pairs_runtime": 0.6267,
+      "eval_qasc_pairs_samples_per_second": 204.245,
+      "eval_qasc_pairs_steps_per_second": 3.191,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_openbookqa_pairs_loss": 1.0837312936782837,
+      "eval_openbookqa_pairs_runtime": 0.5765,
+      "eval_openbookqa_pairs_samples_per_second": 222.02,
+      "eval_openbookqa_pairs_steps_per_second": 3.469,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_msmarco_pairs_loss": 0.6897398233413696,
+      "eval_msmarco_pairs_runtime": 1.2918,
+      "eval_msmarco_pairs_samples_per_second": 99.089,
+      "eval_msmarco_pairs_steps_per_second": 1.548,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_nq_pairs_loss": 0.9603796601295471,
+      "eval_nq_pairs_runtime": 2.3975,
+      "eval_nq_pairs_samples_per_second": 53.39,
+      "eval_nq_pairs_steps_per_second": 0.834,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_trivia_pairs_loss": 1.200446605682373,
+      "eval_trivia_pairs_runtime": 4.4582,
+      "eval_trivia_pairs_samples_per_second": 28.711,
+      "eval_trivia_pairs_steps_per_second": 0.449,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_gooaq_pairs_loss": 1.0353316068649292,
+      "eval_gooaq_pairs_runtime": 0.8765,
+      "eval_gooaq_pairs_samples_per_second": 146.042,
+      "eval_gooaq_pairs_steps_per_second": 2.282,
+      "step": 189
+    },
+    {
+      "epoch": 1.067043048694425,
+      "eval_paws-pos_loss": 0.042069558054208755,
+      "eval_paws-pos_runtime": 0.6909,
+      "eval_paws-pos_samples_per_second": 185.263,
+      "eval_paws-pos_steps_per_second": 2.895,
+      "step": 189
+    },
+    {
+      "epoch": 1.0839802399435428,
+      "grad_norm": 2.979419469833374,
+      "learning_rate": 3.986656614201813e-05,
+      "loss": 0.8724,
+      "step": 192
+    },
+    {
+      "epoch": 1.1009174311926606,
+      "grad_norm": 2.835219144821167,
+      "learning_rate": 3.981196635797361e-05,
+      "loss": 0.9381,
+      "step": 195
+    },
+    {
+      "epoch": 1.1178546224417785,
+      "grad_norm": 3.6650869846343994,
+      "learning_rate": 3.974810189162238e-05,
+      "loss": 0.9617,
+      "step": 198
+    },
+    {
+      "epoch": 1.1347918136908963,
+      "grad_norm": 4.188896656036377,
+      "learning_rate": 3.967501775403343e-05,
+      "loss": 1.0139,
+      "step": 201
+    },
+    {
+      "epoch": 1.1517290049400142,
+      "grad_norm": 3.1624915599823,
+      "learning_rate": 3.959276545421244e-05,
+      "loss": 1.1073,
+      "step": 204
+    },
+    {
+      "epoch": 1.168666196189132,
+      "grad_norm": 3.245002508163452,
+      "learning_rate": 3.950140296279871e-05,
+      "loss": 0.8365,
+      "step": 207
+    },
+    {
+      "epoch": 1.18560338743825,
+      "grad_norm": 4.376185894012451,
+      "learning_rate": 3.9400994671208e-05,
+      "loss": 1.1012,
+      "step": 210
+    },
+    {
+      "epoch": 1.2025405786873677,
+      "grad_norm": 3.236583948135376,
+      "learning_rate": 3.9291611346250066e-05,
+      "loss": 1.0016,
+      "step": 213
+    }
+  ],
+  "logging_steps": 3,
+  "max_steps": 531,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 107,
+  "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": 160,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-214/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:526057a01875662c89243f2f1101012515e0f154fca67a38aba3fb44dcb2d6d0
+size 5688