pszemraj commited on
Commit
4a407cf
1 Parent(s): db8f91c

load model from drive and convert

Browse files
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ checkpoint-*/
README.md ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5
14
+
15
+ This model is a fine-tuned version of [pszemraj/t5-v1_1-large-jflAUGv5-training2](https://huggingface.co/pszemraj/t5-v1_1-large-jflAUGv5-training2) on an unknown dataset.
16
+
17
+ ## Model description
18
+
19
+ More information needed
20
+
21
+ ## Intended uses & limitations
22
+
23
+ More information needed
24
+
25
+ ## Training and evaluation data
26
+
27
+ More information needed
28
+
29
+ ## Training procedure
30
+
31
+ ### Training hyperparameters
32
+
33
+ The following hyperparameters were used during training:
34
+ - learning_rate: 8e-05
35
+ - train_batch_size: 4
36
+ - eval_batch_size: 1
37
+ - seed: 42
38
+ - distributed_type: multi-GPU
39
+ - gradient_accumulation_steps: 32
40
+ - total_train_batch_size: 128
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: cosine
43
+ - lr_scheduler_warmup_ratio: 0.02
44
+ - num_epochs: 1
45
+
46
+ ### Framework versions
47
+
48
+ - Transformers 4.20.1
49
+ - Pytorch 1.11.0+cu113
50
+ - Datasets 2.3.2
51
+ - Tokenizers 0.12.1
config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "pszemraj/t5-v1_1-large-jflAUGv5-training2",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 2816,
7
+ "d_kv": 64,
8
+ "d_model": 1024,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.1,
12
+ "early_stopping": true,
13
+ "eos_token_id": 1,
14
+ "feed_forward_proj": "gated-gelu",
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "max_length": 512,
20
+ "min_length": 8,
21
+ "model_type": "t5",
22
+ "no_repeat_ngram_size": 4,
23
+ "num_beams": 2,
24
+ "num_decoder_layers": 24,
25
+ "num_heads": 16,
26
+ "num_layers": 24,
27
+ "output_past": true,
28
+ "pad_token_id": 0,
29
+ "relative_attention_max_distance": 128,
30
+ "relative_attention_num_buckets": 32,
31
+ "tie_word_embeddings": false,
32
+ "torch_dtype": "float32",
33
+ "transformers_version": "4.20.1",
34
+ "use_cache": false,
35
+ "vocab_size": 32128
36
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aabac79b9c8af99f232493b3dafe804629a8796808d821a05db4da3192fe5e5e
3
+ size 3132672427
t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5_training_metadata.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"output_dir": "/content/drive/MyDrive/Programming/hf-trainer/t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5", "overwrite_output_dir": true, "do_train": false, "do_eval": false, "do_predict": false, "evaluation_strategy": "no", "prediction_loss_only": false, "per_device_train_batch_size": 4, "per_device_eval_batch_size": 1, "per_gpu_train_batch_size": "None", "per_gpu_eval_batch_size": "None", "gradient_accumulation_steps": 32, "eval_accumulation_steps": "None", "eval_delay": 0, "learning_rate": 8e-05, "weight_decay": 0.08, "adam_beta1": 0.9, "adam_beta2": 0.999, "adam_epsilon": 1e-08, "max_grad_norm": 0.6, "num_train_epochs": 1, "max_steps": -1, "lr_scheduler_type": "cosine", "warmup_ratio": 0.02, "warmup_steps": 0, "log_level": -1, "log_level_replica": -1, "log_on_each_node": true, "logging_dir": "/content/drive/MyDrive/Programming/hf-trainer/t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5/logs", "logging_strategy": "steps", "logging_first_step": false, "logging_steps": 2, "logging_nan_inf_filter": true, "save_strategy": "steps", "save_steps": 25, "save_total_limit": 1, "save_on_each_node": false, "no_cuda": false, "seed": 42, "data_seed": "None", "jit_mode_eval": false, "use_ipex": false, "bf16": false, "fp16": true, "fp16_opt_level": "O1", "half_precision_backend": "cuda_amp", "bf16_full_eval": false, "fp16_full_eval": false, "tf32": "None", "local_rank": 0, "xpu_backend": "None", "tpu_num_cores": "None", "tpu_metrics_debug": false, "debug": "[]", "dataloader_drop_last": false, "eval_steps": "None", "dataloader_num_workers": 0, "past_index": -1, "run_name": "/content/drive/MyDrive/Programming/hf-trainer/t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5", "disable_tqdm": false, "remove_unused_columns": true, "label_names": "None", "load_best_model_at_end": false, "metric_for_best_model": "None", "greater_is_better": "None", "ignore_data_skip": false, "sharded_ddp": "[]", "fsdp": "[]", "fsdp_min_num_params": 0, "deepspeed": "/content/ds_config_zero2.json", "label_smoothing_factor": 0.0, "optim": "adamw_hf", "adafactor": false, "group_by_length": false, "length_column_name": "length", "report_to": "['tensorboard']", "ddp_find_unused_parameters": "None", "ddp_bucket_cap_mb": "None", "dataloader_pin_memory": true, "skip_memory_metrics": true, "use_legacy_prediction_loop": false, "push_to_hub": true, "resume_from_checkpoint": "None", "hub_model_id": "t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5", "hub_strategy": "end", "hub_token": "<HUB_TOKEN>", "hub_private_repo": true, "gradient_checkpointing": true, "include_inputs_for_metrics": false, "fp16_backend": "auto", "push_to_hub_model_id": "None", "push_to_hub_organization": "None", "push_to_hub_token": "<PUSH_TO_HUB_TOKEN>", "_n_gpu": 1, "mp_parameters": "", "auto_find_batch_size": false, "full_determinism": false, "torchdynamo": "None", "ray_scope": "last", "sortish_sampler": false, "predict_with_generate": false, "generation_max_length": "None", "generation_num_beams": "None", "train_batch_size": 4, "eval_batch_size": 1, "configs_src": "t5-v1_1-large-jflAUGv5-training2-ft1-jflAUG-v5"}
trainer_state.json ADDED
@@ -0,0 +1,1972 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.4643978008769782,
5
+ "global_step": 652,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.0,
12
+ "learning_rate": 5.517241379310345e-06,
13
+ "loss": 0.4754,
14
+ "step": 2
15
+ },
16
+ {
17
+ "epoch": 0.0,
18
+ "learning_rate": 1.103448275862069e-05,
19
+ "loss": 0.497,
20
+ "step": 4
21
+ },
22
+ {
23
+ "epoch": 0.0,
24
+ "learning_rate": 1.6551724137931037e-05,
25
+ "loss": 0.5361,
26
+ "step": 6
27
+ },
28
+ {
29
+ "epoch": 0.01,
30
+ "learning_rate": 2.206896551724138e-05,
31
+ "loss": 0.4945,
32
+ "step": 8
33
+ },
34
+ {
35
+ "epoch": 0.01,
36
+ "learning_rate": 2.7586206896551727e-05,
37
+ "loss": 0.4847,
38
+ "step": 10
39
+ },
40
+ {
41
+ "epoch": 0.01,
42
+ "learning_rate": 3.310344827586207e-05,
43
+ "loss": 0.4741,
44
+ "step": 12
45
+ },
46
+ {
47
+ "epoch": 0.01,
48
+ "learning_rate": 3.862068965517242e-05,
49
+ "loss": 0.4742,
50
+ "step": 14
51
+ },
52
+ {
53
+ "epoch": 0.01,
54
+ "learning_rate": 4.413793103448276e-05,
55
+ "loss": 0.4544,
56
+ "step": 16
57
+ },
58
+ {
59
+ "epoch": 0.01,
60
+ "learning_rate": 4.9655172413793107e-05,
61
+ "loss": 0.4779,
62
+ "step": 18
63
+ },
64
+ {
65
+ "epoch": 0.01,
66
+ "learning_rate": 5.517241379310345e-05,
67
+ "loss": 0.4374,
68
+ "step": 20
69
+ },
70
+ {
71
+ "epoch": 0.02,
72
+ "learning_rate": 6.068965517241379e-05,
73
+ "loss": 0.5045,
74
+ "step": 22
75
+ },
76
+ {
77
+ "epoch": 0.02,
78
+ "learning_rate": 6.620689655172415e-05,
79
+ "loss": 0.46,
80
+ "step": 24
81
+ },
82
+ {
83
+ "epoch": 0.02,
84
+ "learning_rate": 7.17241379310345e-05,
85
+ "loss": 0.4489,
86
+ "step": 26
87
+ },
88
+ {
89
+ "epoch": 0.02,
90
+ "learning_rate": 7.724137931034484e-05,
91
+ "loss": 0.4535,
92
+ "step": 28
93
+ },
94
+ {
95
+ "epoch": 0.02,
96
+ "learning_rate": 7.999989544228434e-05,
97
+ "loss": 0.4802,
98
+ "step": 30
99
+ },
100
+ {
101
+ "epoch": 0.02,
102
+ "learning_rate": 7.999905898383868e-05,
103
+ "loss": 0.5038,
104
+ "step": 32
105
+ },
106
+ {
107
+ "epoch": 0.02,
108
+ "learning_rate": 7.999738608443897e-05,
109
+ "loss": 0.4804,
110
+ "step": 34
111
+ },
112
+ {
113
+ "epoch": 0.03,
114
+ "learning_rate": 7.999487677906807e-05,
115
+ "loss": 0.4752,
116
+ "step": 36
117
+ },
118
+ {
119
+ "epoch": 0.03,
120
+ "learning_rate": 7.999153112019939e-05,
121
+ "loss": 0.4681,
122
+ "step": 38
123
+ },
124
+ {
125
+ "epoch": 0.03,
126
+ "learning_rate": 7.998734917779568e-05,
127
+ "loss": 0.456,
128
+ "step": 40
129
+ },
130
+ {
131
+ "epoch": 0.03,
132
+ "learning_rate": 7.998233103930771e-05,
133
+ "loss": 0.4724,
134
+ "step": 42
135
+ },
136
+ {
137
+ "epoch": 0.03,
138
+ "learning_rate": 7.99764768096724e-05,
139
+ "loss": 0.4514,
140
+ "step": 44
141
+ },
142
+ {
143
+ "epoch": 0.03,
144
+ "learning_rate": 7.99697866113105e-05,
145
+ "loss": 0.4601,
146
+ "step": 46
147
+ },
148
+ {
149
+ "epoch": 0.03,
150
+ "learning_rate": 7.996226058412426e-05,
151
+ "loss": 0.4754,
152
+ "step": 48
153
+ },
154
+ {
155
+ "epoch": 0.04,
156
+ "learning_rate": 7.995389888549428e-05,
157
+ "loss": 0.4395,
158
+ "step": 50
159
+ },
160
+ {
161
+ "epoch": 0.04,
162
+ "learning_rate": 7.994470169027636e-05,
163
+ "loss": 0.4795,
164
+ "step": 52
165
+ },
166
+ {
167
+ "epoch": 0.04,
168
+ "learning_rate": 7.993466919079781e-05,
169
+ "loss": 0.4506,
170
+ "step": 54
171
+ },
172
+ {
173
+ "epoch": 0.04,
174
+ "learning_rate": 7.992380159685338e-05,
175
+ "loss": 0.5032,
176
+ "step": 56
177
+ },
178
+ {
179
+ "epoch": 0.04,
180
+ "learning_rate": 7.991209913570094e-05,
181
+ "loss": 0.4629,
182
+ "step": 58
183
+ },
184
+ {
185
+ "epoch": 0.04,
186
+ "learning_rate": 7.989956205205669e-05,
187
+ "loss": 0.4653,
188
+ "step": 60
189
+ },
190
+ {
191
+ "epoch": 0.04,
192
+ "learning_rate": 7.988619060809005e-05,
193
+ "loss": 0.4462,
194
+ "step": 62
195
+ },
196
+ {
197
+ "epoch": 0.05,
198
+ "learning_rate": 7.98719850834182e-05,
199
+ "loss": 0.434,
200
+ "step": 64
201
+ },
202
+ {
203
+ "epoch": 0.05,
204
+ "learning_rate": 7.985694577510017e-05,
205
+ "loss": 0.4845,
206
+ "step": 66
207
+ },
208
+ {
209
+ "epoch": 0.05,
210
+ "learning_rate": 7.98410729976307e-05,
211
+ "loss": 0.4586,
212
+ "step": 68
213
+ },
214
+ {
215
+ "epoch": 0.05,
216
+ "learning_rate": 7.982436708293364e-05,
217
+ "loss": 0.4515,
218
+ "step": 70
219
+ },
220
+ {
221
+ "epoch": 0.05,
222
+ "learning_rate": 7.980682838035498e-05,
223
+ "loss": 0.4621,
224
+ "step": 72
225
+ },
226
+ {
227
+ "epoch": 0.05,
228
+ "learning_rate": 7.978845725665557e-05,
229
+ "loss": 0.4524,
230
+ "step": 74
231
+ },
232
+ {
233
+ "epoch": 0.05,
234
+ "learning_rate": 7.976925409600346e-05,
235
+ "loss": 0.4458,
236
+ "step": 76
237
+ },
238
+ {
239
+ "epoch": 0.06,
240
+ "learning_rate": 7.974921929996585e-05,
241
+ "loss": 0.4992,
242
+ "step": 78
243
+ },
244
+ {
245
+ "epoch": 0.06,
246
+ "learning_rate": 7.97283532875007e-05,
247
+ "loss": 0.4931,
248
+ "step": 80
249
+ },
250
+ {
251
+ "epoch": 0.06,
252
+ "learning_rate": 7.970665649494793e-05,
253
+ "loss": 0.469,
254
+ "step": 82
255
+ },
256
+ {
257
+ "epoch": 0.06,
258
+ "learning_rate": 7.968412937602039e-05,
259
+ "loss": 0.5017,
260
+ "step": 84
261
+ },
262
+ {
263
+ "epoch": 0.06,
264
+ "learning_rate": 7.966077240179428e-05,
265
+ "loss": 0.4753,
266
+ "step": 86
267
+ },
268
+ {
269
+ "epoch": 0.06,
270
+ "learning_rate": 7.963658606069931e-05,
271
+ "loss": 0.4618,
272
+ "step": 88
273
+ },
274
+ {
275
+ "epoch": 0.06,
276
+ "learning_rate": 7.961157085850858e-05,
277
+ "loss": 0.4356,
278
+ "step": 90
279
+ },
280
+ {
281
+ "epoch": 0.07,
282
+ "learning_rate": 7.958572731832784e-05,
283
+ "loss": 0.5153,
284
+ "step": 92
285
+ },
286
+ {
287
+ "epoch": 0.07,
288
+ "learning_rate": 7.955905598058471e-05,
289
+ "loss": 0.4933,
290
+ "step": 94
291
+ },
292
+ {
293
+ "epoch": 0.07,
294
+ "learning_rate": 7.95315574030173e-05,
295
+ "loss": 0.4749,
296
+ "step": 96
297
+ },
298
+ {
299
+ "epoch": 0.07,
300
+ "learning_rate": 7.950323216066252e-05,
301
+ "loss": 0.4162,
302
+ "step": 98
303
+ },
304
+ {
305
+ "epoch": 0.07,
306
+ "learning_rate": 7.947408084584417e-05,
307
+ "loss": 0.4894,
308
+ "step": 100
309
+ },
310
+ {
311
+ "epoch": 0.07,
312
+ "learning_rate": 7.944410406816039e-05,
313
+ "loss": 0.4281,
314
+ "step": 102
315
+ },
316
+ {
317
+ "epoch": 0.07,
318
+ "learning_rate": 7.941330245447103e-05,
319
+ "loss": 0.4245,
320
+ "step": 104
321
+ },
322
+ {
323
+ "epoch": 0.08,
324
+ "learning_rate": 7.938167664888458e-05,
325
+ "loss": 0.4468,
326
+ "step": 106
327
+ },
328
+ {
329
+ "epoch": 0.08,
330
+ "learning_rate": 7.934922731274452e-05,
331
+ "loss": 0.4765,
332
+ "step": 108
333
+ },
334
+ {
335
+ "epoch": 0.08,
336
+ "learning_rate": 7.931595512461566e-05,
337
+ "loss": 0.4411,
338
+ "step": 110
339
+ },
340
+ {
341
+ "epoch": 0.08,
342
+ "learning_rate": 7.928186078026991e-05,
343
+ "loss": 0.4727,
344
+ "step": 112
345
+ },
346
+ {
347
+ "epoch": 0.08,
348
+ "learning_rate": 7.924694499267169e-05,
349
+ "loss": 0.4123,
350
+ "step": 114
351
+ },
352
+ {
353
+ "epoch": 0.08,
354
+ "learning_rate": 7.921120849196303e-05,
355
+ "loss": 0.4703,
356
+ "step": 116
357
+ },
358
+ {
359
+ "epoch": 0.08,
360
+ "learning_rate": 7.917465202544833e-05,
361
+ "loss": 0.4674,
362
+ "step": 118
363
+ },
364
+ {
365
+ "epoch": 0.09,
366
+ "learning_rate": 7.913727635757873e-05,
367
+ "loss": 0.4812,
368
+ "step": 120
369
+ },
370
+ {
371
+ "epoch": 0.09,
372
+ "learning_rate": 7.909908226993609e-05,
373
+ "loss": 0.4304,
374
+ "step": 122
375
+ },
376
+ {
377
+ "epoch": 0.09,
378
+ "learning_rate": 7.906007056121667e-05,
379
+ "loss": 0.4449,
380
+ "step": 124
381
+ },
382
+ {
383
+ "epoch": 0.09,
384
+ "learning_rate": 7.902024204721445e-05,
385
+ "loss": 0.4433,
386
+ "step": 126
387
+ },
388
+ {
389
+ "epoch": 0.09,
390
+ "learning_rate": 7.897959756080402e-05,
391
+ "loss": 0.4602,
392
+ "step": 128
393
+ },
394
+ {
395
+ "epoch": 0.09,
396
+ "learning_rate": 7.89381379519232e-05,
397
+ "loss": 0.4698,
398
+ "step": 130
399
+ },
400
+ {
401
+ "epoch": 0.09,
402
+ "learning_rate": 7.889586408755526e-05,
403
+ "loss": 0.3831,
404
+ "step": 132
405
+ },
406
+ {
407
+ "epoch": 0.1,
408
+ "learning_rate": 7.885277685171078e-05,
409
+ "loss": 0.4394,
410
+ "step": 134
411
+ },
412
+ {
413
+ "epoch": 0.1,
414
+ "learning_rate": 7.880887714540917e-05,
415
+ "loss": 0.4302,
416
+ "step": 136
417
+ },
418
+ {
419
+ "epoch": 0.1,
420
+ "learning_rate": 7.876416588665984e-05,
421
+ "loss": 0.4079,
422
+ "step": 138
423
+ },
424
+ {
425
+ "epoch": 0.1,
426
+ "learning_rate": 7.871864401044297e-05,
427
+ "loss": 0.4639,
428
+ "step": 140
429
+ },
430
+ {
431
+ "epoch": 0.1,
432
+ "learning_rate": 7.867231246869002e-05,
433
+ "loss": 0.4498,
434
+ "step": 142
435
+ },
436
+ {
437
+ "epoch": 0.1,
438
+ "learning_rate": 7.862517223026373e-05,
439
+ "loss": 0.4289,
440
+ "step": 144
441
+ },
442
+ {
443
+ "epoch": 0.1,
444
+ "learning_rate": 7.857722428093797e-05,
445
+ "loss": 0.4515,
446
+ "step": 146
447
+ },
448
+ {
449
+ "epoch": 0.11,
450
+ "learning_rate": 7.8528469623377e-05,
451
+ "loss": 0.4322,
452
+ "step": 148
453
+ },
454
+ {
455
+ "epoch": 0.11,
456
+ "learning_rate": 7.847890927711464e-05,
457
+ "loss": 0.434,
458
+ "step": 150
459
+ },
460
+ {
461
+ "epoch": 0.11,
462
+ "learning_rate": 7.842854427853285e-05,
463
+ "loss": 0.4431,
464
+ "step": 152
465
+ },
466
+ {
467
+ "epoch": 0.11,
468
+ "learning_rate": 7.83773756808401e-05,
469
+ "loss": 0.4089,
470
+ "step": 154
471
+ },
472
+ {
473
+ "epoch": 0.11,
474
+ "learning_rate": 7.83254045540493e-05,
475
+ "loss": 0.488,
476
+ "step": 156
477
+ },
478
+ {
479
+ "epoch": 0.11,
480
+ "learning_rate": 7.827263198495554e-05,
481
+ "loss": 0.4526,
482
+ "step": 158
483
+ },
484
+ {
485
+ "epoch": 0.11,
486
+ "learning_rate": 7.821905907711319e-05,
487
+ "loss": 0.4184,
488
+ "step": 160
489
+ },
490
+ {
491
+ "epoch": 0.12,
492
+ "learning_rate": 7.816468695081297e-05,
493
+ "loss": 0.4176,
494
+ "step": 162
495
+ },
496
+ {
497
+ "epoch": 0.12,
498
+ "learning_rate": 7.810951674305845e-05,
499
+ "loss": 0.4562,
500
+ "step": 164
501
+ },
502
+ {
503
+ "epoch": 0.12,
504
+ "learning_rate": 7.805354960754233e-05,
505
+ "loss": 0.4403,
506
+ "step": 166
507
+ },
508
+ {
509
+ "epoch": 0.12,
510
+ "learning_rate": 7.799678671462222e-05,
511
+ "loss": 0.4238,
512
+ "step": 168
513
+ },
514
+ {
515
+ "epoch": 0.12,
516
+ "learning_rate": 7.793922925129626e-05,
517
+ "loss": 0.4911,
518
+ "step": 170
519
+ },
520
+ {
521
+ "epoch": 0.12,
522
+ "learning_rate": 7.788087842117826e-05,
523
+ "loss": 0.4704,
524
+ "step": 172
525
+ },
526
+ {
527
+ "epoch": 0.12,
528
+ "learning_rate": 7.782173544447253e-05,
529
+ "loss": 0.4355,
530
+ "step": 174
531
+ },
532
+ {
533
+ "epoch": 0.13,
534
+ "learning_rate": 7.776180155794836e-05,
535
+ "loss": 0.4341,
536
+ "step": 176
537
+ },
538
+ {
539
+ "epoch": 0.13,
540
+ "learning_rate": 7.770107801491414e-05,
541
+ "loss": 0.4198,
542
+ "step": 178
543
+ },
544
+ {
545
+ "epoch": 0.13,
546
+ "learning_rate": 7.763956608519123e-05,
547
+ "loss": 0.4756,
548
+ "step": 180
549
+ },
550
+ {
551
+ "epoch": 0.13,
552
+ "learning_rate": 7.757726705508732e-05,
553
+ "loss": 0.4315,
554
+ "step": 182
555
+ },
556
+ {
557
+ "epoch": 0.13,
558
+ "learning_rate": 7.751418222736954e-05,
559
+ "loss": 0.4167,
560
+ "step": 184
561
+ },
562
+ {
563
+ "epoch": 0.13,
564
+ "learning_rate": 7.745031292123727e-05,
565
+ "loss": 0.4316,
566
+ "step": 186
567
+ },
568
+ {
569
+ "epoch": 0.13,
570
+ "learning_rate": 7.738566047229454e-05,
571
+ "loss": 0.4103,
572
+ "step": 188
573
+ },
574
+ {
575
+ "epoch": 0.14,
576
+ "learning_rate": 7.732022623252203e-05,
577
+ "loss": 0.4527,
578
+ "step": 190
579
+ },
580
+ {
581
+ "epoch": 0.14,
582
+ "learning_rate": 7.725401157024887e-05,
583
+ "loss": 0.4587,
584
+ "step": 192
585
+ },
586
+ {
587
+ "epoch": 0.14,
588
+ "learning_rate": 7.718701787012406e-05,
589
+ "loss": 0.457,
590
+ "step": 194
591
+ },
592
+ {
593
+ "epoch": 0.14,
594
+ "learning_rate": 7.711924653308737e-05,
595
+ "loss": 0.4463,
596
+ "step": 196
597
+ },
598
+ {
599
+ "epoch": 0.14,
600
+ "learning_rate": 7.705069897634022e-05,
601
+ "loss": 0.3972,
602
+ "step": 198
603
+ },
604
+ {
605
+ "epoch": 0.14,
606
+ "learning_rate": 7.69813766333159e-05,
607
+ "loss": 0.4375,
608
+ "step": 200
609
+ },
610
+ {
611
+ "epoch": 0.14,
612
+ "learning_rate": 7.691128095364969e-05,
613
+ "loss": 0.4472,
614
+ "step": 202
615
+ },
616
+ {
617
+ "epoch": 0.15,
618
+ "learning_rate": 7.684041340314852e-05,
619
+ "loss": 0.4198,
620
+ "step": 204
621
+ },
622
+ {
623
+ "epoch": 0.15,
624
+ "learning_rate": 7.676877546376028e-05,
625
+ "loss": 0.4365,
626
+ "step": 206
627
+ },
628
+ {
629
+ "epoch": 0.15,
630
+ "learning_rate": 7.669636863354287e-05,
631
+ "loss": 0.4393,
632
+ "step": 208
633
+ },
634
+ {
635
+ "epoch": 0.15,
636
+ "learning_rate": 7.662319442663286e-05,
637
+ "loss": 0.373,
638
+ "step": 210
639
+ },
640
+ {
641
+ "epoch": 0.15,
642
+ "learning_rate": 7.654925437321386e-05,
643
+ "loss": 0.4099,
644
+ "step": 212
645
+ },
646
+ {
647
+ "epoch": 0.15,
648
+ "learning_rate": 7.647455001948443e-05,
649
+ "loss": 0.4473,
650
+ "step": 214
651
+ },
652
+ {
653
+ "epoch": 0.15,
654
+ "learning_rate": 7.639908292762585e-05,
655
+ "loss": 0.4578,
656
+ "step": 216
657
+ },
658
+ {
659
+ "epoch": 0.16,
660
+ "learning_rate": 7.632285467576943e-05,
661
+ "loss": 0.4263,
662
+ "step": 218
663
+ },
664
+ {
665
+ "epoch": 0.16,
666
+ "learning_rate": 7.624586685796343e-05,
667
+ "loss": 0.402,
668
+ "step": 220
669
+ },
670
+ {
671
+ "epoch": 0.16,
672
+ "learning_rate": 7.616812108413986e-05,
673
+ "loss": 0.4347,
674
+ "step": 222
675
+ },
676
+ {
677
+ "epoch": 0.16,
678
+ "learning_rate": 7.608961898008066e-05,
679
+ "loss": 0.45,
680
+ "step": 224
681
+ },
682
+ {
683
+ "epoch": 0.16,
684
+ "learning_rate": 7.601036218738382e-05,
685
+ "loss": 0.4445,
686
+ "step": 226
687
+ },
688
+ {
689
+ "epoch": 0.16,
690
+ "learning_rate": 7.593035236342903e-05,
691
+ "loss": 0.421,
692
+ "step": 228
693
+ },
694
+ {
695
+ "epoch": 0.16,
696
+ "learning_rate": 7.5849591181343e-05,
697
+ "loss": 0.4048,
698
+ "step": 230
699
+ },
700
+ {
701
+ "epoch": 0.17,
702
+ "learning_rate": 7.576808032996444e-05,
703
+ "loss": 0.4372,
704
+ "step": 232
705
+ },
706
+ {
707
+ "epoch": 0.17,
708
+ "learning_rate": 7.568582151380881e-05,
709
+ "loss": 0.3985,
710
+ "step": 234
711
+ },
712
+ {
713
+ "epoch": 0.17,
714
+ "learning_rate": 7.560281645303266e-05,
715
+ "loss": 0.4387,
716
+ "step": 236
717
+ },
718
+ {
719
+ "epoch": 0.17,
720
+ "learning_rate": 7.551906688339762e-05,
721
+ "loss": 0.3997,
722
+ "step": 238
723
+ },
724
+ {
725
+ "epoch": 0.17,
726
+ "learning_rate": 7.543457455623413e-05,
727
+ "loss": 0.3969,
728
+ "step": 240
729
+ },
730
+ {
731
+ "epoch": 0.17,
732
+ "learning_rate": 7.534934123840486e-05,
733
+ "loss": 0.4212,
734
+ "step": 242
735
+ },
736
+ {
737
+ "epoch": 0.17,
738
+ "learning_rate": 7.526336871226763e-05,
739
+ "loss": 0.4026,
740
+ "step": 244
741
+ },
742
+ {
743
+ "epoch": 0.18,
744
+ "learning_rate": 7.51766587756383e-05,
745
+ "loss": 0.424,
746
+ "step": 246
747
+ },
748
+ {
749
+ "epoch": 0.18,
750
+ "learning_rate": 7.50892132417531e-05,
751
+ "loss": 0.4037,
752
+ "step": 248
753
+ },
754
+ {
755
+ "epoch": 0.18,
756
+ "learning_rate": 7.500103393923068e-05,
757
+ "loss": 0.395,
758
+ "step": 250
759
+ },
760
+ {
761
+ "epoch": 0.18,
762
+ "learning_rate": 7.491212271203391e-05,
763
+ "loss": 0.385,
764
+ "step": 252
765
+ },
766
+ {
767
+ "epoch": 0.18,
768
+ "learning_rate": 7.482248141943133e-05,
769
+ "loss": 0.4292,
770
+ "step": 254
771
+ },
772
+ {
773
+ "epoch": 0.18,
774
+ "learning_rate": 7.473211193595825e-05,
775
+ "loss": 0.4002,
776
+ "step": 256
777
+ },
778
+ {
779
+ "epoch": 0.18,
780
+ "learning_rate": 7.464101615137756e-05,
781
+ "loss": 0.4208,
782
+ "step": 258
783
+ },
784
+ {
785
+ "epoch": 0.19,
786
+ "learning_rate": 7.454919597064017e-05,
787
+ "loss": 0.3817,
788
+ "step": 260
789
+ },
790
+ {
791
+ "epoch": 0.19,
792
+ "learning_rate": 7.445665331384526e-05,
793
+ "loss": 0.3668,
794
+ "step": 262
795
+ },
796
+ {
797
+ "epoch": 0.19,
798
+ "learning_rate": 7.436339011620007e-05,
799
+ "loss": 0.4409,
800
+ "step": 264
801
+ },
802
+ {
803
+ "epoch": 0.19,
804
+ "learning_rate": 7.426940832797941e-05,
805
+ "loss": 0.3886,
806
+ "step": 266
807
+ },
808
+ {
809
+ "epoch": 0.19,
810
+ "learning_rate": 7.417470991448494e-05,
811
+ "loss": 0.4372,
812
+ "step": 268
813
+ },
814
+ {
815
+ "epoch": 0.19,
816
+ "learning_rate": 7.407929685600403e-05,
817
+ "loss": 0.4362,
818
+ "step": 270
819
+ },
820
+ {
821
+ "epoch": 0.19,
822
+ "learning_rate": 7.398317114776836e-05,
823
+ "loss": 0.4164,
824
+ "step": 272
825
+ },
826
+ {
827
+ "epoch": 0.2,
828
+ "learning_rate": 7.388633479991219e-05,
829
+ "loss": 0.4498,
830
+ "step": 274
831
+ },
832
+ {
833
+ "epoch": 0.2,
834
+ "learning_rate": 7.378878983743037e-05,
835
+ "loss": 0.3623,
836
+ "step": 276
837
+ },
838
+ {
839
+ "epoch": 0.2,
840
+ "learning_rate": 7.369053830013588e-05,
841
+ "loss": 0.4217,
842
+ "step": 278
843
+ },
844
+ {
845
+ "epoch": 0.2,
846
+ "learning_rate": 7.35915822426173e-05,
847
+ "loss": 0.4554,
848
+ "step": 280
849
+ },
850
+ {
851
+ "epoch": 0.2,
852
+ "learning_rate": 7.349192373419583e-05,
853
+ "loss": 0.4505,
854
+ "step": 282
855
+ },
856
+ {
857
+ "epoch": 0.2,
858
+ "learning_rate": 7.339156485888191e-05,
859
+ "loss": 0.3735,
860
+ "step": 284
861
+ },
862
+ {
863
+ "epoch": 0.2,
864
+ "learning_rate": 7.329050771533176e-05,
865
+ "loss": 0.4257,
866
+ "step": 286
867
+ },
868
+ {
869
+ "epoch": 0.21,
870
+ "learning_rate": 7.31887544168034e-05,
871
+ "loss": 0.469,
872
+ "step": 288
873
+ },
874
+ {
875
+ "epoch": 0.21,
876
+ "learning_rate": 7.308630709111261e-05,
877
+ "loss": 0.3918,
878
+ "step": 290
879
+ },
880
+ {
881
+ "epoch": 0.21,
882
+ "learning_rate": 7.29831678805882e-05,
883
+ "loss": 0.4332,
884
+ "step": 292
885
+ },
886
+ {
887
+ "epoch": 0.21,
888
+ "learning_rate": 7.287933894202743e-05,
889
+ "loss": 0.4651,
890
+ "step": 294
891
+ },
892
+ {
893
+ "epoch": 0.21,
894
+ "learning_rate": 7.277482244665079e-05,
895
+ "loss": 0.3642,
896
+ "step": 296
897
+ },
898
+ {
899
+ "epoch": 0.21,
900
+ "learning_rate": 7.266962058005659e-05,
901
+ "loss": 0.4475,
902
+ "step": 298
903
+ },
904
+ {
905
+ "epoch": 0.21,
906
+ "learning_rate": 7.256373554217536e-05,
907
+ "loss": 0.4169,
908
+ "step": 300
909
+ },
910
+ {
911
+ "epoch": 0.22,
912
+ "learning_rate": 7.245716954722374e-05,
913
+ "loss": 0.4228,
914
+ "step": 302
915
+ },
916
+ {
917
+ "epoch": 0.22,
918
+ "learning_rate": 7.234992482365821e-05,
919
+ "loss": 0.4214,
920
+ "step": 304
921
+ },
922
+ {
923
+ "epoch": 0.22,
924
+ "learning_rate": 7.22420036141285e-05,
925
+ "loss": 0.4339,
926
+ "step": 306
927
+ },
928
+ {
929
+ "epoch": 0.22,
930
+ "learning_rate": 7.21334081754307e-05,
931
+ "loss": 0.4068,
932
+ "step": 308
933
+ },
934
+ {
935
+ "epoch": 0.22,
936
+ "learning_rate": 7.202414077846003e-05,
937
+ "loss": 0.4048,
938
+ "step": 310
939
+ },
940
+ {
941
+ "epoch": 0.22,
942
+ "learning_rate": 7.191420370816339e-05,
943
+ "loss": 0.4318,
944
+ "step": 312
945
+ },
946
+ {
947
+ "epoch": 0.22,
948
+ "learning_rate": 7.180359926349158e-05,
949
+ "loss": 0.3897,
950
+ "step": 314
951
+ },
952
+ {
953
+ "epoch": 0.23,
954
+ "learning_rate": 7.169232975735115e-05,
955
+ "loss": 0.4575,
956
+ "step": 316
957
+ },
958
+ {
959
+ "epoch": 0.23,
960
+ "learning_rate": 7.158039751655619e-05,
961
+ "loss": 0.442,
962
+ "step": 318
963
+ },
964
+ {
965
+ "epoch": 0.23,
966
+ "learning_rate": 7.146780488177948e-05,
967
+ "loss": 0.3831,
968
+ "step": 320
969
+ },
970
+ {
971
+ "epoch": 0.23,
972
+ "learning_rate": 7.135455420750371e-05,
973
+ "loss": 0.3654,
974
+ "step": 322
975
+ },
976
+ {
977
+ "epoch": 0.23,
978
+ "learning_rate": 7.124064786197213e-05,
979
+ "loss": 0.4198,
980
+ "step": 324
981
+ },
982
+ {
983
+ "epoch": 0.23,
984
+ "learning_rate": 7.112608822713908e-05,
985
+ "loss": 0.4102,
986
+ "step": 326
987
+ },
988
+ {
989
+ "epoch": 0.23,
990
+ "learning_rate": 7.10108776986202e-05,
991
+ "loss": 0.375,
992
+ "step": 328
993
+ },
994
+ {
995
+ "epoch": 0.24,
996
+ "learning_rate": 7.089501868564224e-05,
997
+ "loss": 0.4253,
998
+ "step": 330
999
+ },
1000
+ {
1001
+ "epoch": 0.24,
1002
+ "learning_rate": 7.077851361099281e-05,
1003
+ "loss": 0.4087,
1004
+ "step": 332
1005
+ },
1006
+ {
1007
+ "epoch": 0.24,
1008
+ "learning_rate": 7.066136491096961e-05,
1009
+ "loss": 0.4172,
1010
+ "step": 334
1011
+ },
1012
+ {
1013
+ "epoch": 0.24,
1014
+ "learning_rate": 7.054357503532953e-05,
1015
+ "loss": 0.4306,
1016
+ "step": 336
1017
+ },
1018
+ {
1019
+ "epoch": 0.24,
1020
+ "learning_rate": 7.042514644723741e-05,
1021
+ "loss": 0.3946,
1022
+ "step": 338
1023
+ },
1024
+ {
1025
+ "epoch": 0.24,
1026
+ "learning_rate": 7.030608162321455e-05,
1027
+ "loss": 0.4295,
1028
+ "step": 340
1029
+ },
1030
+ {
1031
+ "epoch": 0.24,
1032
+ "learning_rate": 7.018638305308693e-05,
1033
+ "loss": 0.4249,
1034
+ "step": 342
1035
+ },
1036
+ {
1037
+ "epoch": 0.25,
1038
+ "learning_rate": 7.006605323993301e-05,
1039
+ "loss": 0.3931,
1040
+ "step": 344
1041
+ },
1042
+ {
1043
+ "epoch": 0.25,
1044
+ "learning_rate": 6.994509470003167e-05,
1045
+ "loss": 0.3852,
1046
+ "step": 346
1047
+ },
1048
+ {
1049
+ "epoch": 0.25,
1050
+ "learning_rate": 6.982350996280926e-05,
1051
+ "loss": 0.3973,
1052
+ "step": 348
1053
+ },
1054
+ {
1055
+ "epoch": 0.25,
1056
+ "learning_rate": 6.970130157078697e-05,
1057
+ "loss": 0.422,
1058
+ "step": 350
1059
+ },
1060
+ {
1061
+ "epoch": 0.25,
1062
+ "learning_rate": 6.957847207952752e-05,
1063
+ "loss": 0.3904,
1064
+ "step": 352
1065
+ },
1066
+ {
1067
+ "epoch": 0.25,
1068
+ "learning_rate": 6.945502405758174e-05,
1069
+ "loss": 0.3854,
1070
+ "step": 354
1071
+ },
1072
+ {
1073
+ "epoch": 0.25,
1074
+ "learning_rate": 6.933096008643489e-05,
1075
+ "loss": 0.3958,
1076
+ "step": 356
1077
+ },
1078
+ {
1079
+ "epoch": 0.25,
1080
+ "learning_rate": 6.92062827604527e-05,
1081
+ "loss": 0.4068,
1082
+ "step": 358
1083
+ },
1084
+ {
1085
+ "epoch": 0.26,
1086
+ "learning_rate": 6.9080994686827e-05,
1087
+ "loss": 0.3632,
1088
+ "step": 360
1089
+ },
1090
+ {
1091
+ "epoch": 0.26,
1092
+ "learning_rate": 6.895509848552134e-05,
1093
+ "loss": 0.4087,
1094
+ "step": 362
1095
+ },
1096
+ {
1097
+ "epoch": 0.26,
1098
+ "learning_rate": 6.882859678921612e-05,
1099
+ "loss": 0.3684,
1100
+ "step": 364
1101
+ },
1102
+ {
1103
+ "epoch": 0.26,
1104
+ "learning_rate": 6.870149224325358e-05,
1105
+ "loss": 0.4083,
1106
+ "step": 366
1107
+ },
1108
+ {
1109
+ "epoch": 0.26,
1110
+ "learning_rate": 6.85737875055824e-05,
1111
+ "loss": 0.4025,
1112
+ "step": 368
1113
+ },
1114
+ {
1115
+ "epoch": 0.26,
1116
+ "learning_rate": 6.844548524670226e-05,
1117
+ "loss": 0.3936,
1118
+ "step": 370
1119
+ },
1120
+ {
1121
+ "epoch": 0.26,
1122
+ "learning_rate": 6.831658814960786e-05,
1123
+ "loss": 0.4363,
1124
+ "step": 372
1125
+ },
1126
+ {
1127
+ "epoch": 0.27,
1128
+ "learning_rate": 6.818709890973289e-05,
1129
+ "loss": 0.4119,
1130
+ "step": 374
1131
+ },
1132
+ {
1133
+ "epoch": 0.27,
1134
+ "learning_rate": 6.805702023489361e-05,
1135
+ "loss": 0.3917,
1136
+ "step": 376
1137
+ },
1138
+ {
1139
+ "epoch": 0.27,
1140
+ "learning_rate": 6.792635484523231e-05,
1141
+ "loss": 0.4127,
1142
+ "step": 378
1143
+ },
1144
+ {
1145
+ "epoch": 0.27,
1146
+ "learning_rate": 6.779510547316034e-05,
1147
+ "loss": 0.3575,
1148
+ "step": 380
1149
+ },
1150
+ {
1151
+ "epoch": 0.27,
1152
+ "learning_rate": 6.766327486330103e-05,
1153
+ "loss": 0.3646,
1154
+ "step": 382
1155
+ },
1156
+ {
1157
+ "epoch": 0.27,
1158
+ "learning_rate": 6.753086577243225e-05,
1159
+ "loss": 0.4003,
1160
+ "step": 384
1161
+ },
1162
+ {
1163
+ "epoch": 0.27,
1164
+ "learning_rate": 6.739788096942879e-05,
1165
+ "loss": 0.4276,
1166
+ "step": 386
1167
+ },
1168
+ {
1169
+ "epoch": 0.28,
1170
+ "learning_rate": 6.726432323520447e-05,
1171
+ "loss": 0.3895,
1172
+ "step": 388
1173
+ },
1174
+ {
1175
+ "epoch": 0.28,
1176
+ "learning_rate": 6.713019536265395e-05,
1177
+ "loss": 0.392,
1178
+ "step": 390
1179
+ },
1180
+ {
1181
+ "epoch": 0.28,
1182
+ "learning_rate": 6.699550015659437e-05,
1183
+ "loss": 0.3909,
1184
+ "step": 392
1185
+ },
1186
+ {
1187
+ "epoch": 0.28,
1188
+ "learning_rate": 6.686024043370667e-05,
1189
+ "loss": 0.39,
1190
+ "step": 394
1191
+ },
1192
+ {
1193
+ "epoch": 0.28,
1194
+ "learning_rate": 6.672441902247666e-05,
1195
+ "loss": 0.3835,
1196
+ "step": 396
1197
+ },
1198
+ {
1199
+ "epoch": 0.28,
1200
+ "learning_rate": 6.658803876313592e-05,
1201
+ "loss": 0.4331,
1202
+ "step": 398
1203
+ },
1204
+ {
1205
+ "epoch": 0.28,
1206
+ "learning_rate": 6.645110250760243e-05,
1207
+ "loss": 0.3894,
1208
+ "step": 400
1209
+ },
1210
+ {
1211
+ "epoch": 0.29,
1212
+ "learning_rate": 6.631361311942085e-05,
1213
+ "loss": 0.3964,
1214
+ "step": 402
1215
+ },
1216
+ {
1217
+ "epoch": 0.29,
1218
+ "learning_rate": 6.617557347370266e-05,
1219
+ "loss": 0.3844,
1220
+ "step": 404
1221
+ },
1222
+ {
1223
+ "epoch": 0.29,
1224
+ "learning_rate": 6.603698645706614e-05,
1225
+ "loss": 0.3761,
1226
+ "step": 406
1227
+ },
1228
+ {
1229
+ "epoch": 0.29,
1230
+ "learning_rate": 6.589785496757587e-05,
1231
+ "loss": 0.399,
1232
+ "step": 408
1233
+ },
1234
+ {
1235
+ "epoch": 0.29,
1236
+ "learning_rate": 6.575818191468217e-05,
1237
+ "loss": 0.4046,
1238
+ "step": 410
1239
+ },
1240
+ {
1241
+ "epoch": 0.29,
1242
+ "learning_rate": 6.561797021916031e-05,
1243
+ "loss": 0.402,
1244
+ "step": 412
1245
+ },
1246
+ {
1247
+ "epoch": 0.29,
1248
+ "learning_rate": 6.547722281304935e-05,
1249
+ "loss": 0.3824,
1250
+ "step": 414
1251
+ },
1252
+ {
1253
+ "epoch": 0.3,
1254
+ "learning_rate": 6.533594263959095e-05,
1255
+ "loss": 0.4048,
1256
+ "step": 416
1257
+ },
1258
+ {
1259
+ "epoch": 0.3,
1260
+ "learning_rate": 6.519413265316762e-05,
1261
+ "loss": 0.382,
1262
+ "step": 418
1263
+ },
1264
+ {
1265
+ "epoch": 0.3,
1266
+ "learning_rate": 6.50517958192412e-05,
1267
+ "loss": 0.3952,
1268
+ "step": 420
1269
+ },
1270
+ {
1271
+ "epoch": 0.3,
1272
+ "learning_rate": 6.49089351142906e-05,
1273
+ "loss": 0.3711,
1274
+ "step": 422
1275
+ },
1276
+ {
1277
+ "epoch": 0.3,
1278
+ "learning_rate": 6.476555352574973e-05,
1279
+ "loss": 0.3902,
1280
+ "step": 424
1281
+ },
1282
+ {
1283
+ "epoch": 0.3,
1284
+ "learning_rate": 6.462165405194492e-05,
1285
+ "loss": 0.4164,
1286
+ "step": 426
1287
+ },
1288
+ {
1289
+ "epoch": 0.3,
1290
+ "learning_rate": 6.447723970203232e-05,
1291
+ "loss": 0.4021,
1292
+ "step": 428
1293
+ },
1294
+ {
1295
+ "epoch": 0.31,
1296
+ "learning_rate": 6.433231349593487e-05,
1297
+ "loss": 0.3853,
1298
+ "step": 430
1299
+ },
1300
+ {
1301
+ "epoch": 0.31,
1302
+ "learning_rate": 6.418687846427922e-05,
1303
+ "loss": 0.385,
1304
+ "step": 432
1305
+ },
1306
+ {
1307
+ "epoch": 0.31,
1308
+ "learning_rate": 6.404093764833235e-05,
1309
+ "loss": 0.4017,
1310
+ "step": 434
1311
+ },
1312
+ {
1313
+ "epoch": 0.31,
1314
+ "learning_rate": 6.389449409993791e-05,
1315
+ "loss": 0.3751,
1316
+ "step": 436
1317
+ },
1318
+ {
1319
+ "epoch": 0.31,
1320
+ "learning_rate": 6.374755088145253e-05,
1321
+ "loss": 0.3937,
1322
+ "step": 438
1323
+ },
1324
+ {
1325
+ "epoch": 0.31,
1326
+ "learning_rate": 6.360011106568158e-05,
1327
+ "loss": 0.3854,
1328
+ "step": 440
1329
+ },
1330
+ {
1331
+ "epoch": 0.31,
1332
+ "learning_rate": 6.345217773581512e-05,
1333
+ "loss": 0.4066,
1334
+ "step": 442
1335
+ },
1336
+ {
1337
+ "epoch": 0.32,
1338
+ "learning_rate": 6.330375398536333e-05,
1339
+ "loss": 0.4047,
1340
+ "step": 444
1341
+ },
1342
+ {
1343
+ "epoch": 0.32,
1344
+ "learning_rate": 6.315484291809181e-05,
1345
+ "loss": 0.3834,
1346
+ "step": 446
1347
+ },
1348
+ {
1349
+ "epoch": 0.32,
1350
+ "learning_rate": 6.300544764795669e-05,
1351
+ "loss": 0.4043,
1352
+ "step": 448
1353
+ },
1354
+ {
1355
+ "epoch": 0.32,
1356
+ "learning_rate": 6.285557129903954e-05,
1357
+ "loss": 0.3703,
1358
+ "step": 450
1359
+ },
1360
+ {
1361
+ "epoch": 0.32,
1362
+ "learning_rate": 6.270521700548196e-05,
1363
+ "loss": 0.3748,
1364
+ "step": 452
1365
+ },
1366
+ {
1367
+ "epoch": 0.32,
1368
+ "learning_rate": 6.255438791142017e-05,
1369
+ "loss": 0.4021,
1370
+ "step": 454
1371
+ },
1372
+ {
1373
+ "epoch": 0.32,
1374
+ "learning_rate": 6.240308717091913e-05,
1375
+ "loss": 0.3973,
1376
+ "step": 456
1377
+ },
1378
+ {
1379
+ "epoch": 0.33,
1380
+ "learning_rate": 6.22513179479067e-05,
1381
+ "loss": 0.3821,
1382
+ "step": 458
1383
+ },
1384
+ {
1385
+ "epoch": 0.33,
1386
+ "learning_rate": 6.209908341610736e-05,
1387
+ "loss": 0.4095,
1388
+ "step": 460
1389
+ },
1390
+ {
1391
+ "epoch": 0.33,
1392
+ "learning_rate": 6.194638675897591e-05,
1393
+ "loss": 0.4244,
1394
+ "step": 462
1395
+ },
1396
+ {
1397
+ "epoch": 0.33,
1398
+ "learning_rate": 6.179323116963095e-05,
1399
+ "loss": 0.3658,
1400
+ "step": 464
1401
+ },
1402
+ {
1403
+ "epoch": 0.33,
1404
+ "learning_rate": 6.163961985078798e-05,
1405
+ "loss": 0.3782,
1406
+ "step": 466
1407
+ },
1408
+ {
1409
+ "epoch": 0.33,
1410
+ "learning_rate": 6.148555601469252e-05,
1411
+ "loss": 0.4021,
1412
+ "step": 468
1413
+ },
1414
+ {
1415
+ "epoch": 0.33,
1416
+ "learning_rate": 6.133104288305293e-05,
1417
+ "loss": 0.3933,
1418
+ "step": 470
1419
+ },
1420
+ {
1421
+ "epoch": 0.34,
1422
+ "learning_rate": 6.1176083686973e-05,
1423
+ "loss": 0.3814,
1424
+ "step": 472
1425
+ },
1426
+ {
1427
+ "epoch": 0.34,
1428
+ "learning_rate": 6.1020681666884406e-05,
1429
+ "loss": 0.3991,
1430
+ "step": 474
1431
+ },
1432
+ {
1433
+ "epoch": 0.34,
1434
+ "learning_rate": 6.086484007247894e-05,
1435
+ "loss": 0.4124,
1436
+ "step": 476
1437
+ },
1438
+ {
1439
+ "epoch": 0.34,
1440
+ "learning_rate": 6.070856216264057e-05,
1441
+ "loss": 0.369,
1442
+ "step": 478
1443
+ },
1444
+ {
1445
+ "epoch": 0.34,
1446
+ "learning_rate": 6.055185120537728e-05,
1447
+ "loss": 0.379,
1448
+ "step": 480
1449
+ },
1450
+ {
1451
+ "epoch": 0.34,
1452
+ "learning_rate": 6.039471047775273e-05,
1453
+ "loss": 0.4135,
1454
+ "step": 482
1455
+ },
1456
+ {
1457
+ "epoch": 0.34,
1458
+ "learning_rate": 6.023714326581772e-05,
1459
+ "loss": 0.3477,
1460
+ "step": 484
1461
+ },
1462
+ {
1463
+ "epoch": 0.35,
1464
+ "learning_rate": 6.0079152864541506e-05,
1465
+ "loss": 0.4139,
1466
+ "step": 486
1467
+ },
1468
+ {
1469
+ "epoch": 0.35,
1470
+ "learning_rate": 5.992074257774285e-05,
1471
+ "loss": 0.3711,
1472
+ "step": 488
1473
+ },
1474
+ {
1475
+ "epoch": 0.35,
1476
+ "learning_rate": 5.976191571802097e-05,
1477
+ "loss": 0.4052,
1478
+ "step": 490
1479
+ },
1480
+ {
1481
+ "epoch": 0.35,
1482
+ "learning_rate": 5.960267560668625e-05,
1483
+ "loss": 0.402,
1484
+ "step": 492
1485
+ },
1486
+ {
1487
+ "epoch": 0.35,
1488
+ "learning_rate": 5.9443025573690795e-05,
1489
+ "loss": 0.3992,
1490
+ "step": 494
1491
+ },
1492
+ {
1493
+ "epoch": 0.35,
1494
+ "learning_rate": 5.928296895755882e-05,
1495
+ "loss": 0.3895,
1496
+ "step": 496
1497
+ },
1498
+ {
1499
+ "epoch": 0.35,
1500
+ "learning_rate": 5.9122509105316736e-05,
1501
+ "loss": 0.4004,
1502
+ "step": 498
1503
+ },
1504
+ {
1505
+ "epoch": 0.36,
1506
+ "learning_rate": 5.896164937242331e-05,
1507
+ "loss": 0.3736,
1508
+ "step": 500
1509
+ },
1510
+ {
1511
+ "epoch": 0.36,
1512
+ "learning_rate": 5.880039312269939e-05,
1513
+ "loss": 0.3452,
1514
+ "step": 502
1515
+ },
1516
+ {
1517
+ "epoch": 0.36,
1518
+ "learning_rate": 5.863874372825757e-05,
1519
+ "loss": 0.3806,
1520
+ "step": 504
1521
+ },
1522
+ {
1523
+ "epoch": 0.36,
1524
+ "learning_rate": 5.847670456943174e-05,
1525
+ "loss": 0.3684,
1526
+ "step": 506
1527
+ },
1528
+ {
1529
+ "epoch": 0.36,
1530
+ "learning_rate": 5.831427903470631e-05,
1531
+ "loss": 0.3891,
1532
+ "step": 508
1533
+ },
1534
+ {
1535
+ "epoch": 0.36,
1536
+ "learning_rate": 5.815147052064543e-05,
1537
+ "loss": 0.3882,
1538
+ "step": 510
1539
+ },
1540
+ {
1541
+ "epoch": 0.36,
1542
+ "learning_rate": 5.798828243182191e-05,
1543
+ "loss": 0.3889,
1544
+ "step": 512
1545
+ },
1546
+ {
1547
+ "epoch": 0.37,
1548
+ "learning_rate": 5.7824718180746034e-05,
1549
+ "loss": 0.3882,
1550
+ "step": 514
1551
+ },
1552
+ {
1553
+ "epoch": 0.37,
1554
+ "learning_rate": 5.766078118779424e-05,
1555
+ "loss": 0.3863,
1556
+ "step": 516
1557
+ },
1558
+ {
1559
+ "epoch": 0.37,
1560
+ "learning_rate": 5.749647488113755e-05,
1561
+ "loss": 0.3629,
1562
+ "step": 518
1563
+ },
1564
+ {
1565
+ "epoch": 0.37,
1566
+ "learning_rate": 5.7331802696669885e-05,
1567
+ "loss": 0.3666,
1568
+ "step": 520
1569
+ },
1570
+ {
1571
+ "epoch": 0.37,
1572
+ "learning_rate": 5.716676807793624e-05,
1573
+ "loss": 0.3499,
1574
+ "step": 522
1575
+ },
1576
+ {
1577
+ "epoch": 0.37,
1578
+ "learning_rate": 5.7001374476060646e-05,
1579
+ "loss": 0.4004,
1580
+ "step": 524
1581
+ },
1582
+ {
1583
+ "epoch": 0.37,
1584
+ "learning_rate": 5.683562534967401e-05,
1585
+ "loss": 0.3468,
1586
+ "step": 526
1587
+ },
1588
+ {
1589
+ "epoch": 0.38,
1590
+ "learning_rate": 5.666952416484183e-05,
1591
+ "loss": 0.3729,
1592
+ "step": 528
1593
+ },
1594
+ {
1595
+ "epoch": 0.38,
1596
+ "learning_rate": 5.650307439499164e-05,
1597
+ "loss": 0.3763,
1598
+ "step": 530
1599
+ },
1600
+ {
1601
+ "epoch": 0.38,
1602
+ "learning_rate": 5.6336279520840445e-05,
1603
+ "loss": 0.3543,
1604
+ "step": 532
1605
+ },
1606
+ {
1607
+ "epoch": 0.38,
1608
+ "learning_rate": 5.61691430303219e-05,
1609
+ "loss": 0.4355,
1610
+ "step": 534
1611
+ },
1612
+ {
1613
+ "epoch": 0.38,
1614
+ "learning_rate": 5.600166841851333e-05,
1615
+ "loss": 0.3764,
1616
+ "step": 536
1617
+ },
1618
+ {
1619
+ "epoch": 0.38,
1620
+ "learning_rate": 5.583385918756277e-05,
1621
+ "loss": 0.3594,
1622
+ "step": 538
1623
+ },
1624
+ {
1625
+ "epoch": 0.38,
1626
+ "learning_rate": 5.5665718846615584e-05,
1627
+ "loss": 0.3589,
1628
+ "step": 540
1629
+ },
1630
+ {
1631
+ "epoch": 0.39,
1632
+ "learning_rate": 5.549725091174117e-05,
1633
+ "loss": 0.3726,
1634
+ "step": 542
1635
+ },
1636
+ {
1637
+ "epoch": 0.39,
1638
+ "learning_rate": 5.532845890585941e-05,
1639
+ "loss": 0.3803,
1640
+ "step": 544
1641
+ },
1642
+ {
1643
+ "epoch": 0.39,
1644
+ "learning_rate": 5.5159346358667e-05,
1645
+ "loss": 0.3661,
1646
+ "step": 546
1647
+ },
1648
+ {
1649
+ "epoch": 0.39,
1650
+ "learning_rate": 5.4989916806563655e-05,
1651
+ "loss": 0.3902,
1652
+ "step": 548
1653
+ },
1654
+ {
1655
+ "epoch": 0.39,
1656
+ "learning_rate": 5.482017379257812e-05,
1657
+ "loss": 0.3953,
1658
+ "step": 550
1659
+ },
1660
+ {
1661
+ "epoch": 0.39,
1662
+ "learning_rate": 5.465012086629411e-05,
1663
+ "loss": 0.3633,
1664
+ "step": 552
1665
+ },
1666
+ {
1667
+ "epoch": 0.39,
1668
+ "learning_rate": 5.447976158377611e-05,
1669
+ "loss": 0.3559,
1670
+ "step": 554
1671
+ },
1672
+ {
1673
+ "epoch": 0.4,
1674
+ "learning_rate": 5.430909950749491e-05,
1675
+ "loss": 0.3633,
1676
+ "step": 556
1677
+ },
1678
+ {
1679
+ "epoch": 0.4,
1680
+ "learning_rate": 5.413813820625324e-05,
1681
+ "loss": 0.3553,
1682
+ "step": 558
1683
+ },
1684
+ {
1685
+ "epoch": 0.4,
1686
+ "learning_rate": 5.3966881255111024e-05,
1687
+ "loss": 0.3908,
1688
+ "step": 560
1689
+ },
1690
+ {
1691
+ "epoch": 0.4,
1692
+ "learning_rate": 5.3795332235310744e-05,
1693
+ "loss": 0.3674,
1694
+ "step": 562
1695
+ },
1696
+ {
1697
+ "epoch": 0.4,
1698
+ "learning_rate": 5.3623494734202415e-05,
1699
+ "loss": 0.4049,
1700
+ "step": 564
1701
+ },
1702
+ {
1703
+ "epoch": 0.4,
1704
+ "learning_rate": 5.345137234516864e-05,
1705
+ "loss": 0.351,
1706
+ "step": 566
1707
+ },
1708
+ {
1709
+ "epoch": 0.4,
1710
+ "learning_rate": 5.327896866754953e-05,
1711
+ "loss": 0.3518,
1712
+ "step": 568
1713
+ },
1714
+ {
1715
+ "epoch": 0.41,
1716
+ "learning_rate": 5.310628730656726e-05,
1717
+ "loss": 0.3458,
1718
+ "step": 570
1719
+ },
1720
+ {
1721
+ "epoch": 0.41,
1722
+ "learning_rate": 5.293333187325086e-05,
1723
+ "loss": 0.3756,
1724
+ "step": 572
1725
+ },
1726
+ {
1727
+ "epoch": 0.41,
1728
+ "learning_rate": 5.276010598436061e-05,
1729
+ "loss": 0.3508,
1730
+ "step": 574
1731
+ },
1732
+ {
1733
+ "epoch": 0.41,
1734
+ "learning_rate": 5.2586613262312414e-05,
1735
+ "loss": 0.3665,
1736
+ "step": 576
1737
+ },
1738
+ {
1739
+ "epoch": 0.41,
1740
+ "learning_rate": 5.241285733510209e-05,
1741
+ "loss": 0.3502,
1742
+ "step": 578
1743
+ },
1744
+ {
1745
+ "epoch": 0.41,
1746
+ "learning_rate": 5.223884183622943e-05,
1747
+ "loss": 0.3846,
1748
+ "step": 580
1749
+ },
1750
+ {
1751
+ "epoch": 0.41,
1752
+ "learning_rate": 5.20645704046223e-05,
1753
+ "loss": 0.3475,
1754
+ "step": 582
1755
+ },
1756
+ {
1757
+ "epoch": 0.42,
1758
+ "learning_rate": 5.189004668456051e-05,
1759
+ "loss": 0.4049,
1760
+ "step": 584
1761
+ },
1762
+ {
1763
+ "epoch": 0.42,
1764
+ "learning_rate": 5.171527432559957e-05,
1765
+ "loss": 0.3801,
1766
+ "step": 586
1767
+ },
1768
+ {
1769
+ "epoch": 0.42,
1770
+ "learning_rate": 5.1540256982494425e-05,
1771
+ "loss": 0.4386,
1772
+ "step": 588
1773
+ },
1774
+ {
1775
+ "epoch": 0.42,
1776
+ "learning_rate": 5.1364998315123024e-05,
1777
+ "loss": 0.3385,
1778
+ "step": 590
1779
+ },
1780
+ {
1781
+ "epoch": 0.42,
1782
+ "learning_rate": 5.118950198840977e-05,
1783
+ "loss": 0.3704,
1784
+ "step": 592
1785
+ },
1786
+ {
1787
+ "epoch": 0.42,
1788
+ "learning_rate": 5.101377167224885e-05,
1789
+ "loss": 0.3842,
1790
+ "step": 594
1791
+ },
1792
+ {
1793
+ "epoch": 0.42,
1794
+ "learning_rate": 5.0837811041427534e-05,
1795
+ "loss": 0.3807,
1796
+ "step": 596
1797
+ },
1798
+ {
1799
+ "epoch": 0.43,
1800
+ "learning_rate": 5.0661623775549376e-05,
1801
+ "loss": 0.377,
1802
+ "step": 598
1803
+ },
1804
+ {
1805
+ "epoch": 0.43,
1806
+ "learning_rate": 5.048521355895714e-05,
1807
+ "loss": 0.3896,
1808
+ "step": 600
1809
+ },
1810
+ {
1811
+ "epoch": 0.43,
1812
+ "learning_rate": 5.030858408065587e-05,
1813
+ "loss": 0.3721,
1814
+ "step": 602
1815
+ },
1816
+ {
1817
+ "epoch": 0.43,
1818
+ "learning_rate": 5.01317390342357e-05,
1819
+ "loss": 0.3744,
1820
+ "step": 604
1821
+ },
1822
+ {
1823
+ "epoch": 0.43,
1824
+ "learning_rate": 4.995468211779459e-05,
1825
+ "loss": 0.4038,
1826
+ "step": 606
1827
+ },
1828
+ {
1829
+ "epoch": 0.43,
1830
+ "learning_rate": 4.977741703386107e-05,
1831
+ "loss": 0.3491,
1832
+ "step": 608
1833
+ },
1834
+ {
1835
+ "epoch": 0.43,
1836
+ "learning_rate": 4.9599947489316725e-05,
1837
+ "loss": 0.3655,
1838
+ "step": 610
1839
+ },
1840
+ {
1841
+ "epoch": 0.44,
1842
+ "learning_rate": 4.942227719531875e-05,
1843
+ "loss": 0.3962,
1844
+ "step": 612
1845
+ },
1846
+ {
1847
+ "epoch": 0.44,
1848
+ "learning_rate": 4.924440986722231e-05,
1849
+ "loss": 0.3642,
1850
+ "step": 614
1851
+ },
1852
+ {
1853
+ "epoch": 0.44,
1854
+ "learning_rate": 4.906634922450285e-05,
1855
+ "loss": 0.3703,
1856
+ "step": 616
1857
+ },
1858
+ {
1859
+ "epoch": 0.44,
1860
+ "learning_rate": 4.888809899067829e-05,
1861
+ "loss": 0.3628,
1862
+ "step": 618
1863
+ },
1864
+ {
1865
+ "epoch": 0.44,
1866
+ "learning_rate": 4.870966289323123e-05,
1867
+ "loss": 0.3639,
1868
+ "step": 620
1869
+ },
1870
+ {
1871
+ "epoch": 0.44,
1872
+ "learning_rate": 4.8531044663530944e-05,
1873
+ "loss": 0.3929,
1874
+ "step": 622
1875
+ },
1876
+ {
1877
+ "epoch": 0.44,
1878
+ "learning_rate": 4.835224803675537e-05,
1879
+ "loss": 0.3438,
1880
+ "step": 624
1881
+ },
1882
+ {
1883
+ "epoch": 0.45,
1884
+ "learning_rate": 4.817327675181297e-05,
1885
+ "loss": 0.3349,
1886
+ "step": 626
1887
+ },
1888
+ {
1889
+ "epoch": 0.45,
1890
+ "learning_rate": 4.799413455126461e-05,
1891
+ "loss": 0.3395,
1892
+ "step": 628
1893
+ },
1894
+ {
1895
+ "epoch": 0.45,
1896
+ "learning_rate": 4.7814825181245254e-05,
1897
+ "loss": 0.3754,
1898
+ "step": 630
1899
+ },
1900
+ {
1901
+ "epoch": 0.45,
1902
+ "learning_rate": 4.763535239138562e-05,
1903
+ "loss": 0.3264,
1904
+ "step": 632
1905
+ },
1906
+ {
1907
+ "epoch": 0.45,
1908
+ "learning_rate": 4.74557199347338e-05,
1909
+ "loss": 0.3946,
1910
+ "step": 634
1911
+ },
1912
+ {
1913
+ "epoch": 0.45,
1914
+ "learning_rate": 4.7275931567676735e-05,
1915
+ "loss": 0.3699,
1916
+ "step": 636
1917
+ },
1918
+ {
1919
+ "epoch": 0.45,
1920
+ "learning_rate": 4.709599104986172e-05,
1921
+ "loss": 0.3708,
1922
+ "step": 638
1923
+ },
1924
+ {
1925
+ "epoch": 0.46,
1926
+ "learning_rate": 4.691590214411772e-05,
1927
+ "loss": 0.3557,
1928
+ "step": 640
1929
+ },
1930
+ {
1931
+ "epoch": 0.46,
1932
+ "learning_rate": 4.6735668616376734e-05,
1933
+ "loss": 0.3537,
1934
+ "step": 642
1935
+ },
1936
+ {
1937
+ "epoch": 0.46,
1938
+ "learning_rate": 4.6555294235595047e-05,
1939
+ "loss": 0.3601,
1940
+ "step": 644
1941
+ },
1942
+ {
1943
+ "epoch": 0.46,
1944
+ "learning_rate": 4.637478277367436e-05,
1945
+ "loss": 0.366,
1946
+ "step": 646
1947
+ },
1948
+ {
1949
+ "epoch": 0.46,
1950
+ "learning_rate": 4.6194138005382945e-05,
1951
+ "loss": 0.3928,
1952
+ "step": 648
1953
+ },
1954
+ {
1955
+ "epoch": 0.46,
1956
+ "learning_rate": 4.6013363708276754e-05,
1957
+ "loss": 0.3209,
1958
+ "step": 650
1959
+ },
1960
+ {
1961
+ "epoch": 0.46,
1962
+ "learning_rate": 4.5832463662620353e-05,
1963
+ "loss": 0.3796,
1964
+ "step": 652
1965
+ }
1966
+ ],
1967
+ "max_steps": 1403,
1968
+ "num_train_epochs": 1,
1969
+ "total_flos": 3.846934282119414e+17,
1970
+ "trial_name": null,
1971
+ "trial_params": null
1972
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9715a94023820b5bb9d8a534f094cdd86d4cdccc47f1614154402c625f96dc73
3
+ size 4527