Commit ·
4e13d83
1
Parent(s): d6ff5b6
added checkpoints
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/trainer_state.json +217 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/trainer_state.json +319 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/trainer_state.json +421 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/trainer_state.json +523 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/trainer_state.json +625 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/trainer_state.json +727 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/config.json +40 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/optimizer.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/pytorch_model.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/rng_state.pth +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/scheduler.pt +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/trainer_state.json +829 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/training_args.bin +3 -0
- rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-67338/config.json +40 -0
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ec7b17b24be98bb5061bee26e48fcb0342f48a66630e356ef3575d4d7cba6d3
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45fd50b290bd60448e72ce386cbd24817b01c712099ea20c4888d14be5d08d06
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e6c4fae2dbc4e09dcb1916b3513181a62899bf27dcfddffe76449df6be937a2
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45c6f90078a2450ac8516906119b963c86e82b14a8e8f9482b75b9f2e26282b4
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/trainer_state.json
ADDED
|
@@ -0,0 +1,217 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 14964,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
}
|
| 209 |
+
],
|
| 210 |
+
"logging_steps": 500,
|
| 211 |
+
"max_steps": 74820,
|
| 212 |
+
"num_train_epochs": 10,
|
| 213 |
+
"save_steps": 500,
|
| 214 |
+
"total_flos": 1.7008214784542912e+17,
|
| 215 |
+
"trial_name": null,
|
| 216 |
+
"trial_params": null
|
| 217 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-14964/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fd2e501a8385c447d794fe6ee9b83ef4c2e239115db3ba191ebc165a10e527f
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c233ab490e4a16384978560e9ce243b94e4d0edde66e9c7736ac72bc328cbed
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26c94452baca72f151e8fd71f227f974658fc15f84f8df12aa691d214f2d3bb1
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95aa11d132c66709e53ab1861363db19e03cdd43abe90abb88ecfb1d00f88ed7
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/trainer_state.json
ADDED
|
@@ -0,0 +1,319 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 22446,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
}
|
| 311 |
+
],
|
| 312 |
+
"logging_steps": 500,
|
| 313 |
+
"max_steps": 74820,
|
| 314 |
+
"num_train_epochs": 10,
|
| 315 |
+
"save_steps": 500,
|
| 316 |
+
"total_flos": 2.5512322176814368e+17,
|
| 317 |
+
"trial_name": null,
|
| 318 |
+
"trial_params": null
|
| 319 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-22446/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5a688150a58135e72dadfc5d4565bc907692c77456484bef730d739ff3a3399
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ad549c3c9a145d0ba0d3601a1b8871ffef6ec50f3964391cf383c6aeb10a44da
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9bec1569f2b2b19949906563cbd6e2d8a69b35607a8366b6695d838b750331b
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5c5e1d1870ca9edfabe43cd9960ac818f48f088d52b8f2b3a331c9e576d215a
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 29928,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.01,
|
| 313 |
+
"learning_rate": 1.4079655543595265e-05,
|
| 314 |
+
"loss": 0.0074,
|
| 315 |
+
"step": 22500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.07,
|
| 319 |
+
"learning_rate": 1.3945102260495158e-05,
|
| 320 |
+
"loss": 0.0051,
|
| 321 |
+
"step": 23000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.14,
|
| 325 |
+
"learning_rate": 1.3810548977395049e-05,
|
| 326 |
+
"loss": 0.0052,
|
| 327 |
+
"step": 23500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.21,
|
| 331 |
+
"learning_rate": 1.3675995694294942e-05,
|
| 332 |
+
"loss": 0.005,
|
| 333 |
+
"step": 24000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 1.3541442411194833e-05,
|
| 338 |
+
"loss": 0.0056,
|
| 339 |
+
"step": 24500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.34,
|
| 343 |
+
"learning_rate": 1.3406889128094726e-05,
|
| 344 |
+
"loss": 0.0049,
|
| 345 |
+
"step": 25000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.41,
|
| 349 |
+
"learning_rate": 1.3272335844994619e-05,
|
| 350 |
+
"loss": 0.0054,
|
| 351 |
+
"step": 25500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.48,
|
| 355 |
+
"learning_rate": 1.3137782561894512e-05,
|
| 356 |
+
"loss": 0.0052,
|
| 357 |
+
"step": 26000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.54,
|
| 361 |
+
"learning_rate": 1.3003229278794403e-05,
|
| 362 |
+
"loss": 0.0055,
|
| 363 |
+
"step": 26500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.61,
|
| 367 |
+
"learning_rate": 1.2868675995694295e-05,
|
| 368 |
+
"loss": 0.0053,
|
| 369 |
+
"step": 27000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.68,
|
| 373 |
+
"learning_rate": 1.2734122712594188e-05,
|
| 374 |
+
"loss": 0.0055,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.74,
|
| 379 |
+
"learning_rate": 1.259956942949408e-05,
|
| 380 |
+
"loss": 0.0056,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.81,
|
| 385 |
+
"learning_rate": 1.2465016146393974e-05,
|
| 386 |
+
"loss": 0.0056,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.88,
|
| 391 |
+
"learning_rate": 1.2330462863293867e-05,
|
| 392 |
+
"loss": 0.0056,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.94,
|
| 397 |
+
"learning_rate": 1.219590958019376e-05,
|
| 398 |
+
"loss": 0.0056,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"eval_accuracy": 0.9940216536800652,
|
| 404 |
+
"eval_f1": 0.0,
|
| 405 |
+
"eval_loss": 0.023068198934197426,
|
| 406 |
+
"eval_precision": 0.0,
|
| 407 |
+
"eval_recall": 0.0,
|
| 408 |
+
"eval_runtime": 81.8028,
|
| 409 |
+
"eval_samples_per_second": 616.153,
|
| 410 |
+
"eval_steps_per_second": 4.816,
|
| 411 |
+
"step": 29928
|
| 412 |
+
}
|
| 413 |
+
],
|
| 414 |
+
"logging_steps": 500,
|
| 415 |
+
"max_steps": 74820,
|
| 416 |
+
"num_train_epochs": 10,
|
| 417 |
+
"save_steps": 500,
|
| 418 |
+
"total_flos": 3.4016429569085824e+17,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-29928/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d46bdaf7365344d25ab56a273686b443074ac5df5b3da18af552adfb8bb0159
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:420b30ad532637403259d1699e64db702f87677d6643105e5e5d9d86ebb4f92e
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21ea456c1c79bc9fb04e070c6d6156fad4a9b07018a34e6d964fb53c806593e2
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0dcee0eaa0686cccc9bf89fbdc18b90fe7a8add00f6fc71d0c5923474a451192
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/trainer_state.json
ADDED
|
@@ -0,0 +1,523 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 37410,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.01,
|
| 313 |
+
"learning_rate": 1.4079655543595265e-05,
|
| 314 |
+
"loss": 0.0074,
|
| 315 |
+
"step": 22500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.07,
|
| 319 |
+
"learning_rate": 1.3945102260495158e-05,
|
| 320 |
+
"loss": 0.0051,
|
| 321 |
+
"step": 23000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.14,
|
| 325 |
+
"learning_rate": 1.3810548977395049e-05,
|
| 326 |
+
"loss": 0.0052,
|
| 327 |
+
"step": 23500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.21,
|
| 331 |
+
"learning_rate": 1.3675995694294942e-05,
|
| 332 |
+
"loss": 0.005,
|
| 333 |
+
"step": 24000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 1.3541442411194833e-05,
|
| 338 |
+
"loss": 0.0056,
|
| 339 |
+
"step": 24500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.34,
|
| 343 |
+
"learning_rate": 1.3406889128094726e-05,
|
| 344 |
+
"loss": 0.0049,
|
| 345 |
+
"step": 25000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.41,
|
| 349 |
+
"learning_rate": 1.3272335844994619e-05,
|
| 350 |
+
"loss": 0.0054,
|
| 351 |
+
"step": 25500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.48,
|
| 355 |
+
"learning_rate": 1.3137782561894512e-05,
|
| 356 |
+
"loss": 0.0052,
|
| 357 |
+
"step": 26000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.54,
|
| 361 |
+
"learning_rate": 1.3003229278794403e-05,
|
| 362 |
+
"loss": 0.0055,
|
| 363 |
+
"step": 26500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.61,
|
| 367 |
+
"learning_rate": 1.2868675995694295e-05,
|
| 368 |
+
"loss": 0.0053,
|
| 369 |
+
"step": 27000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.68,
|
| 373 |
+
"learning_rate": 1.2734122712594188e-05,
|
| 374 |
+
"loss": 0.0055,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.74,
|
| 379 |
+
"learning_rate": 1.259956942949408e-05,
|
| 380 |
+
"loss": 0.0056,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.81,
|
| 385 |
+
"learning_rate": 1.2465016146393974e-05,
|
| 386 |
+
"loss": 0.0056,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.88,
|
| 391 |
+
"learning_rate": 1.2330462863293867e-05,
|
| 392 |
+
"loss": 0.0056,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.94,
|
| 397 |
+
"learning_rate": 1.219590958019376e-05,
|
| 398 |
+
"loss": 0.0056,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"eval_accuracy": 0.9940216536800652,
|
| 404 |
+
"eval_f1": 0.0,
|
| 405 |
+
"eval_loss": 0.023068198934197426,
|
| 406 |
+
"eval_precision": 0.0,
|
| 407 |
+
"eval_recall": 0.0,
|
| 408 |
+
"eval_runtime": 81.8028,
|
| 409 |
+
"eval_samples_per_second": 616.153,
|
| 410 |
+
"eval_steps_per_second": 4.816,
|
| 411 |
+
"step": 29928
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.01,
|
| 415 |
+
"learning_rate": 1.206135629709365e-05,
|
| 416 |
+
"loss": 0.0053,
|
| 417 |
+
"step": 30000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.08,
|
| 421 |
+
"learning_rate": 1.1926803013993542e-05,
|
| 422 |
+
"loss": 0.0035,
|
| 423 |
+
"step": 30500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.14,
|
| 427 |
+
"learning_rate": 1.1792249730893435e-05,
|
| 428 |
+
"loss": 0.0037,
|
| 429 |
+
"step": 31000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.21,
|
| 433 |
+
"learning_rate": 1.1657696447793328e-05,
|
| 434 |
+
"loss": 0.0039,
|
| 435 |
+
"step": 31500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.28,
|
| 439 |
+
"learning_rate": 1.152314316469322e-05,
|
| 440 |
+
"loss": 0.0037,
|
| 441 |
+
"step": 32000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.34,
|
| 445 |
+
"learning_rate": 1.1388589881593113e-05,
|
| 446 |
+
"loss": 0.0038,
|
| 447 |
+
"step": 32500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.41,
|
| 451 |
+
"learning_rate": 1.1254036598493004e-05,
|
| 452 |
+
"loss": 0.004,
|
| 453 |
+
"step": 33000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.48,
|
| 457 |
+
"learning_rate": 1.1119483315392897e-05,
|
| 458 |
+
"loss": 0.0039,
|
| 459 |
+
"step": 33500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.54,
|
| 463 |
+
"learning_rate": 1.0984930032292788e-05,
|
| 464 |
+
"loss": 0.004,
|
| 465 |
+
"step": 34000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.61,
|
| 469 |
+
"learning_rate": 1.0850376749192681e-05,
|
| 470 |
+
"loss": 0.0041,
|
| 471 |
+
"step": 34500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.68,
|
| 475 |
+
"learning_rate": 1.0715823466092574e-05,
|
| 476 |
+
"loss": 0.0042,
|
| 477 |
+
"step": 35000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.74,
|
| 481 |
+
"learning_rate": 1.0581270182992467e-05,
|
| 482 |
+
"loss": 0.004,
|
| 483 |
+
"step": 35500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.81,
|
| 487 |
+
"learning_rate": 1.0446716899892357e-05,
|
| 488 |
+
"loss": 0.0042,
|
| 489 |
+
"step": 36000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.88,
|
| 493 |
+
"learning_rate": 1.031216361679225e-05,
|
| 494 |
+
"loss": 0.0041,
|
| 495 |
+
"step": 36500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 4.95,
|
| 499 |
+
"learning_rate": 1.0177610333692143e-05,
|
| 500 |
+
"loss": 0.0042,
|
| 501 |
+
"step": 37000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 5.0,
|
| 505 |
+
"eval_accuracy": 0.9940509055275022,
|
| 506 |
+
"eval_f1": 0.0,
|
| 507 |
+
"eval_loss": 0.025542501360177994,
|
| 508 |
+
"eval_precision": 0.0,
|
| 509 |
+
"eval_recall": 0.0,
|
| 510 |
+
"eval_runtime": 71.0096,
|
| 511 |
+
"eval_samples_per_second": 709.806,
|
| 512 |
+
"eval_steps_per_second": 5.549,
|
| 513 |
+
"step": 37410
|
| 514 |
+
}
|
| 515 |
+
],
|
| 516 |
+
"logging_steps": 500,
|
| 517 |
+
"max_steps": 74820,
|
| 518 |
+
"num_train_epochs": 10,
|
| 519 |
+
"save_steps": 500,
|
| 520 |
+
"total_flos": 4.252053696135728e+17,
|
| 521 |
+
"trial_name": null,
|
| 522 |
+
"trial_params": null
|
| 523 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-37410/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9de640f5776860cd313a47097c975e864bb42ca47fc51730f89698d9d2857cc4
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bab6353e25df2b251f3b7462ad687d29b3d68929b15b559c28369de6db587918
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2ee5b3486670d33998adce168a823ad85a897e28c706d6b3b4c7d317ac3008b
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04b8e9929c5c026cdb0866b798ad37fddb86cbbceb958e03b72203a2e6c7d4dd
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/trainer_state.json
ADDED
|
@@ -0,0 +1,625 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 44892,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.01,
|
| 313 |
+
"learning_rate": 1.4079655543595265e-05,
|
| 314 |
+
"loss": 0.0074,
|
| 315 |
+
"step": 22500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.07,
|
| 319 |
+
"learning_rate": 1.3945102260495158e-05,
|
| 320 |
+
"loss": 0.0051,
|
| 321 |
+
"step": 23000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.14,
|
| 325 |
+
"learning_rate": 1.3810548977395049e-05,
|
| 326 |
+
"loss": 0.0052,
|
| 327 |
+
"step": 23500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.21,
|
| 331 |
+
"learning_rate": 1.3675995694294942e-05,
|
| 332 |
+
"loss": 0.005,
|
| 333 |
+
"step": 24000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 1.3541442411194833e-05,
|
| 338 |
+
"loss": 0.0056,
|
| 339 |
+
"step": 24500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.34,
|
| 343 |
+
"learning_rate": 1.3406889128094726e-05,
|
| 344 |
+
"loss": 0.0049,
|
| 345 |
+
"step": 25000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.41,
|
| 349 |
+
"learning_rate": 1.3272335844994619e-05,
|
| 350 |
+
"loss": 0.0054,
|
| 351 |
+
"step": 25500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.48,
|
| 355 |
+
"learning_rate": 1.3137782561894512e-05,
|
| 356 |
+
"loss": 0.0052,
|
| 357 |
+
"step": 26000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.54,
|
| 361 |
+
"learning_rate": 1.3003229278794403e-05,
|
| 362 |
+
"loss": 0.0055,
|
| 363 |
+
"step": 26500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.61,
|
| 367 |
+
"learning_rate": 1.2868675995694295e-05,
|
| 368 |
+
"loss": 0.0053,
|
| 369 |
+
"step": 27000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.68,
|
| 373 |
+
"learning_rate": 1.2734122712594188e-05,
|
| 374 |
+
"loss": 0.0055,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.74,
|
| 379 |
+
"learning_rate": 1.259956942949408e-05,
|
| 380 |
+
"loss": 0.0056,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.81,
|
| 385 |
+
"learning_rate": 1.2465016146393974e-05,
|
| 386 |
+
"loss": 0.0056,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.88,
|
| 391 |
+
"learning_rate": 1.2330462863293867e-05,
|
| 392 |
+
"loss": 0.0056,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.94,
|
| 397 |
+
"learning_rate": 1.219590958019376e-05,
|
| 398 |
+
"loss": 0.0056,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"eval_accuracy": 0.9940216536800652,
|
| 404 |
+
"eval_f1": 0.0,
|
| 405 |
+
"eval_loss": 0.023068198934197426,
|
| 406 |
+
"eval_precision": 0.0,
|
| 407 |
+
"eval_recall": 0.0,
|
| 408 |
+
"eval_runtime": 81.8028,
|
| 409 |
+
"eval_samples_per_second": 616.153,
|
| 410 |
+
"eval_steps_per_second": 4.816,
|
| 411 |
+
"step": 29928
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.01,
|
| 415 |
+
"learning_rate": 1.206135629709365e-05,
|
| 416 |
+
"loss": 0.0053,
|
| 417 |
+
"step": 30000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.08,
|
| 421 |
+
"learning_rate": 1.1926803013993542e-05,
|
| 422 |
+
"loss": 0.0035,
|
| 423 |
+
"step": 30500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.14,
|
| 427 |
+
"learning_rate": 1.1792249730893435e-05,
|
| 428 |
+
"loss": 0.0037,
|
| 429 |
+
"step": 31000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.21,
|
| 433 |
+
"learning_rate": 1.1657696447793328e-05,
|
| 434 |
+
"loss": 0.0039,
|
| 435 |
+
"step": 31500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.28,
|
| 439 |
+
"learning_rate": 1.152314316469322e-05,
|
| 440 |
+
"loss": 0.0037,
|
| 441 |
+
"step": 32000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.34,
|
| 445 |
+
"learning_rate": 1.1388589881593113e-05,
|
| 446 |
+
"loss": 0.0038,
|
| 447 |
+
"step": 32500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.41,
|
| 451 |
+
"learning_rate": 1.1254036598493004e-05,
|
| 452 |
+
"loss": 0.004,
|
| 453 |
+
"step": 33000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.48,
|
| 457 |
+
"learning_rate": 1.1119483315392897e-05,
|
| 458 |
+
"loss": 0.0039,
|
| 459 |
+
"step": 33500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.54,
|
| 463 |
+
"learning_rate": 1.0984930032292788e-05,
|
| 464 |
+
"loss": 0.004,
|
| 465 |
+
"step": 34000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.61,
|
| 469 |
+
"learning_rate": 1.0850376749192681e-05,
|
| 470 |
+
"loss": 0.0041,
|
| 471 |
+
"step": 34500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.68,
|
| 475 |
+
"learning_rate": 1.0715823466092574e-05,
|
| 476 |
+
"loss": 0.0042,
|
| 477 |
+
"step": 35000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.74,
|
| 481 |
+
"learning_rate": 1.0581270182992467e-05,
|
| 482 |
+
"loss": 0.004,
|
| 483 |
+
"step": 35500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.81,
|
| 487 |
+
"learning_rate": 1.0446716899892357e-05,
|
| 488 |
+
"loss": 0.0042,
|
| 489 |
+
"step": 36000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.88,
|
| 493 |
+
"learning_rate": 1.031216361679225e-05,
|
| 494 |
+
"loss": 0.0041,
|
| 495 |
+
"step": 36500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 4.95,
|
| 499 |
+
"learning_rate": 1.0177610333692143e-05,
|
| 500 |
+
"loss": 0.0042,
|
| 501 |
+
"step": 37000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 5.0,
|
| 505 |
+
"eval_accuracy": 0.9940509055275022,
|
| 506 |
+
"eval_f1": 0.0,
|
| 507 |
+
"eval_loss": 0.025542501360177994,
|
| 508 |
+
"eval_precision": 0.0,
|
| 509 |
+
"eval_recall": 0.0,
|
| 510 |
+
"eval_runtime": 71.0096,
|
| 511 |
+
"eval_samples_per_second": 709.806,
|
| 512 |
+
"eval_steps_per_second": 5.549,
|
| 513 |
+
"step": 37410
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 5.01,
|
| 517 |
+
"learning_rate": 1.0043057050592036e-05,
|
| 518 |
+
"loss": 0.0038,
|
| 519 |
+
"step": 37500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.08,
|
| 523 |
+
"learning_rate": 9.908503767491927e-06,
|
| 524 |
+
"loss": 0.0029,
|
| 525 |
+
"step": 38000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.15,
|
| 529 |
+
"learning_rate": 9.77395048439182e-06,
|
| 530 |
+
"loss": 0.003,
|
| 531 |
+
"step": 38500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 5.21,
|
| 535 |
+
"learning_rate": 9.639397201291713e-06,
|
| 536 |
+
"loss": 0.003,
|
| 537 |
+
"step": 39000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.28,
|
| 541 |
+
"learning_rate": 9.504843918191604e-06,
|
| 542 |
+
"loss": 0.003,
|
| 543 |
+
"step": 39500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.35,
|
| 547 |
+
"learning_rate": 9.370290635091497e-06,
|
| 548 |
+
"loss": 0.0029,
|
| 549 |
+
"step": 40000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.41,
|
| 553 |
+
"learning_rate": 9.23573735199139e-06,
|
| 554 |
+
"loss": 0.0031,
|
| 555 |
+
"step": 40500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.48,
|
| 559 |
+
"learning_rate": 9.101184068891282e-06,
|
| 560 |
+
"loss": 0.0031,
|
| 561 |
+
"step": 41000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.55,
|
| 565 |
+
"learning_rate": 8.966630785791175e-06,
|
| 566 |
+
"loss": 0.0031,
|
| 567 |
+
"step": 41500
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.61,
|
| 571 |
+
"learning_rate": 8.832077502691066e-06,
|
| 572 |
+
"loss": 0.0032,
|
| 573 |
+
"step": 42000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 5.68,
|
| 577 |
+
"learning_rate": 8.697524219590959e-06,
|
| 578 |
+
"loss": 0.003,
|
| 579 |
+
"step": 42500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.75,
|
| 583 |
+
"learning_rate": 8.56297093649085e-06,
|
| 584 |
+
"loss": 0.0032,
|
| 585 |
+
"step": 43000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.81,
|
| 589 |
+
"learning_rate": 8.428417653390743e-06,
|
| 590 |
+
"loss": 0.0033,
|
| 591 |
+
"step": 43500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 5.88,
|
| 595 |
+
"learning_rate": 8.293864370290636e-06,
|
| 596 |
+
"loss": 0.0032,
|
| 597 |
+
"step": 44000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 5.95,
|
| 601 |
+
"learning_rate": 8.159311087190527e-06,
|
| 602 |
+
"loss": 0.0032,
|
| 603 |
+
"step": 44500
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.0,
|
| 607 |
+
"eval_accuracy": 0.9940655314512207,
|
| 608 |
+
"eval_f1": 0.0,
|
| 609 |
+
"eval_loss": 0.031192485243082047,
|
| 610 |
+
"eval_precision": 0.0,
|
| 611 |
+
"eval_recall": 0.0,
|
| 612 |
+
"eval_runtime": 81.6971,
|
| 613 |
+
"eval_samples_per_second": 616.95,
|
| 614 |
+
"eval_steps_per_second": 4.823,
|
| 615 |
+
"step": 44892
|
| 616 |
+
}
|
| 617 |
+
],
|
| 618 |
+
"logging_steps": 500,
|
| 619 |
+
"max_steps": 74820,
|
| 620 |
+
"num_train_epochs": 10,
|
| 621 |
+
"save_steps": 500,
|
| 622 |
+
"total_flos": 5.1024644353628736e+17,
|
| 623 |
+
"trial_name": null,
|
| 624 |
+
"trial_params": null
|
| 625 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-44892/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a2c94b3fa4105f0848e039f53ebc926d5c50bab3ec3ef6ee702b2cac182fc88
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e93ea760e69150887aa70367fc7c023cde354ea5bbdc7677ff27694947e6896b
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff9f9ba597613c9b6de2929cd52809617964b278efcc8d2d7f05547c143c6c43
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58cdca43612bcf5318fcbf9957a0b062eac9129103ad04c51d2784c587eaceab
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/trainer_state.json
ADDED
|
@@ -0,0 +1,727 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 52374,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.01,
|
| 313 |
+
"learning_rate": 1.4079655543595265e-05,
|
| 314 |
+
"loss": 0.0074,
|
| 315 |
+
"step": 22500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.07,
|
| 319 |
+
"learning_rate": 1.3945102260495158e-05,
|
| 320 |
+
"loss": 0.0051,
|
| 321 |
+
"step": 23000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.14,
|
| 325 |
+
"learning_rate": 1.3810548977395049e-05,
|
| 326 |
+
"loss": 0.0052,
|
| 327 |
+
"step": 23500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.21,
|
| 331 |
+
"learning_rate": 1.3675995694294942e-05,
|
| 332 |
+
"loss": 0.005,
|
| 333 |
+
"step": 24000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 1.3541442411194833e-05,
|
| 338 |
+
"loss": 0.0056,
|
| 339 |
+
"step": 24500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.34,
|
| 343 |
+
"learning_rate": 1.3406889128094726e-05,
|
| 344 |
+
"loss": 0.0049,
|
| 345 |
+
"step": 25000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.41,
|
| 349 |
+
"learning_rate": 1.3272335844994619e-05,
|
| 350 |
+
"loss": 0.0054,
|
| 351 |
+
"step": 25500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.48,
|
| 355 |
+
"learning_rate": 1.3137782561894512e-05,
|
| 356 |
+
"loss": 0.0052,
|
| 357 |
+
"step": 26000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.54,
|
| 361 |
+
"learning_rate": 1.3003229278794403e-05,
|
| 362 |
+
"loss": 0.0055,
|
| 363 |
+
"step": 26500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.61,
|
| 367 |
+
"learning_rate": 1.2868675995694295e-05,
|
| 368 |
+
"loss": 0.0053,
|
| 369 |
+
"step": 27000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.68,
|
| 373 |
+
"learning_rate": 1.2734122712594188e-05,
|
| 374 |
+
"loss": 0.0055,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.74,
|
| 379 |
+
"learning_rate": 1.259956942949408e-05,
|
| 380 |
+
"loss": 0.0056,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.81,
|
| 385 |
+
"learning_rate": 1.2465016146393974e-05,
|
| 386 |
+
"loss": 0.0056,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.88,
|
| 391 |
+
"learning_rate": 1.2330462863293867e-05,
|
| 392 |
+
"loss": 0.0056,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.94,
|
| 397 |
+
"learning_rate": 1.219590958019376e-05,
|
| 398 |
+
"loss": 0.0056,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"eval_accuracy": 0.9940216536800652,
|
| 404 |
+
"eval_f1": 0.0,
|
| 405 |
+
"eval_loss": 0.023068198934197426,
|
| 406 |
+
"eval_precision": 0.0,
|
| 407 |
+
"eval_recall": 0.0,
|
| 408 |
+
"eval_runtime": 81.8028,
|
| 409 |
+
"eval_samples_per_second": 616.153,
|
| 410 |
+
"eval_steps_per_second": 4.816,
|
| 411 |
+
"step": 29928
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.01,
|
| 415 |
+
"learning_rate": 1.206135629709365e-05,
|
| 416 |
+
"loss": 0.0053,
|
| 417 |
+
"step": 30000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.08,
|
| 421 |
+
"learning_rate": 1.1926803013993542e-05,
|
| 422 |
+
"loss": 0.0035,
|
| 423 |
+
"step": 30500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.14,
|
| 427 |
+
"learning_rate": 1.1792249730893435e-05,
|
| 428 |
+
"loss": 0.0037,
|
| 429 |
+
"step": 31000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.21,
|
| 433 |
+
"learning_rate": 1.1657696447793328e-05,
|
| 434 |
+
"loss": 0.0039,
|
| 435 |
+
"step": 31500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.28,
|
| 439 |
+
"learning_rate": 1.152314316469322e-05,
|
| 440 |
+
"loss": 0.0037,
|
| 441 |
+
"step": 32000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.34,
|
| 445 |
+
"learning_rate": 1.1388589881593113e-05,
|
| 446 |
+
"loss": 0.0038,
|
| 447 |
+
"step": 32500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.41,
|
| 451 |
+
"learning_rate": 1.1254036598493004e-05,
|
| 452 |
+
"loss": 0.004,
|
| 453 |
+
"step": 33000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.48,
|
| 457 |
+
"learning_rate": 1.1119483315392897e-05,
|
| 458 |
+
"loss": 0.0039,
|
| 459 |
+
"step": 33500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.54,
|
| 463 |
+
"learning_rate": 1.0984930032292788e-05,
|
| 464 |
+
"loss": 0.004,
|
| 465 |
+
"step": 34000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.61,
|
| 469 |
+
"learning_rate": 1.0850376749192681e-05,
|
| 470 |
+
"loss": 0.0041,
|
| 471 |
+
"step": 34500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.68,
|
| 475 |
+
"learning_rate": 1.0715823466092574e-05,
|
| 476 |
+
"loss": 0.0042,
|
| 477 |
+
"step": 35000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.74,
|
| 481 |
+
"learning_rate": 1.0581270182992467e-05,
|
| 482 |
+
"loss": 0.004,
|
| 483 |
+
"step": 35500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.81,
|
| 487 |
+
"learning_rate": 1.0446716899892357e-05,
|
| 488 |
+
"loss": 0.0042,
|
| 489 |
+
"step": 36000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.88,
|
| 493 |
+
"learning_rate": 1.031216361679225e-05,
|
| 494 |
+
"loss": 0.0041,
|
| 495 |
+
"step": 36500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 4.95,
|
| 499 |
+
"learning_rate": 1.0177610333692143e-05,
|
| 500 |
+
"loss": 0.0042,
|
| 501 |
+
"step": 37000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 5.0,
|
| 505 |
+
"eval_accuracy": 0.9940509055275022,
|
| 506 |
+
"eval_f1": 0.0,
|
| 507 |
+
"eval_loss": 0.025542501360177994,
|
| 508 |
+
"eval_precision": 0.0,
|
| 509 |
+
"eval_recall": 0.0,
|
| 510 |
+
"eval_runtime": 71.0096,
|
| 511 |
+
"eval_samples_per_second": 709.806,
|
| 512 |
+
"eval_steps_per_second": 5.549,
|
| 513 |
+
"step": 37410
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 5.01,
|
| 517 |
+
"learning_rate": 1.0043057050592036e-05,
|
| 518 |
+
"loss": 0.0038,
|
| 519 |
+
"step": 37500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.08,
|
| 523 |
+
"learning_rate": 9.908503767491927e-06,
|
| 524 |
+
"loss": 0.0029,
|
| 525 |
+
"step": 38000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.15,
|
| 529 |
+
"learning_rate": 9.77395048439182e-06,
|
| 530 |
+
"loss": 0.003,
|
| 531 |
+
"step": 38500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 5.21,
|
| 535 |
+
"learning_rate": 9.639397201291713e-06,
|
| 536 |
+
"loss": 0.003,
|
| 537 |
+
"step": 39000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.28,
|
| 541 |
+
"learning_rate": 9.504843918191604e-06,
|
| 542 |
+
"loss": 0.003,
|
| 543 |
+
"step": 39500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.35,
|
| 547 |
+
"learning_rate": 9.370290635091497e-06,
|
| 548 |
+
"loss": 0.0029,
|
| 549 |
+
"step": 40000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.41,
|
| 553 |
+
"learning_rate": 9.23573735199139e-06,
|
| 554 |
+
"loss": 0.0031,
|
| 555 |
+
"step": 40500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.48,
|
| 559 |
+
"learning_rate": 9.101184068891282e-06,
|
| 560 |
+
"loss": 0.0031,
|
| 561 |
+
"step": 41000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.55,
|
| 565 |
+
"learning_rate": 8.966630785791175e-06,
|
| 566 |
+
"loss": 0.0031,
|
| 567 |
+
"step": 41500
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.61,
|
| 571 |
+
"learning_rate": 8.832077502691066e-06,
|
| 572 |
+
"loss": 0.0032,
|
| 573 |
+
"step": 42000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 5.68,
|
| 577 |
+
"learning_rate": 8.697524219590959e-06,
|
| 578 |
+
"loss": 0.003,
|
| 579 |
+
"step": 42500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.75,
|
| 583 |
+
"learning_rate": 8.56297093649085e-06,
|
| 584 |
+
"loss": 0.0032,
|
| 585 |
+
"step": 43000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.81,
|
| 589 |
+
"learning_rate": 8.428417653390743e-06,
|
| 590 |
+
"loss": 0.0033,
|
| 591 |
+
"step": 43500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 5.88,
|
| 595 |
+
"learning_rate": 8.293864370290636e-06,
|
| 596 |
+
"loss": 0.0032,
|
| 597 |
+
"step": 44000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 5.95,
|
| 601 |
+
"learning_rate": 8.159311087190527e-06,
|
| 602 |
+
"loss": 0.0032,
|
| 603 |
+
"step": 44500
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.0,
|
| 607 |
+
"eval_accuracy": 0.9940655314512207,
|
| 608 |
+
"eval_f1": 0.0,
|
| 609 |
+
"eval_loss": 0.031192485243082047,
|
| 610 |
+
"eval_precision": 0.0,
|
| 611 |
+
"eval_recall": 0.0,
|
| 612 |
+
"eval_runtime": 81.6971,
|
| 613 |
+
"eval_samples_per_second": 616.95,
|
| 614 |
+
"eval_steps_per_second": 4.823,
|
| 615 |
+
"step": 44892
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.01,
|
| 619 |
+
"learning_rate": 8.02475780409042e-06,
|
| 620 |
+
"loss": 0.003,
|
| 621 |
+
"step": 45000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.08,
|
| 625 |
+
"learning_rate": 7.890204520990313e-06,
|
| 626 |
+
"loss": 0.0022,
|
| 627 |
+
"step": 45500
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.15,
|
| 631 |
+
"learning_rate": 7.755651237890205e-06,
|
| 632 |
+
"loss": 0.0024,
|
| 633 |
+
"step": 46000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.21,
|
| 637 |
+
"learning_rate": 7.621097954790098e-06,
|
| 638 |
+
"loss": 0.0022,
|
| 639 |
+
"step": 46500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.28,
|
| 643 |
+
"learning_rate": 7.486544671689991e-06,
|
| 644 |
+
"loss": 0.0024,
|
| 645 |
+
"step": 47000
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.35,
|
| 649 |
+
"learning_rate": 7.351991388589882e-06,
|
| 650 |
+
"loss": 0.0025,
|
| 651 |
+
"step": 47500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.42,
|
| 655 |
+
"learning_rate": 7.217438105489775e-06,
|
| 656 |
+
"loss": 0.0025,
|
| 657 |
+
"step": 48000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 6.48,
|
| 661 |
+
"learning_rate": 7.082884822389667e-06,
|
| 662 |
+
"loss": 0.0025,
|
| 663 |
+
"step": 48500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 6.55,
|
| 667 |
+
"learning_rate": 6.948331539289559e-06,
|
| 668 |
+
"loss": 0.0026,
|
| 669 |
+
"step": 49000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 6.62,
|
| 673 |
+
"learning_rate": 6.813778256189451e-06,
|
| 674 |
+
"loss": 0.0025,
|
| 675 |
+
"step": 49500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 6.68,
|
| 679 |
+
"learning_rate": 6.679224973089344e-06,
|
| 680 |
+
"loss": 0.0026,
|
| 681 |
+
"step": 50000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.75,
|
| 685 |
+
"learning_rate": 6.544671689989236e-06,
|
| 686 |
+
"loss": 0.0025,
|
| 687 |
+
"step": 50500
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 6.82,
|
| 691 |
+
"learning_rate": 6.410118406889129e-06,
|
| 692 |
+
"loss": 0.0025,
|
| 693 |
+
"step": 51000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 6.88,
|
| 697 |
+
"learning_rate": 6.275565123789022e-06,
|
| 698 |
+
"loss": 0.0025,
|
| 699 |
+
"step": 51500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 6.95,
|
| 703 |
+
"learning_rate": 6.141011840688913e-06,
|
| 704 |
+
"loss": 0.0025,
|
| 705 |
+
"step": 52000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.0,
|
| 709 |
+
"eval_accuracy": 0.9939119592521766,
|
| 710 |
+
"eval_f1": 0.0,
|
| 711 |
+
"eval_loss": 0.03196028992533684,
|
| 712 |
+
"eval_precision": 0.0,
|
| 713 |
+
"eval_recall": 0.0,
|
| 714 |
+
"eval_runtime": 71.5312,
|
| 715 |
+
"eval_samples_per_second": 704.63,
|
| 716 |
+
"eval_steps_per_second": 5.508,
|
| 717 |
+
"step": 52374
|
| 718 |
+
}
|
| 719 |
+
],
|
| 720 |
+
"logging_steps": 500,
|
| 721 |
+
"max_steps": 74820,
|
| 722 |
+
"num_train_epochs": 10,
|
| 723 |
+
"save_steps": 500,
|
| 724 |
+
"total_flos": 5.95287517459002e+17,
|
| 725 |
+
"trial_name": null,
|
| 726 |
+
"trial_params": null
|
| 727 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-52374/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4d99c525306a2f5415aced53b4f26a4f948a39d19ab91beee79f05a3c0d2dc91
|
| 3 |
+
size 1418281093
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4584cb25d54c7bc100b750d5185dab318806f87dc469c72235b959bd7ea4852c
|
| 3 |
+
size 709125289
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d630713282a7e689b0e08e8ab7654d39f94f47b8f456bac345c21dba19706091
|
| 3 |
+
size 14575
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67fdbe2ad084cae31bd619a47ba1b70ba128ff67dd25af4b47b1bd149313fe24
|
| 3 |
+
size 627
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/trainer_state.json
ADDED
|
@@ -0,0 +1,829 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 59856,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 2e-05,
|
| 14 |
+
"loss": 0.1264,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.9865446716899895e-05,
|
| 20 |
+
"loss": 0.0227,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9730893433799788e-05,
|
| 26 |
+
"loss": 0.0203,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 1.959634015069968e-05,
|
| 32 |
+
"loss": 0.0197,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 1.946178686759957e-05,
|
| 38 |
+
"loss": 0.0185,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 1.9327233584499463e-05,
|
| 44 |
+
"loss": 0.0177,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 1.9192680301399356e-05,
|
| 50 |
+
"loss": 0.0181,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 1.905812701829925e-05,
|
| 56 |
+
"loss": 0.0171,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 1.8923573735199142e-05,
|
| 62 |
+
"loss": 0.0174,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 1.8789020452099035e-05,
|
| 68 |
+
"loss": 0.017,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.74,
|
| 73 |
+
"learning_rate": 1.8654467168998925e-05,
|
| 74 |
+
"loss": 0.0168,
|
| 75 |
+
"step": 5500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 1.8519913885898818e-05,
|
| 80 |
+
"loss": 0.0161,
|
| 81 |
+
"step": 6000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 1.838536060279871e-05,
|
| 86 |
+
"loss": 0.0156,
|
| 87 |
+
"step": 6500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.94,
|
| 91 |
+
"learning_rate": 1.8250807319698604e-05,
|
| 92 |
+
"loss": 0.0167,
|
| 93 |
+
"step": 7000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.9943580499255906,
|
| 98 |
+
"eval_f1": 0.0,
|
| 99 |
+
"eval_loss": 0.015657523646950722,
|
| 100 |
+
"eval_precision": 0.0,
|
| 101 |
+
"eval_recall": 0.0,
|
| 102 |
+
"eval_runtime": 71.099,
|
| 103 |
+
"eval_samples_per_second": 708.913,
|
| 104 |
+
"eval_steps_per_second": 5.542,
|
| 105 |
+
"step": 7482
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.0,
|
| 109 |
+
"learning_rate": 1.8116254036598497e-05,
|
| 110 |
+
"loss": 0.0152,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.07,
|
| 115 |
+
"learning_rate": 1.7981700753498386e-05,
|
| 116 |
+
"loss": 0.0119,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.14,
|
| 121 |
+
"learning_rate": 1.784714747039828e-05,
|
| 122 |
+
"loss": 0.0116,
|
| 123 |
+
"step": 8500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.2,
|
| 127 |
+
"learning_rate": 1.7712594187298172e-05,
|
| 128 |
+
"loss": 0.0116,
|
| 129 |
+
"step": 9000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.27,
|
| 133 |
+
"learning_rate": 1.7578040904198065e-05,
|
| 134 |
+
"loss": 0.0114,
|
| 135 |
+
"step": 9500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.34,
|
| 139 |
+
"learning_rate": 1.7443487621097955e-05,
|
| 140 |
+
"loss": 0.0114,
|
| 141 |
+
"step": 10000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.4,
|
| 145 |
+
"learning_rate": 1.7308934337997848e-05,
|
| 146 |
+
"loss": 0.0114,
|
| 147 |
+
"step": 10500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.47,
|
| 151 |
+
"learning_rate": 1.717438105489774e-05,
|
| 152 |
+
"loss": 0.0118,
|
| 153 |
+
"step": 11000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.54,
|
| 157 |
+
"learning_rate": 1.7039827771797634e-05,
|
| 158 |
+
"loss": 0.0119,
|
| 159 |
+
"step": 11500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.6,
|
| 163 |
+
"learning_rate": 1.6905274488697524e-05,
|
| 164 |
+
"loss": 0.0111,
|
| 165 |
+
"step": 12000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.67,
|
| 169 |
+
"learning_rate": 1.6770721205597417e-05,
|
| 170 |
+
"loss": 0.0118,
|
| 171 |
+
"step": 12500
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.74,
|
| 175 |
+
"learning_rate": 1.663616792249731e-05,
|
| 176 |
+
"loss": 0.0116,
|
| 177 |
+
"step": 13000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.8,
|
| 181 |
+
"learning_rate": 1.6501614639397203e-05,
|
| 182 |
+
"loss": 0.0112,
|
| 183 |
+
"step": 13500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.87,
|
| 187 |
+
"learning_rate": 1.6367061356297096e-05,
|
| 188 |
+
"loss": 0.0114,
|
| 189 |
+
"step": 14000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.94,
|
| 193 |
+
"learning_rate": 1.6232508073196985e-05,
|
| 194 |
+
"loss": 0.0115,
|
| 195 |
+
"step": 14500
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0,
|
| 199 |
+
"eval_accuracy": 0.9940472490465726,
|
| 200 |
+
"eval_f1": 0.0,
|
| 201 |
+
"eval_loss": 0.016565397381782532,
|
| 202 |
+
"eval_precision": 0.0,
|
| 203 |
+
"eval_recall": 0.0,
|
| 204 |
+
"eval_runtime": 82.6741,
|
| 205 |
+
"eval_samples_per_second": 609.659,
|
| 206 |
+
"eval_steps_per_second": 4.766,
|
| 207 |
+
"step": 14964
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.0,
|
| 211 |
+
"learning_rate": 1.6097954790096878e-05,
|
| 212 |
+
"loss": 0.0109,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.07,
|
| 217 |
+
"learning_rate": 1.596340150699677e-05,
|
| 218 |
+
"loss": 0.0077,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.14,
|
| 223 |
+
"learning_rate": 1.5828848223896664e-05,
|
| 224 |
+
"loss": 0.0075,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.21,
|
| 229 |
+
"learning_rate": 1.5694294940796557e-05,
|
| 230 |
+
"loss": 0.0076,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.27,
|
| 235 |
+
"learning_rate": 1.555974165769645e-05,
|
| 236 |
+
"loss": 0.0076,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.34,
|
| 241 |
+
"learning_rate": 1.542518837459634e-05,
|
| 242 |
+
"loss": 0.0077,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.41,
|
| 247 |
+
"learning_rate": 1.5290635091496233e-05,
|
| 248 |
+
"loss": 0.0077,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.47,
|
| 253 |
+
"learning_rate": 1.5156081808396126e-05,
|
| 254 |
+
"loss": 0.0076,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.54,
|
| 259 |
+
"learning_rate": 1.5021528525296019e-05,
|
| 260 |
+
"loss": 0.0078,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.61,
|
| 265 |
+
"learning_rate": 1.4886975242195912e-05,
|
| 266 |
+
"loss": 0.008,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.67,
|
| 271 |
+
"learning_rate": 1.4752421959095805e-05,
|
| 272 |
+
"loss": 0.0077,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.74,
|
| 277 |
+
"learning_rate": 1.4617868675995694e-05,
|
| 278 |
+
"loss": 0.0078,
|
| 279 |
+
"step": 20500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.81,
|
| 283 |
+
"learning_rate": 1.4483315392895587e-05,
|
| 284 |
+
"loss": 0.0081,
|
| 285 |
+
"step": 21000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.87,
|
| 289 |
+
"learning_rate": 1.434876210979548e-05,
|
| 290 |
+
"loss": 0.0081,
|
| 291 |
+
"step": 21500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.94,
|
| 295 |
+
"learning_rate": 1.4214208826695373e-05,
|
| 296 |
+
"loss": 0.0078,
|
| 297 |
+
"step": 22000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.9940838138558689,
|
| 302 |
+
"eval_f1": 0.0,
|
| 303 |
+
"eval_loss": 0.020097261294722557,
|
| 304 |
+
"eval_precision": 0.0,
|
| 305 |
+
"eval_recall": 0.0,
|
| 306 |
+
"eval_runtime": 71.581,
|
| 307 |
+
"eval_samples_per_second": 704.139,
|
| 308 |
+
"eval_steps_per_second": 5.504,
|
| 309 |
+
"step": 22446
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.01,
|
| 313 |
+
"learning_rate": 1.4079655543595265e-05,
|
| 314 |
+
"loss": 0.0074,
|
| 315 |
+
"step": 22500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.07,
|
| 319 |
+
"learning_rate": 1.3945102260495158e-05,
|
| 320 |
+
"loss": 0.0051,
|
| 321 |
+
"step": 23000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.14,
|
| 325 |
+
"learning_rate": 1.3810548977395049e-05,
|
| 326 |
+
"loss": 0.0052,
|
| 327 |
+
"step": 23500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 3.21,
|
| 331 |
+
"learning_rate": 1.3675995694294942e-05,
|
| 332 |
+
"loss": 0.005,
|
| 333 |
+
"step": 24000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.27,
|
| 337 |
+
"learning_rate": 1.3541442411194833e-05,
|
| 338 |
+
"loss": 0.0056,
|
| 339 |
+
"step": 24500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 3.34,
|
| 343 |
+
"learning_rate": 1.3406889128094726e-05,
|
| 344 |
+
"loss": 0.0049,
|
| 345 |
+
"step": 25000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 3.41,
|
| 349 |
+
"learning_rate": 1.3272335844994619e-05,
|
| 350 |
+
"loss": 0.0054,
|
| 351 |
+
"step": 25500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.48,
|
| 355 |
+
"learning_rate": 1.3137782561894512e-05,
|
| 356 |
+
"loss": 0.0052,
|
| 357 |
+
"step": 26000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 3.54,
|
| 361 |
+
"learning_rate": 1.3003229278794403e-05,
|
| 362 |
+
"loss": 0.0055,
|
| 363 |
+
"step": 26500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.61,
|
| 367 |
+
"learning_rate": 1.2868675995694295e-05,
|
| 368 |
+
"loss": 0.0053,
|
| 369 |
+
"step": 27000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.68,
|
| 373 |
+
"learning_rate": 1.2734122712594188e-05,
|
| 374 |
+
"loss": 0.0055,
|
| 375 |
+
"step": 27500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.74,
|
| 379 |
+
"learning_rate": 1.259956942949408e-05,
|
| 380 |
+
"loss": 0.0056,
|
| 381 |
+
"step": 28000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.81,
|
| 385 |
+
"learning_rate": 1.2465016146393974e-05,
|
| 386 |
+
"loss": 0.0056,
|
| 387 |
+
"step": 28500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.88,
|
| 391 |
+
"learning_rate": 1.2330462863293867e-05,
|
| 392 |
+
"loss": 0.0056,
|
| 393 |
+
"step": 29000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.94,
|
| 397 |
+
"learning_rate": 1.219590958019376e-05,
|
| 398 |
+
"loss": 0.0056,
|
| 399 |
+
"step": 29500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.0,
|
| 403 |
+
"eval_accuracy": 0.9940216536800652,
|
| 404 |
+
"eval_f1": 0.0,
|
| 405 |
+
"eval_loss": 0.023068198934197426,
|
| 406 |
+
"eval_precision": 0.0,
|
| 407 |
+
"eval_recall": 0.0,
|
| 408 |
+
"eval_runtime": 81.8028,
|
| 409 |
+
"eval_samples_per_second": 616.153,
|
| 410 |
+
"eval_steps_per_second": 4.816,
|
| 411 |
+
"step": 29928
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.01,
|
| 415 |
+
"learning_rate": 1.206135629709365e-05,
|
| 416 |
+
"loss": 0.0053,
|
| 417 |
+
"step": 30000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.08,
|
| 421 |
+
"learning_rate": 1.1926803013993542e-05,
|
| 422 |
+
"loss": 0.0035,
|
| 423 |
+
"step": 30500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.14,
|
| 427 |
+
"learning_rate": 1.1792249730893435e-05,
|
| 428 |
+
"loss": 0.0037,
|
| 429 |
+
"step": 31000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.21,
|
| 433 |
+
"learning_rate": 1.1657696447793328e-05,
|
| 434 |
+
"loss": 0.0039,
|
| 435 |
+
"step": 31500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.28,
|
| 439 |
+
"learning_rate": 1.152314316469322e-05,
|
| 440 |
+
"loss": 0.0037,
|
| 441 |
+
"step": 32000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.34,
|
| 445 |
+
"learning_rate": 1.1388589881593113e-05,
|
| 446 |
+
"loss": 0.0038,
|
| 447 |
+
"step": 32500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.41,
|
| 451 |
+
"learning_rate": 1.1254036598493004e-05,
|
| 452 |
+
"loss": 0.004,
|
| 453 |
+
"step": 33000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.48,
|
| 457 |
+
"learning_rate": 1.1119483315392897e-05,
|
| 458 |
+
"loss": 0.0039,
|
| 459 |
+
"step": 33500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.54,
|
| 463 |
+
"learning_rate": 1.0984930032292788e-05,
|
| 464 |
+
"loss": 0.004,
|
| 465 |
+
"step": 34000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.61,
|
| 469 |
+
"learning_rate": 1.0850376749192681e-05,
|
| 470 |
+
"loss": 0.0041,
|
| 471 |
+
"step": 34500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 4.68,
|
| 475 |
+
"learning_rate": 1.0715823466092574e-05,
|
| 476 |
+
"loss": 0.0042,
|
| 477 |
+
"step": 35000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 4.74,
|
| 481 |
+
"learning_rate": 1.0581270182992467e-05,
|
| 482 |
+
"loss": 0.004,
|
| 483 |
+
"step": 35500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.81,
|
| 487 |
+
"learning_rate": 1.0446716899892357e-05,
|
| 488 |
+
"loss": 0.0042,
|
| 489 |
+
"step": 36000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.88,
|
| 493 |
+
"learning_rate": 1.031216361679225e-05,
|
| 494 |
+
"loss": 0.0041,
|
| 495 |
+
"step": 36500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 4.95,
|
| 499 |
+
"learning_rate": 1.0177610333692143e-05,
|
| 500 |
+
"loss": 0.0042,
|
| 501 |
+
"step": 37000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 5.0,
|
| 505 |
+
"eval_accuracy": 0.9940509055275022,
|
| 506 |
+
"eval_f1": 0.0,
|
| 507 |
+
"eval_loss": 0.025542501360177994,
|
| 508 |
+
"eval_precision": 0.0,
|
| 509 |
+
"eval_recall": 0.0,
|
| 510 |
+
"eval_runtime": 71.0096,
|
| 511 |
+
"eval_samples_per_second": 709.806,
|
| 512 |
+
"eval_steps_per_second": 5.549,
|
| 513 |
+
"step": 37410
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 5.01,
|
| 517 |
+
"learning_rate": 1.0043057050592036e-05,
|
| 518 |
+
"loss": 0.0038,
|
| 519 |
+
"step": 37500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.08,
|
| 523 |
+
"learning_rate": 9.908503767491927e-06,
|
| 524 |
+
"loss": 0.0029,
|
| 525 |
+
"step": 38000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 5.15,
|
| 529 |
+
"learning_rate": 9.77395048439182e-06,
|
| 530 |
+
"loss": 0.003,
|
| 531 |
+
"step": 38500
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 5.21,
|
| 535 |
+
"learning_rate": 9.639397201291713e-06,
|
| 536 |
+
"loss": 0.003,
|
| 537 |
+
"step": 39000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.28,
|
| 541 |
+
"learning_rate": 9.504843918191604e-06,
|
| 542 |
+
"loss": 0.003,
|
| 543 |
+
"step": 39500
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 5.35,
|
| 547 |
+
"learning_rate": 9.370290635091497e-06,
|
| 548 |
+
"loss": 0.0029,
|
| 549 |
+
"step": 40000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 5.41,
|
| 553 |
+
"learning_rate": 9.23573735199139e-06,
|
| 554 |
+
"loss": 0.0031,
|
| 555 |
+
"step": 40500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 5.48,
|
| 559 |
+
"learning_rate": 9.101184068891282e-06,
|
| 560 |
+
"loss": 0.0031,
|
| 561 |
+
"step": 41000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.55,
|
| 565 |
+
"learning_rate": 8.966630785791175e-06,
|
| 566 |
+
"loss": 0.0031,
|
| 567 |
+
"step": 41500
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.61,
|
| 571 |
+
"learning_rate": 8.832077502691066e-06,
|
| 572 |
+
"loss": 0.0032,
|
| 573 |
+
"step": 42000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 5.68,
|
| 577 |
+
"learning_rate": 8.697524219590959e-06,
|
| 578 |
+
"loss": 0.003,
|
| 579 |
+
"step": 42500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 5.75,
|
| 583 |
+
"learning_rate": 8.56297093649085e-06,
|
| 584 |
+
"loss": 0.0032,
|
| 585 |
+
"step": 43000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 5.81,
|
| 589 |
+
"learning_rate": 8.428417653390743e-06,
|
| 590 |
+
"loss": 0.0033,
|
| 591 |
+
"step": 43500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 5.88,
|
| 595 |
+
"learning_rate": 8.293864370290636e-06,
|
| 596 |
+
"loss": 0.0032,
|
| 597 |
+
"step": 44000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 5.95,
|
| 601 |
+
"learning_rate": 8.159311087190527e-06,
|
| 602 |
+
"loss": 0.0032,
|
| 603 |
+
"step": 44500
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.0,
|
| 607 |
+
"eval_accuracy": 0.9940655314512207,
|
| 608 |
+
"eval_f1": 0.0,
|
| 609 |
+
"eval_loss": 0.031192485243082047,
|
| 610 |
+
"eval_precision": 0.0,
|
| 611 |
+
"eval_recall": 0.0,
|
| 612 |
+
"eval_runtime": 81.6971,
|
| 613 |
+
"eval_samples_per_second": 616.95,
|
| 614 |
+
"eval_steps_per_second": 4.823,
|
| 615 |
+
"step": 44892
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 6.01,
|
| 619 |
+
"learning_rate": 8.02475780409042e-06,
|
| 620 |
+
"loss": 0.003,
|
| 621 |
+
"step": 45000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.08,
|
| 625 |
+
"learning_rate": 7.890204520990313e-06,
|
| 626 |
+
"loss": 0.0022,
|
| 627 |
+
"step": 45500
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.15,
|
| 631 |
+
"learning_rate": 7.755651237890205e-06,
|
| 632 |
+
"loss": 0.0024,
|
| 633 |
+
"step": 46000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.21,
|
| 637 |
+
"learning_rate": 7.621097954790098e-06,
|
| 638 |
+
"loss": 0.0022,
|
| 639 |
+
"step": 46500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.28,
|
| 643 |
+
"learning_rate": 7.486544671689991e-06,
|
| 644 |
+
"loss": 0.0024,
|
| 645 |
+
"step": 47000
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.35,
|
| 649 |
+
"learning_rate": 7.351991388589882e-06,
|
| 650 |
+
"loss": 0.0025,
|
| 651 |
+
"step": 47500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.42,
|
| 655 |
+
"learning_rate": 7.217438105489775e-06,
|
| 656 |
+
"loss": 0.0025,
|
| 657 |
+
"step": 48000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 6.48,
|
| 661 |
+
"learning_rate": 7.082884822389667e-06,
|
| 662 |
+
"loss": 0.0025,
|
| 663 |
+
"step": 48500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 6.55,
|
| 667 |
+
"learning_rate": 6.948331539289559e-06,
|
| 668 |
+
"loss": 0.0026,
|
| 669 |
+
"step": 49000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 6.62,
|
| 673 |
+
"learning_rate": 6.813778256189451e-06,
|
| 674 |
+
"loss": 0.0025,
|
| 675 |
+
"step": 49500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 6.68,
|
| 679 |
+
"learning_rate": 6.679224973089344e-06,
|
| 680 |
+
"loss": 0.0026,
|
| 681 |
+
"step": 50000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.75,
|
| 685 |
+
"learning_rate": 6.544671689989236e-06,
|
| 686 |
+
"loss": 0.0025,
|
| 687 |
+
"step": 50500
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 6.82,
|
| 691 |
+
"learning_rate": 6.410118406889129e-06,
|
| 692 |
+
"loss": 0.0025,
|
| 693 |
+
"step": 51000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 6.88,
|
| 697 |
+
"learning_rate": 6.275565123789022e-06,
|
| 698 |
+
"loss": 0.0025,
|
| 699 |
+
"step": 51500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 6.95,
|
| 703 |
+
"learning_rate": 6.141011840688913e-06,
|
| 704 |
+
"loss": 0.0025,
|
| 705 |
+
"step": 52000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 7.0,
|
| 709 |
+
"eval_accuracy": 0.9939119592521766,
|
| 710 |
+
"eval_f1": 0.0,
|
| 711 |
+
"eval_loss": 0.03196028992533684,
|
| 712 |
+
"eval_precision": 0.0,
|
| 713 |
+
"eval_recall": 0.0,
|
| 714 |
+
"eval_runtime": 71.5312,
|
| 715 |
+
"eval_samples_per_second": 704.63,
|
| 716 |
+
"eval_steps_per_second": 5.508,
|
| 717 |
+
"step": 52374
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 7.02,
|
| 721 |
+
"learning_rate": 6.006458557588806e-06,
|
| 722 |
+
"loss": 0.0023,
|
| 723 |
+
"step": 52500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 7.08,
|
| 727 |
+
"learning_rate": 5.871905274488698e-06,
|
| 728 |
+
"loss": 0.002,
|
| 729 |
+
"step": 53000
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 7.15,
|
| 733 |
+
"learning_rate": 5.73735199138859e-06,
|
| 734 |
+
"loss": 0.0019,
|
| 735 |
+
"step": 53500
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 7.22,
|
| 739 |
+
"learning_rate": 5.602798708288482e-06,
|
| 740 |
+
"loss": 0.002,
|
| 741 |
+
"step": 54000
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 7.28,
|
| 745 |
+
"learning_rate": 5.468245425188375e-06,
|
| 746 |
+
"loss": 0.002,
|
| 747 |
+
"step": 54500
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 7.35,
|
| 751 |
+
"learning_rate": 5.333692142088267e-06,
|
| 752 |
+
"loss": 0.0019,
|
| 753 |
+
"step": 55000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 7.42,
|
| 757 |
+
"learning_rate": 5.19913885898816e-06,
|
| 758 |
+
"loss": 0.002,
|
| 759 |
+
"step": 55500
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.48,
|
| 763 |
+
"learning_rate": 5.064585575888053e-06,
|
| 764 |
+
"loss": 0.0018,
|
| 765 |
+
"step": 56000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 7.55,
|
| 769 |
+
"learning_rate": 4.930032292787945e-06,
|
| 770 |
+
"loss": 0.0022,
|
| 771 |
+
"step": 56500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 7.62,
|
| 775 |
+
"learning_rate": 4.795479009687837e-06,
|
| 776 |
+
"loss": 0.002,
|
| 777 |
+
"step": 57000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 7.69,
|
| 781 |
+
"learning_rate": 4.660925726587729e-06,
|
| 782 |
+
"loss": 0.002,
|
| 783 |
+
"step": 57500
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 7.75,
|
| 787 |
+
"learning_rate": 4.526372443487621e-06,
|
| 788 |
+
"loss": 0.0021,
|
| 789 |
+
"step": 58000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 7.82,
|
| 793 |
+
"learning_rate": 4.391819160387513e-06,
|
| 794 |
+
"loss": 0.0021,
|
| 795 |
+
"step": 58500
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 7.89,
|
| 799 |
+
"learning_rate": 4.2572658772874056e-06,
|
| 800 |
+
"loss": 0.0021,
|
| 801 |
+
"step": 59000
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 7.95,
|
| 805 |
+
"learning_rate": 4.1227125941872986e-06,
|
| 806 |
+
"loss": 0.0023,
|
| 807 |
+
"step": 59500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 8.0,
|
| 811 |
+
"eval_accuracy": 0.9939704629470505,
|
| 812 |
+
"eval_f1": 0.0,
|
| 813 |
+
"eval_loss": 0.03371906280517578,
|
| 814 |
+
"eval_precision": 0.0,
|
| 815 |
+
"eval_recall": 0.0,
|
| 816 |
+
"eval_runtime": 81.4913,
|
| 817 |
+
"eval_samples_per_second": 618.508,
|
| 818 |
+
"eval_steps_per_second": 4.835,
|
| 819 |
+
"step": 59856
|
| 820 |
+
}
|
| 821 |
+
],
|
| 822 |
+
"logging_steps": 500,
|
| 823 |
+
"max_steps": 74820,
|
| 824 |
+
"num_train_epochs": 10,
|
| 825 |
+
"save_steps": 500,
|
| 826 |
+
"total_flos": 6.803285913817166e+17,
|
| 827 |
+
"trial_name": null,
|
| 828 |
+
"trial_params": null
|
| 829 |
+
}
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-59856/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f713488e20afd5141b39f9aaa67feba7da0a360f752dbaa308013d8d5212f449
|
| 3 |
+
size 4091
|
rubert-base-cased-conversational-512-tatoeba_dataset/02-09-2023-11-01-00/checkpoint-67338/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DeepPavlov/rubert-base-cased-conversational",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForTokenClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"directionality": "bidi",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "O",
|
| 14 |
+
"1": "U"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"O": 0,
|
| 20 |
+
"U": 1
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"max_position_embeddings": 512,
|
| 24 |
+
"model_type": "bert",
|
| 25 |
+
"num_attention_heads": 12,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"output_past": true,
|
| 28 |
+
"pad_token_id": 0,
|
| 29 |
+
"pooler_fc_size": 768,
|
| 30 |
+
"pooler_num_attention_heads": 12,
|
| 31 |
+
"pooler_num_fc_layers": 3,
|
| 32 |
+
"pooler_size_per_head": 128,
|
| 33 |
+
"pooler_type": "first_token_transform",
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.32.1",
|
| 37 |
+
"type_vocab_size": 2,
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 119547
|
| 40 |
+
}
|