train with TPUv2 - wikipedia
Browse files- all_results.json +8 -0
- config.json +27 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +505 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 3.5848853368245197,
|
| 4 |
+
"train_runtime": 32321.6149,
|
| 5 |
+
"train_samples": 256484,
|
| 6 |
+
"train_samples_per_second": 158.707,
|
| 7 |
+
"train_steps_per_second": 1.24
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"gradient_checkpointing": false,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 256,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 1024,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 128,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 4,
|
| 19 |
+
"num_hidden_layers": 12,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"torch_dtype": "float32",
|
| 23 |
+
"transformers_version": "4.12.5",
|
| 24 |
+
"type_vocab_size": 1,
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 50265
|
| 27 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bb191cde388180541b9140ff4ae41d313abe0f48b4d3dc2e8582ca078ab7009
|
| 3 |
+
size 90051474
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "special_tokens_map_file": null, "name_or_path": "./roberta-small", "tokenizer_class": "RobertaTokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"train_loss": 3.5848853368245197,
|
| 4 |
+
"train_runtime": 32321.6149,
|
| 5 |
+
"train_samples": 256484,
|
| 6 |
+
"train_samples_per_second": 158.707,
|
| 7 |
+
"train_steps_per_second": 1.24
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,505 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 40080,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 9.875249500998003e-05,
|
| 13 |
+
"loss": 7.8671,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.5,
|
| 18 |
+
"learning_rate": 9.750499001996009e-05,
|
| 19 |
+
"loss": 7.0941,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.75,
|
| 24 |
+
"learning_rate": 9.625748502994012e-05,
|
| 25 |
+
"loss": 6.7944,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"learning_rate": 9.500998003992016e-05,
|
| 31 |
+
"loss": 6.526,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"learning_rate": 9.37624750499002e-05,
|
| 37 |
+
"loss": 6.2583,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.5,
|
| 42 |
+
"learning_rate": 9.251497005988024e-05,
|
| 43 |
+
"loss": 6.0239,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.75,
|
| 48 |
+
"learning_rate": 9.126746506986029e-05,
|
| 49 |
+
"loss": 5.7969,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"learning_rate": 9.001996007984033e-05,
|
| 55 |
+
"loss": 5.5919,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.25,
|
| 60 |
+
"learning_rate": 8.877245508982036e-05,
|
| 61 |
+
"loss": 5.3796,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.5,
|
| 66 |
+
"learning_rate": 8.752495009980041e-05,
|
| 67 |
+
"loss": 5.1887,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.74,
|
| 72 |
+
"learning_rate": 8.627744510978044e-05,
|
| 73 |
+
"loss": 5.0026,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.99,
|
| 78 |
+
"learning_rate": 8.502994011976048e-05,
|
| 79 |
+
"loss": 4.843,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.24,
|
| 84 |
+
"learning_rate": 8.378243512974053e-05,
|
| 85 |
+
"loss": 4.6736,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.49,
|
| 90 |
+
"learning_rate": 8.253493013972057e-05,
|
| 91 |
+
"loss": 4.553,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.74,
|
| 96 |
+
"learning_rate": 8.12874251497006e-05,
|
| 97 |
+
"loss": 4.4224,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.99,
|
| 102 |
+
"learning_rate": 8.003992015968065e-05,
|
| 103 |
+
"loss": 4.3097,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.24,
|
| 108 |
+
"learning_rate": 7.879241516966068e-05,
|
| 109 |
+
"loss": 4.1975,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.49,
|
| 114 |
+
"learning_rate": 7.754491017964072e-05,
|
| 115 |
+
"loss": 4.0994,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.74,
|
| 120 |
+
"learning_rate": 7.629740518962077e-05,
|
| 121 |
+
"loss": 4.0258,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.99,
|
| 126 |
+
"learning_rate": 7.50499001996008e-05,
|
| 127 |
+
"loss": 3.9443,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.24,
|
| 132 |
+
"learning_rate": 7.380239520958085e-05,
|
| 133 |
+
"loss": 3.8616,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 5.49,
|
| 138 |
+
"learning_rate": 7.255489021956088e-05,
|
| 139 |
+
"loss": 3.8086,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.74,
|
| 144 |
+
"learning_rate": 7.130738522954092e-05,
|
| 145 |
+
"loss": 3.7464,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.99,
|
| 150 |
+
"learning_rate": 7.005988023952096e-05,
|
| 151 |
+
"loss": 3.6925,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 6.24,
|
| 156 |
+
"learning_rate": 6.8812375249501e-05,
|
| 157 |
+
"loss": 3.6216,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.49,
|
| 162 |
+
"learning_rate": 6.756487025948103e-05,
|
| 163 |
+
"loss": 3.585,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.74,
|
| 168 |
+
"learning_rate": 6.631736526946109e-05,
|
| 169 |
+
"loss": 3.5383,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.99,
|
| 174 |
+
"learning_rate": 6.506986027944112e-05,
|
| 175 |
+
"loss": 3.4925,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 7.24,
|
| 180 |
+
"learning_rate": 6.382235528942116e-05,
|
| 181 |
+
"loss": 3.4524,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 7.49,
|
| 186 |
+
"learning_rate": 6.25748502994012e-05,
|
| 187 |
+
"loss": 3.3985,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 7.73,
|
| 192 |
+
"learning_rate": 6.132734530938125e-05,
|
| 193 |
+
"loss": 3.3684,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 7.98,
|
| 198 |
+
"learning_rate": 6.007984031936128e-05,
|
| 199 |
+
"loss": 3.3273,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.23,
|
| 204 |
+
"learning_rate": 5.8832335329341316e-05,
|
| 205 |
+
"loss": 3.2898,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 8.48,
|
| 210 |
+
"learning_rate": 5.758483033932136e-05,
|
| 211 |
+
"loss": 3.2554,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.73,
|
| 216 |
+
"learning_rate": 5.6337325349301393e-05,
|
| 217 |
+
"loss": 3.2215,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 8.98,
|
| 222 |
+
"learning_rate": 5.508982035928144e-05,
|
| 223 |
+
"loss": 3.2006,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 9.23,
|
| 228 |
+
"learning_rate": 5.384231536926148e-05,
|
| 229 |
+
"loss": 3.1627,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.48,
|
| 234 |
+
"learning_rate": 5.259481037924152e-05,
|
| 235 |
+
"loss": 3.1391,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 9.73,
|
| 240 |
+
"learning_rate": 5.1347305389221555e-05,
|
| 241 |
+
"loss": 3.1167,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 9.98,
|
| 246 |
+
"learning_rate": 5.0099800399201604e-05,
|
| 247 |
+
"loss": 3.108,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 10.23,
|
| 252 |
+
"learning_rate": 4.885229540918164e-05,
|
| 253 |
+
"loss": 3.0782,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 10.48,
|
| 258 |
+
"learning_rate": 4.7604790419161675e-05,
|
| 259 |
+
"loss": 3.054,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 10.73,
|
| 264 |
+
"learning_rate": 4.635728542914172e-05,
|
| 265 |
+
"loss": 3.0302,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 10.98,
|
| 270 |
+
"learning_rate": 4.510978043912176e-05,
|
| 271 |
+
"loss": 3.0129,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 11.23,
|
| 276 |
+
"learning_rate": 4.3862275449101795e-05,
|
| 277 |
+
"loss": 2.9978,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 11.48,
|
| 282 |
+
"learning_rate": 4.261477045908184e-05,
|
| 283 |
+
"loss": 2.9658,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 11.73,
|
| 288 |
+
"learning_rate": 4.136726546906188e-05,
|
| 289 |
+
"loss": 2.9536,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 11.98,
|
| 294 |
+
"learning_rate": 4.0119760479041915e-05,
|
| 295 |
+
"loss": 2.937,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 12.23,
|
| 300 |
+
"learning_rate": 3.887225548902196e-05,
|
| 301 |
+
"loss": 2.9144,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 12.48,
|
| 306 |
+
"learning_rate": 3.7624750499002e-05,
|
| 307 |
+
"loss": 2.9045,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 12.72,
|
| 312 |
+
"learning_rate": 3.637724550898204e-05,
|
| 313 |
+
"loss": 2.8814,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 12.97,
|
| 318 |
+
"learning_rate": 3.512974051896208e-05,
|
| 319 |
+
"loss": 2.8749,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 13.22,
|
| 324 |
+
"learning_rate": 3.388223552894212e-05,
|
| 325 |
+
"loss": 2.8333,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 13.47,
|
| 330 |
+
"learning_rate": 3.263473053892216e-05,
|
| 331 |
+
"loss": 2.8402,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 13.72,
|
| 336 |
+
"learning_rate": 3.13872255489022e-05,
|
| 337 |
+
"loss": 2.8363,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 13.97,
|
| 342 |
+
"learning_rate": 3.013972055888224e-05,
|
| 343 |
+
"loss": 2.8235,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 14.22,
|
| 348 |
+
"learning_rate": 2.8892215568862274e-05,
|
| 349 |
+
"loss": 2.8038,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 14.47,
|
| 354 |
+
"learning_rate": 2.7644710578842313e-05,
|
| 355 |
+
"loss": 2.7948,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 14.72,
|
| 360 |
+
"learning_rate": 2.6397205588822355e-05,
|
| 361 |
+
"loss": 2.7762,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 14.97,
|
| 366 |
+
"learning_rate": 2.5149700598802394e-05,
|
| 367 |
+
"loss": 2.7711,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 15.22,
|
| 372 |
+
"learning_rate": 2.3902195608782436e-05,
|
| 373 |
+
"loss": 2.7616,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 15.47,
|
| 378 |
+
"learning_rate": 2.2654690618762475e-05,
|
| 379 |
+
"loss": 2.7325,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 15.72,
|
| 384 |
+
"learning_rate": 2.1407185628742514e-05,
|
| 385 |
+
"loss": 2.7418,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 15.97,
|
| 390 |
+
"learning_rate": 2.0159680638722556e-05,
|
| 391 |
+
"loss": 2.7358,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 16.22,
|
| 396 |
+
"learning_rate": 1.8912175648702595e-05,
|
| 397 |
+
"loss": 2.7215,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 16.47,
|
| 402 |
+
"learning_rate": 1.7664670658682637e-05,
|
| 403 |
+
"loss": 2.7199,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 16.72,
|
| 408 |
+
"learning_rate": 1.6417165668662676e-05,
|
| 409 |
+
"loss": 2.7016,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 16.97,
|
| 414 |
+
"learning_rate": 1.5169660678642716e-05,
|
| 415 |
+
"loss": 2.6966,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 17.22,
|
| 420 |
+
"learning_rate": 1.3922155688622754e-05,
|
| 421 |
+
"loss": 2.6957,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 17.47,
|
| 426 |
+
"learning_rate": 1.2674650698602794e-05,
|
| 427 |
+
"loss": 2.7005,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 17.71,
|
| 432 |
+
"learning_rate": 1.1427145708582835e-05,
|
| 433 |
+
"loss": 2.6819,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 17.96,
|
| 438 |
+
"learning_rate": 1.0179640718562875e-05,
|
| 439 |
+
"loss": 2.6823,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 18.21,
|
| 444 |
+
"learning_rate": 8.932135728542916e-06,
|
| 445 |
+
"loss": 2.6879,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 18.46,
|
| 450 |
+
"learning_rate": 7.684630738522954e-06,
|
| 451 |
+
"loss": 2.6588,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 18.71,
|
| 456 |
+
"learning_rate": 6.437125748502994e-06,
|
| 457 |
+
"loss": 2.6521,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 18.96,
|
| 462 |
+
"learning_rate": 5.189620758483034e-06,
|
| 463 |
+
"loss": 2.6667,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 19.21,
|
| 468 |
+
"learning_rate": 3.942115768463074e-06,
|
| 469 |
+
"loss": 2.6727,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 19.46,
|
| 474 |
+
"learning_rate": 2.6946107784431138e-06,
|
| 475 |
+
"loss": 2.6702,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 19.71,
|
| 480 |
+
"learning_rate": 1.4471057884231539e-06,
|
| 481 |
+
"loss": 2.6484,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 19.96,
|
| 486 |
+
"learning_rate": 1.996007984031936e-07,
|
| 487 |
+
"loss": 2.6556,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 20.0,
|
| 492 |
+
"step": 40080,
|
| 493 |
+
"total_flos": 2362586579435520.0,
|
| 494 |
+
"train_loss": 3.5848853368245197,
|
| 495 |
+
"train_runtime": 32321.6149,
|
| 496 |
+
"train_samples_per_second": 158.707,
|
| 497 |
+
"train_steps_per_second": 1.24
|
| 498 |
+
}
|
| 499 |
+
],
|
| 500 |
+
"max_steps": 40080,
|
| 501 |
+
"num_train_epochs": 20,
|
| 502 |
+
"total_flos": 2362586579435520.0,
|
| 503 |
+
"trial_name": null,
|
| 504 |
+
"trial_params": null
|
| 505 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:555f40070f1833ee4a18a0d4f28a10b78a5e59bf7c845402deef05e140f19cb3
|
| 3 |
+
size 2863
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|