Commit
·
9a9fd01
1
Parent(s):
5983545
Upload 9 files
Browse files- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- source.spm +0 -0
- special_tokens_map.json +5 -0
- target.spm +0 -0
- tokenizer_config.json +13 -0
- trainer_state.json +544 -0
- vocab.json +0 -0
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:483c11f2f7859f42f43940df7ded7b3713b46c6ebf41e3d23048f987bd68d3d1
|
| 3 |
+
size 14575
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:547e4ccf154b7229f6b0a3a28ce9a9ab912111a947498b98b93ffe9a563a9669
|
| 3 |
+
size 557
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b88405c57dbc51e9dec4f7ee277c9d9c2595d1c04838590704924d2e7b081463
|
| 3 |
+
size 627
|
source.spm
ADDED
|
Binary file (790 kB). View file
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
target.spm
ADDED
|
Binary file (815 kB). View file
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"model_max_length": 512,
|
| 4 |
+
"name_or_path": "Helsinki-NLP/opus-mt-en-sv",
|
| 5 |
+
"pad_token": "<pad>",
|
| 6 |
+
"separate_vocabs": false,
|
| 7 |
+
"source_lang": "en",
|
| 8 |
+
"sp_model_kwargs": {},
|
| 9 |
+
"special_tokens_map_file": null,
|
| 10 |
+
"target_lang": "sv",
|
| 11 |
+
"tokenizer_class": "MarianTokenizer",
|
| 12 |
+
"unk_token": "<unk>"
|
| 13 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,544 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.998342410608572,
|
| 5 |
+
"global_step": 38000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 4.940800378877576e-06,
|
| 13 |
+
"loss": 0.9254,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.24,
|
| 18 |
+
"learning_rate": 4.88160075775515e-06,
|
| 19 |
+
"loss": 0.8688,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.36,
|
| 24 |
+
"learning_rate": 4.8224011366327265e-06,
|
| 25 |
+
"loss": 0.8181,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.47,
|
| 30 |
+
"learning_rate": 4.763201515510301e-06,
|
| 31 |
+
"loss": 0.8026,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.59,
|
| 36 |
+
"learning_rate": 4.704120293630121e-06,
|
| 37 |
+
"loss": 0.7869,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.71,
|
| 42 |
+
"learning_rate": 4.644920672507696e-06,
|
| 43 |
+
"loss": 0.7747,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.83,
|
| 48 |
+
"learning_rate": 4.5857210513852715e-06,
|
| 49 |
+
"loss": 0.7666,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.95,
|
| 54 |
+
"learning_rate": 4.526639829505092e-06,
|
| 55 |
+
"loss": 0.751,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.0,
|
| 60 |
+
"eval_bleu": 60.550102386870705,
|
| 61 |
+
"eval_loss": 0.6660878658294678,
|
| 62 |
+
"eval_runtime": 642.2167,
|
| 63 |
+
"eval_samples_per_second": 17.536,
|
| 64 |
+
"eval_steps_per_second": 4.385,
|
| 65 |
+
"step": 4223
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.07,
|
| 69 |
+
"learning_rate": 4.4674402083826666e-06,
|
| 70 |
+
"loss": 0.7241,
|
| 71 |
+
"step": 4500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.18,
|
| 75 |
+
"learning_rate": 4.408240587260242e-06,
|
| 76 |
+
"loss": 0.7178,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.3,
|
| 81 |
+
"learning_rate": 4.349040966137817e-06,
|
| 82 |
+
"loss": 0.7086,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.42,
|
| 87 |
+
"learning_rate": 4.289841345015393e-06,
|
| 88 |
+
"loss": 0.7029,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.54,
|
| 93 |
+
"learning_rate": 4.230641723892967e-06,
|
| 94 |
+
"loss": 0.7171,
|
| 95 |
+
"step": 6500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.66,
|
| 99 |
+
"learning_rate": 4.1714421027705425e-06,
|
| 100 |
+
"loss": 0.6904,
|
| 101 |
+
"step": 7000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.78,
|
| 105 |
+
"learning_rate": 4.112242481648118e-06,
|
| 106 |
+
"loss": 0.7011,
|
| 107 |
+
"step": 7500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.89,
|
| 111 |
+
"learning_rate": 4.053042860525692e-06,
|
| 112 |
+
"loss": 0.6929,
|
| 113 |
+
"step": 8000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.0,
|
| 117 |
+
"eval_bleu": 61.03917142412645,
|
| 118 |
+
"eval_loss": 0.6421462893486023,
|
| 119 |
+
"eval_runtime": 642.6358,
|
| 120 |
+
"eval_samples_per_second": 17.525,
|
| 121 |
+
"eval_steps_per_second": 4.382,
|
| 122 |
+
"step": 8446
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.01,
|
| 126 |
+
"learning_rate": 3.993961638645513e-06,
|
| 127 |
+
"loss": 0.6995,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.13,
|
| 132 |
+
"learning_rate": 3.934762017523088e-06,
|
| 133 |
+
"loss": 0.6573,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.25,
|
| 138 |
+
"learning_rate": 3.875680795642908e-06,
|
| 139 |
+
"loss": 0.6592,
|
| 140 |
+
"step": 9500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.37,
|
| 144 |
+
"learning_rate": 3.8164811745204834e-06,
|
| 145 |
+
"loss": 0.6658,
|
| 146 |
+
"step": 10000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.49,
|
| 150 |
+
"learning_rate": 3.7572815533980584e-06,
|
| 151 |
+
"loss": 0.6613,
|
| 152 |
+
"step": 10500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.6,
|
| 156 |
+
"learning_rate": 3.6980819322756337e-06,
|
| 157 |
+
"loss": 0.6553,
|
| 158 |
+
"step": 11000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.72,
|
| 162 |
+
"learning_rate": 3.639000710395454e-06,
|
| 163 |
+
"loss": 0.6497,
|
| 164 |
+
"step": 11500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.84,
|
| 168 |
+
"learning_rate": 3.579801089273029e-06,
|
| 169 |
+
"loss": 0.6513,
|
| 170 |
+
"step": 12000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.96,
|
| 174 |
+
"learning_rate": 3.520601468150604e-06,
|
| 175 |
+
"loss": 0.6596,
|
| 176 |
+
"step": 12500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.0,
|
| 180 |
+
"eval_bleu": 61.322127725698586,
|
| 181 |
+
"eval_loss": 0.630657434463501,
|
| 182 |
+
"eval_runtime": 654.7731,
|
| 183 |
+
"eval_samples_per_second": 17.2,
|
| 184 |
+
"eval_steps_per_second": 4.301,
|
| 185 |
+
"step": 12669
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.08,
|
| 189 |
+
"learning_rate": 3.461401847028179e-06,
|
| 190 |
+
"loss": 0.6272,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 3.2,
|
| 195 |
+
"learning_rate": 3.4022022259057545e-06,
|
| 196 |
+
"loss": 0.6378,
|
| 197 |
+
"step": 13500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.32,
|
| 201 |
+
"learning_rate": 3.34300260478333e-06,
|
| 202 |
+
"loss": 0.6432,
|
| 203 |
+
"step": 14000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.43,
|
| 207 |
+
"learning_rate": 3.283802983660905e-06,
|
| 208 |
+
"loss": 0.6345,
|
| 209 |
+
"step": 14500
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.55,
|
| 213 |
+
"learning_rate": 3.2246033625384797e-06,
|
| 214 |
+
"loss": 0.6259,
|
| 215 |
+
"step": 15000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.67,
|
| 219 |
+
"learning_rate": 3.1654037414160555e-06,
|
| 220 |
+
"loss": 0.6261,
|
| 221 |
+
"step": 15500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.79,
|
| 225 |
+
"learning_rate": 3.1062041202936305e-06,
|
| 226 |
+
"loss": 0.6176,
|
| 227 |
+
"step": 16000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.91,
|
| 231 |
+
"learning_rate": 3.0471228984134506e-06,
|
| 232 |
+
"loss": 0.6128,
|
| 233 |
+
"step": 16500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 4.0,
|
| 237 |
+
"eval_bleu": 61.4472635817894,
|
| 238 |
+
"eval_loss": 0.6250163912773132,
|
| 239 |
+
"eval_runtime": 654.825,
|
| 240 |
+
"eval_samples_per_second": 17.198,
|
| 241 |
+
"eval_steps_per_second": 4.3,
|
| 242 |
+
"step": 16892
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.03,
|
| 246 |
+
"learning_rate": 2.9879232772910256e-06,
|
| 247 |
+
"loss": 0.6061,
|
| 248 |
+
"step": 17000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.14,
|
| 252 |
+
"learning_rate": 2.9287236561686005e-06,
|
| 253 |
+
"loss": 0.6037,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.26,
|
| 258 |
+
"learning_rate": 2.8695240350461763e-06,
|
| 259 |
+
"loss": 0.6056,
|
| 260 |
+
"step": 18000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.38,
|
| 264 |
+
"learning_rate": 2.810442813165996e-06,
|
| 265 |
+
"loss": 0.6008,
|
| 266 |
+
"step": 18500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.5,
|
| 270 |
+
"learning_rate": 2.751243192043571e-06,
|
| 271 |
+
"loss": 0.6034,
|
| 272 |
+
"step": 19000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.62,
|
| 276 |
+
"learning_rate": 2.6920435709211463e-06,
|
| 277 |
+
"loss": 0.5962,
|
| 278 |
+
"step": 19500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.74,
|
| 282 |
+
"learning_rate": 2.6328439497987212e-06,
|
| 283 |
+
"loss": 0.5997,
|
| 284 |
+
"step": 20000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.85,
|
| 288 |
+
"learning_rate": 2.573644328676297e-06,
|
| 289 |
+
"loss": 0.6001,
|
| 290 |
+
"step": 20500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.97,
|
| 294 |
+
"learning_rate": 2.5145631067961168e-06,
|
| 295 |
+
"loss": 0.5937,
|
| 296 |
+
"step": 21000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 5.0,
|
| 300 |
+
"eval_bleu": 61.68307480515254,
|
| 301 |
+
"eval_loss": 0.6184483766555786,
|
| 302 |
+
"eval_runtime": 657.8509,
|
| 303 |
+
"eval_samples_per_second": 17.119,
|
| 304 |
+
"eval_steps_per_second": 4.281,
|
| 305 |
+
"step": 21115
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.09,
|
| 309 |
+
"learning_rate": 2.4553634856736917e-06,
|
| 310 |
+
"loss": 0.5837,
|
| 311 |
+
"step": 21500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.21,
|
| 315 |
+
"learning_rate": 2.396163864551267e-06,
|
| 316 |
+
"loss": 0.5807,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 5.33,
|
| 321 |
+
"learning_rate": 2.3369642434288424e-06,
|
| 322 |
+
"loss": 0.5835,
|
| 323 |
+
"step": 22500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 5.45,
|
| 327 |
+
"learning_rate": 2.277883021548662e-06,
|
| 328 |
+
"loss": 0.5901,
|
| 329 |
+
"step": 23000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.56,
|
| 333 |
+
"learning_rate": 2.2186834004262375e-06,
|
| 334 |
+
"loss": 0.5794,
|
| 335 |
+
"step": 23500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 5.68,
|
| 339 |
+
"learning_rate": 2.1596021785460573e-06,
|
| 340 |
+
"loss": 0.5814,
|
| 341 |
+
"step": 24000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.8,
|
| 345 |
+
"learning_rate": 2.1004025574236326e-06,
|
| 346 |
+
"loss": 0.5874,
|
| 347 |
+
"step": 24500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 5.92,
|
| 351 |
+
"learning_rate": 2.041202936301208e-06,
|
| 352 |
+
"loss": 0.5852,
|
| 353 |
+
"step": 25000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 6.0,
|
| 357 |
+
"eval_bleu": 61.62760640999204,
|
| 358 |
+
"eval_loss": 0.6162133812904358,
|
| 359 |
+
"eval_runtime": 658.1016,
|
| 360 |
+
"eval_samples_per_second": 17.113,
|
| 361 |
+
"eval_steps_per_second": 4.279,
|
| 362 |
+
"step": 25338
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 6.04,
|
| 366 |
+
"learning_rate": 1.982003315178783e-06,
|
| 367 |
+
"loss": 0.5692,
|
| 368 |
+
"step": 25500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 6.16,
|
| 372 |
+
"learning_rate": 1.9228036940563583e-06,
|
| 373 |
+
"loss": 0.5773,
|
| 374 |
+
"step": 26000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 6.28,
|
| 378 |
+
"learning_rate": 1.8636040729339332e-06,
|
| 379 |
+
"loss": 0.5652,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 6.39,
|
| 384 |
+
"learning_rate": 1.8044044518115086e-06,
|
| 385 |
+
"loss": 0.5635,
|
| 386 |
+
"step": 27000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.51,
|
| 390 |
+
"learning_rate": 1.7452048306890837e-06,
|
| 391 |
+
"loss": 0.5576,
|
| 392 |
+
"step": 27500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 6.63,
|
| 396 |
+
"learning_rate": 1.686005209566659e-06,
|
| 397 |
+
"loss": 0.5811,
|
| 398 |
+
"step": 28000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 6.75,
|
| 402 |
+
"learning_rate": 1.626923987686479e-06,
|
| 403 |
+
"loss": 0.5702,
|
| 404 |
+
"step": 28500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.87,
|
| 408 |
+
"learning_rate": 1.567724366564054e-06,
|
| 409 |
+
"loss": 0.5691,
|
| 410 |
+
"step": 29000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 6.99,
|
| 414 |
+
"learning_rate": 1.5085247454416293e-06,
|
| 415 |
+
"loss": 0.5768,
|
| 416 |
+
"step": 29500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 7.0,
|
| 420 |
+
"eval_bleu": 61.61706518334981,
|
| 421 |
+
"eval_loss": 0.6142455339431763,
|
| 422 |
+
"eval_runtime": 648.5567,
|
| 423 |
+
"eval_samples_per_second": 17.365,
|
| 424 |
+
"eval_steps_per_second": 4.342,
|
| 425 |
+
"step": 29561
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 7.1,
|
| 429 |
+
"learning_rate": 1.4493251243192045e-06,
|
| 430 |
+
"loss": 0.5516,
|
| 431 |
+
"step": 30000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 7.22,
|
| 435 |
+
"learning_rate": 1.3902439024390246e-06,
|
| 436 |
+
"loss": 0.5561,
|
| 437 |
+
"step": 30500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 7.34,
|
| 441 |
+
"learning_rate": 1.3310442813165996e-06,
|
| 442 |
+
"loss": 0.5586,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.46,
|
| 447 |
+
"learning_rate": 1.2718446601941747e-06,
|
| 448 |
+
"loss": 0.5598,
|
| 449 |
+
"step": 31500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 7.58,
|
| 453 |
+
"learning_rate": 1.212763438313995e-06,
|
| 454 |
+
"loss": 0.5591,
|
| 455 |
+
"step": 32000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 7.7,
|
| 459 |
+
"learning_rate": 1.15356381719157e-06,
|
| 460 |
+
"loss": 0.563,
|
| 461 |
+
"step": 32500
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 7.81,
|
| 465 |
+
"learning_rate": 1.0943641960691452e-06,
|
| 466 |
+
"loss": 0.5492,
|
| 467 |
+
"step": 33000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 7.93,
|
| 471 |
+
"learning_rate": 1.0351645749467203e-06,
|
| 472 |
+
"loss": 0.555,
|
| 473 |
+
"step": 33500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 8.0,
|
| 477 |
+
"eval_bleu": 61.82024377811232,
|
| 478 |
+
"eval_loss": 0.6130707859992981,
|
| 479 |
+
"eval_runtime": 628.3785,
|
| 480 |
+
"eval_samples_per_second": 17.922,
|
| 481 |
+
"eval_steps_per_second": 4.481,
|
| 482 |
+
"step": 33784
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 8.05,
|
| 486 |
+
"learning_rate": 9.759649538242955e-07,
|
| 487 |
+
"loss": 0.5637,
|
| 488 |
+
"step": 34000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 8.17,
|
| 492 |
+
"learning_rate": 9.167653327018707e-07,
|
| 493 |
+
"loss": 0.5569,
|
| 494 |
+
"step": 34500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 8.29,
|
| 498 |
+
"learning_rate": 8.576841108216909e-07,
|
| 499 |
+
"loss": 0.5456,
|
| 500 |
+
"step": 35000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.41,
|
| 504 |
+
"learning_rate": 7.98484489699266e-07,
|
| 505 |
+
"loss": 0.5515,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 8.52,
|
| 510 |
+
"learning_rate": 7.392848685768411e-07,
|
| 511 |
+
"loss": 0.5521,
|
| 512 |
+
"step": 36000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 8.64,
|
| 516 |
+
"learning_rate": 6.800852474544164e-07,
|
| 517 |
+
"loss": 0.5382,
|
| 518 |
+
"step": 36500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 8.76,
|
| 522 |
+
"learning_rate": 6.208856263319915e-07,
|
| 523 |
+
"loss": 0.5494,
|
| 524 |
+
"step": 37000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 8.88,
|
| 528 |
+
"learning_rate": 5.618044044518116e-07,
|
| 529 |
+
"loss": 0.5504,
|
| 530 |
+
"step": 37500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 9.0,
|
| 534 |
+
"learning_rate": 5.026047833293867e-07,
|
| 535 |
+
"loss": 0.5516,
|
| 536 |
+
"step": 38000
|
| 537 |
+
}
|
| 538 |
+
],
|
| 539 |
+
"max_steps": 42230,
|
| 540 |
+
"num_train_epochs": 10,
|
| 541 |
+
"total_flos": 1.95716976869376e+16,
|
| 542 |
+
"trial_name": null,
|
| 543 |
+
"trial_params": null
|
| 544 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|