Commit
·
b40530e
1
Parent(s):
80cb67e
End of training
Browse files- README.md +5 -5
- all_results.json +13 -13
- eval_results.json +8 -8
- train_results.json +6 -6
- trainer_state.json +288 -42
README.md
CHANGED
|
@@ -14,7 +14,7 @@ model-index:
|
|
| 14 |
name: Masked Language Modeling
|
| 15 |
type: fill-mask
|
| 16 |
dataset:
|
| 17 |
-
name: wikitext
|
| 18 |
type: wikitext
|
| 19 |
config: wikitext-2-raw-v1
|
| 20 |
split: validation
|
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -30,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# mlm
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the wikitext dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 1.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 14 |
name: Masked Language Modeling
|
| 15 |
type: fill-mask
|
| 16 |
dataset:
|
| 17 |
+
name: wikitext wikitext-2-raw-v1
|
| 18 |
type: wikitext
|
| 19 |
config: wikitext-2-raw-v1
|
| 20 |
split: validation
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.7302927161334241
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 30 |
|
| 31 |
# mlm
|
| 32 |
|
| 33 |
+
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the wikitext wikitext-2-raw-v1 dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.2468
|
| 36 |
+
- Accuracy: 0.7303
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"perplexity":
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples":
|
| 13 |
-
"train_samples_per_second":
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.7302927161334241,
|
| 4 |
+
"eval_loss": 1.2467830181121826,
|
| 5 |
+
"eval_runtime": 14.0489,
|
| 6 |
+
"eval_samples": 496,
|
| 7 |
+
"eval_samples_per_second": 35.305,
|
| 8 |
+
"eval_steps_per_second": 4.413,
|
| 9 |
+
"perplexity": 3.479132628765549,
|
| 10 |
+
"train_loss": 1.4182749218410915,
|
| 11 |
+
"train_runtime": 1171.1026,
|
| 12 |
+
"train_samples": 4798,
|
| 13 |
+
"train_samples_per_second": 12.291,
|
| 14 |
+
"train_steps_per_second": 0.384
|
| 15 |
}
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"perplexity":
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 0.7302927161334241,
|
| 4 |
+
"eval_loss": 1.2467830181121826,
|
| 5 |
+
"eval_runtime": 14.0489,
|
| 6 |
+
"eval_samples": 496,
|
| 7 |
+
"eval_samples_per_second": 35.305,
|
| 8 |
+
"eval_steps_per_second": 4.413,
|
| 9 |
+
"perplexity": 3.479132628765549
|
| 10 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.4182749218410915,
|
| 4 |
+
"train_runtime": 1171.1026,
|
| 5 |
+
"train_samples": 4798,
|
| 6 |
+
"train_samples_per_second": 12.291,
|
| 7 |
+
"train_steps_per_second": 0.384
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,79 +1,325 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": ".\\output\\checkpoint-
|
| 4 |
-
"epoch": 0
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 1.
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss":
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"eval_samples_per_second": 2.366,
|
| 59 |
-
"eval_steps_per_second": 0.296,
|
| 60 |
-
"step": 72
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch": 0.
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"
|
| 67 |
-
|
| 68 |
-
|
| 69 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
}
|
| 71 |
],
|
| 72 |
"logging_steps": 10,
|
| 73 |
-
"max_steps":
|
| 74 |
-
"num_train_epochs":
|
| 75 |
"save_steps": 100.0,
|
| 76 |
-
"total_flos":
|
| 77 |
"trial_name": null,
|
| 78 |
"trial_params": null
|
| 79 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7278010101558682,
|
| 3 |
+
"best_model_checkpoint": ".\\output\\checkpoint-450",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 450,
|
| 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": 6e-06,
|
| 14 |
+
"loss": 1.8646,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 20 |
+
"loss": 1.7999,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.2,
|
| 25 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 26 |
+
"loss": 1.6122,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.27,
|
| 31 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 32 |
+
"loss": 1.5558,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.33,
|
| 37 |
+
"learning_rate": 2.9703703703703707e-05,
|
| 38 |
+
"loss": 1.4972,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.4,
|
| 43 |
+
"learning_rate": 2.8962962962962965e-05,
|
| 44 |
+
"loss": 1.482,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.47,
|
| 49 |
+
"learning_rate": 2.8222222222222223e-05,
|
| 50 |
+
"loss": 1.4642,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.53,
|
| 55 |
+
"learning_rate": 2.7481481481481482e-05,
|
| 56 |
+
"loss": 1.4455,
|
| 57 |
+
"step": 80
|
|
|
|
|
|
|
|
|
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.6,
|
| 61 |
+
"learning_rate": 2.6740740740740743e-05,
|
| 62 |
+
"loss": 1.4228,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.67,
|
| 67 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 68 |
+
"loss": 1.4093,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.73,
|
| 73 |
+
"learning_rate": 2.525925925925926e-05,
|
| 74 |
+
"loss": 1.3687,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.8,
|
| 79 |
+
"learning_rate": 2.451851851851852e-05,
|
| 80 |
+
"loss": 1.4491,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.87,
|
| 85 |
+
"learning_rate": 2.377777777777778e-05,
|
| 86 |
+
"loss": 1.4152,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.93,
|
| 91 |
+
"learning_rate": 2.303703703703704e-05,
|
| 92 |
+
"loss": 1.4672,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"learning_rate": 2.2296296296296297e-05,
|
| 98 |
+
"loss": 1.3758,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.0,
|
| 103 |
+
"eval_accuracy": 0.7276595744680852,
|
| 104 |
+
"eval_loss": 1.2825729846954346,
|
| 105 |
+
"eval_runtime": 14.8174,
|
| 106 |
+
"eval_samples_per_second": 33.474,
|
| 107 |
+
"eval_steps_per_second": 4.184,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.07,
|
| 112 |
+
"learning_rate": 2.155555555555556e-05,
|
| 113 |
+
"loss": 1.372,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.13,
|
| 118 |
+
"learning_rate": 2.0814814814814817e-05,
|
| 119 |
+
"loss": 1.4247,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.2,
|
| 124 |
+
"learning_rate": 2.0074074074074075e-05,
|
| 125 |
+
"loss": 1.3985,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.27,
|
| 130 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 131 |
+
"loss": 1.3752,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.33,
|
| 136 |
+
"learning_rate": 1.8592592592592595e-05,
|
| 137 |
+
"loss": 1.4129,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.4,
|
| 142 |
+
"learning_rate": 1.7851851851851853e-05,
|
| 143 |
+
"loss": 1.3847,
|
| 144 |
+
"step": 210
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.47,
|
| 148 |
+
"learning_rate": 1.7111111111111112e-05,
|
| 149 |
+
"loss": 1.3855,
|
| 150 |
+
"step": 220
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.53,
|
| 154 |
+
"learning_rate": 1.6370370370370374e-05,
|
| 155 |
+
"loss": 1.3916,
|
| 156 |
+
"step": 230
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.6,
|
| 160 |
+
"learning_rate": 1.5629629629629632e-05,
|
| 161 |
+
"loss": 1.4184,
|
| 162 |
+
"step": 240
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.67,
|
| 166 |
+
"learning_rate": 1.4888888888888888e-05,
|
| 167 |
+
"loss": 1.3827,
|
| 168 |
+
"step": 250
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.73,
|
| 172 |
+
"learning_rate": 1.4148148148148148e-05,
|
| 173 |
+
"loss": 1.3742,
|
| 174 |
+
"step": 260
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.8,
|
| 178 |
+
"learning_rate": 1.3407407407407408e-05,
|
| 179 |
+
"loss": 1.4016,
|
| 180 |
+
"step": 270
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.87,
|
| 184 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 185 |
+
"loss": 1.3888,
|
| 186 |
+
"step": 280
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.93,
|
| 190 |
+
"learning_rate": 1.1925925925925927e-05,
|
| 191 |
+
"loss": 1.3748,
|
| 192 |
+
"step": 290
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.0,
|
| 196 |
+
"learning_rate": 1.1185185185185185e-05,
|
| 197 |
+
"loss": 1.3763,
|
| 198 |
+
"step": 300
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.0,
|
| 202 |
+
"eval_accuracy": 0.727228590694538,
|
| 203 |
+
"eval_loss": 1.2746729850769043,
|
| 204 |
+
"eval_runtime": 14.4425,
|
| 205 |
+
"eval_samples_per_second": 34.343,
|
| 206 |
+
"eval_steps_per_second": 4.293,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.07,
|
| 211 |
+
"learning_rate": 1.0444444444444445e-05,
|
| 212 |
+
"loss": 1.3781,
|
| 213 |
+
"step": 310
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.13,
|
| 217 |
+
"learning_rate": 9.703703703703703e-06,
|
| 218 |
+
"loss": 1.4019,
|
| 219 |
+
"step": 320
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.2,
|
| 223 |
+
"learning_rate": 8.962962962962963e-06,
|
| 224 |
+
"loss": 1.3862,
|
| 225 |
+
"step": 330
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.27,
|
| 229 |
+
"learning_rate": 8.222222222222222e-06,
|
| 230 |
+
"loss": 1.3784,
|
| 231 |
+
"step": 340
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.33,
|
| 235 |
+
"learning_rate": 7.481481481481482e-06,
|
| 236 |
+
"loss": 1.3407,
|
| 237 |
+
"step": 350
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.4,
|
| 241 |
+
"learning_rate": 6.740740740740741e-06,
|
| 242 |
+
"loss": 1.3496,
|
| 243 |
+
"step": 360
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.47,
|
| 247 |
+
"learning_rate": 6e-06,
|
| 248 |
+
"loss": 1.3681,
|
| 249 |
+
"step": 370
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.53,
|
| 253 |
+
"learning_rate": 5.25925925925926e-06,
|
| 254 |
+
"loss": 1.3453,
|
| 255 |
+
"step": 380
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.6,
|
| 259 |
+
"learning_rate": 4.518518518518519e-06,
|
| 260 |
+
"loss": 1.3294,
|
| 261 |
+
"step": 390
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.67,
|
| 265 |
+
"learning_rate": 3.7777777777777777e-06,
|
| 266 |
+
"loss": 1.3361,
|
| 267 |
+
"step": 400
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.73,
|
| 271 |
+
"learning_rate": 3.0370370370370372e-06,
|
| 272 |
+
"loss": 1.3615,
|
| 273 |
+
"step": 410
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.8,
|
| 277 |
+
"learning_rate": 2.2962962962962964e-06,
|
| 278 |
+
"loss": 1.3104,
|
| 279 |
+
"step": 420
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.87,
|
| 283 |
+
"learning_rate": 1.5555555555555556e-06,
|
| 284 |
+
"loss": 1.3395,
|
| 285 |
+
"step": 430
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.93,
|
| 289 |
+
"learning_rate": 8.148148148148149e-07,
|
| 290 |
+
"loss": 1.3499,
|
| 291 |
+
"step": 440
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.0,
|
| 295 |
+
"learning_rate": 7.407407407407407e-08,
|
| 296 |
+
"loss": 1.3558,
|
| 297 |
+
"step": 450
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.0,
|
| 301 |
+
"eval_accuracy": 0.7278010101558682,
|
| 302 |
+
"eval_loss": 1.260701298713684,
|
| 303 |
+
"eval_runtime": 14.198,
|
| 304 |
+
"eval_samples_per_second": 34.935,
|
| 305 |
+
"eval_steps_per_second": 4.367,
|
| 306 |
+
"step": 450
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 3.0,
|
| 310 |
+
"step": 450,
|
| 311 |
+
"total_flos": 3789443078682624.0,
|
| 312 |
+
"train_loss": 1.4182749218410915,
|
| 313 |
+
"train_runtime": 1171.1026,
|
| 314 |
+
"train_samples_per_second": 12.291,
|
| 315 |
+
"train_steps_per_second": 0.384
|
| 316 |
}
|
| 317 |
],
|
| 318 |
"logging_steps": 10,
|
| 319 |
+
"max_steps": 450,
|
| 320 |
+
"num_train_epochs": 3,
|
| 321 |
"save_steps": 100.0,
|
| 322 |
+
"total_flos": 3789443078682624.0,
|
| 323 |
"trial_name": null,
|
| 324 |
"trial_params": null
|
| 325 |
}
|