Commit
·
81dbdf0
1
Parent(s):
8f41ba2
🍻 cheers
Browse files- README.md +4 -3
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/Dec04_22-01-11_LAPTOP-PH299HA6/events.out.tfevents.1701745624.LAPTOP-PH299HA6.22060.2 +3 -0
- train_results.json +4 -4
- trainer_state.json +862 -103
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# food-recognition
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# food-recognition
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the food-recognition dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 4.2128
|
| 22 |
+
- Accuracy: 0.3025
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 4.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 3.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.30254545454545456,
|
| 4 |
+
"eval_loss": 4.212757110595703,
|
| 5 |
+
"eval_runtime": 1754.6048,
|
| 6 |
+
"eval_samples_per_second": 3.135,
|
| 7 |
+
"eval_steps_per_second": 0.392,
|
| 8 |
+
"train_loss": 0.32274967552227685,
|
| 9 |
+
"train_runtime": 41589.0325,
|
| 10 |
+
"train_samples_per_second": 0.529,
|
| 11 |
+
"train_steps_per_second": 0.033
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 4.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 3.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.30254545454545456,
|
| 4 |
+
"eval_loss": 4.212757110595703,
|
| 5 |
+
"eval_runtime": 1754.6048,
|
| 6 |
+
"eval_samples_per_second": 3.135,
|
| 7 |
+
"eval_steps_per_second": 0.392
|
| 8 |
}
|
runs/Dec04_22-01-11_LAPTOP-PH299HA6/events.out.tfevents.1701745624.LAPTOP-PH299HA6.22060.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5e8fc363fd273f09fd1ed4112bbfdcc8b44073d402ceb0434dead759ac7fa22
|
| 3 |
+
size 363
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"train_loss": 0.32274967552227685,
|
| 4 |
+
"train_runtime": 41589.0325,
|
| 5 |
+
"train_samples_per_second": 0.529,
|
| 6 |
+
"train_steps_per_second": 0.033
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,208 +1,967 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 4.
|
| 3 |
"best_model_checkpoint": "food-recognition\\checkpoint-100",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 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": 0.
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 0.
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 0.
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 0.
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 0.
|
| 38 |
-
"loss":
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 0.
|
| 44 |
-
"loss":
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch":
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss":
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch":
|
| 55 |
-
"learning_rate": 0.
|
| 56 |
-
"loss":
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"learning_rate": 0.
|
| 62 |
-
"loss":
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch":
|
| 67 |
-
"learning_rate": 0.
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch":
|
| 73 |
-
"eval_accuracy": 0.
|
| 74 |
-
"eval_loss": 4.
|
| 75 |
-
"eval_runtime":
|
| 76 |
-
"eval_samples_per_second": 3.
|
| 77 |
-
"eval_steps_per_second": 0.
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch":
|
| 82 |
-
"learning_rate": 0.
|
| 83 |
-
"loss": 0.
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch":
|
| 88 |
-
"learning_rate": 0.
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"learning_rate": 0.
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch":
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"learning_rate":
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch":
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"learning_rate":
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"learning_rate":
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"learning_rate":
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 4.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second": 3.
|
| 146 |
-
"eval_steps_per_second": 0.
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch":
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch":
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch":
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch":
|
| 175 |
-
"learning_rate":
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"learning_rate":
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 270
|
| 190 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
{
|
| 192 |
"epoch": 4.0,
|
| 193 |
-
"step":
|
| 194 |
-
"total_flos":
|
| 195 |
-
"train_loss": 0.
|
| 196 |
-
"train_runtime":
|
| 197 |
-
"train_samples_per_second":
|
| 198 |
-
"train_steps_per_second": 0.
|
| 199 |
}
|
| 200 |
],
|
| 201 |
"logging_steps": 10,
|
| 202 |
-
"max_steps":
|
| 203 |
"num_train_epochs": 4,
|
| 204 |
"save_steps": 100,
|
| 205 |
-
"total_flos":
|
| 206 |
"trial_name": null,
|
| 207 |
"trial_params": null
|
| 208 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 4.212757110595703,
|
| 3 |
"best_model_checkpoint": "food-recognition\\checkpoint-100",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 1376,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 0.00019854651162790697,
|
| 14 |
+
"loss": 4.0108,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"learning_rate": 0.00019709302325581396,
|
| 20 |
+
"loss": 3.1386,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.09,
|
| 25 |
+
"learning_rate": 0.00019563953488372092,
|
| 26 |
+
"loss": 2.5458,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.12,
|
| 31 |
+
"learning_rate": 0.0001941860465116279,
|
| 32 |
+
"loss": 2.1429,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.15,
|
| 37 |
+
"learning_rate": 0.0001927325581395349,
|
| 38 |
+
"loss": 1.7718,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.17,
|
| 43 |
+
"learning_rate": 0.0001912790697674419,
|
| 44 |
+
"loss": 1.4674,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.2,
|
| 49 |
+
"learning_rate": 0.00018982558139534885,
|
| 50 |
+
"loss": 1.2851,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.23,
|
| 55 |
+
"learning_rate": 0.00018837209302325584,
|
| 56 |
+
"loss": 1.0539,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.26,
|
| 61 |
+
"learning_rate": 0.0001869186046511628,
|
| 62 |
+
"loss": 1.0388,
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.29,
|
| 67 |
+
"learning_rate": 0.00018546511627906976,
|
| 68 |
+
"loss": 0.8835,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.29,
|
| 73 |
+
"eval_accuracy": 0.30254545454545456,
|
| 74 |
+
"eval_loss": 4.212757110595703,
|
| 75 |
+
"eval_runtime": 1673.5767,
|
| 76 |
+
"eval_samples_per_second": 3.286,
|
| 77 |
+
"eval_steps_per_second": 0.411,
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 0.32,
|
| 82 |
+
"learning_rate": 0.00018401162790697675,
|
| 83 |
+
"loss": 0.8656,
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.35,
|
| 88 |
+
"learning_rate": 0.0001825581395348837,
|
| 89 |
+
"loss": 0.8376,
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 0.38,
|
| 94 |
+
"learning_rate": 0.00018110465116279073,
|
| 95 |
+
"loss": 0.8526,
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 0.41,
|
| 100 |
+
"learning_rate": 0.0001796511627906977,
|
| 101 |
+
"loss": 0.713,
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.44,
|
| 106 |
+
"learning_rate": 0.00017819767441860468,
|
| 107 |
+
"loss": 0.6217,
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.47,
|
| 112 |
+
"learning_rate": 0.00017674418604651164,
|
| 113 |
+
"loss": 0.6687,
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.49,
|
| 118 |
+
"learning_rate": 0.0001752906976744186,
|
| 119 |
+
"loss": 0.5977,
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 0.52,
|
| 124 |
+
"learning_rate": 0.0001738372093023256,
|
| 125 |
+
"loss": 0.5849,
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 0.55,
|
| 130 |
+
"learning_rate": 0.00017238372093023255,
|
| 131 |
+
"loss": 0.7185,
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 0.58,
|
| 136 |
+
"learning_rate": 0.00017093023255813954,
|
| 137 |
+
"loss": 0.4912,
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 0.58,
|
| 142 |
+
"eval_accuracy": 0.32345454545454544,
|
| 143 |
+
"eval_loss": 4.7445783615112305,
|
| 144 |
+
"eval_runtime": 1710.3127,
|
| 145 |
+
"eval_samples_per_second": 3.216,
|
| 146 |
+
"eval_steps_per_second": 0.402,
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.61,
|
| 151 |
+
"learning_rate": 0.0001694767441860465,
|
| 152 |
+
"loss": 0.5981,
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.64,
|
| 157 |
+
"learning_rate": 0.00016802325581395352,
|
| 158 |
+
"loss": 0.58,
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 0.67,
|
| 163 |
+
"learning_rate": 0.00016656976744186048,
|
| 164 |
+
"loss": 0.5884,
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 0.7,
|
| 169 |
+
"learning_rate": 0.00016511627906976747,
|
| 170 |
+
"loss": 0.4872,
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.73,
|
| 175 |
+
"learning_rate": 0.00016366279069767443,
|
| 176 |
+
"loss": 0.4864,
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.76,
|
| 181 |
+
"learning_rate": 0.0001622093023255814,
|
| 182 |
+
"loss": 0.5104,
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 0.78,
|
| 187 |
+
"learning_rate": 0.00016075581395348838,
|
| 188 |
+
"loss": 0.5299,
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.81,
|
| 193 |
+
"learning_rate": 0.00015930232558139534,
|
| 194 |
+
"loss": 0.3989,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.84,
|
| 199 |
+
"learning_rate": 0.00015784883720930233,
|
| 200 |
+
"loss": 0.4193,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.87,
|
| 205 |
+
"learning_rate": 0.0001563953488372093,
|
| 206 |
+
"loss": 0.4643,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.87,
|
| 211 |
+
"eval_accuracy": 0.3189090909090909,
|
| 212 |
+
"eval_loss": 5.088333606719971,
|
| 213 |
+
"eval_runtime": 1716.3981,
|
| 214 |
+
"eval_samples_per_second": 3.204,
|
| 215 |
+
"eval_steps_per_second": 0.401,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.9,
|
| 220 |
+
"learning_rate": 0.0001549418604651163,
|
| 221 |
+
"loss": 0.586,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.93,
|
| 226 |
+
"learning_rate": 0.00015348837209302327,
|
| 227 |
+
"loss": 0.5347,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.96,
|
| 232 |
+
"learning_rate": 0.00015203488372093026,
|
| 233 |
+
"loss": 0.4394,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.99,
|
| 238 |
+
"learning_rate": 0.00015058139534883722,
|
| 239 |
+
"loss": 0.4341,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.02,
|
| 244 |
+
"learning_rate": 0.00014912790697674418,
|
| 245 |
+
"loss": 0.3118,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.05,
|
| 250 |
+
"learning_rate": 0.00014767441860465117,
|
| 251 |
+
"loss": 0.2853,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.08,
|
| 256 |
+
"learning_rate": 0.00014622093023255813,
|
| 257 |
+
"loss": 0.2566,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.1,
|
| 262 |
+
"learning_rate": 0.00014476744186046512,
|
| 263 |
+
"loss": 0.2474,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.13,
|
| 268 |
+
"learning_rate": 0.0001433139534883721,
|
| 269 |
+
"loss": 0.1435,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.16,
|
| 274 |
+
"learning_rate": 0.0001418604651162791,
|
| 275 |
+
"loss": 0.247,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.16,
|
| 280 |
+
"eval_accuracy": 0.33290909090909093,
|
| 281 |
+
"eval_loss": 5.445866107940674,
|
| 282 |
+
"eval_runtime": 1717.1708,
|
| 283 |
+
"eval_samples_per_second": 3.203,
|
| 284 |
+
"eval_steps_per_second": 0.401,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 1.19,
|
| 289 |
+
"learning_rate": 0.00014040697674418606,
|
| 290 |
+
"loss": 0.2051,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.22,
|
| 295 |
+
"learning_rate": 0.00013895348837209302,
|
| 296 |
+
"loss": 0.3008,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.25,
|
| 301 |
+
"learning_rate": 0.0001375,
|
| 302 |
+
"loss": 0.2073,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.28,
|
| 307 |
+
"learning_rate": 0.00013604651162790697,
|
| 308 |
+
"loss": 0.2971,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.31,
|
| 313 |
+
"learning_rate": 0.00013459302325581396,
|
| 314 |
+
"loss": 0.2173,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.34,
|
| 319 |
+
"learning_rate": 0.00013313953488372092,
|
| 320 |
+
"loss": 0.2396,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.37,
|
| 325 |
+
"learning_rate": 0.0001316860465116279,
|
| 326 |
+
"loss": 0.2487,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 1.4,
|
| 331 |
+
"learning_rate": 0.0001302325581395349,
|
| 332 |
+
"loss": 0.2698,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 1.42,
|
| 337 |
+
"learning_rate": 0.00012877906976744189,
|
| 338 |
+
"loss": 0.2713,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.45,
|
| 343 |
+
"learning_rate": 0.00012732558139534885,
|
| 344 |
+
"loss": 0.2823,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.45,
|
| 349 |
+
"eval_accuracy": 0.3341818181818182,
|
| 350 |
+
"eval_loss": 5.750086784362793,
|
| 351 |
+
"eval_runtime": 1698.6764,
|
| 352 |
+
"eval_samples_per_second": 3.238,
|
| 353 |
+
"eval_steps_per_second": 0.405,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.48,
|
| 358 |
+
"learning_rate": 0.0001258720930232558,
|
| 359 |
+
"loss": 0.2623,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.51,
|
| 364 |
+
"learning_rate": 0.0001244186046511628,
|
| 365 |
+
"loss": 0.2341,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.54,
|
| 370 |
+
"learning_rate": 0.00012296511627906976,
|
| 371 |
+
"loss": 0.2091,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.57,
|
| 376 |
+
"learning_rate": 0.00012151162790697675,
|
| 377 |
+
"loss": 0.1789,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.6,
|
| 382 |
+
"learning_rate": 0.00012005813953488372,
|
| 383 |
+
"loss": 0.2676,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.63,
|
| 388 |
+
"learning_rate": 0.00011860465116279071,
|
| 389 |
+
"loss": 0.2987,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.66,
|
| 394 |
+
"learning_rate": 0.00011715116279069769,
|
| 395 |
+
"loss": 0.1935,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.69,
|
| 400 |
+
"learning_rate": 0.00011569767441860466,
|
| 401 |
+
"loss": 0.1657,
|
| 402 |
+
"step": 580
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.72,
|
| 406 |
+
"learning_rate": 0.00011424418604651164,
|
| 407 |
+
"loss": 0.2439,
|
| 408 |
+
"step": 590
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.74,
|
| 412 |
+
"learning_rate": 0.00011279069767441861,
|
| 413 |
+
"loss": 0.2834,
|
| 414 |
+
"step": 600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.74,
|
| 418 |
+
"eval_accuracy": 0.3332727272727273,
|
| 419 |
+
"eval_loss": 5.845569610595703,
|
| 420 |
+
"eval_runtime": 1687.9666,
|
| 421 |
+
"eval_samples_per_second": 3.258,
|
| 422 |
+
"eval_steps_per_second": 0.408,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.77,
|
| 427 |
+
"learning_rate": 0.00011133720930232559,
|
| 428 |
+
"loss": 0.1667,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.8,
|
| 433 |
+
"learning_rate": 0.00010988372093023256,
|
| 434 |
+
"loss": 0.1833,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.83,
|
| 439 |
+
"learning_rate": 0.00010843023255813954,
|
| 440 |
+
"loss": 0.2945,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.86,
|
| 445 |
+
"learning_rate": 0.00010697674418604651,
|
| 446 |
+
"loss": 0.1885,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.89,
|
| 451 |
+
"learning_rate": 0.0001055232558139535,
|
| 452 |
+
"loss": 0.1704,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.92,
|
| 457 |
+
"learning_rate": 0.00010406976744186048,
|
| 458 |
+
"loss": 0.1836,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.95,
|
| 463 |
+
"learning_rate": 0.00010261627906976745,
|
| 464 |
+
"loss": 0.1789,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.98,
|
| 469 |
+
"learning_rate": 0.00010116279069767443,
|
| 470 |
+
"loss": 0.1852,
|
| 471 |
+
"step": 680
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.01,
|
| 475 |
+
"learning_rate": 9.97093023255814e-05,
|
| 476 |
+
"loss": 0.2062,
|
| 477 |
+
"step": 690
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.03,
|
| 481 |
+
"learning_rate": 9.825581395348838e-05,
|
| 482 |
+
"loss": 0.0732,
|
| 483 |
+
"step": 700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.03,
|
| 487 |
+
"eval_accuracy": 0.336,
|
| 488 |
+
"eval_loss": 5.888232707977295,
|
| 489 |
+
"eval_runtime": 1707.0859,
|
| 490 |
+
"eval_samples_per_second": 3.222,
|
| 491 |
+
"eval_steps_per_second": 0.403,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.06,
|
| 496 |
+
"learning_rate": 9.680232558139535e-05,
|
| 497 |
+
"loss": 0.0611,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.09,
|
| 502 |
+
"learning_rate": 9.534883720930233e-05,
|
| 503 |
+
"loss": 0.1223,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.12,
|
| 508 |
+
"learning_rate": 9.38953488372093e-05,
|
| 509 |
+
"loss": 0.0292,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.15,
|
| 514 |
+
"learning_rate": 9.244186046511628e-05,
|
| 515 |
+
"loss": 0.0757,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.18,
|
| 520 |
+
"learning_rate": 9.098837209302325e-05,
|
| 521 |
+
"loss": 0.0237,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.21,
|
| 526 |
+
"learning_rate": 8.953488372093024e-05,
|
| 527 |
+
"loss": 0.0236,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.24,
|
| 532 |
+
"learning_rate": 8.808139534883722e-05,
|
| 533 |
+
"loss": 0.0447,
|
| 534 |
+
"step": 770
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.27,
|
| 538 |
+
"learning_rate": 8.662790697674419e-05,
|
| 539 |
+
"loss": 0.0557,
|
| 540 |
+
"step": 780
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.3,
|
| 544 |
+
"learning_rate": 8.517441860465117e-05,
|
| 545 |
+
"loss": 0.1056,
|
| 546 |
+
"step": 790
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.33,
|
| 550 |
+
"learning_rate": 8.372093023255814e-05,
|
| 551 |
+
"loss": 0.0235,
|
| 552 |
+
"step": 800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.33,
|
| 556 |
+
"eval_accuracy": 0.34,
|
| 557 |
+
"eval_loss": 6.1062750816345215,
|
| 558 |
+
"eval_runtime": 1712.0663,
|
| 559 |
+
"eval_samples_per_second": 3.212,
|
| 560 |
+
"eval_steps_per_second": 0.402,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.35,
|
| 565 |
+
"learning_rate": 8.226744186046512e-05,
|
| 566 |
+
"loss": 0.0777,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 2.38,
|
| 571 |
+
"learning_rate": 8.081395348837209e-05,
|
| 572 |
+
"loss": 0.0933,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 2.41,
|
| 577 |
+
"learning_rate": 7.936046511627907e-05,
|
| 578 |
+
"loss": 0.062,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.44,
|
| 583 |
+
"learning_rate": 7.790697674418606e-05,
|
| 584 |
+
"loss": 0.0831,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.47,
|
| 589 |
+
"learning_rate": 7.645348837209303e-05,
|
| 590 |
+
"loss": 0.0892,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.5,
|
| 595 |
+
"learning_rate": 7.500000000000001e-05,
|
| 596 |
+
"loss": 0.0926,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.53,
|
| 601 |
+
"learning_rate": 7.354651162790698e-05,
|
| 602 |
+
"loss": 0.0547,
|
| 603 |
+
"step": 870
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.56,
|
| 607 |
+
"learning_rate": 7.209302325581396e-05,
|
| 608 |
+
"loss": 0.0636,
|
| 609 |
+
"step": 880
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.59,
|
| 613 |
+
"learning_rate": 7.063953488372093e-05,
|
| 614 |
+
"loss": 0.0314,
|
| 615 |
+
"step": 890
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 2.62,
|
| 619 |
+
"learning_rate": 6.918604651162791e-05,
|
| 620 |
+
"loss": 0.046,
|
| 621 |
+
"step": 900
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.62,
|
| 625 |
+
"eval_accuracy": 0.3438181818181818,
|
| 626 |
+
"eval_loss": 6.208089351654053,
|
| 627 |
+
"eval_runtime": 1691.5718,
|
| 628 |
+
"eval_samples_per_second": 3.251,
|
| 629 |
+
"eval_steps_per_second": 0.407,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.65,
|
| 634 |
+
"learning_rate": 6.773255813953488e-05,
|
| 635 |
+
"loss": 0.0698,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.67,
|
| 640 |
+
"learning_rate": 6.627906976744186e-05,
|
| 641 |
+
"loss": 0.0936,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.7,
|
| 646 |
+
"learning_rate": 6.482558139534885e-05,
|
| 647 |
+
"loss": 0.0746,
|
| 648 |
+
"step": 930
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.73,
|
| 652 |
+
"learning_rate": 6.337209302325582e-05,
|
| 653 |
+
"loss": 0.0679,
|
| 654 |
+
"step": 940
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.76,
|
| 658 |
+
"learning_rate": 6.19186046511628e-05,
|
| 659 |
+
"loss": 0.0414,
|
| 660 |
+
"step": 950
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.79,
|
| 664 |
+
"learning_rate": 6.0465116279069765e-05,
|
| 665 |
+
"loss": 0.1188,
|
| 666 |
+
"step": 960
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 2.82,
|
| 670 |
+
"learning_rate": 5.9011627906976754e-05,
|
| 671 |
+
"loss": 0.1271,
|
| 672 |
+
"step": 970
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.85,
|
| 676 |
+
"learning_rate": 5.755813953488373e-05,
|
| 677 |
+
"loss": 0.074,
|
| 678 |
+
"step": 980
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.88,
|
| 682 |
+
"learning_rate": 5.61046511627907e-05,
|
| 683 |
+
"loss": 0.0997,
|
| 684 |
+
"step": 990
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.91,
|
| 688 |
+
"learning_rate": 5.465116279069767e-05,
|
| 689 |
+
"loss": 0.0393,
|
| 690 |
+
"step": 1000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.91,
|
| 694 |
+
"eval_accuracy": 0.3410909090909091,
|
| 695 |
+
"eval_loss": 6.3317975997924805,
|
| 696 |
+
"eval_runtime": 1724.2418,
|
| 697 |
+
"eval_samples_per_second": 3.19,
|
| 698 |
+
"eval_steps_per_second": 0.399,
|
| 699 |
+
"step": 1000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.94,
|
| 703 |
+
"learning_rate": 5.319767441860465e-05,
|
| 704 |
+
"loss": 0.0475,
|
| 705 |
+
"step": 1010
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.97,
|
| 709 |
+
"learning_rate": 5.1744186046511636e-05,
|
| 710 |
+
"loss": 0.0535,
|
| 711 |
+
"step": 1020
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.99,
|
| 715 |
+
"learning_rate": 5.0290697674418605e-05,
|
| 716 |
+
"loss": 0.0215,
|
| 717 |
+
"step": 1030
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.02,
|
| 721 |
+
"learning_rate": 4.883720930232558e-05,
|
| 722 |
+
"loss": 0.0466,
|
| 723 |
+
"step": 1040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 3.05,
|
| 727 |
+
"learning_rate": 4.738372093023256e-05,
|
| 728 |
+
"loss": 0.04,
|
| 729 |
+
"step": 1050
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.08,
|
| 733 |
+
"learning_rate": 4.593023255813954e-05,
|
| 734 |
+
"loss": 0.0538,
|
| 735 |
+
"step": 1060
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 3.11,
|
| 739 |
+
"learning_rate": 4.447674418604651e-05,
|
| 740 |
+
"loss": 0.0111,
|
| 741 |
+
"step": 1070
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.14,
|
| 745 |
+
"learning_rate": 4.302325581395349e-05,
|
| 746 |
+
"loss": 0.014,
|
| 747 |
+
"step": 1080
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.17,
|
| 751 |
+
"learning_rate": 4.156976744186047e-05,
|
| 752 |
+
"loss": 0.0145,
|
| 753 |
+
"step": 1090
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.2,
|
| 757 |
+
"learning_rate": 4.0116279069767444e-05,
|
| 758 |
+
"loss": 0.009,
|
| 759 |
+
"step": 1100
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.2,
|
| 763 |
+
"eval_accuracy": 0.34563636363636363,
|
| 764 |
+
"eval_loss": 6.358022689819336,
|
| 765 |
+
"eval_runtime": 1703.5661,
|
| 766 |
+
"eval_samples_per_second": 3.229,
|
| 767 |
+
"eval_steps_per_second": 0.404,
|
| 768 |
+
"step": 1100
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.23,
|
| 772 |
+
"learning_rate": 3.866279069767442e-05,
|
| 773 |
+
"loss": 0.0094,
|
| 774 |
+
"step": 1110
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 3.26,
|
| 778 |
+
"learning_rate": 3.7209302325581394e-05,
|
| 779 |
+
"loss": 0.0397,
|
| 780 |
+
"step": 1120
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 3.28,
|
| 784 |
+
"learning_rate": 3.5755813953488376e-05,
|
| 785 |
+
"loss": 0.031,
|
| 786 |
+
"step": 1130
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 3.31,
|
| 790 |
+
"learning_rate": 3.430232558139535e-05,
|
| 791 |
+
"loss": 0.0163,
|
| 792 |
+
"step": 1140
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 3.34,
|
| 796 |
+
"learning_rate": 3.284883720930233e-05,
|
| 797 |
+
"loss": 0.0431,
|
| 798 |
+
"step": 1150
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 3.37,
|
| 802 |
+
"learning_rate": 3.13953488372093e-05,
|
| 803 |
+
"loss": 0.0171,
|
| 804 |
+
"step": 1160
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.4,
|
| 808 |
+
"learning_rate": 2.9941860465116277e-05,
|
| 809 |
+
"loss": 0.0116,
|
| 810 |
+
"step": 1170
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 3.43,
|
| 814 |
+
"learning_rate": 2.848837209302326e-05,
|
| 815 |
+
"loss": 0.0091,
|
| 816 |
+
"step": 1180
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.46,
|
| 820 |
+
"learning_rate": 2.703488372093023e-05,
|
| 821 |
+
"loss": 0.0119,
|
| 822 |
+
"step": 1190
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 3.49,
|
| 826 |
+
"learning_rate": 2.5581395348837212e-05,
|
| 827 |
+
"loss": 0.0285,
|
| 828 |
+
"step": 1200
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 3.49,
|
| 832 |
+
"eval_accuracy": 0.34345454545454546,
|
| 833 |
+
"eval_loss": 6.4274139404296875,
|
| 834 |
+
"eval_runtime": 1709.043,
|
| 835 |
+
"eval_samples_per_second": 3.218,
|
| 836 |
+
"eval_steps_per_second": 0.403,
|
| 837 |
+
"step": 1200
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.52,
|
| 841 |
+
"learning_rate": 2.4127906976744188e-05,
|
| 842 |
+
"loss": 0.0093,
|
| 843 |
+
"step": 1210
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 3.55,
|
| 847 |
+
"learning_rate": 2.2674418604651163e-05,
|
| 848 |
+
"loss": 0.0082,
|
| 849 |
+
"step": 1220
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 3.58,
|
| 853 |
+
"learning_rate": 2.1220930232558138e-05,
|
| 854 |
+
"loss": 0.0191,
|
| 855 |
+
"step": 1230
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 3.6,
|
| 859 |
+
"learning_rate": 1.9767441860465116e-05,
|
| 860 |
+
"loss": 0.0081,
|
| 861 |
+
"step": 1240
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 3.63,
|
| 865 |
+
"learning_rate": 1.831395348837209e-05,
|
| 866 |
+
"loss": 0.0457,
|
| 867 |
+
"step": 1250
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 3.66,
|
| 871 |
+
"learning_rate": 1.686046511627907e-05,
|
| 872 |
+
"loss": 0.0211,
|
| 873 |
+
"step": 1260
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 3.69,
|
| 877 |
+
"learning_rate": 1.5406976744186045e-05,
|
| 878 |
+
"loss": 0.0095,
|
| 879 |
+
"step": 1270
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.72,
|
| 883 |
+
"learning_rate": 1.3953488372093024e-05,
|
| 884 |
+
"loss": 0.0076,
|
| 885 |
+
"step": 1280
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 3.75,
|
| 889 |
+
"learning_rate": 1.25e-05,
|
| 890 |
+
"loss": 0.0077,
|
| 891 |
+
"step": 1290
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 3.78,
|
| 895 |
+
"learning_rate": 1.1046511627906977e-05,
|
| 896 |
+
"loss": 0.0095,
|
| 897 |
+
"step": 1300
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 3.78,
|
| 901 |
+
"eval_accuracy": 0.346,
|
| 902 |
+
"eval_loss": 6.443375587463379,
|
| 903 |
+
"eval_runtime": 1699.0765,
|
| 904 |
+
"eval_samples_per_second": 3.237,
|
| 905 |
+
"eval_steps_per_second": 0.405,
|
| 906 |
+
"step": 1300
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 3.81,
|
| 910 |
+
"learning_rate": 9.593023255813954e-06,
|
| 911 |
+
"loss": 0.0076,
|
| 912 |
+
"step": 1310
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 3.84,
|
| 916 |
+
"learning_rate": 8.139534883720931e-06,
|
| 917 |
+
"loss": 0.0163,
|
| 918 |
+
"step": 1320
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 3.87,
|
| 922 |
+
"learning_rate": 6.686046511627907e-06,
|
| 923 |
+
"loss": 0.0272,
|
| 924 |
+
"step": 1330
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 3.9,
|
| 928 |
+
"learning_rate": 5.232558139534884e-06,
|
| 929 |
+
"loss": 0.0123,
|
| 930 |
+
"step": 1340
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 3.92,
|
| 934 |
+
"learning_rate": 3.7790697674418603e-06,
|
| 935 |
+
"loss": 0.0077,
|
| 936 |
+
"step": 1350
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 3.95,
|
| 940 |
+
"learning_rate": 2.325581395348837e-06,
|
| 941 |
+
"loss": 0.0285,
|
| 942 |
+
"step": 1360
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 3.98,
|
| 946 |
+
"learning_rate": 8.72093023255814e-07,
|
| 947 |
+
"loss": 0.0073,
|
| 948 |
+
"step": 1370
|
| 949 |
+
},
|
| 950 |
{
|
| 951 |
"epoch": 4.0,
|
| 952 |
+
"step": 1376,
|
| 953 |
+
"total_flos": 1.7063364713472e+18,
|
| 954 |
+
"train_loss": 0.32274967552227685,
|
| 955 |
+
"train_runtime": 41589.0325,
|
| 956 |
+
"train_samples_per_second": 0.529,
|
| 957 |
+
"train_steps_per_second": 0.033
|
| 958 |
}
|
| 959 |
],
|
| 960 |
"logging_steps": 10,
|
| 961 |
+
"max_steps": 1376,
|
| 962 |
"num_train_epochs": 4,
|
| 963 |
"save_steps": 100,
|
| 964 |
+
"total_flos": 1.7063364713472e+18,
|
| 965 |
"trial_name": null,
|
| 966 |
"trial_params": null
|
| 967 |
}
|