End of training
Browse files- README.md +11 -46
- all_results.json +11 -11
- config.json +4 -14
- eval_results.json +6 -6
- model.safetensors +2 -2
- runs/Sep01_13-00-27_c4735777ea3c/events.out.tfevents.1725195641.c4735777ea3c.36.0 +3 -0
- runs/Sep01_13-01-10_c4735777ea3c/events.out.tfevents.1725195681.c4735777ea3c.36.1 +3 -0
- runs/Sep01_13-01-10_c4735777ea3c/events.out.tfevents.1725200277.c4735777ea3c.36.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +569 -863
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -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
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
@@ -61,52 +61,17 @@ The following hyperparameters were used during training:
|
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
- lr_scheduler_warmup_ratio: 0.1
|
| 64 |
-
- num_epochs:
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
| 68 |
-
| Training Loss | Epoch
|
| 69 |
-
|
| 70 |
-
|
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.4073 | 6.0 | 189 | 0.4398 | 0.8462 |
|
| 76 |
-
| 0.4035 | 6.9841 | 220 | 0.4284 | 0.8487 |
|
| 77 |
-
| 0.3609 | 8.0 | 252 | 0.3886 | 0.8542 |
|
| 78 |
-
| 0.3196 | 8.9841 | 283 | 0.4561 | 0.8432 |
|
| 79 |
-
| 0.2723 | 10.0 | 315 | 0.3703 | 0.8697 |
|
| 80 |
-
| 0.2521 | 10.9841 | 346 | 0.3639 | 0.8722 |
|
| 81 |
-
| 0.2644 | 12.0 | 378 | 0.3288 | 0.8832 |
|
| 82 |
-
| 0.2282 | 12.9841 | 409 | 0.3625 | 0.8712 |
|
| 83 |
-
| 0.2435 | 14.0 | 441 | 0.3175 | 0.8962 |
|
| 84 |
-
| 0.2051 | 14.9841 | 472 | 0.3649 | 0.8707 |
|
| 85 |
-
| 0.1922 | 16.0 | 504 | 0.3022 | 0.8952 |
|
| 86 |
-
| 0.1824 | 16.9841 | 535 | 0.3596 | 0.8752 |
|
| 87 |
-
| 0.1799 | 18.0 | 567 | 0.3293 | 0.8942 |
|
| 88 |
-
| 0.1562 | 18.9841 | 598 | 0.3204 | 0.8992 |
|
| 89 |
-
| 0.1582 | 20.0 | 630 | 0.3467 | 0.8837 |
|
| 90 |
-
| 0.1516 | 20.9841 | 661 | 0.3247 | 0.8942 |
|
| 91 |
-
| 0.1285 | 22.0 | 693 | 0.3304 | 0.8912 |
|
| 92 |
-
| 0.1454 | 22.9841 | 724 | 0.3031 | 0.8957 |
|
| 93 |
-
| 0.1548 | 24.0 | 756 | 0.3086 | 0.8992 |
|
| 94 |
-
| 0.1041 | 24.9841 | 787 | 0.2945 | 0.9021 |
|
| 95 |
-
| 0.1161 | 26.0 | 819 | 0.2968 | 0.9106 |
|
| 96 |
-
| 0.1141 | 26.9841 | 850 | 0.2805 | 0.9096 |
|
| 97 |
-
| 0.1078 | 28.0 | 882 | 0.3178 | 0.9011 |
|
| 98 |
-
| 0.1192 | 28.9841 | 913 | 0.3182 | 0.9041 |
|
| 99 |
-
| 0.0977 | 30.0 | 945 | 0.3000 | 0.9061 |
|
| 100 |
-
| 0.1011 | 30.9841 | 976 | 0.3065 | 0.9041 |
|
| 101 |
-
| 0.0865 | 32.0 | 1008 | 0.3193 | 0.9051 |
|
| 102 |
-
| 0.0845 | 32.9841 | 1039 | 0.3047 | 0.9121 |
|
| 103 |
-
| 0.0823 | 34.0 | 1071 | 0.3037 | 0.9116 |
|
| 104 |
-
| 0.0809 | 34.9841 | 1102 | 0.3329 | 0.9011 |
|
| 105 |
-
| 0.0789 | 36.0 | 1134 | 0.3215 | 0.9121 |
|
| 106 |
-
| 0.0724 | 36.9841 | 1165 | 0.3273 | 0.9096 |
|
| 107 |
-
| 0.0722 | 38.0 | 1197 | 0.3096 | 0.9091 |
|
| 108 |
-
| 0.0811 | 38.9841 | 1228 | 0.3206 | 0.9126 |
|
| 109 |
-
| 0.0659 | 39.3651 | 1240 | 0.3216 | 0.9126 |
|
| 110 |
|
| 111 |
|
| 112 |
### Framework versions
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.994671729544341
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.0256
|
| 36 |
+
- Accuracy: 0.9947
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 61 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 62 |
- lr_scheduler_type: linear
|
| 63 |
- lr_scheduler_warmup_ratio: 0.1
|
| 64 |
+
- num_epochs: 5
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
| 68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
+
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 70 |
+
| 0.0268 | 0.9990 | 255 | 0.0256 | 0.9947 |
|
| 71 |
+
| 0.0167 | 1.9980 | 510 | 0.0275 | 0.9947 |
|
| 72 |
+
| 0.0177 | 2.9971 | 765 | 0.0268 | 0.9936 |
|
| 73 |
+
| 0.0158 | 4.0 | 1021 | 0.0238 | 0.9945 |
|
| 74 |
+
| 0.0112 | 4.9951 | 1275 | 0.0259 | 0.9944 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
|
| 76 |
|
| 77 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.995102840352596,
|
| 3 |
+
"eval_accuracy": 0.994671729544341,
|
| 4 |
+
"eval_loss": 0.025641364976763725,
|
| 5 |
+
"eval_runtime": 102.6469,
|
| 6 |
+
"eval_samples_per_second": 159.07,
|
| 7 |
+
"eval_steps_per_second": 2.494,
|
| 8 |
+
"total_flos": 8.109813238393209e+18,
|
| 9 |
+
"train_loss": 0.030976083857171675,
|
| 10 |
+
"train_runtime": 4473.2861,
|
| 11 |
+
"train_samples_per_second": 73.002,
|
| 12 |
+
"train_steps_per_second": 0.285
|
| 13 |
}
|
config.json
CHANGED
|
@@ -17,24 +17,14 @@
|
|
| 17 |
"hidden_dropout_prob": 0.0,
|
| 18 |
"hidden_size": 768,
|
| 19 |
"id2label": {
|
| 20 |
-
"0": "
|
| 21 |
-
"1": "
|
| 22 |
-
"2": "bkl",
|
| 23 |
-
"3": "df",
|
| 24 |
-
"4": "mel",
|
| 25 |
-
"5": "nv",
|
| 26 |
-
"6": "vasc"
|
| 27 |
},
|
| 28 |
"image_size": 224,
|
| 29 |
"initializer_range": 0.02,
|
| 30 |
"label2id": {
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
"bkl": 2,
|
| 34 |
-
"df": 3,
|
| 35 |
-
"mel": 4,
|
| 36 |
-
"nv": 5,
|
| 37 |
-
"vasc": 6
|
| 38 |
},
|
| 39 |
"layer_norm_eps": 1e-05,
|
| 40 |
"mlp_ratio": 4.0,
|
|
|
|
| 17 |
"hidden_dropout_prob": 0.0,
|
| 18 |
"hidden_size": 768,
|
| 19 |
"id2label": {
|
| 20 |
+
"0": "images_class_0",
|
| 21 |
+
"1": "images_class_1"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 22 |
},
|
| 23 |
"image_size": 224,
|
| 24 |
"initializer_range": 0.02,
|
| 25 |
"label2id": {
|
| 26 |
+
"images_class_0": 0,
|
| 27 |
+
"images_class_1": 1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
},
|
| 29 |
"layer_norm_eps": 1e-05,
|
| 30 |
"mlp_ratio": 4.0,
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.995102840352596,
|
| 3 |
+
"eval_accuracy": 0.994671729544341,
|
| 4 |
+
"eval_loss": 0.025641364976763725,
|
| 5 |
+
"eval_runtime": 102.6469,
|
| 6 |
+
"eval_samples_per_second": 159.07,
|
| 7 |
+
"eval_steps_per_second": 2.494
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e99d284193dff48e7408acb1e292b749fb80dfded99dd7a9cac006f9cedcd63
|
| 3 |
+
size 110342832
|
runs/Sep01_13-00-27_c4735777ea3c/events.out.tfevents.1725195641.c4735777ea3c.36.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:460c1abd62a221991453d26571558712f0be7b82e9d8844149ce04890437fb34
|
| 3 |
+
size 5415
|
runs/Sep01_13-01-10_c4735777ea3c/events.out.tfevents.1725195681.c4735777ea3c.36.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a546d17d197aa389b0ba06ae202fe68078c9f0d6a0c76655956e79a7e7f0e5f
|
| 3 |
+
size 34131
|
runs/Sep01_13-01-10_c4735777ea3c/events.out.tfevents.1725200277.c4735777ea3c.36.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ede407413d2c9b7ea17e049da582d1cee5f855b6df7e41b4c24169224716570a
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.995102840352596,
|
| 3 |
+
"total_flos": 8.109813238393209e+18,
|
| 4 |
+
"train_loss": 0.030976083857171675,
|
| 5 |
+
"train_runtime": 4473.2861,
|
| 6 |
+
"train_samples_per_second": 73.002,
|
| 7 |
+
"train_steps_per_second": 0.285
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1255 +1,961 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-
|
| 4 |
-
"epoch":
|
| 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 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 1.
|
| 29 |
-
"loss":
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"
|
| 35 |
-
"
|
| 36 |
-
"
|
| 37 |
-
"eval_samples_per_second": 58.445,
|
| 38 |
-
"eval_steps_per_second": 0.934,
|
| 39 |
-
"step": 31
|
| 40 |
-
},
|
| 41 |
-
{
|
| 42 |
-
"epoch": 1.2698412698412698,
|
| 43 |
-
"grad_norm": 6.524947166442871,
|
| 44 |
-
"learning_rate": 1.6129032258064517e-05,
|
| 45 |
-
"loss": 0.929,
|
| 46 |
"step": 40
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch":
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss": 0.
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch":
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"learning_rate": 2.
|
| 59 |
-
"loss": 0.
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"epoch":
|
| 64 |
-
"
|
| 65 |
-
"
|
| 66 |
-
"
|
| 67 |
-
"eval_samples_per_second": 80.502,
|
| 68 |
-
"eval_steps_per_second": 1.286,
|
| 69 |
-
"step": 63
|
| 70 |
-
},
|
| 71 |
-
{
|
| 72 |
-
"epoch": 2.2222222222222223,
|
| 73 |
-
"grad_norm": 3.6991262435913086,
|
| 74 |
-
"learning_rate": 2.822580645161291e-05,
|
| 75 |
-
"loss": 0.7149,
|
| 76 |
"step": 70
|
| 77 |
},
|
| 78 |
{
|
| 79 |
-
"epoch":
|
| 80 |
-
"grad_norm":
|
| 81 |
-
"learning_rate": 3.
|
| 82 |
-
"loss": 0.
|
| 83 |
"step": 80
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"grad_norm":
|
| 88 |
-
"learning_rate": 3.
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 90
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch":
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"
|
| 97 |
-
"eval_samples_per_second": 80.58,
|
| 98 |
-
"eval_steps_per_second": 1.287,
|
| 99 |
-
"step": 94
|
| 100 |
-
},
|
| 101 |
-
{
|
| 102 |
-
"epoch": 3.1746031746031744,
|
| 103 |
-
"grad_norm": 9.863725662231445,
|
| 104 |
-
"learning_rate": 4.032258064516129e-05,
|
| 105 |
-
"loss": 0.5768,
|
| 106 |
"step": 100
|
| 107 |
},
|
| 108 |
{
|
| 109 |
-
"epoch":
|
| 110 |
-
"grad_norm":
|
| 111 |
-
"learning_rate": 4.
|
| 112 |
-
"loss": 0.
|
| 113 |
"step": 110
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch":
|
| 117 |
-
"grad_norm":
|
| 118 |
-
"learning_rate": 4.
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 120
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"
|
| 125 |
-
"
|
| 126 |
-
"
|
| 127 |
-
"eval_samples_per_second": 81.79,
|
| 128 |
-
"eval_steps_per_second": 1.307,
|
| 129 |
-
"step": 126
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 4.1269841269841265,
|
| 133 |
-
"grad_norm": 6.913752555847168,
|
| 134 |
-
"learning_rate": 4.973118279569893e-05,
|
| 135 |
-
"loss": 0.4906,
|
| 136 |
"step": 130
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch":
|
| 140 |
-
"grad_norm":
|
| 141 |
-
"learning_rate": 4.
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 140
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch":
|
| 147 |
-
"grad_norm":
|
| 148 |
-
"learning_rate": 4.
|
| 149 |
-
"loss": 0.
|
| 150 |
"step": 150
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"epoch":
|
| 154 |
-
"
|
| 155 |
-
"
|
| 156 |
-
"
|
| 157 |
-
"eval_samples_per_second": 80.831,
|
| 158 |
-
"eval_steps_per_second": 1.291,
|
| 159 |
-
"step": 157
|
| 160 |
-
},
|
| 161 |
-
{
|
| 162 |
-
"epoch": 5.079365079365079,
|
| 163 |
-
"grad_norm": 4.5059404373168945,
|
| 164 |
-
"learning_rate": 4.8387096774193554e-05,
|
| 165 |
-
"loss": 0.4574,
|
| 166 |
"step": 160
|
| 167 |
},
|
| 168 |
{
|
| 169 |
-
"epoch":
|
| 170 |
-
"grad_norm":
|
| 171 |
-
"learning_rate": 4.
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 170
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"grad_norm":
|
| 178 |
-
"learning_rate": 4.
|
| 179 |
-
"loss": 0.
|
| 180 |
"step": 180
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch":
|
| 184 |
-
"
|
| 185 |
-
"
|
| 186 |
-
"
|
| 187 |
-
"eval_samples_per_second": 82.513,
|
| 188 |
-
"eval_steps_per_second": 1.318,
|
| 189 |
-
"step": 189
|
| 190 |
-
},
|
| 191 |
-
{
|
| 192 |
-
"epoch": 6.031746031746032,
|
| 193 |
-
"grad_norm": 4.483707904815674,
|
| 194 |
-
"learning_rate": 4.704301075268818e-05,
|
| 195 |
-
"loss": 0.4442,
|
| 196 |
"step": 190
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"epoch":
|
| 200 |
-
"grad_norm":
|
| 201 |
-
"learning_rate": 4.
|
| 202 |
-
"loss": 0.
|
| 203 |
"step": 200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch":
|
| 207 |
-
"grad_norm":
|
| 208 |
-
"learning_rate": 4.
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 210
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch":
|
| 214 |
-
"grad_norm":
|
| 215 |
-
"learning_rate": 4.
|
| 216 |
-
"loss": 0.
|
| 217 |
-
"step": 220
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"epoch": 6.984126984126984,
|
| 221 |
-
"eval_accuracy": 0.8487269096355466,
|
| 222 |
-
"eval_loss": 0.4284241795539856,
|
| 223 |
-
"eval_runtime": 24.7921,
|
| 224 |
-
"eval_samples_per_second": 80.792,
|
| 225 |
-
"eval_steps_per_second": 1.291,
|
| 226 |
"step": 220
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch":
|
| 230 |
-
"grad_norm":
|
| 231 |
-
"learning_rate": 4.
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 230
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"grad_norm":
|
| 238 |
-
"learning_rate": 4.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 240
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch":
|
| 244 |
-
"grad_norm":
|
| 245 |
-
"learning_rate": 4.
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 250
|
| 248 |
},
|
| 249 |
{
|
| 250 |
-
"epoch":
|
| 251 |
-
"eval_accuracy": 0.
|
| 252 |
-
"eval_loss": 0.
|
| 253 |
-
"eval_runtime":
|
| 254 |
-
"eval_samples_per_second":
|
| 255 |
-
"eval_steps_per_second": 1.
|
| 256 |
-
"step":
|
| 257 |
},
|
| 258 |
{
|
| 259 |
-
"epoch":
|
| 260 |
-
"grad_norm":
|
| 261 |
-
"learning_rate": 4.
|
| 262 |
-
"loss": 0.
|
| 263 |
"step": 260
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"epoch":
|
| 267 |
-
"grad_norm":
|
| 268 |
-
"learning_rate": 4.
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 270
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"grad_norm":
|
| 275 |
-
"learning_rate": 4.
|
| 276 |
-
"loss": 0.
|
| 277 |
"step": 280
|
| 278 |
},
|
| 279 |
{
|
| 280 |
-
"epoch":
|
| 281 |
-
"
|
| 282 |
-
"
|
| 283 |
-
"
|
| 284 |
-
"eval_samples_per_second": 79.458,
|
| 285 |
-
"eval_steps_per_second": 1.269,
|
| 286 |
-
"step": 283
|
| 287 |
-
},
|
| 288 |
-
{
|
| 289 |
-
"epoch": 9.206349206349206,
|
| 290 |
-
"grad_norm": 5.623856544494629,
|
| 291 |
-
"learning_rate": 4.256272401433692e-05,
|
| 292 |
-
"loss": 0.3088,
|
| 293 |
"step": 290
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"epoch":
|
| 297 |
-
"grad_norm":
|
| 298 |
-
"learning_rate": 4.
|
| 299 |
-
"loss": 0.
|
| 300 |
"step": 300
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"epoch":
|
| 304 |
-
"grad_norm":
|
| 305 |
-
"learning_rate": 4.
|
| 306 |
-
"loss": 0.
|
| 307 |
"step": 310
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch":
|
| 311 |
-
"
|
| 312 |
-
"
|
| 313 |
-
"
|
| 314 |
-
"eval_samples_per_second": 80.749,
|
| 315 |
-
"eval_steps_per_second": 1.29,
|
| 316 |
-
"step": 315
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 10.158730158730158,
|
| 320 |
-
"grad_norm": 6.234573841094971,
|
| 321 |
-
"learning_rate": 4.121863799283154e-05,
|
| 322 |
-
"loss": 0.2525,
|
| 323 |
"step": 320
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"grad_norm":
|
| 328 |
-
"learning_rate": 4.
|
| 329 |
-
"loss": 0.
|
| 330 |
"step": 330
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"epoch":
|
| 334 |
-
"grad_norm":
|
| 335 |
-
"learning_rate": 4.
|
| 336 |
-
"loss": 0.
|
| 337 |
"step": 340
|
| 338 |
},
|
| 339 |
{
|
| 340 |
-
"epoch":
|
| 341 |
-
"
|
| 342 |
-
"
|
| 343 |
-
"
|
| 344 |
-
"eval_samples_per_second": 80.97,
|
| 345 |
-
"eval_steps_per_second": 1.294,
|
| 346 |
-
"step": 346
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 11.11111111111111,
|
| 350 |
-
"grad_norm": 6.113388538360596,
|
| 351 |
-
"learning_rate": 3.987455197132617e-05,
|
| 352 |
-
"loss": 0.271,
|
| 353 |
"step": 350
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate": 3.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 360
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm":
|
| 365 |
-
"learning_rate": 3.
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 370
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"
|
| 372 |
-
"
|
| 373 |
-
"
|
| 374 |
-
"eval_samples_per_second": 80.495,
|
| 375 |
-
"eval_steps_per_second": 1.286,
|
| 376 |
-
"step": 378
|
| 377 |
-
},
|
| 378 |
-
{
|
| 379 |
-
"epoch": 12.063492063492063,
|
| 380 |
-
"grad_norm": 6.271592140197754,
|
| 381 |
-
"learning_rate": 3.8530465949820786e-05,
|
| 382 |
-
"loss": 0.2441,
|
| 383 |
"step": 380
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch":
|
| 387 |
-
"grad_norm":
|
| 388 |
-
"learning_rate": 3.
|
| 389 |
-
"loss": 0.
|
| 390 |
"step": 390
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch":
|
| 394 |
-
"grad_norm":
|
| 395 |
-
"learning_rate": 3.
|
| 396 |
-
"loss": 0.
|
| 397 |
"step": 400
|
| 398 |
},
|
| 399 |
{
|
| 400 |
-
"epoch":
|
| 401 |
-
"
|
| 402 |
-
"
|
| 403 |
-
"
|
| 404 |
-
"eval_samples_per_second": 81.633,
|
| 405 |
-
"eval_steps_per_second": 1.304,
|
| 406 |
-
"step": 409
|
| 407 |
-
},
|
| 408 |
-
{
|
| 409 |
-
"epoch": 13.015873015873016,
|
| 410 |
-
"grad_norm": 7.433064937591553,
|
| 411 |
-
"learning_rate": 3.718637992831541e-05,
|
| 412 |
-
"loss": 0.2257,
|
| 413 |
"step": 410
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"grad_norm":
|
| 418 |
-
"learning_rate": 3.
|
| 419 |
-
"loss": 0.
|
| 420 |
"step": 420
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"epoch":
|
| 424 |
-
"grad_norm":
|
| 425 |
-
"learning_rate": 3.
|
| 426 |
-
"loss": 0.
|
| 427 |
"step": 430
|
| 428 |
},
|
| 429 |
{
|
| 430 |
-
"epoch":
|
| 431 |
-
"grad_norm":
|
| 432 |
-
"learning_rate": 3.
|
| 433 |
-
"loss": 0.
|
| 434 |
"step": 440
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"
|
| 439 |
-
"
|
| 440 |
-
"
|
| 441 |
-
"eval_samples_per_second": 81.874,
|
| 442 |
-
"eval_steps_per_second": 1.308,
|
| 443 |
-
"step": 441
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 14.285714285714286,
|
| 447 |
-
"grad_norm": 8.859041213989258,
|
| 448 |
-
"learning_rate": 3.539426523297491e-05,
|
| 449 |
-
"loss": 0.2068,
|
| 450 |
"step": 450
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"epoch":
|
| 454 |
-
"grad_norm":
|
| 455 |
-
"learning_rate": 3.
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 460
|
| 458 |
},
|
| 459 |
{
|
| 460 |
-
"epoch":
|
| 461 |
-
"grad_norm":
|
| 462 |
-
"learning_rate": 3.
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 470
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch":
|
| 468 |
-
"
|
| 469 |
-
"
|
| 470 |
-
"
|
| 471 |
-
"eval_samples_per_second": 81.831,
|
| 472 |
-
"eval_steps_per_second": 1.307,
|
| 473 |
-
"step": 472
|
| 474 |
-
},
|
| 475 |
-
{
|
| 476 |
-
"epoch": 15.238095238095237,
|
| 477 |
-
"grad_norm": 5.328235149383545,
|
| 478 |
-
"learning_rate": 3.405017921146954e-05,
|
| 479 |
-
"loss": 0.2018,
|
| 480 |
"step": 480
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"epoch":
|
| 484 |
-
"grad_norm":
|
| 485 |
-
"learning_rate": 3.
|
| 486 |
-
"loss": 0.
|
| 487 |
"step": 490
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch":
|
| 491 |
-
"grad_norm":
|
| 492 |
-
"learning_rate": 3.
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 500
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"
|
| 499 |
-
"
|
| 500 |
-
"
|
| 501 |
-
"
|
| 502 |
-
"eval_steps_per_second": 1.284,
|
| 503 |
-
"step": 504
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"
|
| 508 |
-
"
|
| 509 |
-
"
|
|
|
|
|
|
|
| 510 |
"step": 510
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"grad_norm":
|
| 515 |
-
"learning_rate": 3.
|
| 516 |
-
"loss": 0.
|
| 517 |
"step": 520
|
| 518 |
},
|
| 519 |
{
|
| 520 |
-
"epoch":
|
| 521 |
-
"grad_norm":
|
| 522 |
-
"learning_rate": 3.
|
| 523 |
-
"loss": 0.
|
| 524 |
"step": 530
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"
|
| 529 |
-
"
|
| 530 |
-
"
|
| 531 |
-
"eval_samples_per_second": 81.024,
|
| 532 |
-
"eval_steps_per_second": 1.294,
|
| 533 |
-
"step": 535
|
| 534 |
-
},
|
| 535 |
-
{
|
| 536 |
-
"epoch": 17.142857142857142,
|
| 537 |
-
"grad_norm": 7.638742923736572,
|
| 538 |
-
"learning_rate": 3.136200716845878e-05,
|
| 539 |
-
"loss": 0.1841,
|
| 540 |
"step": 540
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"grad_norm":
|
| 545 |
-
"learning_rate": 3.
|
| 546 |
-
"loss": 0.
|
| 547 |
"step": 550
|
| 548 |
},
|
| 549 |
{
|
| 550 |
-
"epoch":
|
| 551 |
-
"grad_norm":
|
| 552 |
-
"learning_rate": 3.
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 560
|
| 555 |
},
|
| 556 |
{
|
| 557 |
-
"epoch":
|
| 558 |
-
"
|
| 559 |
-
"
|
| 560 |
-
"
|
| 561 |
-
"eval_samples_per_second": 80.988,
|
| 562 |
-
"eval_steps_per_second": 1.294,
|
| 563 |
-
"step": 567
|
| 564 |
-
},
|
| 565 |
-
{
|
| 566 |
-
"epoch": 18.095238095238095,
|
| 567 |
-
"grad_norm": 6.038789749145508,
|
| 568 |
-
"learning_rate": 3.0017921146953403e-05,
|
| 569 |
-
"loss": 0.1544,
|
| 570 |
"step": 570
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"epoch":
|
| 574 |
-
"grad_norm":
|
| 575 |
-
"learning_rate":
|
| 576 |
-
"loss": 0.
|
| 577 |
"step": 580
|
| 578 |
},
|
| 579 |
{
|
| 580 |
-
"epoch":
|
| 581 |
-
"grad_norm":
|
| 582 |
-
"learning_rate": 2.
|
| 583 |
-
"loss": 0.
|
| 584 |
"step": 590
|
| 585 |
},
|
| 586 |
{
|
| 587 |
-
"epoch":
|
| 588 |
-
"
|
| 589 |
-
"
|
| 590 |
-
"
|
| 591 |
-
"eval_samples_per_second": 80.94,
|
| 592 |
-
"eval_steps_per_second": 1.293,
|
| 593 |
-
"step": 598
|
| 594 |
-
},
|
| 595 |
-
{
|
| 596 |
-
"epoch": 19.047619047619047,
|
| 597 |
-
"grad_norm": 8.029606819152832,
|
| 598 |
-
"learning_rate": 2.8673835125448028e-05,
|
| 599 |
-
"loss": 0.1582,
|
| 600 |
"step": 600
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"epoch":
|
| 604 |
-
"grad_norm":
|
| 605 |
-
"learning_rate": 2.
|
| 606 |
-
"loss": 0.
|
| 607 |
"step": 610
|
| 608 |
},
|
| 609 |
{
|
| 610 |
-
"epoch":
|
| 611 |
-
"grad_norm":
|
| 612 |
-
"learning_rate": 2.
|
| 613 |
-
"loss": 0.
|
| 614 |
"step": 620
|
| 615 |
},
|
| 616 |
{
|
| 617 |
-
"epoch":
|
| 618 |
-
"grad_norm":
|
| 619 |
-
"learning_rate": 2.
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 630
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch":
|
| 625 |
-
"
|
| 626 |
-
"
|
| 627 |
-
"
|
| 628 |
-
"eval_samples_per_second": 80.161,
|
| 629 |
-
"eval_steps_per_second": 1.281,
|
| 630 |
-
"step": 630
|
| 631 |
-
},
|
| 632 |
-
{
|
| 633 |
-
"epoch": 20.317460317460316,
|
| 634 |
-
"grad_norm": 8.681838989257812,
|
| 635 |
-
"learning_rate": 2.6881720430107527e-05,
|
| 636 |
-
"loss": 0.1645,
|
| 637 |
"step": 640
|
| 638 |
},
|
| 639 |
{
|
| 640 |
-
"epoch":
|
| 641 |
-
"grad_norm":
|
| 642 |
-
"learning_rate": 2.
|
| 643 |
-
"loss": 0.
|
| 644 |
"step": 650
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"grad_norm":
|
| 649 |
-
"learning_rate": 2.
|
| 650 |
-
"loss": 0.
|
| 651 |
"step": 660
|
| 652 |
},
|
| 653 |
{
|
| 654 |
-
"epoch":
|
| 655 |
-
"
|
| 656 |
-
"
|
| 657 |
-
"
|
| 658 |
-
"eval_samples_per_second": 78.727,
|
| 659 |
-
"eval_steps_per_second": 1.258,
|
| 660 |
-
"step": 661
|
| 661 |
-
},
|
| 662 |
-
{
|
| 663 |
-
"epoch": 21.26984126984127,
|
| 664 |
-
"grad_norm": 5.681866645812988,
|
| 665 |
-
"learning_rate": 2.5537634408602152e-05,
|
| 666 |
-
"loss": 0.1411,
|
| 667 |
"step": 670
|
| 668 |
},
|
| 669 |
{
|
| 670 |
-
"epoch":
|
| 671 |
-
"grad_norm":
|
| 672 |
-
"learning_rate": 2.
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 680
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"grad_norm":
|
| 679 |
-
"learning_rate": 2.
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 690
|
| 682 |
},
|
| 683 |
{
|
| 684 |
-
"epoch":
|
| 685 |
-
"
|
| 686 |
-
"
|
| 687 |
-
"
|
| 688 |
-
"eval_samples_per_second": 82.632,
|
| 689 |
-
"eval_steps_per_second": 1.32,
|
| 690 |
-
"step": 693
|
| 691 |
-
},
|
| 692 |
-
{
|
| 693 |
-
"epoch": 22.22222222222222,
|
| 694 |
-
"grad_norm": 5.077579021453857,
|
| 695 |
-
"learning_rate": 2.4193548387096777e-05,
|
| 696 |
-
"loss": 0.1238,
|
| 697 |
"step": 700
|
| 698 |
},
|
| 699 |
{
|
| 700 |
-
"epoch":
|
| 701 |
-
"grad_norm":
|
| 702 |
-
"learning_rate": 2.
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 710
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"grad_norm":
|
| 709 |
-
"learning_rate": 2.
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 720
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch":
|
| 715 |
-
"
|
| 716 |
-
"
|
| 717 |
-
"
|
| 718 |
-
"eval_samples_per_second": 81.745,
|
| 719 |
-
"eval_steps_per_second": 1.306,
|
| 720 |
-
"step": 724
|
| 721 |
-
},
|
| 722 |
-
{
|
| 723 |
-
"epoch": 23.174603174603174,
|
| 724 |
-
"grad_norm": 4.609182834625244,
|
| 725 |
-
"learning_rate": 2.28494623655914e-05,
|
| 726 |
-
"loss": 0.1063,
|
| 727 |
"step": 730
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"grad_norm":
|
| 732 |
-
"learning_rate": 2.
|
| 733 |
-
"loss": 0.
|
| 734 |
"step": 740
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate": 2.
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 750
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch":
|
| 745 |
-
"
|
| 746 |
-
"
|
| 747 |
-
"
|
| 748 |
-
"
|
| 749 |
-
"eval_steps_per_second": 1.314,
|
| 750 |
-
"step": 756
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"epoch":
|
| 754 |
-
"
|
| 755 |
-
"
|
| 756 |
-
"
|
| 757 |
-
"
|
|
|
|
|
|
|
| 758 |
},
|
| 759 |
{
|
| 760 |
-
"epoch":
|
| 761 |
-
"grad_norm":
|
| 762 |
-
"learning_rate": 2.
|
| 763 |
-
"loss": 0.
|
| 764 |
"step": 770
|
| 765 |
},
|
| 766 |
{
|
| 767 |
-
"epoch":
|
| 768 |
-
"grad_norm":
|
| 769 |
-
"learning_rate": 2.
|
| 770 |
-
"loss": 0.
|
| 771 |
"step": 780
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"epoch":
|
| 775 |
-
"
|
| 776 |
-
"
|
| 777 |
-
"
|
| 778 |
-
"eval_samples_per_second": 81.309,
|
| 779 |
-
"eval_steps_per_second": 1.299,
|
| 780 |
-
"step": 787
|
| 781 |
-
},
|
| 782 |
-
{
|
| 783 |
-
"epoch": 25.07936507936508,
|
| 784 |
-
"grad_norm": 3.8356785774230957,
|
| 785 |
-
"learning_rate": 2.0161290322580645e-05,
|
| 786 |
-
"loss": 0.1131,
|
| 787 |
"step": 790
|
| 788 |
},
|
| 789 |
{
|
| 790 |
-
"epoch":
|
| 791 |
-
"grad_norm":
|
| 792 |
-
"learning_rate":
|
| 793 |
-
"loss": 0.
|
| 794 |
"step": 800
|
| 795 |
},
|
| 796 |
{
|
| 797 |
-
"epoch":
|
| 798 |
-
"grad_norm":
|
| 799 |
-
"learning_rate":
|
| 800 |
-
"loss": 0.
|
| 801 |
"step": 810
|
| 802 |
},
|
| 803 |
{
|
| 804 |
-
"epoch":
|
| 805 |
-
"
|
| 806 |
-
"
|
| 807 |
-
"
|
| 808 |
-
"eval_samples_per_second": 83.047,
|
| 809 |
-
"eval_steps_per_second": 1.327,
|
| 810 |
-
"step": 819
|
| 811 |
-
},
|
| 812 |
-
{
|
| 813 |
-
"epoch": 26.03174603174603,
|
| 814 |
-
"grad_norm": 8.177849769592285,
|
| 815 |
-
"learning_rate": 1.881720430107527e-05,
|
| 816 |
-
"loss": 0.1178,
|
| 817 |
"step": 820
|
| 818 |
},
|
| 819 |
{
|
| 820 |
-
"epoch":
|
| 821 |
-
"grad_norm":
|
| 822 |
-
"learning_rate": 1.
|
| 823 |
-
"loss": 0.
|
| 824 |
"step": 830
|
| 825 |
},
|
| 826 |
{
|
| 827 |
-
"epoch":
|
| 828 |
-
"grad_norm":
|
| 829 |
-
"learning_rate": 1.
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 840
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch":
|
| 835 |
-
"grad_norm":
|
| 836 |
-
"learning_rate": 1.
|
| 837 |
-
"loss": 0.
|
| 838 |
-
"step": 850
|
| 839 |
-
},
|
| 840 |
-
{
|
| 841 |
-
"epoch": 26.984126984126984,
|
| 842 |
-
"eval_accuracy": 0.90963554667998,
|
| 843 |
-
"eval_loss": 0.2805466949939728,
|
| 844 |
-
"eval_runtime": 24.6132,
|
| 845 |
-
"eval_samples_per_second": 81.379,
|
| 846 |
-
"eval_steps_per_second": 1.3,
|
| 847 |
"step": 850
|
| 848 |
},
|
| 849 |
{
|
| 850 |
-
"epoch":
|
| 851 |
-
"grad_norm":
|
| 852 |
-
"learning_rate": 1.
|
| 853 |
-
"loss": 0.
|
| 854 |
"step": 860
|
| 855 |
},
|
| 856 |
{
|
| 857 |
-
"epoch":
|
| 858 |
-
"grad_norm":
|
| 859 |
-
"learning_rate": 1.
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 870
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"grad_norm":
|
| 866 |
-
"learning_rate": 1.
|
| 867 |
-
"loss": 0.
|
| 868 |
"step": 880
|
| 869 |
},
|
| 870 |
{
|
| 871 |
-
"epoch":
|
| 872 |
-
"
|
| 873 |
-
"
|
| 874 |
-
"
|
| 875 |
-
"eval_samples_per_second": 82.989,
|
| 876 |
-
"eval_steps_per_second": 1.326,
|
| 877 |
-
"step": 882
|
| 878 |
-
},
|
| 879 |
-
{
|
| 880 |
-
"epoch": 28.253968253968253,
|
| 881 |
-
"grad_norm": 3.115154266357422,
|
| 882 |
-
"learning_rate": 1.568100358422939e-05,
|
| 883 |
-
"loss": 0.0964,
|
| 884 |
"step": 890
|
| 885 |
},
|
| 886 |
{
|
| 887 |
-
"epoch":
|
| 888 |
-
"grad_norm":
|
| 889 |
-
"learning_rate": 1.
|
| 890 |
-
"loss": 0.
|
| 891 |
"step": 900
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch":
|
| 895 |
-
"grad_norm":
|
| 896 |
-
"learning_rate": 1.
|
| 897 |
-
"loss": 0.
|
| 898 |
"step": 910
|
| 899 |
},
|
| 900 |
{
|
| 901 |
-
"epoch":
|
| 902 |
-
"
|
| 903 |
-
"
|
| 904 |
-
"
|
| 905 |
-
"eval_samples_per_second": 82.794,
|
| 906 |
-
"eval_steps_per_second": 1.323,
|
| 907 |
-
"step": 913
|
| 908 |
-
},
|
| 909 |
-
{
|
| 910 |
-
"epoch": 29.206349206349206,
|
| 911 |
-
"grad_norm": 6.668429374694824,
|
| 912 |
-
"learning_rate": 1.4336917562724014e-05,
|
| 913 |
-
"loss": 0.1,
|
| 914 |
"step": 920
|
| 915 |
},
|
| 916 |
{
|
| 917 |
-
"epoch":
|
| 918 |
-
"grad_norm":
|
| 919 |
-
"learning_rate": 1.
|
| 920 |
-
"loss": 0.
|
| 921 |
"step": 930
|
| 922 |
},
|
| 923 |
{
|
| 924 |
-
"epoch":
|
| 925 |
-
"grad_norm":
|
| 926 |
-
"learning_rate": 1.
|
| 927 |
-
"loss": 0.
|
| 928 |
"step": 940
|
| 929 |
},
|
| 930 |
{
|
| 931 |
-
"epoch":
|
| 932 |
-
"
|
| 933 |
-
"
|
| 934 |
-
"
|
| 935 |
-
"eval_samples_per_second": 81.922,
|
| 936 |
-
"eval_steps_per_second": 1.309,
|
| 937 |
-
"step": 945
|
| 938 |
-
},
|
| 939 |
-
{
|
| 940 |
-
"epoch": 30.158730158730158,
|
| 941 |
-
"grad_norm": 4.359825611114502,
|
| 942 |
-
"learning_rate": 1.2992831541218639e-05,
|
| 943 |
-
"loss": 0.0855,
|
| 944 |
"step": 950
|
| 945 |
},
|
| 946 |
{
|
| 947 |
-
"epoch":
|
| 948 |
-
"grad_norm":
|
| 949 |
-
"learning_rate": 1.
|
| 950 |
-
"loss": 0.
|
| 951 |
"step": 960
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"epoch":
|
| 955 |
-
"grad_norm":
|
| 956 |
-
"learning_rate": 1.
|
| 957 |
-
"loss": 0.
|
| 958 |
"step": 970
|
| 959 |
},
|
| 960 |
{
|
| 961 |
-
"epoch":
|
| 962 |
-
"
|
| 963 |
-
"
|
| 964 |
-
"
|
| 965 |
-
"eval_samples_per_second": 55.03,
|
| 966 |
-
"eval_steps_per_second": 0.879,
|
| 967 |
-
"step": 976
|
| 968 |
-
},
|
| 969 |
-
{
|
| 970 |
-
"epoch": 31.11111111111111,
|
| 971 |
-
"grad_norm": 5.782038688659668,
|
| 972 |
-
"learning_rate": 1.1648745519713262e-05,
|
| 973 |
-
"loss": 0.0967,
|
| 974 |
"step": 980
|
| 975 |
},
|
| 976 |
{
|
| 977 |
-
"epoch":
|
| 978 |
-
"grad_norm":
|
| 979 |
-
"learning_rate": 1.
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 990
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch":
|
| 985 |
-
"grad_norm":
|
| 986 |
-
"learning_rate": 1.
|
| 987 |
-
"loss": 0.
|
| 988 |
"step": 1000
|
| 989 |
},
|
| 990 |
{
|
| 991 |
-
"epoch":
|
| 992 |
-
"
|
| 993 |
-
"
|
| 994 |
-
"
|
| 995 |
-
"eval_samples_per_second": 80.456,
|
| 996 |
-
"eval_steps_per_second": 1.285,
|
| 997 |
-
"step": 1008
|
| 998 |
-
},
|
| 999 |
-
{
|
| 1000 |
-
"epoch": 32.06349206349206,
|
| 1001 |
-
"grad_norm": 4.185471534729004,
|
| 1002 |
-
"learning_rate": 1.0304659498207886e-05,
|
| 1003 |
-
"loss": 0.0821,
|
| 1004 |
"step": 1010
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
-
"epoch":
|
| 1008 |
-
"grad_norm":
|
| 1009 |
-
"learning_rate":
|
| 1010 |
-
"loss": 0.
|
| 1011 |
"step": 1020
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"epoch":
|
| 1015 |
-
"
|
| 1016 |
-
"
|
| 1017 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1018 |
"step": 1030
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
-
"epoch":
|
| 1022 |
-
"
|
| 1023 |
-
"
|
| 1024 |
-
"
|
| 1025 |
-
"eval_samples_per_second": 81.364,
|
| 1026 |
-
"eval_steps_per_second": 1.3,
|
| 1027 |
-
"step": 1039
|
| 1028 |
-
},
|
| 1029 |
-
{
|
| 1030 |
-
"epoch": 33.01587301587302,
|
| 1031 |
-
"grad_norm": 3.8526599407196045,
|
| 1032 |
-
"learning_rate": 8.960573476702509e-06,
|
| 1033 |
-
"loss": 0.0979,
|
| 1034 |
"step": 1040
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
-
"epoch":
|
| 1038 |
-
"grad_norm":
|
| 1039 |
-
"learning_rate":
|
| 1040 |
-
"loss": 0.
|
| 1041 |
"step": 1050
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
-
"epoch":
|
| 1045 |
-
"grad_norm":
|
| 1046 |
-
"learning_rate":
|
| 1047 |
-
"loss": 0.
|
| 1048 |
"step": 1060
|
| 1049 |
},
|
| 1050 |
{
|
| 1051 |
-
"epoch":
|
| 1052 |
-
"grad_norm":
|
| 1053 |
-
"learning_rate":
|
| 1054 |
-
"loss": 0.
|
| 1055 |
"step": 1070
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
-
"epoch":
|
| 1059 |
-
"
|
| 1060 |
-
"
|
| 1061 |
-
"
|
| 1062 |
-
"eval_samples_per_second": 81.198,
|
| 1063 |
-
"eval_steps_per_second": 1.297,
|
| 1064 |
-
"step": 1071
|
| 1065 |
-
},
|
| 1066 |
-
{
|
| 1067 |
-
"epoch": 34.285714285714285,
|
| 1068 |
-
"grad_norm": 4.8823771476745605,
|
| 1069 |
-
"learning_rate": 7.168458781362007e-06,
|
| 1070 |
-
"loss": 0.0866,
|
| 1071 |
"step": 1080
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
-
"epoch":
|
| 1075 |
-
"grad_norm":
|
| 1076 |
-
"learning_rate":
|
| 1077 |
-
"loss": 0.
|
| 1078 |
"step": 1090
|
| 1079 |
},
|
| 1080 |
{
|
| 1081 |
-
"epoch":
|
| 1082 |
-
"grad_norm":
|
| 1083 |
-
"learning_rate":
|
| 1084 |
-
"loss": 0.
|
| 1085 |
"step": 1100
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
-
"epoch":
|
| 1089 |
-
"
|
| 1090 |
-
"
|
| 1091 |
-
"
|
| 1092 |
-
"eval_samples_per_second": 81.82,
|
| 1093 |
-
"eval_steps_per_second": 1.307,
|
| 1094 |
-
"step": 1102
|
| 1095 |
-
},
|
| 1096 |
-
{
|
| 1097 |
-
"epoch": 35.23809523809524,
|
| 1098 |
-
"grad_norm": 3.9122490882873535,
|
| 1099 |
-
"learning_rate": 5.824372759856631e-06,
|
| 1100 |
-
"loss": 0.0704,
|
| 1101 |
"step": 1110
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
-
"epoch":
|
| 1105 |
-
"grad_norm":
|
| 1106 |
-
"learning_rate":
|
| 1107 |
-
"loss": 0.
|
| 1108 |
"step": 1120
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
-
"epoch":
|
| 1112 |
-
"grad_norm":
|
| 1113 |
-
"learning_rate":
|
| 1114 |
-
"loss": 0.
|
| 1115 |
"step": 1130
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
-
"epoch":
|
| 1119 |
-
"
|
| 1120 |
-
"
|
| 1121 |
-
"
|
| 1122 |
-
"eval_samples_per_second": 82.549,
|
| 1123 |
-
"eval_steps_per_second": 1.319,
|
| 1124 |
-
"step": 1134
|
| 1125 |
-
},
|
| 1126 |
-
{
|
| 1127 |
-
"epoch": 36.19047619047619,
|
| 1128 |
-
"grad_norm": 3.785637140274048,
|
| 1129 |
-
"learning_rate": 4.4802867383512545e-06,
|
| 1130 |
-
"loss": 0.0764,
|
| 1131 |
"step": 1140
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
-
"epoch":
|
| 1135 |
-
"grad_norm":
|
| 1136 |
-
"learning_rate":
|
| 1137 |
-
"loss": 0.
|
| 1138 |
"step": 1150
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
-
"epoch":
|
| 1142 |
-
"grad_norm":
|
| 1143 |
-
"learning_rate":
|
| 1144 |
-
"loss": 0.
|
| 1145 |
"step": 1160
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
-
"epoch":
|
| 1149 |
-
"
|
| 1150 |
-
"
|
| 1151 |
-
"
|
| 1152 |
-
"eval_samples_per_second": 81.864,
|
| 1153 |
-
"eval_steps_per_second": 1.308,
|
| 1154 |
-
"step": 1165
|
| 1155 |
-
},
|
| 1156 |
-
{
|
| 1157 |
-
"epoch": 37.142857142857146,
|
| 1158 |
-
"grad_norm": 5.3057050704956055,
|
| 1159 |
-
"learning_rate": 3.1362007168458782e-06,
|
| 1160 |
-
"loss": 0.0763,
|
| 1161 |
"step": 1170
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
-
"epoch":
|
| 1165 |
-
"grad_norm":
|
| 1166 |
-
"learning_rate":
|
| 1167 |
-
"loss": 0.
|
| 1168 |
"step": 1180
|
| 1169 |
},
|
| 1170 |
{
|
| 1171 |
-
"epoch":
|
| 1172 |
-
"grad_norm":
|
| 1173 |
-
"learning_rate":
|
| 1174 |
-
"loss": 0.
|
| 1175 |
"step": 1190
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
-
"epoch":
|
| 1179 |
-
"
|
| 1180 |
-
"
|
| 1181 |
-
"
|
| 1182 |
-
"eval_samples_per_second": 81.201,
|
| 1183 |
-
"eval_steps_per_second": 1.297,
|
| 1184 |
-
"step": 1197
|
| 1185 |
-
},
|
| 1186 |
-
{
|
| 1187 |
-
"epoch": 38.095238095238095,
|
| 1188 |
-
"grad_norm": 4.277989387512207,
|
| 1189 |
-
"learning_rate": 1.7921146953405017e-06,
|
| 1190 |
-
"loss": 0.0793,
|
| 1191 |
"step": 1200
|
| 1192 |
},
|
| 1193 |
{
|
| 1194 |
-
"epoch":
|
| 1195 |
-
"grad_norm":
|
| 1196 |
-
"learning_rate":
|
| 1197 |
-
"loss": 0.
|
| 1198 |
"step": 1210
|
| 1199 |
},
|
| 1200 |
{
|
| 1201 |
-
"epoch":
|
| 1202 |
-
"grad_norm":
|
| 1203 |
-
"learning_rate":
|
| 1204 |
-
"loss": 0.
|
| 1205 |
"step": 1220
|
| 1206 |
},
|
| 1207 |
{
|
| 1208 |
-
"epoch":
|
| 1209 |
-
"
|
| 1210 |
-
"
|
| 1211 |
-
"
|
| 1212 |
-
"eval_samples_per_second": 80.873,
|
| 1213 |
-
"eval_steps_per_second": 1.292,
|
| 1214 |
-
"step": 1228
|
| 1215 |
-
},
|
| 1216 |
-
{
|
| 1217 |
-
"epoch": 39.04761904761905,
|
| 1218 |
-
"grad_norm": 4.357304573059082,
|
| 1219 |
-
"learning_rate": 4.4802867383512544e-07,
|
| 1220 |
-
"loss": 0.0855,
|
| 1221 |
"step": 1230
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
-
"epoch":
|
| 1225 |
-
"grad_norm":
|
| 1226 |
-
"learning_rate":
|
| 1227 |
-
"loss": 0.
|
| 1228 |
"step": 1240
|
| 1229 |
},
|
| 1230 |
{
|
| 1231 |
-
"epoch":
|
| 1232 |
-
"
|
| 1233 |
-
"
|
| 1234 |
-
"
|
| 1235 |
-
"
|
| 1236 |
-
|
| 1237 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1238 |
},
|
| 1239 |
{
|
| 1240 |
-
"epoch":
|
| 1241 |
-
"step":
|
| 1242 |
-
"total_flos":
|
| 1243 |
-
"train_loss": 0.
|
| 1244 |
-
"train_runtime":
|
| 1245 |
-
"train_samples_per_second":
|
| 1246 |
-
"train_steps_per_second": 0.
|
| 1247 |
}
|
| 1248 |
],
|
| 1249 |
"logging_steps": 10,
|
| 1250 |
-
"max_steps":
|
| 1251 |
"num_input_tokens_seen": 0,
|
| 1252 |
-
"num_train_epochs":
|
| 1253 |
"save_steps": 500,
|
| 1254 |
"stateful_callbacks": {
|
| 1255 |
"TrainerControl": {
|
|
@@ -1263,7 +969,7 @@
|
|
| 1263 |
"attributes": {}
|
| 1264 |
}
|
| 1265 |
},
|
| 1266 |
-
"total_flos":
|
| 1267 |
"train_batch_size": 64,
|
| 1268 |
"trial_name": null,
|
| 1269 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.994671729544341,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-255",
|
| 4 |
+
"epoch": 4.995102840352596,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1275,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.039177277179236046,
|
| 13 |
+
"grad_norm": 9.329336166381836,
|
| 14 |
+
"learning_rate": 3.90625e-06,
|
| 15 |
+
"loss": 0.7,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.07835455435847209,
|
| 20 |
+
"grad_norm": 4.470575332641602,
|
| 21 |
+
"learning_rate": 7.8125e-06,
|
| 22 |
+
"loss": 0.4115,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.11753183153770813,
|
| 27 |
+
"grad_norm": 1.0650607347488403,
|
| 28 |
+
"learning_rate": 1.171875e-05,
|
| 29 |
+
"loss": 0.1272,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.15670910871694418,
|
| 34 |
+
"grad_norm": 0.7108937501907349,
|
| 35 |
+
"learning_rate": 1.5625e-05,
|
| 36 |
+
"loss": 0.0479,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.1958863858961802,
|
| 41 |
+
"grad_norm": 1.0625723600387573,
|
| 42 |
+
"learning_rate": 1.953125e-05,
|
| 43 |
+
"loss": 0.0359,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.23506366307541626,
|
| 48 |
+
"grad_norm": 0.1246357336640358,
|
| 49 |
+
"learning_rate": 2.34375e-05,
|
| 50 |
+
"loss": 0.0227,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.2742409402546523,
|
| 55 |
+
"grad_norm": 0.5148888826370239,
|
| 56 |
+
"learning_rate": 2.734375e-05,
|
| 57 |
+
"loss": 0.0295,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.31341821743388837,
|
| 62 |
+
"grad_norm": 0.9743920564651489,
|
| 63 |
+
"learning_rate": 3.125e-05,
|
| 64 |
+
"loss": 0.0294,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.3525954946131244,
|
| 69 |
+
"grad_norm": 0.31461623311042786,
|
| 70 |
+
"learning_rate": 3.5156250000000004e-05,
|
| 71 |
+
"loss": 0.0289,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.3917727717923604,
|
| 76 |
+
"grad_norm": 1.0115059614181519,
|
| 77 |
+
"learning_rate": 3.90625e-05,
|
| 78 |
+
"loss": 0.0391,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.4309500489715965,
|
| 83 |
+
"grad_norm": 0.08507003635168076,
|
| 84 |
+
"learning_rate": 4.2968750000000004e-05,
|
| 85 |
+
"loss": 0.018,
|
| 86 |
"step": 110
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.4701273261508325,
|
| 90 |
+
"grad_norm": 0.1573338657617569,
|
| 91 |
+
"learning_rate": 4.6875e-05,
|
| 92 |
+
"loss": 0.0214,
|
| 93 |
"step": 120
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.5093046033300686,
|
| 97 |
+
"grad_norm": 0.4784347712993622,
|
| 98 |
+
"learning_rate": 4.9912816041848304e-05,
|
| 99 |
+
"loss": 0.0264,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
"step": 130
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.5484818805093046,
|
| 104 |
+
"grad_norm": 0.2329588085412979,
|
| 105 |
+
"learning_rate": 4.94768962510898e-05,
|
| 106 |
+
"loss": 0.0277,
|
| 107 |
"step": 140
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.5876591576885406,
|
| 111 |
+
"grad_norm": 0.05806314945220947,
|
| 112 |
+
"learning_rate": 4.90409764603313e-05,
|
| 113 |
+
"loss": 0.0221,
|
| 114 |
"step": 150
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.6268364348677767,
|
| 118 |
+
"grad_norm": 1.9614442586898804,
|
| 119 |
+
"learning_rate": 4.86050566695728e-05,
|
| 120 |
+
"loss": 0.0311,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
"step": 160
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.6660137120470128,
|
| 125 |
+
"grad_norm": 1.0261849164962769,
|
| 126 |
+
"learning_rate": 4.8169136878814306e-05,
|
| 127 |
+
"loss": 0.0334,
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 0.7051909892262488,
|
| 132 |
+
"grad_norm": 0.30455902218818665,
|
| 133 |
+
"learning_rate": 4.7733217088055796e-05,
|
| 134 |
+
"loss": 0.0271,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.7443682664054848,
|
| 139 |
+
"grad_norm": 0.38795918226242065,
|
| 140 |
+
"learning_rate": 4.72972972972973e-05,
|
| 141 |
+
"loss": 0.018,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 142 |
"step": 190
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.7835455435847208,
|
| 146 |
+
"grad_norm": 0.25275129079818726,
|
| 147 |
+
"learning_rate": 4.68613775065388e-05,
|
| 148 |
+
"loss": 0.0331,
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.8227228207639569,
|
| 153 |
+
"grad_norm": 0.15323331952095032,
|
| 154 |
+
"learning_rate": 4.642545771578029e-05,
|
| 155 |
+
"loss": 0.0242,
|
| 156 |
"step": 210
|
| 157 |
},
|
| 158 |
{
|
| 159 |
+
"epoch": 0.861900097943193,
|
| 160 |
+
"grad_norm": 0.4740907847881317,
|
| 161 |
+
"learning_rate": 4.59895379250218e-05,
|
| 162 |
+
"loss": 0.0232,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 163 |
"step": 220
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"epoch": 0.901077375122429,
|
| 167 |
+
"grad_norm": 0.5308133363723755,
|
| 168 |
+
"learning_rate": 4.55536181342633e-05,
|
| 169 |
+
"loss": 0.0177,
|
| 170 |
"step": 230
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"epoch": 0.940254652301665,
|
| 174 |
+
"grad_norm": 0.985102653503418,
|
| 175 |
+
"learning_rate": 4.51176983435048e-05,
|
| 176 |
+
"loss": 0.0301,
|
| 177 |
"step": 240
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.9794319294809011,
|
| 181 |
+
"grad_norm": 0.6651228070259094,
|
| 182 |
+
"learning_rate": 4.4681778552746294e-05,
|
| 183 |
+
"loss": 0.0268,
|
| 184 |
"step": 250
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.9990205680705191,
|
| 188 |
+
"eval_accuracy": 0.994671729544341,
|
| 189 |
+
"eval_loss": 0.025641364976763725,
|
| 190 |
+
"eval_runtime": 164.878,
|
| 191 |
+
"eval_samples_per_second": 99.031,
|
| 192 |
+
"eval_steps_per_second": 1.553,
|
| 193 |
+
"step": 255
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"epoch": 1.018609206660137,
|
| 197 |
+
"grad_norm": 0.157534658908844,
|
| 198 |
+
"learning_rate": 4.42458587619878e-05,
|
| 199 |
+
"loss": 0.0175,
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 1.0577864838393731,
|
| 204 |
+
"grad_norm": 0.3950434923171997,
|
| 205 |
+
"learning_rate": 4.3809938971229295e-05,
|
| 206 |
+
"loss": 0.0205,
|
| 207 |
"step": 270
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 1.0969637610186092,
|
| 211 |
+
"grad_norm": 0.771115243434906,
|
| 212 |
+
"learning_rate": 4.337401918047079e-05,
|
| 213 |
+
"loss": 0.032,
|
| 214 |
"step": 280
|
| 215 |
},
|
| 216 |
{
|
| 217 |
+
"epoch": 1.1361410381978452,
|
| 218 |
+
"grad_norm": 0.24406306445598602,
|
| 219 |
+
"learning_rate": 4.2938099389712295e-05,
|
| 220 |
+
"loss": 0.0261,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 221 |
"step": 290
|
| 222 |
},
|
| 223 |
{
|
| 224 |
+
"epoch": 1.1753183153770812,
|
| 225 |
+
"grad_norm": 0.5867096781730652,
|
| 226 |
+
"learning_rate": 4.250217959895379e-05,
|
| 227 |
+
"loss": 0.0297,
|
| 228 |
"step": 300
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 1.2144955925563172,
|
| 232 |
+
"grad_norm": 0.3914402723312378,
|
| 233 |
+
"learning_rate": 4.2066259808195296e-05,
|
| 234 |
+
"loss": 0.0245,
|
| 235 |
"step": 310
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 1.2536728697355533,
|
| 239 |
+
"grad_norm": 0.2989954352378845,
|
| 240 |
+
"learning_rate": 4.163034001743679e-05,
|
| 241 |
+
"loss": 0.0275,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 242 |
"step": 320
|
| 243 |
},
|
| 244 |
{
|
| 245 |
+
"epoch": 1.2928501469147895,
|
| 246 |
+
"grad_norm": 0.46654826402664185,
|
| 247 |
+
"learning_rate": 4.119442022667829e-05,
|
| 248 |
+
"loss": 0.031,
|
| 249 |
"step": 330
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 1.3320274240940255,
|
| 253 |
+
"grad_norm": 0.17611253261566162,
|
| 254 |
+
"learning_rate": 4.0758500435919793e-05,
|
| 255 |
+
"loss": 0.0247,
|
| 256 |
"step": 340
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 1.3712047012732616,
|
| 260 |
+
"grad_norm": 0.31176942586898804,
|
| 261 |
+
"learning_rate": 4.032258064516129e-05,
|
| 262 |
+
"loss": 0.0242,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 263 |
"step": 350
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 1.4103819784524976,
|
| 267 |
+
"grad_norm": 0.5007725954055786,
|
| 268 |
+
"learning_rate": 3.9886660854402794e-05,
|
| 269 |
+
"loss": 0.033,
|
| 270 |
"step": 360
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 1.4495592556317336,
|
| 274 |
+
"grad_norm": 0.6336463093757629,
|
| 275 |
+
"learning_rate": 3.945074106364429e-05,
|
| 276 |
+
"loss": 0.0127,
|
| 277 |
"step": 370
|
| 278 |
},
|
| 279 |
{
|
| 280 |
+
"epoch": 1.4887365328109696,
|
| 281 |
+
"grad_norm": 0.349540650844574,
|
| 282 |
+
"learning_rate": 3.9014821272885795e-05,
|
| 283 |
+
"loss": 0.0327,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 284 |
"step": 380
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 1.5279138099902057,
|
| 288 |
+
"grad_norm": 0.48749464750289917,
|
| 289 |
+
"learning_rate": 3.857890148212729e-05,
|
| 290 |
+
"loss": 0.0247,
|
| 291 |
"step": 390
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 1.5670910871694417,
|
| 295 |
+
"grad_norm": 0.21512505412101746,
|
| 296 |
+
"learning_rate": 3.814298169136879e-05,
|
| 297 |
+
"loss": 0.0244,
|
| 298 |
"step": 400
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 1.606268364348678,
|
| 302 |
+
"grad_norm": 0.1843574196100235,
|
| 303 |
+
"learning_rate": 3.770706190061029e-05,
|
| 304 |
+
"loss": 0.0238,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 305 |
"step": 410
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 1.645445641527914,
|
| 309 |
+
"grad_norm": 0.17209237813949585,
|
| 310 |
+
"learning_rate": 3.727114210985179e-05,
|
| 311 |
+
"loss": 0.0159,
|
| 312 |
"step": 420
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 1.68462291870715,
|
| 316 |
+
"grad_norm": 0.17134279012680054,
|
| 317 |
+
"learning_rate": 3.6835222319093286e-05,
|
| 318 |
+
"loss": 0.0147,
|
| 319 |
"step": 430
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 1.723800195886386,
|
| 323 |
+
"grad_norm": 0.3783511519432068,
|
| 324 |
+
"learning_rate": 3.639930252833479e-05,
|
| 325 |
+
"loss": 0.026,
|
| 326 |
"step": 440
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 1.762977473065622,
|
| 330 |
+
"grad_norm": 0.20875829458236694,
|
| 331 |
+
"learning_rate": 3.596338273757629e-05,
|
| 332 |
+
"loss": 0.0169,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 333 |
"step": 450
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 1.802154750244858,
|
| 337 |
+
"grad_norm": 0.061811015009880066,
|
| 338 |
+
"learning_rate": 3.552746294681778e-05,
|
| 339 |
+
"loss": 0.0115,
|
| 340 |
"step": 460
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 1.841332027424094,
|
| 344 |
+
"grad_norm": 0.5182923078536987,
|
| 345 |
+
"learning_rate": 3.509154315605929e-05,
|
| 346 |
+
"loss": 0.0217,
|
| 347 |
"step": 470
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 1.88050930460333,
|
| 351 |
+
"grad_norm": 0.3887523114681244,
|
| 352 |
+
"learning_rate": 3.465562336530079e-05,
|
| 353 |
+
"loss": 0.0189,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 354 |
"step": 480
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 1.9196865817825661,
|
| 358 |
+
"grad_norm": 0.13711315393447876,
|
| 359 |
+
"learning_rate": 3.421970357454228e-05,
|
| 360 |
+
"loss": 0.0132,
|
| 361 |
"step": 490
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 1.9588638589618022,
|
| 365 |
+
"grad_norm": 0.3663609027862549,
|
| 366 |
+
"learning_rate": 3.3783783783783784e-05,
|
| 367 |
+
"loss": 0.025,
|
| 368 |
"step": 500
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.9980411361410382,
|
| 372 |
+
"grad_norm": 0.28243473172187805,
|
| 373 |
+
"learning_rate": 3.334786399302529e-05,
|
| 374 |
+
"loss": 0.0167,
|
| 375 |
+
"step": 510
|
|
|
|
|
|
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 1.9980411361410382,
|
| 379 |
+
"eval_accuracy": 0.994671729544341,
|
| 380 |
+
"eval_loss": 0.027501454576849937,
|
| 381 |
+
"eval_runtime": 109.9626,
|
| 382 |
+
"eval_samples_per_second": 148.487,
|
| 383 |
+
"eval_steps_per_second": 2.328,
|
| 384 |
"step": 510
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 2.037218413320274,
|
| 388 |
+
"grad_norm": 0.20075927674770355,
|
| 389 |
+
"learning_rate": 3.2911944202266785e-05,
|
| 390 |
+
"loss": 0.0189,
|
| 391 |
"step": 520
|
| 392 |
},
|
| 393 |
{
|
| 394 |
+
"epoch": 2.0763956904995102,
|
| 395 |
+
"grad_norm": 0.5090253353118896,
|
| 396 |
+
"learning_rate": 3.247602441150828e-05,
|
| 397 |
+
"loss": 0.0158,
|
| 398 |
"step": 530
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 2.1155729676787463,
|
| 402 |
+
"grad_norm": 0.20879769325256348,
|
| 403 |
+
"learning_rate": 3.2040104620749785e-05,
|
| 404 |
+
"loss": 0.0198,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 405 |
"step": 540
|
| 406 |
},
|
| 407 |
{
|
| 408 |
+
"epoch": 2.1547502448579823,
|
| 409 |
+
"grad_norm": 0.7945526242256165,
|
| 410 |
+
"learning_rate": 3.160418482999128e-05,
|
| 411 |
+
"loss": 0.0244,
|
| 412 |
"step": 550
|
| 413 |
},
|
| 414 |
{
|
| 415 |
+
"epoch": 2.1939275220372183,
|
| 416 |
+
"grad_norm": 0.49813878536224365,
|
| 417 |
+
"learning_rate": 3.116826503923278e-05,
|
| 418 |
+
"loss": 0.0325,
|
| 419 |
"step": 560
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 2.2331047992164543,
|
| 423 |
+
"grad_norm": 0.417555034160614,
|
| 424 |
+
"learning_rate": 3.073234524847428e-05,
|
| 425 |
+
"loss": 0.0245,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 426 |
"step": 570
|
| 427 |
},
|
| 428 |
{
|
| 429 |
+
"epoch": 2.2722820763956904,
|
| 430 |
+
"grad_norm": 0.13541927933692932,
|
| 431 |
+
"learning_rate": 3.0296425457715783e-05,
|
| 432 |
+
"loss": 0.0176,
|
| 433 |
"step": 580
|
| 434 |
},
|
| 435 |
{
|
| 436 |
+
"epoch": 2.3114593535749264,
|
| 437 |
+
"grad_norm": 0.22694525122642517,
|
| 438 |
+
"learning_rate": 2.9860505666957283e-05,
|
| 439 |
+
"loss": 0.0199,
|
| 440 |
"step": 590
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 2.3506366307541624,
|
| 444 |
+
"grad_norm": 0.06209595128893852,
|
| 445 |
+
"learning_rate": 2.942458587619878e-05,
|
| 446 |
+
"loss": 0.0127,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 447 |
"step": 600
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 2.389813907933399,
|
| 451 |
+
"grad_norm": 0.4733225703239441,
|
| 452 |
+
"learning_rate": 2.898866608544028e-05,
|
| 453 |
+
"loss": 0.0334,
|
| 454 |
"step": 610
|
| 455 |
},
|
| 456 |
{
|
| 457 |
+
"epoch": 2.4289911851126345,
|
| 458 |
+
"grad_norm": 0.37968209385871887,
|
| 459 |
+
"learning_rate": 2.855274629468178e-05,
|
| 460 |
+
"loss": 0.028,
|
| 461 |
"step": 620
|
| 462 |
},
|
| 463 |
{
|
| 464 |
+
"epoch": 2.468168462291871,
|
| 465 |
+
"grad_norm": 0.1550379991531372,
|
| 466 |
+
"learning_rate": 2.8116826503923278e-05,
|
| 467 |
+
"loss": 0.0227,
|
| 468 |
"step": 630
|
| 469 |
},
|
| 470 |
{
|
| 471 |
+
"epoch": 2.5073457394711065,
|
| 472 |
+
"grad_norm": 0.4125171899795532,
|
| 473 |
+
"learning_rate": 2.7680906713164778e-05,
|
| 474 |
+
"loss": 0.0149,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 475 |
"step": 640
|
| 476 |
},
|
| 477 |
{
|
| 478 |
+
"epoch": 2.546523016650343,
|
| 479 |
+
"grad_norm": 0.6104760766029358,
|
| 480 |
+
"learning_rate": 2.724498692240628e-05,
|
| 481 |
+
"loss": 0.0256,
|
| 482 |
"step": 650
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 2.585700293829579,
|
| 486 |
+
"grad_norm": 0.4356853663921356,
|
| 487 |
+
"learning_rate": 2.6809067131647782e-05,
|
| 488 |
+
"loss": 0.0222,
|
| 489 |
"step": 660
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 2.624877571008815,
|
| 493 |
+
"grad_norm": 0.3686465322971344,
|
| 494 |
+
"learning_rate": 2.6373147340889275e-05,
|
| 495 |
+
"loss": 0.0213,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 496 |
"step": 670
|
| 497 |
},
|
| 498 |
{
|
| 499 |
+
"epoch": 2.664054848188051,
|
| 500 |
+
"grad_norm": 0.30900245904922485,
|
| 501 |
+
"learning_rate": 2.5937227550130776e-05,
|
| 502 |
+
"loss": 0.0226,
|
| 503 |
"step": 680
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 2.703232125367287,
|
| 507 |
+
"grad_norm": 0.1763727217912674,
|
| 508 |
+
"learning_rate": 2.550130775937228e-05,
|
| 509 |
+
"loss": 0.0175,
|
| 510 |
"step": 690
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 2.742409402546523,
|
| 514 |
+
"grad_norm": 0.7033935785293579,
|
| 515 |
+
"learning_rate": 2.5065387968613773e-05,
|
| 516 |
+
"loss": 0.0161,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 517 |
"step": 700
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 2.781586679725759,
|
| 521 |
+
"grad_norm": 0.4877593219280243,
|
| 522 |
+
"learning_rate": 2.4629468177855277e-05,
|
| 523 |
+
"loss": 0.0245,
|
| 524 |
"step": 710
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 2.820763956904995,
|
| 528 |
+
"grad_norm": 0.22476495802402496,
|
| 529 |
+
"learning_rate": 2.4193548387096777e-05,
|
| 530 |
+
"loss": 0.0158,
|
| 531 |
"step": 720
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"epoch": 2.859941234084231,
|
| 535 |
+
"grad_norm": 0.5083233118057251,
|
| 536 |
+
"learning_rate": 2.3757628596338274e-05,
|
| 537 |
+
"loss": 0.02,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 538 |
"step": 730
|
| 539 |
},
|
| 540 |
{
|
| 541 |
+
"epoch": 2.899118511263467,
|
| 542 |
+
"grad_norm": 0.131087526679039,
|
| 543 |
+
"learning_rate": 2.3321708805579774e-05,
|
| 544 |
+
"loss": 0.021,
|
| 545 |
"step": 740
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 2.9382957884427032,
|
| 549 |
+
"grad_norm": 0.2617853581905365,
|
| 550 |
+
"learning_rate": 2.2885789014821274e-05,
|
| 551 |
+
"loss": 0.0204,
|
| 552 |
"step": 750
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 2.9774730656219393,
|
| 556 |
+
"grad_norm": 0.19207285344600677,
|
| 557 |
+
"learning_rate": 2.244986922406277e-05,
|
| 558 |
+
"loss": 0.0177,
|
| 559 |
+
"step": 760
|
|
|
|
|
|
|
| 560 |
},
|
| 561 |
{
|
| 562 |
+
"epoch": 2.997061704211557,
|
| 563 |
+
"eval_accuracy": 0.9935693287604116,
|
| 564 |
+
"eval_loss": 0.026763953268527985,
|
| 565 |
+
"eval_runtime": 112.7695,
|
| 566 |
+
"eval_samples_per_second": 144.791,
|
| 567 |
+
"eval_steps_per_second": 2.27,
|
| 568 |
+
"step": 765
|
| 569 |
},
|
| 570 |
{
|
| 571 |
+
"epoch": 3.0166503428011753,
|
| 572 |
+
"grad_norm": 0.7242885231971741,
|
| 573 |
+
"learning_rate": 2.2013949433304275e-05,
|
| 574 |
+
"loss": 0.0206,
|
| 575 |
"step": 770
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 3.0558276199804113,
|
| 579 |
+
"grad_norm": 0.6204590201377869,
|
| 580 |
+
"learning_rate": 2.1578029642545772e-05,
|
| 581 |
+
"loss": 0.0167,
|
| 582 |
"step": 780
|
| 583 |
},
|
| 584 |
{
|
| 585 |
+
"epoch": 3.0950048971596473,
|
| 586 |
+
"grad_norm": 0.14467577636241913,
|
| 587 |
+
"learning_rate": 2.1142109851787272e-05,
|
| 588 |
+
"loss": 0.0207,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 589 |
"step": 790
|
| 590 |
},
|
| 591 |
{
|
| 592 |
+
"epoch": 3.1341821743388834,
|
| 593 |
+
"grad_norm": 0.3296276032924652,
|
| 594 |
+
"learning_rate": 2.0706190061028772e-05,
|
| 595 |
+
"loss": 0.0177,
|
| 596 |
"step": 800
|
| 597 |
},
|
| 598 |
{
|
| 599 |
+
"epoch": 3.1733594515181194,
|
| 600 |
+
"grad_norm": 0.7694735527038574,
|
| 601 |
+
"learning_rate": 2.0270270270270273e-05,
|
| 602 |
+
"loss": 0.0141,
|
| 603 |
"step": 810
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 3.2125367286973554,
|
| 607 |
+
"grad_norm": 0.35446447134017944,
|
| 608 |
+
"learning_rate": 1.983435047951177e-05,
|
| 609 |
+
"loss": 0.0206,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 610 |
"step": 820
|
| 611 |
},
|
| 612 |
{
|
| 613 |
+
"epoch": 3.2517140058765914,
|
| 614 |
+
"grad_norm": 1.089401125907898,
|
| 615 |
+
"learning_rate": 1.939843068875327e-05,
|
| 616 |
+
"loss": 0.0326,
|
| 617 |
"step": 830
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 3.2908912830558275,
|
| 621 |
+
"grad_norm": 0.21801254153251648,
|
| 622 |
+
"learning_rate": 1.896251089799477e-05,
|
| 623 |
+
"loss": 0.0162,
|
| 624 |
"step": 840
|
| 625 |
},
|
| 626 |
{
|
| 627 |
+
"epoch": 3.3300685602350635,
|
| 628 |
+
"grad_norm": 0.42867550253868103,
|
| 629 |
+
"learning_rate": 1.8526591107236267e-05,
|
| 630 |
+
"loss": 0.0195,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 631 |
"step": 850
|
| 632 |
},
|
| 633 |
{
|
| 634 |
+
"epoch": 3.3692458374143,
|
| 635 |
+
"grad_norm": 0.24043036997318268,
|
| 636 |
+
"learning_rate": 1.809067131647777e-05,
|
| 637 |
+
"loss": 0.0155,
|
| 638 |
"step": 860
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 3.4084231145935355,
|
| 642 |
+
"grad_norm": 0.4469415545463562,
|
| 643 |
+
"learning_rate": 1.7654751525719268e-05,
|
| 644 |
+
"loss": 0.0193,
|
| 645 |
"step": 870
|
| 646 |
},
|
| 647 |
{
|
| 648 |
+
"epoch": 3.447600391772772,
|
| 649 |
+
"grad_norm": 0.14562171697616577,
|
| 650 |
+
"learning_rate": 1.7218831734960768e-05,
|
| 651 |
+
"loss": 0.0169,
|
| 652 |
"step": 880
|
| 653 |
},
|
| 654 |
{
|
| 655 |
+
"epoch": 3.486777668952008,
|
| 656 |
+
"grad_norm": 0.4943673610687256,
|
| 657 |
+
"learning_rate": 1.6782911944202268e-05,
|
| 658 |
+
"loss": 0.0281,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 659 |
"step": 890
|
| 660 |
},
|
| 661 |
{
|
| 662 |
+
"epoch": 3.525954946131244,
|
| 663 |
+
"grad_norm": 0.5602672696113586,
|
| 664 |
+
"learning_rate": 1.6346992153443765e-05,
|
| 665 |
+
"loss": 0.0252,
|
| 666 |
"step": 900
|
| 667 |
},
|
| 668 |
{
|
| 669 |
+
"epoch": 3.56513222331048,
|
| 670 |
+
"grad_norm": 0.6581624150276184,
|
| 671 |
+
"learning_rate": 1.591107236268527e-05,
|
| 672 |
+
"loss": 0.0202,
|
| 673 |
"step": 910
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"epoch": 3.604309500489716,
|
| 677 |
+
"grad_norm": 0.45326659083366394,
|
| 678 |
+
"learning_rate": 1.5475152571926766e-05,
|
| 679 |
+
"loss": 0.0257,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 680 |
"step": 920
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 3.643486777668952,
|
| 684 |
+
"grad_norm": 1.0565999746322632,
|
| 685 |
+
"learning_rate": 1.5039232781168266e-05,
|
| 686 |
+
"loss": 0.02,
|
| 687 |
"step": 930
|
| 688 |
},
|
| 689 |
{
|
| 690 |
+
"epoch": 3.682664054848188,
|
| 691 |
+
"grad_norm": 0.2989865839481354,
|
| 692 |
+
"learning_rate": 1.4603312990409764e-05,
|
| 693 |
+
"loss": 0.0163,
|
| 694 |
"step": 940
|
| 695 |
},
|
| 696 |
{
|
| 697 |
+
"epoch": 3.721841332027424,
|
| 698 |
+
"grad_norm": 0.09961768984794617,
|
| 699 |
+
"learning_rate": 1.4167393199651266e-05,
|
| 700 |
+
"loss": 0.0121,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 701 |
"step": 950
|
| 702 |
},
|
| 703 |
{
|
| 704 |
+
"epoch": 3.76101860920666,
|
| 705 |
+
"grad_norm": 0.6661494970321655,
|
| 706 |
+
"learning_rate": 1.3731473408892765e-05,
|
| 707 |
+
"loss": 0.0123,
|
| 708 |
"step": 960
|
| 709 |
},
|
| 710 |
{
|
| 711 |
+
"epoch": 3.8001958863858962,
|
| 712 |
+
"grad_norm": 0.4994729459285736,
|
| 713 |
+
"learning_rate": 1.3295553618134264e-05,
|
| 714 |
+
"loss": 0.0261,
|
| 715 |
"step": 970
|
| 716 |
},
|
| 717 |
{
|
| 718 |
+
"epoch": 3.8393731635651323,
|
| 719 |
+
"grad_norm": 0.5843683481216431,
|
| 720 |
+
"learning_rate": 1.2859633827375764e-05,
|
| 721 |
+
"loss": 0.0222,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 722 |
"step": 980
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 3.8785504407443683,
|
| 726 |
+
"grad_norm": 0.30400168895721436,
|
| 727 |
+
"learning_rate": 1.2423714036617264e-05,
|
| 728 |
+
"loss": 0.0217,
|
| 729 |
"step": 990
|
| 730 |
},
|
| 731 |
{
|
| 732 |
+
"epoch": 3.9177277179236043,
|
| 733 |
+
"grad_norm": 0.1775442361831665,
|
| 734 |
+
"learning_rate": 1.1987794245858763e-05,
|
| 735 |
+
"loss": 0.0194,
|
| 736 |
"step": 1000
|
| 737 |
},
|
| 738 |
{
|
| 739 |
+
"epoch": 3.9569049951028403,
|
| 740 |
+
"grad_norm": 0.754188060760498,
|
| 741 |
+
"learning_rate": 1.1551874455100261e-05,
|
| 742 |
+
"loss": 0.0235,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 743 |
"step": 1010
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 3.9960822722820764,
|
| 747 |
+
"grad_norm": 0.2706276774406433,
|
| 748 |
+
"learning_rate": 1.1115954664341762e-05,
|
| 749 |
+
"loss": 0.0158,
|
| 750 |
"step": 1020
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"epoch": 4.0,
|
| 754 |
+
"eval_accuracy": 0.9944879960803528,
|
| 755 |
+
"eval_loss": 0.023838121443986893,
|
| 756 |
+
"eval_runtime": 105.6855,
|
| 757 |
+
"eval_samples_per_second": 154.496,
|
| 758 |
+
"eval_steps_per_second": 2.422,
|
| 759 |
+
"step": 1021
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 4.035259549461313,
|
| 763 |
+
"grad_norm": 0.7036624550819397,
|
| 764 |
+
"learning_rate": 1.0680034873583262e-05,
|
| 765 |
+
"loss": 0.0212,
|
| 766 |
"step": 1030
|
| 767 |
},
|
| 768 |
{
|
| 769 |
+
"epoch": 4.074436826640548,
|
| 770 |
+
"grad_norm": 0.3211575746536255,
|
| 771 |
+
"learning_rate": 1.024411508282476e-05,
|
| 772 |
+
"loss": 0.0158,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 773 |
"step": 1040
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 4.113614103819785,
|
| 777 |
+
"grad_norm": 0.28554221987724304,
|
| 778 |
+
"learning_rate": 9.80819529206626e-06,
|
| 779 |
+
"loss": 0.0166,
|
| 780 |
"step": 1050
|
| 781 |
},
|
| 782 |
{
|
| 783 |
+
"epoch": 4.1527913809990205,
|
| 784 |
+
"grad_norm": 0.23619802296161652,
|
| 785 |
+
"learning_rate": 9.372275501307761e-06,
|
| 786 |
+
"loss": 0.022,
|
| 787 |
"step": 1060
|
| 788 |
},
|
| 789 |
{
|
| 790 |
+
"epoch": 4.191968658178257,
|
| 791 |
+
"grad_norm": 0.24213068187236786,
|
| 792 |
+
"learning_rate": 8.93635571054926e-06,
|
| 793 |
+
"loss": 0.0201,
|
| 794 |
"step": 1070
|
| 795 |
},
|
| 796 |
{
|
| 797 |
+
"epoch": 4.2311459353574925,
|
| 798 |
+
"grad_norm": 0.5181974172592163,
|
| 799 |
+
"learning_rate": 8.500435919790758e-06,
|
| 800 |
+
"loss": 0.0289,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 801 |
"step": 1080
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"epoch": 4.270323212536729,
|
| 805 |
+
"grad_norm": 0.9876229166984558,
|
| 806 |
+
"learning_rate": 8.064516129032258e-06,
|
| 807 |
+
"loss": 0.0198,
|
| 808 |
"step": 1090
|
| 809 |
},
|
| 810 |
{
|
| 811 |
+
"epoch": 4.3095004897159646,
|
| 812 |
+
"grad_norm": 1.2217401266098022,
|
| 813 |
+
"learning_rate": 7.628596338273758e-06,
|
| 814 |
+
"loss": 0.0197,
|
| 815 |
"step": 1100
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 4.348677766895201,
|
| 819 |
+
"grad_norm": 0.6681068539619446,
|
| 820 |
+
"learning_rate": 7.192676547515258e-06,
|
| 821 |
+
"loss": 0.0217,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 822 |
"step": 1110
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 4.387855044074437,
|
| 826 |
+
"grad_norm": 0.2516974210739136,
|
| 827 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 828 |
+
"loss": 0.015,
|
| 829 |
"step": 1120
|
| 830 |
},
|
| 831 |
{
|
| 832 |
+
"epoch": 4.427032321253673,
|
| 833 |
+
"grad_norm": 0.6431485414505005,
|
| 834 |
+
"learning_rate": 6.320836965998257e-06,
|
| 835 |
+
"loss": 0.0165,
|
| 836 |
"step": 1130
|
| 837 |
},
|
| 838 |
{
|
| 839 |
+
"epoch": 4.466209598432909,
|
| 840 |
+
"grad_norm": 0.8847171664237976,
|
| 841 |
+
"learning_rate": 5.884917175239756e-06,
|
| 842 |
+
"loss": 0.0169,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 843 |
"step": 1140
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"epoch": 4.505386875612145,
|
| 847 |
+
"grad_norm": 0.13626788556575775,
|
| 848 |
+
"learning_rate": 5.448997384481256e-06,
|
| 849 |
+
"loss": 0.0135,
|
| 850 |
"step": 1150
|
| 851 |
},
|
| 852 |
{
|
| 853 |
+
"epoch": 4.544564152791381,
|
| 854 |
+
"grad_norm": 0.19009195268154144,
|
| 855 |
+
"learning_rate": 5.013077593722755e-06,
|
| 856 |
+
"loss": 0.0082,
|
| 857 |
"step": 1160
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 4.583741429970617,
|
| 861 |
+
"grad_norm": 0.5194115042686462,
|
| 862 |
+
"learning_rate": 4.577157802964255e-06,
|
| 863 |
+
"loss": 0.0186,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 864 |
"step": 1170
|
| 865 |
},
|
| 866 |
{
|
| 867 |
+
"epoch": 4.622918707149853,
|
| 868 |
+
"grad_norm": 0.3963209092617035,
|
| 869 |
+
"learning_rate": 4.141238012205754e-06,
|
| 870 |
+
"loss": 0.0211,
|
| 871 |
"step": 1180
|
| 872 |
},
|
| 873 |
{
|
| 874 |
+
"epoch": 4.662095984329089,
|
| 875 |
+
"grad_norm": 0.392490953207016,
|
| 876 |
+
"learning_rate": 3.7053182214472536e-06,
|
| 877 |
+
"loss": 0.0121,
|
| 878 |
"step": 1190
|
| 879 |
},
|
| 880 |
{
|
| 881 |
+
"epoch": 4.701273261508325,
|
| 882 |
+
"grad_norm": 0.2609846293926239,
|
| 883 |
+
"learning_rate": 3.2693984306887534e-06,
|
| 884 |
+
"loss": 0.024,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 885 |
"step": 1200
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 4.740450538687561,
|
| 889 |
+
"grad_norm": 0.9114863872528076,
|
| 890 |
+
"learning_rate": 2.8334786399302533e-06,
|
| 891 |
+
"loss": 0.022,
|
| 892 |
"step": 1210
|
| 893 |
},
|
| 894 |
{
|
| 895 |
+
"epoch": 4.779627815866798,
|
| 896 |
+
"grad_norm": 0.3400118947029114,
|
| 897 |
+
"learning_rate": 2.3975588491717523e-06,
|
| 898 |
+
"loss": 0.0139,
|
| 899 |
"step": 1220
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 4.818805093046033,
|
| 903 |
+
"grad_norm": 0.8354482650756836,
|
| 904 |
+
"learning_rate": 1.961639058413252e-06,
|
| 905 |
+
"loss": 0.0168,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 906 |
"step": 1230
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 4.857982370225269,
|
| 910 |
+
"grad_norm": 0.4545687139034271,
|
| 911 |
+
"learning_rate": 1.5257192676547516e-06,
|
| 912 |
+
"loss": 0.0176,
|
| 913 |
"step": 1240
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"epoch": 4.897159647404505,
|
| 917 |
+
"grad_norm": 0.6158453226089478,
|
| 918 |
+
"learning_rate": 1.0897994768962512e-06,
|
| 919 |
+
"loss": 0.0169,
|
| 920 |
+
"step": 1250
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 4.936336924583742,
|
| 924 |
+
"grad_norm": 0.11892726272344589,
|
| 925 |
+
"learning_rate": 6.538796861377506e-07,
|
| 926 |
+
"loss": 0.0196,
|
| 927 |
+
"step": 1260
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 4.975514201762977,
|
| 931 |
+
"grad_norm": 0.5337911248207092,
|
| 932 |
+
"learning_rate": 2.179598953792502e-07,
|
| 933 |
+
"loss": 0.0112,
|
| 934 |
+
"step": 1270
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 4.995102840352596,
|
| 938 |
+
"eval_accuracy": 0.9943655071043606,
|
| 939 |
+
"eval_loss": 0.02594069205224514,
|
| 940 |
+
"eval_runtime": 104.7525,
|
| 941 |
+
"eval_samples_per_second": 155.872,
|
| 942 |
+
"eval_steps_per_second": 2.444,
|
| 943 |
+
"step": 1275
|
| 944 |
},
|
| 945 |
{
|
| 946 |
+
"epoch": 4.995102840352596,
|
| 947 |
+
"step": 1275,
|
| 948 |
+
"total_flos": 8.109813238393209e+18,
|
| 949 |
+
"train_loss": 0.030976083857171675,
|
| 950 |
+
"train_runtime": 4473.2861,
|
| 951 |
+
"train_samples_per_second": 73.002,
|
| 952 |
+
"train_steps_per_second": 0.285
|
| 953 |
}
|
| 954 |
],
|
| 955 |
"logging_steps": 10,
|
| 956 |
+
"max_steps": 1275,
|
| 957 |
"num_input_tokens_seen": 0,
|
| 958 |
+
"num_train_epochs": 5,
|
| 959 |
"save_steps": 500,
|
| 960 |
"stateful_callbacks": {
|
| 961 |
"TrainerControl": {
|
|
|
|
| 969 |
"attributes": {}
|
| 970 |
}
|
| 971 |
},
|
| 972 |
+
"total_flos": 8.109813238393209e+18,
|
| 973 |
"train_batch_size": 64,
|
| 974 |
"trial_name": null,
|
| 975 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c28f4f40287fbfa917bd2f4f97b7457f8e5f3e4b2a551453613ec650e854b4e
|
| 3 |
size 5240
|