Training in progress, epoch 0
Browse files- all_results.json +6 -6
- model.safetensors +1 -1
- runs/Dec09_18-28-48_c9ba084c85bf/events.out.tfevents.1733768937.c9ba084c85bf.5668.2 +3 -0
- train_results.json +6 -6
- trainer_state.json +516 -31
- training_args.bin +1 -1
all_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": 38.4,
|
| 3 |
+
"total_flos": 7.390182645274706e+17,
|
| 4 |
+
"train_loss": 0.35739548206329347,
|
| 5 |
+
"train_runtime": 377.1425,
|
| 6 |
+
"train_samples_per_second": 82.091,
|
| 7 |
+
"train_steps_per_second": 0.636
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 110342832
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8011d54913d9f20509ab0301a10e2d821326033939c98c88c668d04ab0123bc
|
| 3 |
size 110342832
|
runs/Dec09_18-28-48_c9ba084c85bf/events.out.tfevents.1733768937.c9ba084c85bf.5668.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a08ceba8255e097d3f1ffdfea3e668b4eabf4fe9b52a25be8ba56a5496bb31f
|
| 3 |
+
size 5688
|
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": 38.4,
|
| 3 |
+
"total_flos": 7.390182645274706e+17,
|
| 4 |
+
"train_loss": 0.35739548206329347,
|
| 5 |
+
"train_runtime": 377.1425,
|
| 6 |
+
"train_samples_per_second": 82.091,
|
| 7 |
+
"train_steps_per_second": 0.636
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,61 +1,546 @@
|
|
| 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": 0.96,
|
| 13 |
-
"eval_accuracy": 0.
|
| 14 |
-
"eval_loss": 0.
|
| 15 |
-
"eval_runtime": 0.
|
| 16 |
-
"eval_samples_per_second":
|
| 17 |
-
"eval_steps_per_second": 6.
|
| 18 |
"step": 6
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 1.6,
|
| 22 |
-
"grad_norm":
|
| 23 |
-
"learning_rate": 2.
|
| 24 |
-
"loss": 0.
|
| 25 |
"step": 10
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 1.92,
|
| 29 |
-
"eval_accuracy": 0.
|
| 30 |
-
"eval_loss": 0.
|
| 31 |
-
"eval_runtime": 0.
|
| 32 |
-
"eval_samples_per_second":
|
| 33 |
-
"eval_steps_per_second": 6.
|
| 34 |
"step": 12
|
| 35 |
},
|
| 36 |
{
|
| 37 |
"epoch": 2.88,
|
| 38 |
"eval_accuracy": 0.686046511627907,
|
| 39 |
-
"eval_loss": 0.
|
| 40 |
-
"eval_runtime": 0.
|
| 41 |
-
"eval_samples_per_second":
|
| 42 |
-
"eval_steps_per_second": 6.
|
| 43 |
"step": 18
|
| 44 |
},
|
| 45 |
{
|
| 46 |
-
"epoch": 2
|
| 47 |
-
"
|
| 48 |
-
"
|
| 49 |
-
"
|
| 50 |
-
"
|
| 51 |
-
|
| 52 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
],
|
| 55 |
"logging_steps": 10,
|
| 56 |
-
"max_steps":
|
| 57 |
"num_input_tokens_seen": 0,
|
| 58 |
-
"num_train_epochs":
|
| 59 |
"save_steps": 500,
|
| 60 |
"stateful_callbacks": {
|
| 61 |
"TrainerControl": {
|
|
@@ -69,7 +554,7 @@
|
|
| 69 |
"attributes": {}
|
| 70 |
}
|
| 71 |
},
|
| 72 |
-
"total_flos":
|
| 73 |
"train_batch_size": 32,
|
| 74 |
"trial_name": null,
|
| 75 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8023255813953488,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-100",
|
| 4 |
+
"epoch": 38.4,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 240,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.96,
|
| 13 |
+
"eval_accuracy": 0.686046511627907,
|
| 14 |
+
"eval_loss": 0.5580831170082092,
|
| 15 |
+
"eval_runtime": 0.4689,
|
| 16 |
+
"eval_samples_per_second": 183.412,
|
| 17 |
+
"eval_steps_per_second": 6.398,
|
| 18 |
"step": 6
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 1.6,
|
| 22 |
+
"grad_norm": 11.4143648147583,
|
| 23 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 24 |
+
"loss": 0.5082,
|
| 25 |
"step": 10
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"epoch": 1.92,
|
| 29 |
+
"eval_accuracy": 0.6627906976744186,
|
| 30 |
+
"eval_loss": 0.5865343809127808,
|
| 31 |
+
"eval_runtime": 0.4656,
|
| 32 |
+
"eval_samples_per_second": 184.706,
|
| 33 |
+
"eval_steps_per_second": 6.443,
|
| 34 |
"step": 12
|
| 35 |
},
|
| 36 |
{
|
| 37 |
"epoch": 2.88,
|
| 38 |
"eval_accuracy": 0.686046511627907,
|
| 39 |
+
"eval_loss": 0.5983394980430603,
|
| 40 |
+
"eval_runtime": 0.457,
|
| 41 |
+
"eval_samples_per_second": 188.165,
|
| 42 |
+
"eval_steps_per_second": 6.564,
|
| 43 |
"step": 18
|
| 44 |
},
|
| 45 |
{
|
| 46 |
+
"epoch": 3.2,
|
| 47 |
+
"grad_norm": 43.284976959228516,
|
| 48 |
+
"learning_rate": 4.166666666666667e-05,
|
| 49 |
+
"loss": 0.4618,
|
| 50 |
+
"step": 20
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_accuracy": 0.6744186046511628,
|
| 55 |
+
"eval_loss": 0.6791186928749084,
|
| 56 |
+
"eval_runtime": 0.4897,
|
| 57 |
+
"eval_samples_per_second": 175.601,
|
| 58 |
+
"eval_steps_per_second": 6.126,
|
| 59 |
+
"step": 25
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.8,
|
| 63 |
+
"grad_norm": 52.168453216552734,
|
| 64 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 65 |
+
"loss": 0.3901,
|
| 66 |
+
"step": 30
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.96,
|
| 70 |
+
"eval_accuracy": 0.7325581395348837,
|
| 71 |
+
"eval_loss": 0.5641953945159912,
|
| 72 |
+
"eval_runtime": 0.4579,
|
| 73 |
+
"eval_samples_per_second": 187.798,
|
| 74 |
+
"eval_steps_per_second": 6.551,
|
| 75 |
+
"step": 31
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.92,
|
| 79 |
+
"eval_accuracy": 0.7093023255813954,
|
| 80 |
+
"eval_loss": 0.5044412612915039,
|
| 81 |
+
"eval_runtime": 0.4414,
|
| 82 |
+
"eval_samples_per_second": 194.839,
|
| 83 |
+
"eval_steps_per_second": 6.797,
|
| 84 |
+
"step": 37
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.4,
|
| 88 |
+
"grad_norm": 144.19395446777344,
|
| 89 |
+
"learning_rate": 4.62962962962963e-05,
|
| 90 |
+
"loss": 0.4175,
|
| 91 |
+
"step": 40
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 6.88,
|
| 95 |
+
"eval_accuracy": 0.6976744186046512,
|
| 96 |
+
"eval_loss": 0.5285409688949585,
|
| 97 |
+
"eval_runtime": 0.4518,
|
| 98 |
+
"eval_samples_per_second": 190.363,
|
| 99 |
+
"eval_steps_per_second": 6.641,
|
| 100 |
+
"step": 43
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 8.0,
|
| 104 |
+
"grad_norm": 22.712488174438477,
|
| 105 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 106 |
+
"loss": 0.4308,
|
| 107 |
+
"step": 50
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 8.0,
|
| 111 |
+
"eval_accuracy": 0.7093023255813954,
|
| 112 |
+
"eval_loss": 0.5152430534362793,
|
| 113 |
+
"eval_runtime": 0.4733,
|
| 114 |
+
"eval_samples_per_second": 181.688,
|
| 115 |
+
"eval_steps_per_second": 6.338,
|
| 116 |
+
"step": 50
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 8.96,
|
| 120 |
+
"eval_accuracy": 0.7209302325581395,
|
| 121 |
+
"eval_loss": 0.5627985000610352,
|
| 122 |
+
"eval_runtime": 0.4653,
|
| 123 |
+
"eval_samples_per_second": 184.819,
|
| 124 |
+
"eval_steps_per_second": 6.447,
|
| 125 |
+
"step": 56
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 9.6,
|
| 129 |
+
"grad_norm": 19.49258041381836,
|
| 130 |
+
"learning_rate": 4.166666666666667e-05,
|
| 131 |
+
"loss": 0.3998,
|
| 132 |
+
"step": 60
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.92,
|
| 136 |
+
"eval_accuracy": 0.7674418604651163,
|
| 137 |
+
"eval_loss": 0.5401325821876526,
|
| 138 |
+
"eval_runtime": 0.4492,
|
| 139 |
+
"eval_samples_per_second": 191.442,
|
| 140 |
+
"eval_steps_per_second": 6.678,
|
| 141 |
+
"step": 62
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.88,
|
| 145 |
+
"eval_accuracy": 0.7790697674418605,
|
| 146 |
+
"eval_loss": 0.5199148654937744,
|
| 147 |
+
"eval_runtime": 0.4492,
|
| 148 |
+
"eval_samples_per_second": 191.438,
|
| 149 |
+
"eval_steps_per_second": 6.678,
|
| 150 |
+
"step": 68
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 11.2,
|
| 154 |
+
"grad_norm": 16.36634063720703,
|
| 155 |
+
"learning_rate": 3.935185185185186e-05,
|
| 156 |
+
"loss": 0.3682,
|
| 157 |
+
"step": 70
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 12.0,
|
| 161 |
+
"eval_accuracy": 0.7906976744186046,
|
| 162 |
+
"eval_loss": 0.5042721033096313,
|
| 163 |
+
"eval_runtime": 0.4513,
|
| 164 |
+
"eval_samples_per_second": 190.572,
|
| 165 |
+
"eval_steps_per_second": 6.648,
|
| 166 |
+
"step": 75
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.8,
|
| 170 |
+
"grad_norm": 20.526153564453125,
|
| 171 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 172 |
+
"loss": 0.3528,
|
| 173 |
+
"step": 80
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 12.96,
|
| 177 |
+
"eval_accuracy": 0.7790697674418605,
|
| 178 |
+
"eval_loss": 0.47959619760513306,
|
| 179 |
+
"eval_runtime": 0.4571,
|
| 180 |
+
"eval_samples_per_second": 188.126,
|
| 181 |
+
"eval_steps_per_second": 6.563,
|
| 182 |
+
"step": 81
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.92,
|
| 186 |
+
"eval_accuracy": 0.7790697674418605,
|
| 187 |
+
"eval_loss": 0.4937508702278137,
|
| 188 |
+
"eval_runtime": 0.4685,
|
| 189 |
+
"eval_samples_per_second": 183.563,
|
| 190 |
+
"eval_steps_per_second": 6.403,
|
| 191 |
+
"step": 87
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 14.4,
|
| 195 |
+
"grad_norm": 24.452070236206055,
|
| 196 |
+
"learning_rate": 3.472222222222222e-05,
|
| 197 |
+
"loss": 0.3324,
|
| 198 |
+
"step": 90
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 14.88,
|
| 202 |
+
"eval_accuracy": 0.7558139534883721,
|
| 203 |
+
"eval_loss": 0.4879016876220703,
|
| 204 |
+
"eval_runtime": 0.461,
|
| 205 |
+
"eval_samples_per_second": 186.569,
|
| 206 |
+
"eval_steps_per_second": 6.508,
|
| 207 |
+
"step": 93
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 16.0,
|
| 211 |
+
"grad_norm": 20.879064559936523,
|
| 212 |
+
"learning_rate": 3.240740740740741e-05,
|
| 213 |
+
"loss": 0.3579,
|
| 214 |
+
"step": 100
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.0,
|
| 218 |
+
"eval_accuracy": 0.8023255813953488,
|
| 219 |
+
"eval_loss": 0.4972393214702606,
|
| 220 |
+
"eval_runtime": 0.4593,
|
| 221 |
+
"eval_samples_per_second": 187.231,
|
| 222 |
+
"eval_steps_per_second": 6.531,
|
| 223 |
+
"step": 100
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.96,
|
| 227 |
+
"eval_accuracy": 0.7674418604651163,
|
| 228 |
+
"eval_loss": 0.45788562297821045,
|
| 229 |
+
"eval_runtime": 0.4452,
|
| 230 |
+
"eval_samples_per_second": 193.187,
|
| 231 |
+
"eval_steps_per_second": 6.739,
|
| 232 |
+
"step": 106
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 17.6,
|
| 236 |
+
"grad_norm": 30.157981872558594,
|
| 237 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 238 |
+
"loss": 0.3566,
|
| 239 |
+
"step": 110
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 17.92,
|
| 243 |
+
"eval_accuracy": 0.7790697674418605,
|
| 244 |
+
"eval_loss": 0.48914143443107605,
|
| 245 |
+
"eval_runtime": 0.4757,
|
| 246 |
+
"eval_samples_per_second": 180.776,
|
| 247 |
+
"eval_steps_per_second": 6.306,
|
| 248 |
+
"step": 112
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.88,
|
| 252 |
+
"eval_accuracy": 0.8023255813953488,
|
| 253 |
+
"eval_loss": 0.4653853178024292,
|
| 254 |
+
"eval_runtime": 0.4882,
|
| 255 |
+
"eval_samples_per_second": 176.154,
|
| 256 |
+
"eval_steps_per_second": 6.145,
|
| 257 |
+
"step": 118
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 19.2,
|
| 261 |
+
"grad_norm": 18.67655372619629,
|
| 262 |
+
"learning_rate": 2.777777777777778e-05,
|
| 263 |
+
"loss": 0.3382,
|
| 264 |
+
"step": 120
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 20.0,
|
| 268 |
+
"eval_accuracy": 0.7906976744186046,
|
| 269 |
+
"eval_loss": 0.46721386909484863,
|
| 270 |
+
"eval_runtime": 0.4644,
|
| 271 |
+
"eval_samples_per_second": 185.204,
|
| 272 |
+
"eval_steps_per_second": 6.461,
|
| 273 |
+
"step": 125
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 20.8,
|
| 277 |
+
"grad_norm": 16.18861198425293,
|
| 278 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 279 |
+
"loss": 0.3534,
|
| 280 |
+
"step": 130
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.96,
|
| 284 |
+
"eval_accuracy": 0.7790697674418605,
|
| 285 |
+
"eval_loss": 0.4687628149986267,
|
| 286 |
+
"eval_runtime": 0.4743,
|
| 287 |
+
"eval_samples_per_second": 181.318,
|
| 288 |
+
"eval_steps_per_second": 6.325,
|
| 289 |
+
"step": 131
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 21.92,
|
| 293 |
+
"eval_accuracy": 0.7558139534883721,
|
| 294 |
+
"eval_loss": 0.48910030722618103,
|
| 295 |
+
"eval_runtime": 0.4909,
|
| 296 |
+
"eval_samples_per_second": 175.174,
|
| 297 |
+
"eval_steps_per_second": 6.111,
|
| 298 |
+
"step": 137
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 22.4,
|
| 302 |
+
"grad_norm": 16.807844161987305,
|
| 303 |
+
"learning_rate": 2.314814814814815e-05,
|
| 304 |
+
"loss": 0.3462,
|
| 305 |
+
"step": 140
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 22.88,
|
| 309 |
+
"eval_accuracy": 0.7441860465116279,
|
| 310 |
+
"eval_loss": 0.502510130405426,
|
| 311 |
+
"eval_runtime": 0.4792,
|
| 312 |
+
"eval_samples_per_second": 179.482,
|
| 313 |
+
"eval_steps_per_second": 6.261,
|
| 314 |
+
"step": 143
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 24.0,
|
| 318 |
+
"grad_norm": 37.028865814208984,
|
| 319 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 320 |
+
"loss": 0.3208,
|
| 321 |
+
"step": 150
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 24.0,
|
| 325 |
+
"eval_accuracy": 0.7674418604651163,
|
| 326 |
+
"eval_loss": 0.5026240944862366,
|
| 327 |
+
"eval_runtime": 0.4788,
|
| 328 |
+
"eval_samples_per_second": 179.607,
|
| 329 |
+
"eval_steps_per_second": 6.265,
|
| 330 |
+
"step": 150
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.96,
|
| 334 |
+
"eval_accuracy": 0.7674418604651163,
|
| 335 |
+
"eval_loss": 0.493563175201416,
|
| 336 |
+
"eval_runtime": 0.4917,
|
| 337 |
+
"eval_samples_per_second": 174.896,
|
| 338 |
+
"eval_steps_per_second": 6.101,
|
| 339 |
+
"step": 156
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 25.6,
|
| 343 |
+
"grad_norm": 10.407114028930664,
|
| 344 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 345 |
+
"loss": 0.3408,
|
| 346 |
+
"step": 160
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.92,
|
| 350 |
+
"eval_accuracy": 0.7790697674418605,
|
| 351 |
+
"eval_loss": 0.46066081523895264,
|
| 352 |
+
"eval_runtime": 0.4807,
|
| 353 |
+
"eval_samples_per_second": 178.913,
|
| 354 |
+
"eval_steps_per_second": 6.241,
|
| 355 |
+
"step": 162
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 26.88,
|
| 359 |
+
"eval_accuracy": 0.7906976744186046,
|
| 360 |
+
"eval_loss": 0.4420366883277893,
|
| 361 |
+
"eval_runtime": 0.4813,
|
| 362 |
+
"eval_samples_per_second": 178.692,
|
| 363 |
+
"eval_steps_per_second": 6.233,
|
| 364 |
+
"step": 168
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 27.2,
|
| 368 |
+
"grad_norm": 10.114925384521484,
|
| 369 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 370 |
+
"loss": 0.333,
|
| 371 |
+
"step": 170
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 28.0,
|
| 375 |
+
"eval_accuracy": 0.7906976744186046,
|
| 376 |
+
"eval_loss": 0.4296189248561859,
|
| 377 |
+
"eval_runtime": 0.4717,
|
| 378 |
+
"eval_samples_per_second": 182.312,
|
| 379 |
+
"eval_steps_per_second": 6.36,
|
| 380 |
+
"step": 175
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 28.8,
|
| 384 |
+
"grad_norm": 16.923192977905273,
|
| 385 |
+
"learning_rate": 1.388888888888889e-05,
|
| 386 |
+
"loss": 0.3169,
|
| 387 |
+
"step": 180
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.96,
|
| 391 |
+
"eval_accuracy": 0.7906976744186046,
|
| 392 |
+
"eval_loss": 0.4503108263015747,
|
| 393 |
+
"eval_runtime": 0.4671,
|
| 394 |
+
"eval_samples_per_second": 184.114,
|
| 395 |
+
"eval_steps_per_second": 6.423,
|
| 396 |
+
"step": 181
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 29.92,
|
| 400 |
+
"eval_accuracy": 0.8023255813953488,
|
| 401 |
+
"eval_loss": 0.4894145727157593,
|
| 402 |
+
"eval_runtime": 0.4851,
|
| 403 |
+
"eval_samples_per_second": 177.272,
|
| 404 |
+
"eval_steps_per_second": 6.184,
|
| 405 |
+
"step": 187
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 30.4,
|
| 409 |
+
"grad_norm": 16.85201072692871,
|
| 410 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 411 |
+
"loss": 0.3267,
|
| 412 |
+
"step": 190
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 30.88,
|
| 416 |
+
"eval_accuracy": 0.7906976744186046,
|
| 417 |
+
"eval_loss": 0.4838241636753082,
|
| 418 |
+
"eval_runtime": 0.4795,
|
| 419 |
+
"eval_samples_per_second": 179.37,
|
| 420 |
+
"eval_steps_per_second": 6.257,
|
| 421 |
+
"step": 193
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 32.0,
|
| 425 |
+
"grad_norm": 19.141088485717773,
|
| 426 |
+
"learning_rate": 9.259259259259259e-06,
|
| 427 |
+
"loss": 0.3114,
|
| 428 |
+
"step": 200
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 32.0,
|
| 432 |
+
"eval_accuracy": 0.7790697674418605,
|
| 433 |
+
"eval_loss": 0.5220069289207458,
|
| 434 |
+
"eval_runtime": 0.5078,
|
| 435 |
+
"eval_samples_per_second": 169.353,
|
| 436 |
+
"eval_steps_per_second": 5.908,
|
| 437 |
+
"step": 200
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 32.96,
|
| 441 |
+
"eval_accuracy": 0.7790697674418605,
|
| 442 |
+
"eval_loss": 0.49199116230010986,
|
| 443 |
+
"eval_runtime": 0.4644,
|
| 444 |
+
"eval_samples_per_second": 185.186,
|
| 445 |
+
"eval_steps_per_second": 6.46,
|
| 446 |
+
"step": 206
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 33.6,
|
| 450 |
+
"grad_norm": 27.817716598510742,
|
| 451 |
+
"learning_rate": 6.944444444444445e-06,
|
| 452 |
+
"loss": 0.3143,
|
| 453 |
+
"step": 210
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.92,
|
| 457 |
+
"eval_accuracy": 0.7441860465116279,
|
| 458 |
+
"eval_loss": 0.48098844289779663,
|
| 459 |
+
"eval_runtime": 0.4803,
|
| 460 |
+
"eval_samples_per_second": 179.049,
|
| 461 |
+
"eval_steps_per_second": 6.246,
|
| 462 |
+
"step": 212
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 34.88,
|
| 466 |
+
"eval_accuracy": 0.7558139534883721,
|
| 467 |
+
"eval_loss": 0.4828723073005676,
|
| 468 |
+
"eval_runtime": 0.492,
|
| 469 |
+
"eval_samples_per_second": 174.781,
|
| 470 |
+
"eval_steps_per_second": 6.097,
|
| 471 |
+
"step": 218
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 35.2,
|
| 475 |
+
"grad_norm": 46.449974060058594,
|
| 476 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 477 |
+
"loss": 0.3034,
|
| 478 |
+
"step": 220
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 36.0,
|
| 482 |
+
"eval_accuracy": 0.7674418604651163,
|
| 483 |
+
"eval_loss": 0.4835253953933716,
|
| 484 |
+
"eval_runtime": 0.496,
|
| 485 |
+
"eval_samples_per_second": 173.392,
|
| 486 |
+
"eval_steps_per_second": 6.049,
|
| 487 |
+
"step": 225
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 36.8,
|
| 491 |
+
"grad_norm": 18.135772705078125,
|
| 492 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 493 |
+
"loss": 0.2944,
|
| 494 |
+
"step": 230
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.96,
|
| 498 |
+
"eval_accuracy": 0.7790697674418605,
|
| 499 |
+
"eval_loss": 0.4811899960041046,
|
| 500 |
+
"eval_runtime": 0.4612,
|
| 501 |
+
"eval_samples_per_second": 186.468,
|
| 502 |
+
"eval_steps_per_second": 6.505,
|
| 503 |
+
"step": 231
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 37.92,
|
| 507 |
+
"eval_accuracy": 0.7674418604651163,
|
| 508 |
+
"eval_loss": 0.47796696424484253,
|
| 509 |
+
"eval_runtime": 0.4674,
|
| 510 |
+
"eval_samples_per_second": 183.98,
|
| 511 |
+
"eval_steps_per_second": 6.418,
|
| 512 |
+
"step": 237
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 38.4,
|
| 516 |
+
"grad_norm": 20.760650634765625,
|
| 517 |
+
"learning_rate": 0.0,
|
| 518 |
+
"loss": 0.3018,
|
| 519 |
+
"step": 240
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 38.4,
|
| 523 |
+
"eval_accuracy": 0.7674418604651163,
|
| 524 |
+
"eval_loss": 0.4778375029563904,
|
| 525 |
+
"eval_runtime": 0.4679,
|
| 526 |
+
"eval_samples_per_second": 183.798,
|
| 527 |
+
"eval_steps_per_second": 6.412,
|
| 528 |
+
"step": 240
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 38.4,
|
| 532 |
+
"step": 240,
|
| 533 |
+
"total_flos": 7.390182645274706e+17,
|
| 534 |
+
"train_loss": 0.35739548206329347,
|
| 535 |
+
"train_runtime": 377.1425,
|
| 536 |
+
"train_samples_per_second": 82.091,
|
| 537 |
+
"train_steps_per_second": 0.636
|
| 538 |
}
|
| 539 |
],
|
| 540 |
"logging_steps": 10,
|
| 541 |
+
"max_steps": 240,
|
| 542 |
"num_input_tokens_seen": 0,
|
| 543 |
+
"num_train_epochs": 40,
|
| 544 |
"save_steps": 500,
|
| 545 |
"stateful_callbacks": {
|
| 546 |
"TrainerControl": {
|
|
|
|
| 554 |
"attributes": {}
|
| 555 |
}
|
| 556 |
},
|
| 557 |
+
"total_flos": 7.390182645274706e+17,
|
| 558 |
"train_batch_size": 32,
|
| 559 |
"trial_name": null,
|
| 560 |
"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:a64f9ef7699538b1f8a977c317ca4296b4f737cdf255e81c2a42b4206d4e0dc8
|
| 3 |
size 5240
|