End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr15_08-24-37_d3d1cd9553d4/events.out.tfevents.1713172101.d3d1cd9553d4.1260.3 +3 -0
- train_results.json +8 -0
- trainer_state.json +709 -0
README.md
CHANGED
|
@@ -32,7 +32,7 @@ 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: 1.0
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 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: 0.0029
|
| 36 |
- Accuracy: 1.0
|
| 37 |
|
| 38 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.96,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.0029437176417559385,
|
| 5 |
+
"eval_runtime": 33.6386,
|
| 6 |
+
"eval_samples_per_second": 53.51,
|
| 7 |
+
"eval_steps_per_second": 1.694,
|
| 8 |
+
"total_flos": 2.8025139821285376e+18,
|
| 9 |
+
"train_loss": 0.13459565003810117,
|
| 10 |
+
"train_runtime": 2248.9328,
|
| 11 |
+
"train_samples_per_second": 50.411,
|
| 12 |
+
"train_steps_per_second": 0.392
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.96,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.0029437176417559385,
|
| 5 |
+
"eval_runtime": 33.6386,
|
| 6 |
+
"eval_samples_per_second": 53.51,
|
| 7 |
+
"eval_steps_per_second": 1.694
|
| 8 |
+
}
|
runs/Apr15_08-24-37_d3d1cd9553d4/events.out.tfevents.1713172101.d3d1cd9553d4.1260.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37a4ceee9101265e89814b07049b9c9461a68d63bcf142d5e770002a2a4b2172
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.96,
|
| 3 |
+
"total_flos": 2.8025139821285376e+18,
|
| 4 |
+
"train_loss": 0.13459565003810117,
|
| 5 |
+
"train_runtime": 2248.9328,
|
| 6 |
+
"train_samples_per_second": 50.411,
|
| 7 |
+
"train_steps_per_second": 0.392
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,709 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-car0001/checkpoint-126",
|
| 4 |
+
"epoch": 6.958579881656805,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 882,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08,
|
| 13 |
+
"grad_norm": 14.615551948547363,
|
| 14 |
+
"learning_rate": 5.617977528089888e-06,
|
| 15 |
+
"loss": 2.2752,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.16,
|
| 20 |
+
"grad_norm": 12.17358684539795,
|
| 21 |
+
"learning_rate": 1.1235955056179776e-05,
|
| 22 |
+
"loss": 2.1418,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.24,
|
| 27 |
+
"grad_norm": 24.358016967773438,
|
| 28 |
+
"learning_rate": 1.6853932584269665e-05,
|
| 29 |
+
"loss": 1.9448,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.32,
|
| 34 |
+
"grad_norm": 22.546871185302734,
|
| 35 |
+
"learning_rate": 2.2471910112359552e-05,
|
| 36 |
+
"loss": 1.6261,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.39,
|
| 41 |
+
"grad_norm": 22.2501220703125,
|
| 42 |
+
"learning_rate": 2.8089887640449443e-05,
|
| 43 |
+
"loss": 1.2392,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.47,
|
| 48 |
+
"grad_norm": 27.82279396057129,
|
| 49 |
+
"learning_rate": 3.370786516853933e-05,
|
| 50 |
+
"loss": 0.8038,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.55,
|
| 55 |
+
"grad_norm": 21.99744987487793,
|
| 56 |
+
"learning_rate": 3.9325842696629214e-05,
|
| 57 |
+
"loss": 0.4458,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.63,
|
| 62 |
+
"grad_norm": 7.488814830780029,
|
| 63 |
+
"learning_rate": 4.4943820224719104e-05,
|
| 64 |
+
"loss": 0.1849,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.71,
|
| 69 |
+
"grad_norm": 23.27031898498535,
|
| 70 |
+
"learning_rate": 4.9936948297604034e-05,
|
| 71 |
+
"loss": 0.1282,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.79,
|
| 76 |
+
"grad_norm": 4.6876349449157715,
|
| 77 |
+
"learning_rate": 4.930643127364439e-05,
|
| 78 |
+
"loss": 0.077,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.87,
|
| 83 |
+
"grad_norm": 8.07778549194336,
|
| 84 |
+
"learning_rate": 4.8675914249684744e-05,
|
| 85 |
+
"loss": 0.0453,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.95,
|
| 90 |
+
"grad_norm": 8.182723045349121,
|
| 91 |
+
"learning_rate": 4.8045397225725096e-05,
|
| 92 |
+
"loss": 0.0328,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.99,
|
| 97 |
+
"eval_accuracy": 1.0,
|
| 98 |
+
"eval_loss": 0.0029437176417559385,
|
| 99 |
+
"eval_runtime": 32.9924,
|
| 100 |
+
"eval_samples_per_second": 54.558,
|
| 101 |
+
"eval_steps_per_second": 1.728,
|
| 102 |
+
"step": 126
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.03,
|
| 106 |
+
"grad_norm": 7.431449890136719,
|
| 107 |
+
"learning_rate": 4.741488020176545e-05,
|
| 108 |
+
"loss": 0.0312,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1,
|
| 113 |
+
"grad_norm": 1.8369030952453613,
|
| 114 |
+
"learning_rate": 4.67843631778058e-05,
|
| 115 |
+
"loss": 0.029,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.18,
|
| 120 |
+
"grad_norm": 3.7118313312530518,
|
| 121 |
+
"learning_rate": 4.615384615384616e-05,
|
| 122 |
+
"loss": 0.0334,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.26,
|
| 127 |
+
"grad_norm": 2.428013324737549,
|
| 128 |
+
"learning_rate": 4.552332912988651e-05,
|
| 129 |
+
"loss": 0.021,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.34,
|
| 134 |
+
"grad_norm": 0.6982976794242859,
|
| 135 |
+
"learning_rate": 4.489281210592686e-05,
|
| 136 |
+
"loss": 0.0186,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.42,
|
| 141 |
+
"grad_norm": 0.09051941335201263,
|
| 142 |
+
"learning_rate": 4.426229508196721e-05,
|
| 143 |
+
"loss": 0.0234,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.5,
|
| 148 |
+
"grad_norm": 1.7318867444992065,
|
| 149 |
+
"learning_rate": 4.3631778058007565e-05,
|
| 150 |
+
"loss": 0.0125,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.58,
|
| 155 |
+
"grad_norm": 5.148034572601318,
|
| 156 |
+
"learning_rate": 4.3001261034047924e-05,
|
| 157 |
+
"loss": 0.0354,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.66,
|
| 162 |
+
"grad_norm": 5.717312812805176,
|
| 163 |
+
"learning_rate": 4.2370744010088275e-05,
|
| 164 |
+
"loss": 0.0255,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.74,
|
| 169 |
+
"grad_norm": 5.518301010131836,
|
| 170 |
+
"learning_rate": 4.174022698612863e-05,
|
| 171 |
+
"loss": 0.0191,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.81,
|
| 176 |
+
"grad_norm": 1.6274638175964355,
|
| 177 |
+
"learning_rate": 4.110970996216898e-05,
|
| 178 |
+
"loss": 0.0316,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.89,
|
| 183 |
+
"grad_norm": 0.7395884394645691,
|
| 184 |
+
"learning_rate": 4.047919293820933e-05,
|
| 185 |
+
"loss": 0.034,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.97,
|
| 190 |
+
"grad_norm": 7.8381757736206055,
|
| 191 |
+
"learning_rate": 3.984867591424969e-05,
|
| 192 |
+
"loss": 0.0107,
|
| 193 |
+
"step": 250
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.0,
|
| 197 |
+
"eval_accuracy": 1.0,
|
| 198 |
+
"eval_loss": 0.00012798496754840016,
|
| 199 |
+
"eval_runtime": 32.2135,
|
| 200 |
+
"eval_samples_per_second": 55.877,
|
| 201 |
+
"eval_steps_per_second": 1.769,
|
| 202 |
+
"step": 253
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.05,
|
| 206 |
+
"grad_norm": 2.6944656372070312,
|
| 207 |
+
"learning_rate": 3.921815889029004e-05,
|
| 208 |
+
"loss": 0.0197,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.13,
|
| 213 |
+
"grad_norm": 0.7401177883148193,
|
| 214 |
+
"learning_rate": 3.858764186633039e-05,
|
| 215 |
+
"loss": 0.0136,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.21,
|
| 220 |
+
"grad_norm": 0.7982012033462524,
|
| 221 |
+
"learning_rate": 3.7957124842370744e-05,
|
| 222 |
+
"loss": 0.0153,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.29,
|
| 227 |
+
"grad_norm": 2.5111560821533203,
|
| 228 |
+
"learning_rate": 3.7326607818411096e-05,
|
| 229 |
+
"loss": 0.0138,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.37,
|
| 234 |
+
"grad_norm": 21.724336624145508,
|
| 235 |
+
"learning_rate": 3.6696090794451454e-05,
|
| 236 |
+
"loss": 0.0231,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.45,
|
| 241 |
+
"grad_norm": 1.90717351436615,
|
| 242 |
+
"learning_rate": 3.6065573770491806e-05,
|
| 243 |
+
"loss": 0.0087,
|
| 244 |
+
"step": 310
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.52,
|
| 248 |
+
"grad_norm": 8.899871826171875,
|
| 249 |
+
"learning_rate": 3.543505674653216e-05,
|
| 250 |
+
"loss": 0.0222,
|
| 251 |
+
"step": 320
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.6,
|
| 255 |
+
"grad_norm": 0.35501202940940857,
|
| 256 |
+
"learning_rate": 3.480453972257251e-05,
|
| 257 |
+
"loss": 0.0101,
|
| 258 |
+
"step": 330
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.68,
|
| 262 |
+
"grad_norm": 0.01468953862786293,
|
| 263 |
+
"learning_rate": 3.417402269861286e-05,
|
| 264 |
+
"loss": 0.0064,
|
| 265 |
+
"step": 340
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.76,
|
| 269 |
+
"grad_norm": 2.1835994720458984,
|
| 270 |
+
"learning_rate": 3.354350567465322e-05,
|
| 271 |
+
"loss": 0.0081,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.84,
|
| 276 |
+
"grad_norm": 0.027698852121829987,
|
| 277 |
+
"learning_rate": 3.291298865069357e-05,
|
| 278 |
+
"loss": 0.0207,
|
| 279 |
+
"step": 360
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 2.92,
|
| 283 |
+
"grad_norm": 1.9383903741836548,
|
| 284 |
+
"learning_rate": 3.2282471626733923e-05,
|
| 285 |
+
"loss": 0.007,
|
| 286 |
+
"step": 370
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.0,
|
| 290 |
+
"grad_norm": 0.31910964846611023,
|
| 291 |
+
"learning_rate": 3.1651954602774275e-05,
|
| 292 |
+
"loss": 0.015,
|
| 293 |
+
"step": 380
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.0,
|
| 297 |
+
"eval_accuracy": 1.0,
|
| 298 |
+
"eval_loss": 0.00020072748884558678,
|
| 299 |
+
"eval_runtime": 32.1614,
|
| 300 |
+
"eval_samples_per_second": 55.968,
|
| 301 |
+
"eval_steps_per_second": 1.772,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.08,
|
| 306 |
+
"grad_norm": 0.042351339012384415,
|
| 307 |
+
"learning_rate": 3.102143757881463e-05,
|
| 308 |
+
"loss": 0.0072,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.16,
|
| 313 |
+
"grad_norm": 3.833139657974243,
|
| 314 |
+
"learning_rate": 3.0390920554854985e-05,
|
| 315 |
+
"loss": 0.0142,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.23,
|
| 320 |
+
"grad_norm": 9.079660415649414,
|
| 321 |
+
"learning_rate": 2.9760403530895337e-05,
|
| 322 |
+
"loss": 0.0151,
|
| 323 |
+
"step": 410
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.31,
|
| 327 |
+
"grad_norm": 1.185364007949829,
|
| 328 |
+
"learning_rate": 2.912988650693569e-05,
|
| 329 |
+
"loss": 0.0094,
|
| 330 |
+
"step": 420
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.39,
|
| 334 |
+
"grad_norm": 0.005675105378031731,
|
| 335 |
+
"learning_rate": 2.849936948297604e-05,
|
| 336 |
+
"loss": 0.0069,
|
| 337 |
+
"step": 430
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.47,
|
| 341 |
+
"grad_norm": 0.7052080035209656,
|
| 342 |
+
"learning_rate": 2.7868852459016392e-05,
|
| 343 |
+
"loss": 0.0066,
|
| 344 |
+
"step": 440
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.55,
|
| 348 |
+
"grad_norm": 0.4254097044467926,
|
| 349 |
+
"learning_rate": 2.723833543505675e-05,
|
| 350 |
+
"loss": 0.0189,
|
| 351 |
+
"step": 450
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.63,
|
| 355 |
+
"grad_norm": 0.8299044370651245,
|
| 356 |
+
"learning_rate": 2.6607818411097103e-05,
|
| 357 |
+
"loss": 0.0045,
|
| 358 |
+
"step": 460
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.71,
|
| 362 |
+
"grad_norm": 1.3662859201431274,
|
| 363 |
+
"learning_rate": 2.5977301387137454e-05,
|
| 364 |
+
"loss": 0.0128,
|
| 365 |
+
"step": 470
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 3.79,
|
| 369 |
+
"grad_norm": 0.009952099993824959,
|
| 370 |
+
"learning_rate": 2.5346784363177806e-05,
|
| 371 |
+
"loss": 0.0086,
|
| 372 |
+
"step": 480
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 3.87,
|
| 376 |
+
"grad_norm": 1.3820610046386719,
|
| 377 |
+
"learning_rate": 2.471626733921816e-05,
|
| 378 |
+
"loss": 0.0144,
|
| 379 |
+
"step": 490
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 3.94,
|
| 383 |
+
"grad_norm": 0.028887655586004257,
|
| 384 |
+
"learning_rate": 2.4085750315258513e-05,
|
| 385 |
+
"loss": 0.0048,
|
| 386 |
+
"step": 500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 4.0,
|
| 390 |
+
"eval_accuracy": 1.0,
|
| 391 |
+
"eval_loss": 5.466423317557201e-06,
|
| 392 |
+
"eval_runtime": 32.5873,
|
| 393 |
+
"eval_samples_per_second": 55.236,
|
| 394 |
+
"eval_steps_per_second": 1.749,
|
| 395 |
+
"step": 507
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.02,
|
| 399 |
+
"grad_norm": 0.012447195127606392,
|
| 400 |
+
"learning_rate": 2.3455233291298868e-05,
|
| 401 |
+
"loss": 0.0123,
|
| 402 |
+
"step": 510
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.1,
|
| 406 |
+
"grad_norm": 5.536173343658447,
|
| 407 |
+
"learning_rate": 2.282471626733922e-05,
|
| 408 |
+
"loss": 0.0118,
|
| 409 |
+
"step": 520
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.18,
|
| 413 |
+
"grad_norm": 0.02774740196764469,
|
| 414 |
+
"learning_rate": 2.2194199243379575e-05,
|
| 415 |
+
"loss": 0.0059,
|
| 416 |
+
"step": 530
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.26,
|
| 420 |
+
"grad_norm": 0.2927396297454834,
|
| 421 |
+
"learning_rate": 2.1563682219419927e-05,
|
| 422 |
+
"loss": 0.0068,
|
| 423 |
+
"step": 540
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.34,
|
| 427 |
+
"grad_norm": 0.0665479302406311,
|
| 428 |
+
"learning_rate": 2.093316519546028e-05,
|
| 429 |
+
"loss": 0.0044,
|
| 430 |
+
"step": 550
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.42,
|
| 434 |
+
"grad_norm": 0.009304400533437729,
|
| 435 |
+
"learning_rate": 2.0302648171500633e-05,
|
| 436 |
+
"loss": 0.0142,
|
| 437 |
+
"step": 560
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.5,
|
| 441 |
+
"grad_norm": 0.6083464026451111,
|
| 442 |
+
"learning_rate": 1.9672131147540985e-05,
|
| 443 |
+
"loss": 0.0148,
|
| 444 |
+
"step": 570
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.58,
|
| 448 |
+
"grad_norm": 0.3781495690345764,
|
| 449 |
+
"learning_rate": 1.904161412358134e-05,
|
| 450 |
+
"loss": 0.0065,
|
| 451 |
+
"step": 580
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 4.65,
|
| 455 |
+
"grad_norm": 0.01165520865470171,
|
| 456 |
+
"learning_rate": 1.8411097099621692e-05,
|
| 457 |
+
"loss": 0.0144,
|
| 458 |
+
"step": 590
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.73,
|
| 462 |
+
"grad_norm": 0.023280547931790352,
|
| 463 |
+
"learning_rate": 1.7780580075662044e-05,
|
| 464 |
+
"loss": 0.0038,
|
| 465 |
+
"step": 600
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.81,
|
| 469 |
+
"grad_norm": 0.9535312652587891,
|
| 470 |
+
"learning_rate": 1.71500630517024e-05,
|
| 471 |
+
"loss": 0.012,
|
| 472 |
+
"step": 610
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 4.89,
|
| 476 |
+
"grad_norm": 0.9200620055198669,
|
| 477 |
+
"learning_rate": 1.651954602774275e-05,
|
| 478 |
+
"loss": 0.0102,
|
| 479 |
+
"step": 620
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 4.97,
|
| 483 |
+
"grad_norm": 1.7359685897827148,
|
| 484 |
+
"learning_rate": 1.5889029003783106e-05,
|
| 485 |
+
"loss": 0.0133,
|
| 486 |
+
"step": 630
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 4.99,
|
| 490 |
+
"eval_accuracy": 1.0,
|
| 491 |
+
"eval_loss": 7.2136931521527e-06,
|
| 492 |
+
"eval_runtime": 32.8061,
|
| 493 |
+
"eval_samples_per_second": 54.868,
|
| 494 |
+
"eval_steps_per_second": 1.737,
|
| 495 |
+
"step": 633
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.05,
|
| 499 |
+
"grad_norm": 0.00256229005753994,
|
| 500 |
+
"learning_rate": 1.5258511979823456e-05,
|
| 501 |
+
"loss": 0.0038,
|
| 502 |
+
"step": 640
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 5.13,
|
| 506 |
+
"grad_norm": 0.02973257750272751,
|
| 507 |
+
"learning_rate": 1.4627994955863808e-05,
|
| 508 |
+
"loss": 0.0085,
|
| 509 |
+
"step": 650
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.21,
|
| 513 |
+
"grad_norm": 0.014365817420184612,
|
| 514 |
+
"learning_rate": 1.3997477931904163e-05,
|
| 515 |
+
"loss": 0.0047,
|
| 516 |
+
"step": 660
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 5.29,
|
| 520 |
+
"grad_norm": 5.261212348937988,
|
| 521 |
+
"learning_rate": 1.3366960907944514e-05,
|
| 522 |
+
"loss": 0.0024,
|
| 523 |
+
"step": 670
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 5.36,
|
| 527 |
+
"grad_norm": 1.2331159114837646,
|
| 528 |
+
"learning_rate": 1.273644388398487e-05,
|
| 529 |
+
"loss": 0.0078,
|
| 530 |
+
"step": 680
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.44,
|
| 534 |
+
"grad_norm": 0.011407382786273956,
|
| 535 |
+
"learning_rate": 1.2105926860025221e-05,
|
| 536 |
+
"loss": 0.0037,
|
| 537 |
+
"step": 690
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 5.52,
|
| 541 |
+
"grad_norm": 0.9521317481994629,
|
| 542 |
+
"learning_rate": 1.1475409836065575e-05,
|
| 543 |
+
"loss": 0.0099,
|
| 544 |
+
"step": 700
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 5.6,
|
| 548 |
+
"grad_norm": 3.931731939315796,
|
| 549 |
+
"learning_rate": 1.0844892812105928e-05,
|
| 550 |
+
"loss": 0.0046,
|
| 551 |
+
"step": 710
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 5.68,
|
| 555 |
+
"grad_norm": 6.321459770202637,
|
| 556 |
+
"learning_rate": 1.021437578814628e-05,
|
| 557 |
+
"loss": 0.006,
|
| 558 |
+
"step": 720
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 5.76,
|
| 562 |
+
"grad_norm": 0.41888654232025146,
|
| 563 |
+
"learning_rate": 9.583858764186633e-06,
|
| 564 |
+
"loss": 0.0013,
|
| 565 |
+
"step": 730
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 5.84,
|
| 569 |
+
"grad_norm": 0.03879016637802124,
|
| 570 |
+
"learning_rate": 8.953341740226987e-06,
|
| 571 |
+
"loss": 0.0064,
|
| 572 |
+
"step": 740
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 5.92,
|
| 576 |
+
"grad_norm": 0.14714492857456207,
|
| 577 |
+
"learning_rate": 8.32282471626734e-06,
|
| 578 |
+
"loss": 0.0072,
|
| 579 |
+
"step": 750
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 6.0,
|
| 583 |
+
"grad_norm": 0.005957466084510088,
|
| 584 |
+
"learning_rate": 7.692307692307694e-06,
|
| 585 |
+
"loss": 0.0032,
|
| 586 |
+
"step": 760
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 6.0,
|
| 590 |
+
"eval_accuracy": 1.0,
|
| 591 |
+
"eval_loss": 1.4342664371724823e-06,
|
| 592 |
+
"eval_runtime": 32.4522,
|
| 593 |
+
"eval_samples_per_second": 55.466,
|
| 594 |
+
"eval_steps_per_second": 1.756,
|
| 595 |
+
"step": 760
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.07,
|
| 599 |
+
"grad_norm": 0.0026133994106203318,
|
| 600 |
+
"learning_rate": 7.061790668348046e-06,
|
| 601 |
+
"loss": 0.0069,
|
| 602 |
+
"step": 770
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.15,
|
| 606 |
+
"grad_norm": 21.232322692871094,
|
| 607 |
+
"learning_rate": 6.431273644388399e-06,
|
| 608 |
+
"loss": 0.013,
|
| 609 |
+
"step": 780
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 6.23,
|
| 613 |
+
"grad_norm": 0.010045124217867851,
|
| 614 |
+
"learning_rate": 5.800756620428751e-06,
|
| 615 |
+
"loss": 0.0047,
|
| 616 |
+
"step": 790
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 6.31,
|
| 620 |
+
"grad_norm": 6.4604997634887695,
|
| 621 |
+
"learning_rate": 5.170239596469105e-06,
|
| 622 |
+
"loss": 0.0203,
|
| 623 |
+
"step": 800
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 6.39,
|
| 627 |
+
"grad_norm": 0.6475911736488342,
|
| 628 |
+
"learning_rate": 4.539722572509458e-06,
|
| 629 |
+
"loss": 0.0028,
|
| 630 |
+
"step": 810
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 6.47,
|
| 634 |
+
"grad_norm": 2.049196720123291,
|
| 635 |
+
"learning_rate": 3.909205548549811e-06,
|
| 636 |
+
"loss": 0.0046,
|
| 637 |
+
"step": 820
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 6.55,
|
| 641 |
+
"grad_norm": 0.4927035868167877,
|
| 642 |
+
"learning_rate": 3.278688524590164e-06,
|
| 643 |
+
"loss": 0.0076,
|
| 644 |
+
"step": 830
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 6.63,
|
| 648 |
+
"grad_norm": 3.101295232772827,
|
| 649 |
+
"learning_rate": 2.648171500630517e-06,
|
| 650 |
+
"loss": 0.0061,
|
| 651 |
+
"step": 840
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.71,
|
| 655 |
+
"grad_norm": 0.005894433241337538,
|
| 656 |
+
"learning_rate": 2.0176544766708702e-06,
|
| 657 |
+
"loss": 0.0062,
|
| 658 |
+
"step": 850
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 6.79,
|
| 662 |
+
"grad_norm": 2.8131444454193115,
|
| 663 |
+
"learning_rate": 1.3871374527112234e-06,
|
| 664 |
+
"loss": 0.0042,
|
| 665 |
+
"step": 860
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.86,
|
| 669 |
+
"grad_norm": 0.9489508867263794,
|
| 670 |
+
"learning_rate": 7.566204287515763e-07,
|
| 671 |
+
"loss": 0.0061,
|
| 672 |
+
"step": 870
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 6.94,
|
| 676 |
+
"grad_norm": 0.0012047379277646542,
|
| 677 |
+
"learning_rate": 1.261034047919294e-07,
|
| 678 |
+
"loss": 0.0097,
|
| 679 |
+
"step": 880
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 6.96,
|
| 683 |
+
"eval_accuracy": 1.0,
|
| 684 |
+
"eval_loss": 1.1116125051557901e-06,
|
| 685 |
+
"eval_runtime": 33.6881,
|
| 686 |
+
"eval_samples_per_second": 53.431,
|
| 687 |
+
"eval_steps_per_second": 1.692,
|
| 688 |
+
"step": 882
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 6.96,
|
| 692 |
+
"step": 882,
|
| 693 |
+
"total_flos": 2.8025139821285376e+18,
|
| 694 |
+
"train_loss": 0.13459565003810117,
|
| 695 |
+
"train_runtime": 2248.9328,
|
| 696 |
+
"train_samples_per_second": 50.411,
|
| 697 |
+
"train_steps_per_second": 0.392
|
| 698 |
+
}
|
| 699 |
+
],
|
| 700 |
+
"logging_steps": 10,
|
| 701 |
+
"max_steps": 882,
|
| 702 |
+
"num_input_tokens_seen": 0,
|
| 703 |
+
"num_train_epochs": 7,
|
| 704 |
+
"save_steps": 500,
|
| 705 |
+
"total_flos": 2.8025139821285376e+18,
|
| 706 |
+
"train_batch_size": 32,
|
| 707 |
+
"trial_name": null,
|
| 708 |
+
"trial_params": null
|
| 709 |
+
}
|