Commit
•
6bf0a96
1
Parent(s):
34019bd
End of training
Browse files- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +2731 -78
all_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 0.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.3966994296890982e+17,
|
| 4 |
+
"train_loss": 3.909515940200563,
|
| 5 |
+
"train_runtime": 39767.0605,
|
| 6 |
+
"train_samples": 15836,
|
| 7 |
+
"train_samples_per_second": 0.398,
|
| 8 |
+
"train_steps_per_second": 0.1
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples":
|
| 7 |
-
"train_samples_per_second": 0.
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.3966994296890982e+17,
|
| 4 |
+
"train_loss": 3.909515940200563,
|
| 5 |
+
"train_runtime": 39767.0605,
|
| 6 |
+
"train_samples": 15836,
|
| 7 |
+
"train_samples_per_second": 0.398,
|
| 8 |
+
"train_steps_per_second": 0.1
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -3,138 +3,2791 @@
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 4.
|
| 15 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 100
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 4.
|
| 22 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
"step": 200
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 4.
|
| 29 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 300
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
"step": 400
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
"step": 500
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
"step": 600
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 58 |
"step": 700
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate":
|
| 71 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
"step": 900
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate":
|
| 78 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
"step": 1000
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
"step": 1100
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"step": 1200
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
-
"learning_rate":
|
| 99 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
"step": 1300
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
-
"learning_rate":
|
| 106 |
-
"loss":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
"step": 1400
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss":
|
| 114 |
-
"step":
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
-
"learning_rate":
|
| 120 |
-
"loss":
|
| 121 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 1.0,
|
| 125 |
-
"step":
|
| 126 |
-
"total_flos": 2.
|
| 127 |
-
"train_loss":
|
| 128 |
-
"train_runtime":
|
| 129 |
-
"train_samples_per_second": 0.
|
| 130 |
-
"train_steps_per_second": 0.
|
| 131 |
}
|
| 132 |
],
|
| 133 |
-
"logging_steps":
|
| 134 |
-
"max_steps":
|
| 135 |
"num_input_tokens_seen": 0,
|
| 136 |
"num_train_epochs": 1,
|
| 137 |
-
"save_steps":
|
| 138 |
"stateful_callbacks": {
|
| 139 |
"TrainerControl": {
|
| 140 |
"args": {
|
|
@@ -147,8 +2800,8 @@
|
|
| 147 |
"attributes": {}
|
| 148 |
}
|
| 149 |
},
|
| 150 |
-
"total_flos": 2.
|
| 151 |
-
"train_batch_size":
|
| 152 |
"trial_name": null,
|
| 153 |
"trial_params": null
|
| 154 |
}
|
|
|
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 3959,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.002525890376357666,
|
| 13 |
+
"grad_norm": 4348443.0,
|
| 14 |
+
"learning_rate": 4.987370548118212e-05,
|
| 15 |
+
"loss": 11.2101,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005051780752715332,
|
| 20 |
+
"grad_norm": 6754872.5,
|
| 21 |
+
"learning_rate": 4.974741096236424e-05,
|
| 22 |
+
"loss": 9.3824,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.007577671129072998,
|
| 27 |
+
"grad_norm": 5965014.0,
|
| 28 |
+
"learning_rate": 4.962111644354635e-05,
|
| 29 |
+
"loss": 7.8958,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.010103561505430665,
|
| 34 |
+
"grad_norm": 2165972.25,
|
| 35 |
+
"learning_rate": 4.949482192472847e-05,
|
| 36 |
+
"loss": 7.421,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01262945188178833,
|
| 41 |
+
"grad_norm": 3444951.25,
|
| 42 |
+
"learning_rate": 4.936852740591059e-05,
|
| 43 |
+
"loss": 7.2076,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.015155342258145996,
|
| 48 |
+
"grad_norm": 1671924.125,
|
| 49 |
+
"learning_rate": 4.92422328870927e-05,
|
| 50 |
+
"loss": 7.121,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.017681232634503663,
|
| 55 |
+
"grad_norm": 1477403.75,
|
| 56 |
+
"learning_rate": 4.911593836827482e-05,
|
| 57 |
+
"loss": 6.8903,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02020712301086133,
|
| 62 |
+
"grad_norm": 1541815.25,
|
| 63 |
+
"learning_rate": 4.898964384945693e-05,
|
| 64 |
+
"loss": 6.7837,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.022733013387218996,
|
| 69 |
+
"grad_norm": 9298988.0,
|
| 70 |
+
"learning_rate": 4.886334933063905e-05,
|
| 71 |
+
"loss": 6.9096,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02525890376357666,
|
| 76 |
+
"grad_norm": 1347703.0,
|
| 77 |
+
"learning_rate": 4.873705481182117e-05,
|
| 78 |
+
"loss": 6.4829,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.027784794139934326,
|
| 83 |
+
"grad_norm": 744675.0625,
|
| 84 |
+
"learning_rate": 4.861076029300329e-05,
|
| 85 |
+
"loss": 6.292,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.030310684516291993,
|
| 90 |
+
"grad_norm": 1124820.25,
|
| 91 |
+
"learning_rate": 4.84844657741854e-05,
|
| 92 |
+
"loss": 5.9801,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.032836574892649656,
|
| 97 |
+
"grad_norm": 926771.5,
|
| 98 |
+
"learning_rate": 4.835817125536752e-05,
|
| 99 |
+
"loss": 6.5963,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.035362465269007326,
|
| 104 |
+
"grad_norm": 929847.5,
|
| 105 |
+
"learning_rate": 4.823187673654964e-05,
|
| 106 |
+
"loss": 6.235,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03788835564536499,
|
| 111 |
+
"grad_norm": 2526786.75,
|
| 112 |
+
"learning_rate": 4.810558221773176e-05,
|
| 113 |
+
"loss": 5.9566,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04041424602172266,
|
| 118 |
+
"grad_norm": 914504.5,
|
| 119 |
+
"learning_rate": 4.797928769891387e-05,
|
| 120 |
+
"loss": 5.9669,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04294013639808032,
|
| 125 |
+
"grad_norm": 795588.8125,
|
| 126 |
+
"learning_rate": 4.785299318009598e-05,
|
| 127 |
+
"loss": 5.7371,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04546602677443799,
|
| 132 |
+
"grad_norm": 954969.25,
|
| 133 |
+
"learning_rate": 4.77266986612781e-05,
|
| 134 |
+
"loss": 5.5412,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.047991917150795656,
|
| 139 |
+
"grad_norm": 964213.375,
|
| 140 |
+
"learning_rate": 4.760040414246022e-05,
|
| 141 |
+
"loss": 5.5661,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05051780752715332,
|
| 146 |
+
"grad_norm": 1348911.125,
|
| 147 |
+
"learning_rate": 4.747410962364234e-05,
|
| 148 |
+
"loss": 5.7872,
|
| 149 |
"step": 200
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 0.05304369790351099,
|
| 153 |
+
"grad_norm": 793627.5,
|
| 154 |
+
"learning_rate": 4.7347815104824456e-05,
|
| 155 |
+
"loss": 5.6837,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05556958827986865,
|
| 160 |
+
"grad_norm": 762008.625,
|
| 161 |
+
"learning_rate": 4.722152058600657e-05,
|
| 162 |
+
"loss": 5.3987,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05809547865622632,
|
| 167 |
+
"grad_norm": 724261.0625,
|
| 168 |
+
"learning_rate": 4.709522606718869e-05,
|
| 169 |
+
"loss": 5.4719,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.060621369032583985,
|
| 174 |
+
"grad_norm": 1137774.75,
|
| 175 |
+
"learning_rate": 4.6968931548370805e-05,
|
| 176 |
+
"loss": 5.4004,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06314725940894166,
|
| 181 |
+
"grad_norm": 752281.3125,
|
| 182 |
+
"learning_rate": 4.6842637029552924e-05,
|
| 183 |
+
"loss": 5.3922,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06567314978529931,
|
| 188 |
+
"grad_norm": 558181.375,
|
| 189 |
+
"learning_rate": 4.6716342510735036e-05,
|
| 190 |
+
"loss": 5.2856,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06819904016165698,
|
| 195 |
+
"grad_norm": 927075.5,
|
| 196 |
+
"learning_rate": 4.6590047991917155e-05,
|
| 197 |
+
"loss": 5.1218,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07072493053801465,
|
| 202 |
+
"grad_norm": 1678542.0,
|
| 203 |
+
"learning_rate": 4.646375347309927e-05,
|
| 204 |
+
"loss": 5.0974,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07325082091437232,
|
| 209 |
+
"grad_norm": 1064729.5,
|
| 210 |
+
"learning_rate": 4.6337458954281386e-05,
|
| 211 |
+
"loss": 5.2749,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07577671129072998,
|
| 216 |
+
"grad_norm": 1193670.375,
|
| 217 |
+
"learning_rate": 4.6211164435463505e-05,
|
| 218 |
+
"loss": 5.1134,
|
| 219 |
"step": 300
|
| 220 |
},
|
| 221 |
{
|
| 222 |
+
"epoch": 0.07830260166708765,
|
| 223 |
+
"grad_norm": 979860.8125,
|
| 224 |
+
"learning_rate": 4.6084869916645617e-05,
|
| 225 |
+
"loss": 4.9763,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08082849204344532,
|
| 230 |
+
"grad_norm": 868298.8125,
|
| 231 |
+
"learning_rate": 4.5958575397827735e-05,
|
| 232 |
+
"loss": 5.0413,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08335438241980297,
|
| 237 |
+
"grad_norm": 1145414.625,
|
| 238 |
+
"learning_rate": 4.5832280879009854e-05,
|
| 239 |
+
"loss": 4.8234,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08588027279616064,
|
| 244 |
+
"grad_norm": 1516276.375,
|
| 245 |
+
"learning_rate": 4.570598636019197e-05,
|
| 246 |
+
"loss": 4.7487,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08840616317251832,
|
| 251 |
+
"grad_norm": 1310860.25,
|
| 252 |
+
"learning_rate": 4.557969184137409e-05,
|
| 253 |
+
"loss": 4.869,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09093205354887599,
|
| 258 |
+
"grad_norm": 961273.9375,
|
| 259 |
+
"learning_rate": 4.5453397322556204e-05,
|
| 260 |
+
"loss": 5.2202,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09345794392523364,
|
| 265 |
+
"grad_norm": 888523.3125,
|
| 266 |
+
"learning_rate": 4.532710280373832e-05,
|
| 267 |
+
"loss": 5.1888,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09598383430159131,
|
| 272 |
+
"grad_norm": 1027426.9375,
|
| 273 |
+
"learning_rate": 4.5200808284920434e-05,
|
| 274 |
+
"loss": 4.8891,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.09850972467794898,
|
| 279 |
+
"grad_norm": 1760436.75,
|
| 280 |
+
"learning_rate": 4.507451376610255e-05,
|
| 281 |
+
"loss": 4.7838,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10103561505430664,
|
| 286 |
+
"grad_norm": 835284.875,
|
| 287 |
+
"learning_rate": 4.4948219247284665e-05,
|
| 288 |
+
"loss": 4.9398,
|
| 289 |
"step": 400
|
| 290 |
},
|
| 291 |
{
|
| 292 |
+
"epoch": 0.10356150543066431,
|
| 293 |
+
"grad_norm": 1165127.125,
|
| 294 |
+
"learning_rate": 4.4821924728466784e-05,
|
| 295 |
+
"loss": 5.0785,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.10608739580702198,
|
| 300 |
+
"grad_norm": 1061038.75,
|
| 301 |
+
"learning_rate": 4.46956302096489e-05,
|
| 302 |
+
"loss": 4.9144,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.10861328618337965,
|
| 307 |
+
"grad_norm": 1058956.875,
|
| 308 |
+
"learning_rate": 4.456933569083102e-05,
|
| 309 |
+
"loss": 4.7658,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1111391765597373,
|
| 314 |
+
"grad_norm": 838523.1875,
|
| 315 |
+
"learning_rate": 4.444304117201314e-05,
|
| 316 |
+
"loss": 4.9675,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11366506693609497,
|
| 321 |
+
"grad_norm": 1297820.5,
|
| 322 |
+
"learning_rate": 4.431674665319525e-05,
|
| 323 |
+
"loss": 4.8902,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11619095731245264,
|
| 328 |
+
"grad_norm": 841045.25,
|
| 329 |
+
"learning_rate": 4.419045213437737e-05,
|
| 330 |
+
"loss": 4.7037,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1187168476888103,
|
| 335 |
+
"grad_norm": 1158120.75,
|
| 336 |
+
"learning_rate": 4.406415761555949e-05,
|
| 337 |
+
"loss": 4.7918,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12124273806516797,
|
| 342 |
+
"grad_norm": 646241.4375,
|
| 343 |
+
"learning_rate": 4.39378630967416e-05,
|
| 344 |
+
"loss": 4.7433,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12376862844152564,
|
| 349 |
+
"grad_norm": 641319.3125,
|
| 350 |
+
"learning_rate": 4.381156857792372e-05,
|
| 351 |
+
"loss": 4.608,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1262945188178833,
|
| 356 |
+
"grad_norm": 1063831.625,
|
| 357 |
+
"learning_rate": 4.368527405910583e-05,
|
| 358 |
+
"loss": 4.8131,
|
| 359 |
"step": 500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
+
"epoch": 0.12882040919424098,
|
| 363 |
+
"grad_norm": 1552814.75,
|
| 364 |
+
"learning_rate": 4.355897954028795e-05,
|
| 365 |
+
"loss": 4.6981,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13134629957059862,
|
| 370 |
+
"grad_norm": 915990.25,
|
| 371 |
+
"learning_rate": 4.343268502147007e-05,
|
| 372 |
+
"loss": 4.8095,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1338721899469563,
|
| 377 |
+
"grad_norm": 838948.625,
|
| 378 |
+
"learning_rate": 4.330639050265219e-05,
|
| 379 |
+
"loss": 4.8535,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.13639808032331396,
|
| 384 |
+
"grad_norm": 697365.9375,
|
| 385 |
+
"learning_rate": 4.31800959838343e-05,
|
| 386 |
+
"loss": 5.0165,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13892397069967163,
|
| 391 |
+
"grad_norm": 853774.25,
|
| 392 |
+
"learning_rate": 4.305380146501642e-05,
|
| 393 |
+
"loss": 4.6141,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1414498610760293,
|
| 398 |
+
"grad_norm": 714720.875,
|
| 399 |
+
"learning_rate": 4.292750694619854e-05,
|
| 400 |
+
"loss": 4.6416,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14397575145238697,
|
| 405 |
+
"grad_norm": 953777.125,
|
| 406 |
+
"learning_rate": 4.280121242738066e-05,
|
| 407 |
+
"loss": 4.8611,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.14650164182874464,
|
| 412 |
+
"grad_norm": 1033054.5,
|
| 413 |
+
"learning_rate": 4.2674917908562776e-05,
|
| 414 |
+
"loss": 4.2724,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1490275322051023,
|
| 419 |
+
"grad_norm": 877196.1875,
|
| 420 |
+
"learning_rate": 4.254862338974489e-05,
|
| 421 |
+
"loss": 4.7521,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15155342258145996,
|
| 426 |
+
"grad_norm": 1127457.0,
|
| 427 |
+
"learning_rate": 4.2422328870927e-05,
|
| 428 |
+
"loss": 4.3,
|
| 429 |
"step": 600
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 0.15407931295781763,
|
| 433 |
+
"grad_norm": 627419.4375,
|
| 434 |
+
"learning_rate": 4.229603435210912e-05,
|
| 435 |
+
"loss": 4.5331,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.1566052033341753,
|
| 440 |
+
"grad_norm": 691871.3125,
|
| 441 |
+
"learning_rate": 4.216973983329124e-05,
|
| 442 |
+
"loss": 4.5695,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.15913109371053297,
|
| 447 |
+
"grad_norm": 637530.0625,
|
| 448 |
+
"learning_rate": 4.2043445314473356e-05,
|
| 449 |
+
"loss": 4.4994,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16165698408689064,
|
| 454 |
+
"grad_norm": 692771.5625,
|
| 455 |
+
"learning_rate": 4.191715079565547e-05,
|
| 456 |
+
"loss": 4.403,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1641828744632483,
|
| 461 |
+
"grad_norm": 738835.1875,
|
| 462 |
+
"learning_rate": 4.179085627683759e-05,
|
| 463 |
+
"loss": 4.5743,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.16670876483960595,
|
| 468 |
+
"grad_norm": 1637430.25,
|
| 469 |
+
"learning_rate": 4.1664561758019706e-05,
|
| 470 |
+
"loss": 4.5377,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.16923465521596362,
|
| 475 |
+
"grad_norm": 760962.875,
|
| 476 |
+
"learning_rate": 4.1538267239201825e-05,
|
| 477 |
+
"loss": 4.5168,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.1717605455923213,
|
| 482 |
+
"grad_norm": 1142129.25,
|
| 483 |
+
"learning_rate": 4.1411972720383937e-05,
|
| 484 |
+
"loss": 4.2554,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17428643596867896,
|
| 489 |
+
"grad_norm": 614473.6875,
|
| 490 |
+
"learning_rate": 4.1285678201566055e-05,
|
| 491 |
+
"loss": 4.8585,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.17681232634503663,
|
| 496 |
+
"grad_norm": 1348435.875,
|
| 497 |
+
"learning_rate": 4.115938368274817e-05,
|
| 498 |
+
"loss": 4.4252,
|
| 499 |
"step": 700
|
| 500 |
},
|
| 501 |
{
|
| 502 |
+
"epoch": 0.1793382167213943,
|
| 503 |
+
"grad_norm": 1183288.25,
|
| 504 |
+
"learning_rate": 4.1033089163930286e-05,
|
| 505 |
+
"loss": 4.4951,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18186410709775197,
|
| 510 |
+
"grad_norm": 579328.4375,
|
| 511 |
+
"learning_rate": 4.0906794645112405e-05,
|
| 512 |
+
"loss": 4.322,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.1843899974741096,
|
| 517 |
+
"grad_norm": 1145040.25,
|
| 518 |
+
"learning_rate": 4.078050012629452e-05,
|
| 519 |
+
"loss": 4.1513,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.18691588785046728,
|
| 524 |
+
"grad_norm": 1142186.875,
|
| 525 |
+
"learning_rate": 4.0654205607476636e-05,
|
| 526 |
+
"loss": 4.1724,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.18944177822682495,
|
| 531 |
+
"grad_norm": 1073306.125,
|
| 532 |
+
"learning_rate": 4.0527911088658754e-05,
|
| 533 |
+
"loss": 4.2603,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19196766860318262,
|
| 538 |
+
"grad_norm": 856185.875,
|
| 539 |
+
"learning_rate": 4.040161656984087e-05,
|
| 540 |
+
"loss": 4.1743,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.1944935589795403,
|
| 545 |
+
"grad_norm": 1259884.75,
|
| 546 |
+
"learning_rate": 4.027532205102299e-05,
|
| 547 |
+
"loss": 4.4978,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.19701944935589796,
|
| 552 |
+
"grad_norm": 988211.375,
|
| 553 |
+
"learning_rate": 4.0149027532205104e-05,
|
| 554 |
+
"loss": 4.4756,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.19954533973225563,
|
| 559 |
+
"grad_norm": 1129890.375,
|
| 560 |
+
"learning_rate": 4.002273301338722e-05,
|
| 561 |
+
"loss": 4.4766,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.20207123010861328,
|
| 566 |
+
"grad_norm": 1127913.0,
|
| 567 |
+
"learning_rate": 3.989643849456934e-05,
|
| 568 |
+
"loss": 4.1696,
|
| 569 |
"step": 800
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 0.20459712048497095,
|
| 573 |
+
"grad_norm": 643195.0625,
|
| 574 |
+
"learning_rate": 3.9770143975751453e-05,
|
| 575 |
+
"loss": 4.2825,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.20712301086132862,
|
| 580 |
+
"grad_norm": 585030.1875,
|
| 581 |
+
"learning_rate": 3.9643849456933565e-05,
|
| 582 |
+
"loss": 4.5449,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.20964890123768629,
|
| 587 |
+
"grad_norm": 964589.9375,
|
| 588 |
+
"learning_rate": 3.9517554938115684e-05,
|
| 589 |
+
"loss": 4.2085,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21217479161404396,
|
| 594 |
+
"grad_norm": 1307139.375,
|
| 595 |
+
"learning_rate": 3.93912604192978e-05,
|
| 596 |
+
"loss": 4.2636,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.21470068199040163,
|
| 601 |
+
"grad_norm": 1642451.125,
|
| 602 |
+
"learning_rate": 3.926496590047992e-05,
|
| 603 |
+
"loss": 4.1194,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2172265723667593,
|
| 608 |
+
"grad_norm": 767172.5625,
|
| 609 |
+
"learning_rate": 3.913867138166204e-05,
|
| 610 |
+
"loss": 3.9216,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.21975246274311694,
|
| 615 |
+
"grad_norm": 1053997.0,
|
| 616 |
+
"learning_rate": 3.901237686284415e-05,
|
| 617 |
+
"loss": 4.4344,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2222783531194746,
|
| 622 |
+
"grad_norm": 1236466.5,
|
| 623 |
+
"learning_rate": 3.888608234402627e-05,
|
| 624 |
+
"loss": 4.5718,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.22480424349583228,
|
| 629 |
+
"grad_norm": 1578047.5,
|
| 630 |
+
"learning_rate": 3.875978782520839e-05,
|
| 631 |
+
"loss": 4.2614,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.22733013387218995,
|
| 636 |
+
"grad_norm": 1341923.75,
|
| 637 |
+
"learning_rate": 3.863349330639051e-05,
|
| 638 |
+
"loss": 4.2356,
|
| 639 |
"step": 900
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 0.22985602424854762,
|
| 643 |
+
"grad_norm": 1259278.75,
|
| 644 |
+
"learning_rate": 3.850719878757262e-05,
|
| 645 |
+
"loss": 4.1374,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2323819146249053,
|
| 650 |
+
"grad_norm": 1208085.125,
|
| 651 |
+
"learning_rate": 3.838090426875473e-05,
|
| 652 |
+
"loss": 4.3741,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.23490780500126293,
|
| 657 |
+
"grad_norm": 713612.375,
|
| 658 |
+
"learning_rate": 3.825460974993685e-05,
|
| 659 |
+
"loss": 4.2538,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2374336953776206,
|
| 664 |
+
"grad_norm": 773853.25,
|
| 665 |
+
"learning_rate": 3.812831523111897e-05,
|
| 666 |
+
"loss": 4.0877,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.23995958575397827,
|
| 671 |
+
"grad_norm": 1925775.125,
|
| 672 |
+
"learning_rate": 3.800202071230109e-05,
|
| 673 |
+
"loss": 3.8277,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.24248547613033594,
|
| 678 |
+
"grad_norm": 814602.125,
|
| 679 |
+
"learning_rate": 3.78757261934832e-05,
|
| 680 |
+
"loss": 4.016,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2450113665066936,
|
| 685 |
+
"grad_norm": 1109952.0,
|
| 686 |
+
"learning_rate": 3.774943167466532e-05,
|
| 687 |
+
"loss": 3.9932,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.24753725688305128,
|
| 692 |
+
"grad_norm": 744110.125,
|
| 693 |
+
"learning_rate": 3.762313715584744e-05,
|
| 694 |
+
"loss": 4.0229,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2500631472594089,
|
| 699 |
+
"grad_norm": 904255.75,
|
| 700 |
+
"learning_rate": 3.749684263702956e-05,
|
| 701 |
+
"loss": 4.4272,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2525890376357666,
|
| 706 |
+
"grad_norm": 829616.3125,
|
| 707 |
+
"learning_rate": 3.7370548118211676e-05,
|
| 708 |
+
"loss": 4.3034,
|
| 709 |
"step": 1000
|
| 710 |
},
|
| 711 |
{
|
| 712 |
+
"epoch": 0.25511492801212426,
|
| 713 |
+
"grad_norm": 1236616.375,
|
| 714 |
+
"learning_rate": 3.724425359939379e-05,
|
| 715 |
+
"loss": 3.9168,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.25764081838848196,
|
| 720 |
+
"grad_norm": 696752.75,
|
| 721 |
+
"learning_rate": 3.711795908057591e-05,
|
| 722 |
+
"loss": 4.2012,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2601667087648396,
|
| 727 |
+
"grad_norm": 680934.875,
|
| 728 |
+
"learning_rate": 3.699166456175802e-05,
|
| 729 |
+
"loss": 4.2596,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.26269259914119725,
|
| 734 |
+
"grad_norm": 1334373.125,
|
| 735 |
+
"learning_rate": 3.686537004294014e-05,
|
| 736 |
+
"loss": 4.461,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.26521848951755495,
|
| 741 |
+
"grad_norm": 1002765.875,
|
| 742 |
+
"learning_rate": 3.6739075524122257e-05,
|
| 743 |
+
"loss": 4.2859,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.2677443798939126,
|
| 748 |
+
"grad_norm": 507567.5,
|
| 749 |
+
"learning_rate": 3.661278100530437e-05,
|
| 750 |
+
"loss": 3.8771,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2702702702702703,
|
| 755 |
+
"grad_norm": 968382.5,
|
| 756 |
+
"learning_rate": 3.648648648648649e-05,
|
| 757 |
+
"loss": 4.1681,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.2727961606466279,
|
| 762 |
+
"grad_norm": 1306089.125,
|
| 763 |
+
"learning_rate": 3.6360191967668606e-05,
|
| 764 |
+
"loss": 4.2425,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.2753220510229856,
|
| 769 |
+
"grad_norm": 1039757.5,
|
| 770 |
+
"learning_rate": 3.6233897448850725e-05,
|
| 771 |
+
"loss": 4.3264,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.27784794139934327,
|
| 776 |
+
"grad_norm": 1464928.5,
|
| 777 |
+
"learning_rate": 3.610760293003284e-05,
|
| 778 |
+
"loss": 4.2827,
|
| 779 |
"step": 1100
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 0.2803738317757009,
|
| 783 |
+
"grad_norm": 824237.6875,
|
| 784 |
+
"learning_rate": 3.5981308411214956e-05,
|
| 785 |
+
"loss": 3.8312,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2828997221520586,
|
| 790 |
+
"grad_norm": 942513.125,
|
| 791 |
+
"learning_rate": 3.5855013892397074e-05,
|
| 792 |
+
"loss": 4.1363,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.28542561252841625,
|
| 797 |
+
"grad_norm": 697307.1875,
|
| 798 |
+
"learning_rate": 3.5728719373579186e-05,
|
| 799 |
+
"loss": 3.9892,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.28795150290477395,
|
| 804 |
+
"grad_norm": 1249596.5,
|
| 805 |
+
"learning_rate": 3.5602424854761305e-05,
|
| 806 |
+
"loss": 4.0763,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2904773932811316,
|
| 811 |
+
"grad_norm": 963218.0625,
|
| 812 |
+
"learning_rate": 3.547613033594342e-05,
|
| 813 |
+
"loss": 4.1946,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.2930032836574893,
|
| 818 |
+
"grad_norm": 1025589.625,
|
| 819 |
+
"learning_rate": 3.5349835817125536e-05,
|
| 820 |
+
"loss": 4.1447,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.29552917403384693,
|
| 825 |
+
"grad_norm": 696336.3125,
|
| 826 |
+
"learning_rate": 3.5223541298307655e-05,
|
| 827 |
+
"loss": 4.1982,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.2980550644102046,
|
| 832 |
+
"grad_norm": 826398.875,
|
| 833 |
+
"learning_rate": 3.5097246779489773e-05,
|
| 834 |
+
"loss": 4.1935,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.30058095478656227,
|
| 839 |
+
"grad_norm": 527454.375,
|
| 840 |
+
"learning_rate": 3.497095226067189e-05,
|
| 841 |
+
"loss": 3.8515,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3031068451629199,
|
| 846 |
+
"grad_norm": 1154428.125,
|
| 847 |
+
"learning_rate": 3.4844657741854004e-05,
|
| 848 |
+
"loss": 4.123,
|
| 849 |
"step": 1200
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 0.3056327355392776,
|
| 853 |
+
"grad_norm": 843105.25,
|
| 854 |
+
"learning_rate": 3.471836322303612e-05,
|
| 855 |
+
"loss": 4.0649,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.30815862591563525,
|
| 860 |
+
"grad_norm": 677728.0,
|
| 861 |
+
"learning_rate": 3.459206870421824e-05,
|
| 862 |
+
"loss": 4.1858,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.31068451629199295,
|
| 867 |
+
"grad_norm": 843633.8125,
|
| 868 |
+
"learning_rate": 3.4465774185400354e-05,
|
| 869 |
+
"loss": 3.9904,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3132104066683506,
|
| 874 |
+
"grad_norm": 1079652.625,
|
| 875 |
+
"learning_rate": 3.433947966658247e-05,
|
| 876 |
+
"loss": 4.1296,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.31573629704470824,
|
| 881 |
+
"grad_norm": 1375458.875,
|
| 882 |
+
"learning_rate": 3.4213185147764585e-05,
|
| 883 |
+
"loss": 3.845,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.31826218742106593,
|
| 888 |
+
"grad_norm": 1572799.75,
|
| 889 |
+
"learning_rate": 3.40868906289467e-05,
|
| 890 |
+
"loss": 3.8965,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3207880777974236,
|
| 895 |
+
"grad_norm": 626954.0625,
|
| 896 |
+
"learning_rate": 3.396059611012882e-05,
|
| 897 |
+
"loss": 4.2417,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3233139681737813,
|
| 902 |
+
"grad_norm": 778080.8125,
|
| 903 |
+
"learning_rate": 3.383430159131094e-05,
|
| 904 |
+
"loss": 4.0585,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3258398585501389,
|
| 909 |
+
"grad_norm": 1268475.375,
|
| 910 |
+
"learning_rate": 3.370800707249305e-05,
|
| 911 |
+
"loss": 3.5957,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3283657489264966,
|
| 916 |
+
"grad_norm": 983964.5,
|
| 917 |
+
"learning_rate": 3.358171255367517e-05,
|
| 918 |
+
"loss": 4.4113,
|
| 919 |
"step": 1300
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 0.33089163930285426,
|
| 923 |
+
"grad_norm": 1136581.5,
|
| 924 |
+
"learning_rate": 3.345541803485729e-05,
|
| 925 |
+
"loss": 4.2545,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3334175296792119,
|
| 930 |
+
"grad_norm": 1313611.375,
|
| 931 |
+
"learning_rate": 3.332912351603941e-05,
|
| 932 |
+
"loss": 4.0466,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3359434200555696,
|
| 937 |
+
"grad_norm": 595656.0625,
|
| 938 |
+
"learning_rate": 3.320282899722152e-05,
|
| 939 |
+
"loss": 4.1121,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.33846931043192724,
|
| 944 |
+
"grad_norm": 1134474.625,
|
| 945 |
+
"learning_rate": 3.307653447840364e-05,
|
| 946 |
+
"loss": 3.8602,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.34099520080828494,
|
| 951 |
+
"grad_norm": 932185.5,
|
| 952 |
+
"learning_rate": 3.295023995958575e-05,
|
| 953 |
+
"loss": 4.0099,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3435210911846426,
|
| 958 |
+
"grad_norm": 992654.875,
|
| 959 |
+
"learning_rate": 3.282394544076787e-05,
|
| 960 |
+
"loss": 3.8907,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3460469815610003,
|
| 965 |
+
"grad_norm": 798724.1875,
|
| 966 |
+
"learning_rate": 3.269765092194999e-05,
|
| 967 |
+
"loss": 3.4421,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.3485728719373579,
|
| 972 |
+
"grad_norm": 1000527.125,
|
| 973 |
+
"learning_rate": 3.257135640313211e-05,
|
| 974 |
+
"loss": 4.0111,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.35109876231371556,
|
| 979 |
+
"grad_norm": 1207395.875,
|
| 980 |
+
"learning_rate": 3.244506188431422e-05,
|
| 981 |
+
"loss": 4.332,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.35362465269007326,
|
| 986 |
+
"grad_norm": 736049.8125,
|
| 987 |
+
"learning_rate": 3.231876736549634e-05,
|
| 988 |
+
"loss": 4.0475,
|
| 989 |
"step": 1400
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 0.3561505430664309,
|
| 993 |
+
"grad_norm": 902288.75,
|
| 994 |
+
"learning_rate": 3.219247284667846e-05,
|
| 995 |
+
"loss": 3.7045,
|
| 996 |
+
"step": 1410
|
| 997 |
},
|
| 998 |
{
|
| 999 |
+
"epoch": 0.3586764334427886,
|
| 1000 |
+
"grad_norm": 1179507.375,
|
| 1001 |
+
"learning_rate": 3.2066178327860577e-05,
|
| 1002 |
+
"loss": 3.7423,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.36120232381914624,
|
| 1007 |
+
"grad_norm": 1319296.25,
|
| 1008 |
+
"learning_rate": 3.193988380904269e-05,
|
| 1009 |
+
"loss": 3.7602,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.36372821419550394,
|
| 1014 |
+
"grad_norm": 1040202.3125,
|
| 1015 |
+
"learning_rate": 3.181358929022481e-05,
|
| 1016 |
+
"loss": 3.871,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3662541045718616,
|
| 1021 |
+
"grad_norm": 892262.9375,
|
| 1022 |
+
"learning_rate": 3.168729477140692e-05,
|
| 1023 |
+
"loss": 4.1026,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3687799949482192,
|
| 1028 |
+
"grad_norm": 1401292.5,
|
| 1029 |
+
"learning_rate": 3.156100025258904e-05,
|
| 1030 |
+
"loss": 3.9609,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.3713058853245769,
|
| 1035 |
+
"grad_norm": 1242199.5,
|
| 1036 |
+
"learning_rate": 3.143470573377116e-05,
|
| 1037 |
+
"loss": 3.9826,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.37383177570093457,
|
| 1042 |
+
"grad_norm": 953926.4375,
|
| 1043 |
+
"learning_rate": 3.130841121495327e-05,
|
| 1044 |
+
"loss": 3.7857,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.37635766607729226,
|
| 1049 |
+
"grad_norm": 1005284.125,
|
| 1050 |
+
"learning_rate": 3.118211669613539e-05,
|
| 1051 |
+
"loss": 3.8462,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.3788835564536499,
|
| 1056 |
+
"grad_norm": 617377.0,
|
| 1057 |
+
"learning_rate": 3.1055822177317506e-05,
|
| 1058 |
+
"loss": 4.0298,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.3814094468300076,
|
| 1063 |
+
"grad_norm": 1349709.0,
|
| 1064 |
+
"learning_rate": 3.0929527658499625e-05,
|
| 1065 |
+
"loss": 4.108,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.38393533720636525,
|
| 1070 |
+
"grad_norm": 863097.375,
|
| 1071 |
+
"learning_rate": 3.0803233139681744e-05,
|
| 1072 |
+
"loss": 3.7545,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.3864612275827229,
|
| 1077 |
+
"grad_norm": 1074415.75,
|
| 1078 |
+
"learning_rate": 3.0676938620863856e-05,
|
| 1079 |
+
"loss": 3.7492,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.3889871179590806,
|
| 1084 |
+
"grad_norm": 1269458.125,
|
| 1085 |
+
"learning_rate": 3.0550644102045975e-05,
|
| 1086 |
+
"loss": 4.0702,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.39151300833543823,
|
| 1091 |
+
"grad_norm": 623983.6875,
|
| 1092 |
+
"learning_rate": 3.042434958322809e-05,
|
| 1093 |
+
"loss": 3.7522,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.3940388987117959,
|
| 1098 |
+
"grad_norm": 847152.8125,
|
| 1099 |
+
"learning_rate": 3.029805506441021e-05,
|
| 1100 |
+
"loss": 3.9036,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.39656478908815357,
|
| 1105 |
+
"grad_norm": 1389634.5,
|
| 1106 |
+
"learning_rate": 3.017176054559232e-05,
|
| 1107 |
+
"loss": 3.7832,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.39909067946451127,
|
| 1112 |
+
"grad_norm": 983121.3125,
|
| 1113 |
+
"learning_rate": 3.004546602677444e-05,
|
| 1114 |
+
"loss": 3.6402,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4016165698408689,
|
| 1119 |
+
"grad_norm": 1020898.8125,
|
| 1120 |
+
"learning_rate": 2.9919171507956555e-05,
|
| 1121 |
+
"loss": 3.6429,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.40414246021722655,
|
| 1126 |
+
"grad_norm": 847203.9375,
|
| 1127 |
+
"learning_rate": 2.9792876989138674e-05,
|
| 1128 |
+
"loss": 3.8984,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.40666835059358425,
|
| 1133 |
+
"grad_norm": 1133185.5,
|
| 1134 |
+
"learning_rate": 2.9666582470320793e-05,
|
| 1135 |
+
"loss": 3.7906,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4091942409699419,
|
| 1140 |
+
"grad_norm": 1127204.625,
|
| 1141 |
+
"learning_rate": 2.9540287951502905e-05,
|
| 1142 |
+
"loss": 4.143,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4117201313462996,
|
| 1147 |
+
"grad_norm": 1315488.625,
|
| 1148 |
+
"learning_rate": 2.9413993432685023e-05,
|
| 1149 |
+
"loss": 3.8023,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.41424602172265723,
|
| 1154 |
+
"grad_norm": 776120.5625,
|
| 1155 |
+
"learning_rate": 2.928769891386714e-05,
|
| 1156 |
+
"loss": 3.4144,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.41677191209901493,
|
| 1161 |
+
"grad_norm": 832920.9375,
|
| 1162 |
+
"learning_rate": 2.9161404395049257e-05,
|
| 1163 |
+
"loss": 3.6427,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.41929780247537257,
|
| 1168 |
+
"grad_norm": 1192992.25,
|
| 1169 |
+
"learning_rate": 2.9035109876231376e-05,
|
| 1170 |
+
"loss": 3.9277,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4218236928517302,
|
| 1175 |
+
"grad_norm": 918210.4375,
|
| 1176 |
+
"learning_rate": 2.8908815357413488e-05,
|
| 1177 |
+
"loss": 3.6729,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4243495832280879,
|
| 1182 |
+
"grad_norm": 924215.25,
|
| 1183 |
+
"learning_rate": 2.8782520838595607e-05,
|
| 1184 |
+
"loss": 3.7461,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.42687547360444555,
|
| 1189 |
+
"grad_norm": 715260.8125,
|
| 1190 |
+
"learning_rate": 2.8656226319777722e-05,
|
| 1191 |
+
"loss": 3.7783,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.42940136398080325,
|
| 1196 |
+
"grad_norm": 1466543.5,
|
| 1197 |
+
"learning_rate": 2.852993180095984e-05,
|
| 1198 |
+
"loss": 3.8766,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4319272543571609,
|
| 1203 |
+
"grad_norm": 669040.375,
|
| 1204 |
+
"learning_rate": 2.8403637282141953e-05,
|
| 1205 |
+
"loss": 3.9012,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4344531447335186,
|
| 1210 |
+
"grad_norm": 2193973.75,
|
| 1211 |
+
"learning_rate": 2.8277342763324072e-05,
|
| 1212 |
+
"loss": 3.3848,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.43697903510987623,
|
| 1217 |
+
"grad_norm": 831080.625,
|
| 1218 |
+
"learning_rate": 2.815104824450619e-05,
|
| 1219 |
+
"loss": 3.6074,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.4395049254862339,
|
| 1224 |
+
"grad_norm": 960923.75,
|
| 1225 |
+
"learning_rate": 2.8024753725688306e-05,
|
| 1226 |
+
"loss": 3.6234,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4420308158625916,
|
| 1231 |
+
"grad_norm": 1489447.5,
|
| 1232 |
+
"learning_rate": 2.7898459206870425e-05,
|
| 1233 |
+
"loss": 3.6699,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.4445567062389492,
|
| 1238 |
+
"grad_norm": 611681.375,
|
| 1239 |
+
"learning_rate": 2.7772164688052537e-05,
|
| 1240 |
+
"loss": 3.389,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4470825966153069,
|
| 1245 |
+
"grad_norm": 2048307.875,
|
| 1246 |
+
"learning_rate": 2.7645870169234656e-05,
|
| 1247 |
+
"loss": 3.8727,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.44960848699166456,
|
| 1252 |
+
"grad_norm": 855010.75,
|
| 1253 |
+
"learning_rate": 2.7519575650416774e-05,
|
| 1254 |
+
"loss": 3.839,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.45213437736802226,
|
| 1259 |
+
"grad_norm": 954567.125,
|
| 1260 |
+
"learning_rate": 2.739328113159889e-05,
|
| 1261 |
+
"loss": 3.507,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4546602677443799,
|
| 1266 |
+
"grad_norm": 1057372.25,
|
| 1267 |
+
"learning_rate": 2.726698661278101e-05,
|
| 1268 |
+
"loss": 4.0177,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.45718615812073754,
|
| 1273 |
+
"grad_norm": 1196149.375,
|
| 1274 |
+
"learning_rate": 2.714069209396312e-05,
|
| 1275 |
+
"loss": 3.6044,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.45971204849709524,
|
| 1280 |
+
"grad_norm": 664877.5,
|
| 1281 |
+
"learning_rate": 2.701439757514524e-05,
|
| 1282 |
+
"loss": 3.7145,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4622379388734529,
|
| 1287 |
+
"grad_norm": 1227852.0,
|
| 1288 |
+
"learning_rate": 2.6888103056327358e-05,
|
| 1289 |
+
"loss": 3.7588,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4647638292498106,
|
| 1294 |
+
"grad_norm": 1403871.0,
|
| 1295 |
+
"learning_rate": 2.6761808537509473e-05,
|
| 1296 |
+
"loss": 3.672,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4672897196261682,
|
| 1301 |
+
"grad_norm": 1831221.875,
|
| 1302 |
+
"learning_rate": 2.663551401869159e-05,
|
| 1303 |
+
"loss": 3.8373,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.46981561000252586,
|
| 1308 |
+
"grad_norm": 1008128.75,
|
| 1309 |
+
"learning_rate": 2.6509219499873704e-05,
|
| 1310 |
+
"loss": 3.6977,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.47234150037888356,
|
| 1315 |
+
"grad_norm": 704558.8125,
|
| 1316 |
+
"learning_rate": 2.6382924981055823e-05,
|
| 1317 |
+
"loss": 3.9759,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.4748673907552412,
|
| 1322 |
+
"grad_norm": 1501172.875,
|
| 1323 |
+
"learning_rate": 2.6256630462237942e-05,
|
| 1324 |
+
"loss": 3.3989,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.4773932811315989,
|
| 1329 |
+
"grad_norm": 1022078.375,
|
| 1330 |
+
"learning_rate": 2.613033594342006e-05,
|
| 1331 |
+
"loss": 3.7731,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.47991917150795654,
|
| 1336 |
+
"grad_norm": 892236.75,
|
| 1337 |
+
"learning_rate": 2.6004041424602173e-05,
|
| 1338 |
+
"loss": 3.6186,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.48244506188431424,
|
| 1343 |
+
"grad_norm": 592257.875,
|
| 1344 |
+
"learning_rate": 2.5877746905784288e-05,
|
| 1345 |
+
"loss": 3.3699,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.4849709522606719,
|
| 1350 |
+
"grad_norm": 761213.9375,
|
| 1351 |
+
"learning_rate": 2.5751452386966407e-05,
|
| 1352 |
+
"loss": 3.6678,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.4874968426370295,
|
| 1357 |
+
"grad_norm": 1076466.0,
|
| 1358 |
+
"learning_rate": 2.5625157868148525e-05,
|
| 1359 |
+
"loss": 3.4251,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.4900227330133872,
|
| 1364 |
+
"grad_norm": 1370577.375,
|
| 1365 |
+
"learning_rate": 2.5498863349330644e-05,
|
| 1366 |
+
"loss": 3.6471,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.49254862338974487,
|
| 1371 |
+
"grad_norm": 682627.5625,
|
| 1372 |
+
"learning_rate": 2.5372568830512756e-05,
|
| 1373 |
+
"loss": 3.4545,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.49507451376610256,
|
| 1378 |
+
"grad_norm": 812808.125,
|
| 1379 |
+
"learning_rate": 2.524627431169487e-05,
|
| 1380 |
+
"loss": 3.64,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.4976004041424602,
|
| 1385 |
+
"grad_norm": 1190569.0,
|
| 1386 |
+
"learning_rate": 2.511997979287699e-05,
|
| 1387 |
+
"loss": 3.5669,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5001262945188178,
|
| 1392 |
+
"grad_norm": 1053372.125,
|
| 1393 |
+
"learning_rate": 2.4993685274059106e-05,
|
| 1394 |
+
"loss": 3.561,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5026521848951756,
|
| 1399 |
+
"grad_norm": 1154567.75,
|
| 1400 |
+
"learning_rate": 2.4867390755241225e-05,
|
| 1401 |
+
"loss": 3.7747,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5051780752715332,
|
| 1406 |
+
"grad_norm": 1105378.875,
|
| 1407 |
+
"learning_rate": 2.4741096236423343e-05,
|
| 1408 |
+
"loss": 3.5388,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5077039656478909,
|
| 1413 |
+
"grad_norm": 975523.9375,
|
| 1414 |
+
"learning_rate": 2.4614801717605455e-05,
|
| 1415 |
+
"loss": 3.7567,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5102298560242485,
|
| 1420 |
+
"grad_norm": 1316863.5,
|
| 1421 |
+
"learning_rate": 2.4488507198787574e-05,
|
| 1422 |
+
"loss": 3.4911,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5127557464006062,
|
| 1427 |
+
"grad_norm": 1235055.375,
|
| 1428 |
+
"learning_rate": 2.436221267996969e-05,
|
| 1429 |
+
"loss": 3.3298,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5152816367769639,
|
| 1434 |
+
"grad_norm": 1190251.5,
|
| 1435 |
+
"learning_rate": 2.4235918161151808e-05,
|
| 1436 |
+
"loss": 3.7792,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5178075271533216,
|
| 1441 |
+
"grad_norm": 1828954.875,
|
| 1442 |
+
"learning_rate": 2.4109623642333924e-05,
|
| 1443 |
+
"loss": 3.3407,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5203334175296792,
|
| 1448 |
+
"grad_norm": 734125.1875,
|
| 1449 |
+
"learning_rate": 2.398332912351604e-05,
|
| 1450 |
+
"loss": 3.6435,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5228593079060369,
|
| 1455 |
+
"grad_norm": 1114272.0,
|
| 1456 |
+
"learning_rate": 2.3857034604698158e-05,
|
| 1457 |
+
"loss": 3.603,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5253851982823945,
|
| 1462 |
+
"grad_norm": 1464147.0,
|
| 1463 |
+
"learning_rate": 2.3730740085880273e-05,
|
| 1464 |
+
"loss": 3.6587,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5279110886587522,
|
| 1469 |
+
"grad_norm": 2074921.0,
|
| 1470 |
+
"learning_rate": 2.3604445567062392e-05,
|
| 1471 |
+
"loss": 3.7505,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5304369790351099,
|
| 1476 |
+
"grad_norm": 1063793.625,
|
| 1477 |
+
"learning_rate": 2.3478151048244507e-05,
|
| 1478 |
+
"loss": 3.8397,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5329628694114675,
|
| 1483 |
+
"grad_norm": 702127.6875,
|
| 1484 |
+
"learning_rate": 2.3351856529426626e-05,
|
| 1485 |
+
"loss": 3.4248,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5354887597878252,
|
| 1490 |
+
"grad_norm": 1237508.5,
|
| 1491 |
+
"learning_rate": 2.3225562010608738e-05,
|
| 1492 |
+
"loss": 3.6119,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5380146501641829,
|
| 1497 |
+
"grad_norm": 678885.625,
|
| 1498 |
+
"learning_rate": 2.3099267491790857e-05,
|
| 1499 |
+
"loss": 3.4523,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5405405405405406,
|
| 1504 |
+
"grad_norm": 1506679.875,
|
| 1505 |
+
"learning_rate": 2.2972972972972976e-05,
|
| 1506 |
+
"loss": 3.4496,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5430664309168982,
|
| 1511 |
+
"grad_norm": 1016625.6875,
|
| 1512 |
+
"learning_rate": 2.284667845415509e-05,
|
| 1513 |
+
"loss": 3.6607,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.5455923212932559,
|
| 1518 |
+
"grad_norm": 916826.3125,
|
| 1519 |
+
"learning_rate": 2.272038393533721e-05,
|
| 1520 |
+
"loss": 3.6147,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5481182116696135,
|
| 1525 |
+
"grad_norm": 1150088.625,
|
| 1526 |
+
"learning_rate": 2.2594089416519322e-05,
|
| 1527 |
+
"loss": 3.3045,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5506441020459713,
|
| 1532 |
+
"grad_norm": 1215418.875,
|
| 1533 |
+
"learning_rate": 2.246779489770144e-05,
|
| 1534 |
+
"loss": 3.6112,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5531699924223289,
|
| 1539 |
+
"grad_norm": 610014.5,
|
| 1540 |
+
"learning_rate": 2.2341500378883556e-05,
|
| 1541 |
+
"loss": 3.5236,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5556958827986865,
|
| 1546 |
+
"grad_norm": 834213.625,
|
| 1547 |
+
"learning_rate": 2.2215205860065675e-05,
|
| 1548 |
+
"loss": 3.4063,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5582217731750442,
|
| 1553 |
+
"grad_norm": 976218.8125,
|
| 1554 |
+
"learning_rate": 2.2088911341247793e-05,
|
| 1555 |
+
"loss": 3.311,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5607476635514018,
|
| 1560 |
+
"grad_norm": 940372.0,
|
| 1561 |
+
"learning_rate": 2.196261682242991e-05,
|
| 1562 |
+
"loss": 3.6945,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5632735539277596,
|
| 1567 |
+
"grad_norm": 903718.9375,
|
| 1568 |
+
"learning_rate": 2.1836322303612024e-05,
|
| 1569 |
+
"loss": 3.3955,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5657994443041172,
|
| 1574 |
+
"grad_norm": 791363.5,
|
| 1575 |
+
"learning_rate": 2.171002778479414e-05,
|
| 1576 |
+
"loss": 3.5137,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5683253346804749,
|
| 1581 |
+
"grad_norm": 958748.625,
|
| 1582 |
+
"learning_rate": 2.158373326597626e-05,
|
| 1583 |
+
"loss": 3.3765,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5708512250568325,
|
| 1588 |
+
"grad_norm": 1431520.25,
|
| 1589 |
+
"learning_rate": 2.1457438747158374e-05,
|
| 1590 |
+
"loss": 3.1066,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.5733771154331903,
|
| 1595 |
+
"grad_norm": 793628.1875,
|
| 1596 |
+
"learning_rate": 2.1331144228340493e-05,
|
| 1597 |
+
"loss": 3.5486,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5759030058095479,
|
| 1602 |
+
"grad_norm": 935731.3125,
|
| 1603 |
+
"learning_rate": 2.1204849709522608e-05,
|
| 1604 |
+
"loss": 3.427,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5784288961859055,
|
| 1609 |
+
"grad_norm": 1024607.625,
|
| 1610 |
+
"learning_rate": 2.1078555190704723e-05,
|
| 1611 |
+
"loss": 3.2279,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5809547865622632,
|
| 1616 |
+
"grad_norm": 1027219.3125,
|
| 1617 |
+
"learning_rate": 2.0952260671886842e-05,
|
| 1618 |
+
"loss": 3.7803,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5834806769386208,
|
| 1623 |
+
"grad_norm": 762017.8125,
|
| 1624 |
+
"learning_rate": 2.0825966153068957e-05,
|
| 1625 |
+
"loss": 3.7272,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5860065673149786,
|
| 1630 |
+
"grad_norm": 1251829.125,
|
| 1631 |
+
"learning_rate": 2.0699671634251076e-05,
|
| 1632 |
+
"loss": 3.6896,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.5885324576913362,
|
| 1637 |
+
"grad_norm": 841272.5,
|
| 1638 |
+
"learning_rate": 2.057337711543319e-05,
|
| 1639 |
+
"loss": 3.3452,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.5910583480676939,
|
| 1644 |
+
"grad_norm": 1126517.625,
|
| 1645 |
+
"learning_rate": 2.0447082596615307e-05,
|
| 1646 |
+
"loss": 3.386,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.5935842384440515,
|
| 1651 |
+
"grad_norm": 982217.3125,
|
| 1652 |
+
"learning_rate": 2.0320788077797426e-05,
|
| 1653 |
+
"loss": 3.7347,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.5961101288204091,
|
| 1658 |
+
"grad_norm": 644312.75,
|
| 1659 |
+
"learning_rate": 2.019449355897954e-05,
|
| 1660 |
+
"loss": 3.1686,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.5986360191967669,
|
| 1665 |
+
"grad_norm": 1135732.625,
|
| 1666 |
+
"learning_rate": 2.006819904016166e-05,
|
| 1667 |
+
"loss": 3.6648,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6011619095731245,
|
| 1672 |
+
"grad_norm": 886122.0,
|
| 1673 |
+
"learning_rate": 1.9941904521343775e-05,
|
| 1674 |
+
"loss": 3.3114,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.6036877999494822,
|
| 1679 |
+
"grad_norm": 1676023.875,
|
| 1680 |
+
"learning_rate": 1.981561000252589e-05,
|
| 1681 |
+
"loss": 3.8232,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6062136903258398,
|
| 1686 |
+
"grad_norm": 1056139.0,
|
| 1687 |
+
"learning_rate": 1.9689315483708006e-05,
|
| 1688 |
+
"loss": 3.3829,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6087395807021975,
|
| 1693 |
+
"grad_norm": 1303276.625,
|
| 1694 |
+
"learning_rate": 1.9563020964890125e-05,
|
| 1695 |
+
"loss": 3.7356,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6112654710785552,
|
| 1700 |
+
"grad_norm": 1168222.5,
|
| 1701 |
+
"learning_rate": 1.9436726446072244e-05,
|
| 1702 |
+
"loss": 3.4672,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6137913614549129,
|
| 1707 |
+
"grad_norm": 974863.5,
|
| 1708 |
+
"learning_rate": 1.931043192725436e-05,
|
| 1709 |
+
"loss": 3.7479,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.6163172518312705,
|
| 1714 |
+
"grad_norm": 727352.375,
|
| 1715 |
+
"learning_rate": 1.9184137408436474e-05,
|
| 1716 |
+
"loss": 3.4183,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6188431422076281,
|
| 1721 |
+
"grad_norm": 1103530.0,
|
| 1722 |
+
"learning_rate": 1.905784288961859e-05,
|
| 1723 |
+
"loss": 3.5818,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6213690325839859,
|
| 1728 |
+
"grad_norm": 866723.3125,
|
| 1729 |
+
"learning_rate": 1.893154837080071e-05,
|
| 1730 |
+
"loss": 3.4123,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6238949229603435,
|
| 1735 |
+
"grad_norm": 1022545.8125,
|
| 1736 |
+
"learning_rate": 1.8805253851982824e-05,
|
| 1737 |
+
"loss": 3.1739,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6264208133367012,
|
| 1742 |
+
"grad_norm": 1114618.75,
|
| 1743 |
+
"learning_rate": 1.8678959333164943e-05,
|
| 1744 |
+
"loss": 3.6795,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6289467037130588,
|
| 1749 |
+
"grad_norm": 1178612.875,
|
| 1750 |
+
"learning_rate": 1.8552664814347058e-05,
|
| 1751 |
+
"loss": 3.3061,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6314725940894165,
|
| 1756 |
+
"grad_norm": 757466.0,
|
| 1757 |
+
"learning_rate": 1.8426370295529173e-05,
|
| 1758 |
+
"loss": 3.4692,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6339984844657742,
|
| 1763 |
+
"grad_norm": 932833.75,
|
| 1764 |
+
"learning_rate": 1.8300075776711292e-05,
|
| 1765 |
+
"loss": 3.3951,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6365243748421319,
|
| 1770 |
+
"grad_norm": 811280.75,
|
| 1771 |
+
"learning_rate": 1.8173781257893408e-05,
|
| 1772 |
+
"loss": 3.42,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6390502652184895,
|
| 1777 |
+
"grad_norm": 828370.875,
|
| 1778 |
+
"learning_rate": 1.8047486739075526e-05,
|
| 1779 |
+
"loss": 3.5068,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6415761555948472,
|
| 1784 |
+
"grad_norm": 796540.6875,
|
| 1785 |
+
"learning_rate": 1.7921192220257642e-05,
|
| 1786 |
+
"loss": 3.4725,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6441020459712048,
|
| 1791 |
+
"grad_norm": 1101170.0,
|
| 1792 |
+
"learning_rate": 1.7794897701439757e-05,
|
| 1793 |
+
"loss": 3.5315,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6466279363475625,
|
| 1798 |
+
"grad_norm": 1084591.375,
|
| 1799 |
+
"learning_rate": 1.7668603182621876e-05,
|
| 1800 |
+
"loss": 3.3357,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6491538267239202,
|
| 1805 |
+
"grad_norm": 846566.375,
|
| 1806 |
+
"learning_rate": 1.754230866380399e-05,
|
| 1807 |
+
"loss": 3.4352,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6516797171002778,
|
| 1812 |
+
"grad_norm": 817152.9375,
|
| 1813 |
+
"learning_rate": 1.741601414498611e-05,
|
| 1814 |
+
"loss": 3.3913,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6542056074766355,
|
| 1819 |
+
"grad_norm": 896931.875,
|
| 1820 |
+
"learning_rate": 1.7289719626168225e-05,
|
| 1821 |
+
"loss": 3.3725,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6567314978529932,
|
| 1826 |
+
"grad_norm": 580791.0,
|
| 1827 |
+
"learning_rate": 1.716342510735034e-05,
|
| 1828 |
+
"loss": 3.3803,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6592573882293509,
|
| 1833 |
+
"grad_norm": 817709.375,
|
| 1834 |
+
"learning_rate": 1.7037130588532456e-05,
|
| 1835 |
+
"loss": 3.4564,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.6617832786057085,
|
| 1840 |
+
"grad_norm": 1013458.75,
|
| 1841 |
+
"learning_rate": 1.6910836069714575e-05,
|
| 1842 |
+
"loss": 3.4659,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6643091689820662,
|
| 1847 |
+
"grad_norm": 855889.0625,
|
| 1848 |
+
"learning_rate": 1.6784541550896694e-05,
|
| 1849 |
+
"loss": 3.1951,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6668350593584238,
|
| 1854 |
+
"grad_norm": 1114807.125,
|
| 1855 |
+
"learning_rate": 1.665824703207881e-05,
|
| 1856 |
+
"loss": 3.2604,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6693609497347816,
|
| 1861 |
+
"grad_norm": 1472619.75,
|
| 1862 |
+
"learning_rate": 1.6531952513260925e-05,
|
| 1863 |
+
"loss": 3.2474,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6718868401111392,
|
| 1868 |
+
"grad_norm": 933670.625,
|
| 1869 |
+
"learning_rate": 1.640565799444304e-05,
|
| 1870 |
+
"loss": 3.2033,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6744127304874968,
|
| 1875 |
+
"grad_norm": 752061.375,
|
| 1876 |
+
"learning_rate": 1.627936347562516e-05,
|
| 1877 |
+
"loss": 3.5689,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6769386208638545,
|
| 1882 |
+
"grad_norm": 1359021.375,
|
| 1883 |
+
"learning_rate": 1.6153068956807274e-05,
|
| 1884 |
+
"loss": 3.289,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6794645112402121,
|
| 1889 |
+
"grad_norm": 755337.0625,
|
| 1890 |
+
"learning_rate": 1.6026774437989393e-05,
|
| 1891 |
+
"loss": 3.4877,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.6819904016165699,
|
| 1896 |
+
"grad_norm": 797617.3125,
|
| 1897 |
+
"learning_rate": 1.5900479919171508e-05,
|
| 1898 |
+
"loss": 3.4128,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6845162919929275,
|
| 1903 |
+
"grad_norm": 754639.75,
|
| 1904 |
+
"learning_rate": 1.5774185400353624e-05,
|
| 1905 |
+
"loss": 3.4954,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6870421823692852,
|
| 1910 |
+
"grad_norm": 853287.1875,
|
| 1911 |
+
"learning_rate": 1.5647890881535742e-05,
|
| 1912 |
+
"loss": 3.2564,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6895680727456428,
|
| 1917 |
+
"grad_norm": 554583.1875,
|
| 1918 |
+
"learning_rate": 1.5521596362717858e-05,
|
| 1919 |
+
"loss": 3.2221,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.6920939631220006,
|
| 1924 |
+
"grad_norm": 722306.8125,
|
| 1925 |
+
"learning_rate": 1.5395301843899977e-05,
|
| 1926 |
+
"loss": 3.1547,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6946198534983582,
|
| 1931 |
+
"grad_norm": 833465.875,
|
| 1932 |
+
"learning_rate": 1.5269007325082092e-05,
|
| 1933 |
+
"loss": 3.1672,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6971457438747158,
|
| 1938 |
+
"grad_norm": 801353.3125,
|
| 1939 |
+
"learning_rate": 1.5142712806264209e-05,
|
| 1940 |
+
"loss": 3.4491,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6996716342510735,
|
| 1945 |
+
"grad_norm": 621898.875,
|
| 1946 |
+
"learning_rate": 1.5016418287446326e-05,
|
| 1947 |
+
"loss": 3.4968,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7021975246274311,
|
| 1952 |
+
"grad_norm": 926238.0625,
|
| 1953 |
+
"learning_rate": 1.4890123768628441e-05,
|
| 1954 |
+
"loss": 3.5613,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7047234150037889,
|
| 1959 |
+
"grad_norm": 991762.4375,
|
| 1960 |
+
"learning_rate": 1.476382924981056e-05,
|
| 1961 |
+
"loss": 3.1534,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7072493053801465,
|
| 1966 |
+
"grad_norm": 876887.3125,
|
| 1967 |
+
"learning_rate": 1.4637534730992674e-05,
|
| 1968 |
+
"loss": 3.1666,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7097751957565042,
|
| 1973 |
+
"grad_norm": 748469.75,
|
| 1974 |
+
"learning_rate": 1.4511240212174793e-05,
|
| 1975 |
+
"loss": 3.3201,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.7123010861328618,
|
| 1980 |
+
"grad_norm": 913840.125,
|
| 1981 |
+
"learning_rate": 1.4384945693356908e-05,
|
| 1982 |
+
"loss": 3.3263,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7148269765092194,
|
| 1987 |
+
"grad_norm": 776508.25,
|
| 1988 |
+
"learning_rate": 1.4258651174539025e-05,
|
| 1989 |
+
"loss": 3.3736,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7173528668855772,
|
| 1994 |
+
"grad_norm": 1003093.6875,
|
| 1995 |
+
"learning_rate": 1.4132356655721144e-05,
|
| 1996 |
+
"loss": 3.3369,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7198787572619348,
|
| 2001 |
+
"grad_norm": 480733.9375,
|
| 2002 |
+
"learning_rate": 1.400606213690326e-05,
|
| 2003 |
+
"loss": 3.4599,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7224046476382925,
|
| 2008 |
+
"grad_norm": 921517.8125,
|
| 2009 |
+
"learning_rate": 1.3879767618085376e-05,
|
| 2010 |
+
"loss": 3.3327,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7249305380146501,
|
| 2015 |
+
"grad_norm": 1106659.0,
|
| 2016 |
+
"learning_rate": 1.3753473099267492e-05,
|
| 2017 |
+
"loss": 3.4347,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7274564283910079,
|
| 2022 |
+
"grad_norm": 506619.90625,
|
| 2023 |
+
"learning_rate": 1.3627178580449609e-05,
|
| 2024 |
+
"loss": 3.2675,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7299823187673655,
|
| 2029 |
+
"grad_norm": 862991.0625,
|
| 2030 |
+
"learning_rate": 1.3500884061631724e-05,
|
| 2031 |
+
"loss": 3.2461,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7325082091437232,
|
| 2036 |
+
"grad_norm": 835895.8125,
|
| 2037 |
+
"learning_rate": 1.3374589542813843e-05,
|
| 2038 |
+
"loss": 3.3976,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7350340995200808,
|
| 2043 |
+
"grad_norm": 699905.9375,
|
| 2044 |
+
"learning_rate": 1.324829502399596e-05,
|
| 2045 |
+
"loss": 3.4889,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.7375599898964385,
|
| 2050 |
+
"grad_norm": 824201.75,
|
| 2051 |
+
"learning_rate": 1.3122000505178075e-05,
|
| 2052 |
+
"loss": 3.548,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7400858802727962,
|
| 2057 |
+
"grad_norm": 871363.75,
|
| 2058 |
+
"learning_rate": 1.2995705986360193e-05,
|
| 2059 |
+
"loss": 3.5843,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7426117706491538,
|
| 2064 |
+
"grad_norm": 627847.1875,
|
| 2065 |
+
"learning_rate": 1.2869411467542308e-05,
|
| 2066 |
+
"loss": 3.1553,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7451376610255115,
|
| 2071 |
+
"grad_norm": 758722.4375,
|
| 2072 |
+
"learning_rate": 1.2743116948724427e-05,
|
| 2073 |
+
"loss": 3.4285,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7476635514018691,
|
| 2078 |
+
"grad_norm": 851608.8125,
|
| 2079 |
+
"learning_rate": 1.2616822429906542e-05,
|
| 2080 |
+
"loss": 3.3232,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.7501894417782268,
|
| 2085 |
+
"grad_norm": 917633.8125,
|
| 2086 |
+
"learning_rate": 1.249052791108866e-05,
|
| 2087 |
+
"loss": 3.1254,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7527153321545845,
|
| 2092 |
+
"grad_norm": 1110027.375,
|
| 2093 |
+
"learning_rate": 1.2364233392270776e-05,
|
| 2094 |
+
"loss": 3.4867,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7552412225309422,
|
| 2099 |
+
"grad_norm": 817171.625,
|
| 2100 |
+
"learning_rate": 1.2237938873452892e-05,
|
| 2101 |
+
"loss": 3.4251,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7577671129072998,
|
| 2106 |
+
"grad_norm": 1074504.25,
|
| 2107 |
+
"learning_rate": 1.211164435463501e-05,
|
| 2108 |
+
"loss": 3.4581,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7602930032836575,
|
| 2113 |
+
"grad_norm": 777395.4375,
|
| 2114 |
+
"learning_rate": 1.1985349835817126e-05,
|
| 2115 |
+
"loss": 3.3139,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7628188936600152,
|
| 2120 |
+
"grad_norm": 657749.75,
|
| 2121 |
+
"learning_rate": 1.1859055316999243e-05,
|
| 2122 |
+
"loss": 3.2542,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7653447840363728,
|
| 2127 |
+
"grad_norm": 536018.1875,
|
| 2128 |
+
"learning_rate": 1.173276079818136e-05,
|
| 2129 |
+
"loss": 3.4512,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7678706744127305,
|
| 2134 |
+
"grad_norm": 1034483.5625,
|
| 2135 |
+
"learning_rate": 1.1606466279363475e-05,
|
| 2136 |
+
"loss": 3.3698,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7703965647890881,
|
| 2141 |
+
"grad_norm": 1303588.5,
|
| 2142 |
+
"learning_rate": 1.1480171760545592e-05,
|
| 2143 |
+
"loss": 3.1435,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7729224551654458,
|
| 2148 |
+
"grad_norm": 961475.75,
|
| 2149 |
+
"learning_rate": 1.135387724172771e-05,
|
| 2150 |
+
"loss": 3.1625,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7754483455418035,
|
| 2155 |
+
"grad_norm": 811610.3125,
|
| 2156 |
+
"learning_rate": 1.1227582722909827e-05,
|
| 2157 |
+
"loss": 3.2445,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7779742359181612,
|
| 2162 |
+
"grad_norm": 573133.25,
|
| 2163 |
+
"learning_rate": 1.1101288204091944e-05,
|
| 2164 |
+
"loss": 3.2385,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7805001262945188,
|
| 2169 |
+
"grad_norm": 719160.0625,
|
| 2170 |
+
"learning_rate": 1.0974993685274059e-05,
|
| 2171 |
+
"loss": 3.4963,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7830260166708765,
|
| 2176 |
+
"grad_norm": 625430.0625,
|
| 2177 |
+
"learning_rate": 1.0848699166456176e-05,
|
| 2178 |
+
"loss": 3.219,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.7855519070472341,
|
| 2183 |
+
"grad_norm": 847258.625,
|
| 2184 |
+
"learning_rate": 1.0722404647638293e-05,
|
| 2185 |
+
"loss": 3.3977,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7880777974235919,
|
| 2190 |
+
"grad_norm": 909602.0625,
|
| 2191 |
+
"learning_rate": 1.0596110128820409e-05,
|
| 2192 |
+
"loss": 3.1271,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.7906036877999495,
|
| 2197 |
+
"grad_norm": 2303105.75,
|
| 2198 |
+
"learning_rate": 1.0469815610002526e-05,
|
| 2199 |
+
"loss": 3.1443,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7931295781763071,
|
| 2204 |
+
"grad_norm": 627608.9375,
|
| 2205 |
+
"learning_rate": 1.0343521091184644e-05,
|
| 2206 |
+
"loss": 3.1124,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.7956554685526648,
|
| 2211 |
+
"grad_norm": 683938.75,
|
| 2212 |
+
"learning_rate": 1.021722657236676e-05,
|
| 2213 |
+
"loss": 3.5188,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.7981813589290225,
|
| 2218 |
+
"grad_norm": 742955.25,
|
| 2219 |
+
"learning_rate": 1.0090932053548877e-05,
|
| 2220 |
+
"loss": 3.3953,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8007072493053802,
|
| 2225 |
+
"grad_norm": 871770.5,
|
| 2226 |
+
"learning_rate": 9.964637534730994e-06,
|
| 2227 |
+
"loss": 3.2334,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8032331396817378,
|
| 2232 |
+
"grad_norm": 751897.5625,
|
| 2233 |
+
"learning_rate": 9.83834301591311e-06,
|
| 2234 |
+
"loss": 3.2724,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8057590300580955,
|
| 2239 |
+
"grad_norm": 740156.9375,
|
| 2240 |
+
"learning_rate": 9.712048497095226e-06,
|
| 2241 |
+
"loss": 3.344,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.8082849204344531,
|
| 2246 |
+
"grad_norm": 634736.5,
|
| 2247 |
+
"learning_rate": 9.585753978277342e-06,
|
| 2248 |
+
"loss": 3.1028,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8108108108108109,
|
| 2253 |
+
"grad_norm": 1092338.25,
|
| 2254 |
+
"learning_rate": 9.45945945945946e-06,
|
| 2255 |
+
"loss": 3.5916,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8133367011871685,
|
| 2260 |
+
"grad_norm": 915638.875,
|
| 2261 |
+
"learning_rate": 9.333164940641578e-06,
|
| 2262 |
+
"loss": 3.126,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8158625915635261,
|
| 2267 |
+
"grad_norm": 919722.3125,
|
| 2268 |
+
"learning_rate": 9.206870421823693e-06,
|
| 2269 |
+
"loss": 3.1032,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8183884819398838,
|
| 2274 |
+
"grad_norm": 791748.625,
|
| 2275 |
+
"learning_rate": 9.08057590300581e-06,
|
| 2276 |
+
"loss": 3.4718,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.8209143723162414,
|
| 2281 |
+
"grad_norm": 607153.8125,
|
| 2282 |
+
"learning_rate": 8.954281384187927e-06,
|
| 2283 |
+
"loss": 2.989,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8234402626925992,
|
| 2288 |
+
"grad_norm": 800489.625,
|
| 2289 |
+
"learning_rate": 8.827986865370043e-06,
|
| 2290 |
+
"loss": 3.1845,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8259661530689568,
|
| 2295 |
+
"grad_norm": 748984.9375,
|
| 2296 |
+
"learning_rate": 8.70169234655216e-06,
|
| 2297 |
+
"loss": 3.2721,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8284920434453145,
|
| 2302 |
+
"grad_norm": 686388.0625,
|
| 2303 |
+
"learning_rate": 8.575397827734277e-06,
|
| 2304 |
+
"loss": 3.178,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8310179338216721,
|
| 2309 |
+
"grad_norm": 634646.5,
|
| 2310 |
+
"learning_rate": 8.449103308916394e-06,
|
| 2311 |
+
"loss": 3.1601,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8335438241980299,
|
| 2316 |
+
"grad_norm": 973350.375,
|
| 2317 |
+
"learning_rate": 8.322808790098511e-06,
|
| 2318 |
+
"loss": 3.1181,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8360697145743875,
|
| 2323 |
+
"grad_norm": 802978.1875,
|
| 2324 |
+
"learning_rate": 8.196514271280626e-06,
|
| 2325 |
+
"loss": 3.1694,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8385956049507451,
|
| 2330 |
+
"grad_norm": 543352.25,
|
| 2331 |
+
"learning_rate": 8.070219752462743e-06,
|
| 2332 |
+
"loss": 3.0763,
|
| 2333 |
+
"step": 3320
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8411214953271028,
|
| 2337 |
+
"grad_norm": 965499.0625,
|
| 2338 |
+
"learning_rate": 7.94392523364486e-06,
|
| 2339 |
+
"loss": 3.3403,
|
| 2340 |
+
"step": 3330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8436473857034604,
|
| 2344 |
+
"grad_norm": 1031962.0,
|
| 2345 |
+
"learning_rate": 7.817630714826977e-06,
|
| 2346 |
+
"loss": 3.1295,
|
| 2347 |
+
"step": 3340
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.8461732760798182,
|
| 2351 |
+
"grad_norm": 900595.875,
|
| 2352 |
+
"learning_rate": 7.691336196009095e-06,
|
| 2353 |
+
"loss": 3.0841,
|
| 2354 |
+
"step": 3350
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8486991664561758,
|
| 2358 |
+
"grad_norm": 878680.5625,
|
| 2359 |
+
"learning_rate": 7.565041677191211e-06,
|
| 2360 |
+
"loss": 3.2706,
|
| 2361 |
+
"step": 3360
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8512250568325335,
|
| 2365 |
+
"grad_norm": 541207.75,
|
| 2366 |
+
"learning_rate": 7.438747158373327e-06,
|
| 2367 |
+
"loss": 3.2398,
|
| 2368 |
+
"step": 3370
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8537509472088911,
|
| 2372 |
+
"grad_norm": 689078.875,
|
| 2373 |
+
"learning_rate": 7.312452639555443e-06,
|
| 2374 |
+
"loss": 3.0991,
|
| 2375 |
+
"step": 3380
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8562768375852488,
|
| 2379 |
+
"grad_norm": 831574.1875,
|
| 2380 |
+
"learning_rate": 7.18615812073756e-06,
|
| 2381 |
+
"loss": 3.2631,
|
| 2382 |
+
"step": 3390
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8588027279616065,
|
| 2386 |
+
"grad_norm": 777584.0,
|
| 2387 |
+
"learning_rate": 7.0598636019196766e-06,
|
| 2388 |
+
"loss": 3.4074,
|
| 2389 |
+
"step": 3400
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8613286183379641,
|
| 2393 |
+
"grad_norm": 742726.625,
|
| 2394 |
+
"learning_rate": 6.9335690831017945e-06,
|
| 2395 |
+
"loss": 3.166,
|
| 2396 |
+
"step": 3410
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8638545087143218,
|
| 2400 |
+
"grad_norm": 580725.0625,
|
| 2401 |
+
"learning_rate": 6.807274564283911e-06,
|
| 2402 |
+
"loss": 3.1598,
|
| 2403 |
+
"step": 3420
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.8663803990906794,
|
| 2407 |
+
"grad_norm": 588559.3125,
|
| 2408 |
+
"learning_rate": 6.680980045466027e-06,
|
| 2409 |
+
"loss": 3.0765,
|
| 2410 |
+
"step": 3430
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8689062894670372,
|
| 2414 |
+
"grad_norm": 512794.3125,
|
| 2415 |
+
"learning_rate": 6.554685526648144e-06,
|
| 2416 |
+
"loss": 3.3832,
|
| 2417 |
+
"step": 3440
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8714321798433948,
|
| 2421 |
+
"grad_norm": 876623.1875,
|
| 2422 |
+
"learning_rate": 6.42839100783026e-06,
|
| 2423 |
+
"loss": 3.4546,
|
| 2424 |
+
"step": 3450
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8739580702197525,
|
| 2428 |
+
"grad_norm": 716818.125,
|
| 2429 |
+
"learning_rate": 6.3020964890123765e-06,
|
| 2430 |
+
"loss": 3.2477,
|
| 2431 |
+
"step": 3460
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8764839605961101,
|
| 2435 |
+
"grad_norm": 1114704.75,
|
| 2436 |
+
"learning_rate": 6.175801970194494e-06,
|
| 2437 |
+
"loss": 3.1437,
|
| 2438 |
+
"step": 3470
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8790098509724678,
|
| 2442 |
+
"grad_norm": 639491.625,
|
| 2443 |
+
"learning_rate": 6.049507451376611e-06,
|
| 2444 |
+
"loss": 3.1617,
|
| 2445 |
+
"step": 3480
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.8815357413488255,
|
| 2449 |
+
"grad_norm": 717235.0625,
|
| 2450 |
+
"learning_rate": 5.923212932558727e-06,
|
| 2451 |
+
"loss": 3.1024,
|
| 2452 |
+
"step": 3490
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8840616317251832,
|
| 2456 |
+
"grad_norm": 736368.0,
|
| 2457 |
+
"learning_rate": 5.796918413740844e-06,
|
| 2458 |
+
"loss": 3.3481,
|
| 2459 |
+
"step": 3500
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.8865875221015408,
|
| 2463 |
+
"grad_norm": 527303.1875,
|
| 2464 |
+
"learning_rate": 5.670623894922961e-06,
|
| 2465 |
+
"loss": 3.0436,
|
| 2466 |
+
"step": 3510
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8891134124778984,
|
| 2470 |
+
"grad_norm": 1408582.375,
|
| 2471 |
+
"learning_rate": 5.544329376105077e-06,
|
| 2472 |
+
"loss": 3.3846,
|
| 2473 |
+
"step": 3520
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8916393028542561,
|
| 2477 |
+
"grad_norm": 557347.875,
|
| 2478 |
+
"learning_rate": 5.418034857287194e-06,
|
| 2479 |
+
"loss": 3.3017,
|
| 2480 |
+
"step": 3530
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8941651932306138,
|
| 2484 |
+
"grad_norm": 533418.25,
|
| 2485 |
+
"learning_rate": 5.2917403384693106e-06,
|
| 2486 |
+
"loss": 2.9057,
|
| 2487 |
+
"step": 3540
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8966910836069715,
|
| 2491 |
+
"grad_norm": 940764.0,
|
| 2492 |
+
"learning_rate": 5.165445819651428e-06,
|
| 2493 |
+
"loss": 2.9862,
|
| 2494 |
+
"step": 3550
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8992169739833291,
|
| 2498 |
+
"grad_norm": 753501.4375,
|
| 2499 |
+
"learning_rate": 5.039151300833544e-06,
|
| 2500 |
+
"loss": 3.0074,
|
| 2501 |
+
"step": 3560
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9017428643596868,
|
| 2505 |
+
"grad_norm": 641765.1875,
|
| 2506 |
+
"learning_rate": 4.912856782015661e-06,
|
| 2507 |
+
"loss": 3.2161,
|
| 2508 |
+
"step": 3570
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.9042687547360445,
|
| 2512 |
+
"grad_norm": 846279.5625,
|
| 2513 |
+
"learning_rate": 4.786562263197777e-06,
|
| 2514 |
+
"loss": 3.1604,
|
| 2515 |
+
"step": 3580
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9067946451124022,
|
| 2519 |
+
"grad_norm": 623566.875,
|
| 2520 |
+
"learning_rate": 4.660267744379894e-06,
|
| 2521 |
+
"loss": 3.1331,
|
| 2522 |
+
"step": 3590
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9093205354887598,
|
| 2526 |
+
"grad_norm": 711430.625,
|
| 2527 |
+
"learning_rate": 4.533973225562011e-06,
|
| 2528 |
+
"loss": 3.2304,
|
| 2529 |
+
"step": 3600
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9118464258651174,
|
| 2533 |
+
"grad_norm": 656724.5,
|
| 2534 |
+
"learning_rate": 4.4076787067441276e-06,
|
| 2535 |
+
"loss": 3.4538,
|
| 2536 |
+
"step": 3610
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9143723162414751,
|
| 2540 |
+
"grad_norm": 595240.0,
|
| 2541 |
+
"learning_rate": 4.281384187926244e-06,
|
| 2542 |
+
"loss": 3.2896,
|
| 2543 |
+
"step": 3620
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.9168982066178328,
|
| 2547 |
+
"grad_norm": 662666.375,
|
| 2548 |
+
"learning_rate": 4.155089669108361e-06,
|
| 2549 |
+
"loss": 3.0247,
|
| 2550 |
+
"step": 3630
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9194240969941905,
|
| 2554 |
+
"grad_norm": 671792.5625,
|
| 2555 |
+
"learning_rate": 4.028795150290478e-06,
|
| 2556 |
+
"loss": 3.2846,
|
| 2557 |
+
"step": 3640
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9219499873705481,
|
| 2561 |
+
"grad_norm": 946814.3125,
|
| 2562 |
+
"learning_rate": 3.902500631472594e-06,
|
| 2563 |
+
"loss": 3.2046,
|
| 2564 |
+
"step": 3650
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9244758777469058,
|
| 2568 |
+
"grad_norm": 795244.75,
|
| 2569 |
+
"learning_rate": 3.776206112654711e-06,
|
| 2570 |
+
"loss": 3.1295,
|
| 2571 |
+
"step": 3660
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9270017681232634,
|
| 2575 |
+
"grad_norm": 670945.125,
|
| 2576 |
+
"learning_rate": 3.649911593836828e-06,
|
| 2577 |
+
"loss": 3.1636,
|
| 2578 |
+
"step": 3670
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9295276584996212,
|
| 2582 |
+
"grad_norm": 752728.5625,
|
| 2583 |
+
"learning_rate": 3.5236170750189446e-06,
|
| 2584 |
+
"loss": 3.3796,
|
| 2585 |
+
"step": 3680
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9320535488759788,
|
| 2589 |
+
"grad_norm": 798886.1875,
|
| 2590 |
+
"learning_rate": 3.3973225562010608e-06,
|
| 2591 |
+
"loss": 3.101,
|
| 2592 |
+
"step": 3690
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9345794392523364,
|
| 2596 |
+
"grad_norm": 470130.46875,
|
| 2597 |
+
"learning_rate": 3.2710280373831774e-06,
|
| 2598 |
+
"loss": 3.3339,
|
| 2599 |
+
"step": 3700
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9371053296286941,
|
| 2603 |
+
"grad_norm": 437737.59375,
|
| 2604 |
+
"learning_rate": 3.1447335185652945e-06,
|
| 2605 |
+
"loss": 3.0232,
|
| 2606 |
+
"step": 3710
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9396312200050517,
|
| 2610 |
+
"grad_norm": 497686.5625,
|
| 2611 |
+
"learning_rate": 3.018438999747411e-06,
|
| 2612 |
+
"loss": 3.2697,
|
| 2613 |
+
"step": 3720
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.9421571103814095,
|
| 2617 |
+
"grad_norm": 851300.5,
|
| 2618 |
+
"learning_rate": 2.892144480929528e-06,
|
| 2619 |
+
"loss": 3.4282,
|
| 2620 |
+
"step": 3730
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9446830007577671,
|
| 2624 |
+
"grad_norm": 715128.5,
|
| 2625 |
+
"learning_rate": 2.7658499621116445e-06,
|
| 2626 |
+
"loss": 3.1052,
|
| 2627 |
+
"step": 3740
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9472088911341248,
|
| 2631 |
+
"grad_norm": 904129.3125,
|
| 2632 |
+
"learning_rate": 2.639555443293761e-06,
|
| 2633 |
+
"loss": 3.1504,
|
| 2634 |
+
"step": 3750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9497347815104824,
|
| 2638 |
+
"grad_norm": 897233.625,
|
| 2639 |
+
"learning_rate": 2.5132609244758778e-06,
|
| 2640 |
+
"loss": 3.2904,
|
| 2641 |
+
"step": 3760
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9522606718868402,
|
| 2645 |
+
"grad_norm": 722250.1875,
|
| 2646 |
+
"learning_rate": 2.3869664056579944e-06,
|
| 2647 |
+
"loss": 3.1392,
|
| 2648 |
+
"step": 3770
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.9547865622631978,
|
| 2652 |
+
"grad_norm": 721083.9375,
|
| 2653 |
+
"learning_rate": 2.2606718868401115e-06,
|
| 2654 |
+
"loss": 3.137,
|
| 2655 |
+
"step": 3780
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.9573124526395554,
|
| 2659 |
+
"grad_norm": 637189.25,
|
| 2660 |
+
"learning_rate": 2.1343773680222277e-06,
|
| 2661 |
+
"loss": 3.1711,
|
| 2662 |
+
"step": 3790
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9598383430159131,
|
| 2666 |
+
"grad_norm": 858492.0,
|
| 2667 |
+
"learning_rate": 2.008082849204345e-06,
|
| 2668 |
+
"loss": 3.2843,
|
| 2669 |
+
"step": 3800
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.9623642333922707,
|
| 2673 |
+
"grad_norm": 714733.625,
|
| 2674 |
+
"learning_rate": 1.8817883303864613e-06,
|
| 2675 |
+
"loss": 3.305,
|
| 2676 |
+
"step": 3810
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9648901237686285,
|
| 2680 |
+
"grad_norm": 655039.75,
|
| 2681 |
+
"learning_rate": 1.7554938115685781e-06,
|
| 2682 |
+
"loss": 3.4754,
|
| 2683 |
+
"step": 3820
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9674160141449861,
|
| 2687 |
+
"grad_norm": 537881.875,
|
| 2688 |
+
"learning_rate": 1.6291992927506948e-06,
|
| 2689 |
+
"loss": 3.1539,
|
| 2690 |
+
"step": 3830
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.9699419045213438,
|
| 2694 |
+
"grad_norm": 654013.3125,
|
| 2695 |
+
"learning_rate": 1.5029047739328114e-06,
|
| 2696 |
+
"loss": 3.3037,
|
| 2697 |
+
"step": 3840
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9724677948977014,
|
| 2701 |
+
"grad_norm": 654474.0,
|
| 2702 |
+
"learning_rate": 1.376610255114928e-06,
|
| 2703 |
+
"loss": 3.1471,
|
| 2704 |
+
"step": 3850
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.974993685274059,
|
| 2708 |
+
"grad_norm": 925252.9375,
|
| 2709 |
+
"learning_rate": 1.2503157362970447e-06,
|
| 2710 |
+
"loss": 3.0344,
|
| 2711 |
+
"step": 3860
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.9775195756504168,
|
| 2715 |
+
"grad_norm": 690294.6875,
|
| 2716 |
+
"learning_rate": 1.1240212174791616e-06,
|
| 2717 |
+
"loss": 3.0319,
|
| 2718 |
+
"step": 3870
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.9800454660267744,
|
| 2722 |
+
"grad_norm": 654045.1875,
|
| 2723 |
+
"learning_rate": 9.977266986612783e-07,
|
| 2724 |
+
"loss": 3.2279,
|
| 2725 |
+
"step": 3880
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.9825713564031321,
|
| 2729 |
+
"grad_norm": 593023.5,
|
| 2730 |
+
"learning_rate": 8.714321798433948e-07,
|
| 2731 |
+
"loss": 3.3599,
|
| 2732 |
+
"step": 3890
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.9850972467794897,
|
| 2736 |
+
"grad_norm": 451262.59375,
|
| 2737 |
+
"learning_rate": 7.451376610255115e-07,
|
| 2738 |
+
"loss": 2.9683,
|
| 2739 |
+
"step": 3900
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.9876231371558475,
|
| 2743 |
+
"grad_norm": 814919.1875,
|
| 2744 |
+
"learning_rate": 6.188431422076282e-07,
|
| 2745 |
+
"loss": 3.4731,
|
| 2746 |
+
"step": 3910
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.9901490275322051,
|
| 2750 |
+
"grad_norm": 596075.0,
|
| 2751 |
+
"learning_rate": 4.925486233897449e-07,
|
| 2752 |
+
"loss": 3.0758,
|
| 2753 |
+
"step": 3920
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.9926749179085628,
|
| 2757 |
+
"grad_norm": 436784.15625,
|
| 2758 |
+
"learning_rate": 3.6625410457186164e-07,
|
| 2759 |
+
"loss": 3.0476,
|
| 2760 |
+
"step": 3930
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.9952008082849204,
|
| 2764 |
+
"grad_norm": 646142.9375,
|
| 2765 |
+
"learning_rate": 2.399595857539783e-07,
|
| 2766 |
+
"loss": 3.0169,
|
| 2767 |
+
"step": 3940
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.997726698661278,
|
| 2771 |
+
"grad_norm": 558096.9375,
|
| 2772 |
+
"learning_rate": 1.1366506693609499e-07,
|
| 2773 |
+
"loss": 3.1317,
|
| 2774 |
+
"step": 3950
|
| 2775 |
},
|
| 2776 |
{
|
| 2777 |
"epoch": 1.0,
|
| 2778 |
+
"step": 3959,
|
| 2779 |
+
"total_flos": 2.3966994296890982e+17,
|
| 2780 |
+
"train_loss": 3.909515940200563,
|
| 2781 |
+
"train_runtime": 39767.0605,
|
| 2782 |
+
"train_samples_per_second": 0.398,
|
| 2783 |
+
"train_steps_per_second": 0.1
|
| 2784 |
}
|
| 2785 |
],
|
| 2786 |
+
"logging_steps": 10,
|
| 2787 |
+
"max_steps": 3959,
|
| 2788 |
"num_input_tokens_seen": 0,
|
| 2789 |
"num_train_epochs": 1,
|
| 2790 |
+
"save_steps": 3959,
|
| 2791 |
"stateful_callbacks": {
|
| 2792 |
"TrainerControl": {
|
| 2793 |
"args": {
|
|
|
|
| 2800 |
"attributes": {}
|
| 2801 |
}
|
| 2802 |
},
|
| 2803 |
+
"total_flos": 2.3966994296890982e+17,
|
| 2804 |
+
"train_batch_size": 4,
|
| 2805 |
"trial_name": null,
|
| 2806 |
"trial_params": null
|
| 2807 |
}
|