Upload trainer_state.json
Browse files- trainer_state.json +139 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,139 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 9552,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 4.738274706867672e-05,
|
| 13 |
+
"loss": 1.5983,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.31,
|
| 18 |
+
"learning_rate": 4.4765494137353433e-05,
|
| 19 |
+
"loss": 1.0414,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.47,
|
| 24 |
+
"learning_rate": 4.2148241206030156e-05,
|
| 25 |
+
"loss": 0.8918,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.63,
|
| 30 |
+
"learning_rate": 3.953098827470687e-05,
|
| 31 |
+
"loss": 0.8029,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.79,
|
| 36 |
+
"learning_rate": 3.691373534338359e-05,
|
| 37 |
+
"loss": 0.7394,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.94,
|
| 42 |
+
"learning_rate": 3.42964824120603e-05,
|
| 43 |
+
"loss": 0.7008,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.1,
|
| 48 |
+
"learning_rate": 3.1679229480737025e-05,
|
| 49 |
+
"loss": 0.6665,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.26,
|
| 54 |
+
"learning_rate": 2.906197654941374e-05,
|
| 55 |
+
"loss": 0.6283,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.41,
|
| 60 |
+
"learning_rate": 2.6444723618090456e-05,
|
| 61 |
+
"loss": 0.6072,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.57,
|
| 66 |
+
"learning_rate": 2.382747068676717e-05,
|
| 67 |
+
"loss": 0.5773,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.73,
|
| 72 |
+
"learning_rate": 2.1210217755443887e-05,
|
| 73 |
+
"loss": 0.5638,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.88,
|
| 78 |
+
"learning_rate": 1.8592964824120602e-05,
|
| 79 |
+
"loss": 0.5515,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.04,
|
| 84 |
+
"learning_rate": 1.597571189279732e-05,
|
| 85 |
+
"loss": 0.529,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.2,
|
| 90 |
+
"learning_rate": 1.3358458961474038e-05,
|
| 91 |
+
"loss": 0.5156,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.36,
|
| 96 |
+
"learning_rate": 1.0741206030150754e-05,
|
| 97 |
+
"loss": 0.501,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.51,
|
| 102 |
+
"learning_rate": 8.123953098827471e-06,
|
| 103 |
+
"loss": 0.4886,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.67,
|
| 108 |
+
"learning_rate": 5.5067001675041875e-06,
|
| 109 |
+
"loss": 0.4748,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.83,
|
| 114 |
+
"learning_rate": 2.8894472361809047e-06,
|
| 115 |
+
"loss": 0.4751,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.98,
|
| 120 |
+
"learning_rate": 2.7219430485762144e-07,
|
| 121 |
+
"loss": 0.4687,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.0,
|
| 126 |
+
"step": 9552,
|
| 127 |
+
"total_flos": 8.908025376224051e+16,
|
| 128 |
+
"train_loss": 0.6736920107549159,
|
| 129 |
+
"train_runtime": 5995.2753,
|
| 130 |
+
"train_samples_per_second": 15.931,
|
| 131 |
+
"train_steps_per_second": 1.593
|
| 132 |
+
}
|
| 133 |
+
],
|
| 134 |
+
"max_steps": 9552,
|
| 135 |
+
"num_train_epochs": 3,
|
| 136 |
+
"total_flos": 8.908025376224051e+16,
|
| 137 |
+
"trial_name": null,
|
| 138 |
+
"trial_params": null
|
| 139 |
+
}
|