Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +903 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,903 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 123,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.024390243902439025,
|
| 13 |
+
"grad_norm": 4.9090986251831055,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.0623,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04878048780487805,
|
| 20 |
+
"grad_norm": 4.635759353637695,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.0554,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07317073170731707,
|
| 27 |
+
"grad_norm": 4.518852233886719,
|
| 28 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 29 |
+
"loss": 0.9605,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0975609756097561,
|
| 34 |
+
"grad_norm": 3.9740612506866455,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 0.9034,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.12195121951219512,
|
| 41 |
+
"grad_norm": 2.2463619709014893,
|
| 42 |
+
"learning_rate": 1.9996515418688493e-05,
|
| 43 |
+
"loss": 0.9062,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14634146341463414,
|
| 48 |
+
"grad_norm": 3.367905616760254,
|
| 49 |
+
"learning_rate": 1.998606410321534e-05,
|
| 50 |
+
"loss": 0.915,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.17073170731707318,
|
| 55 |
+
"grad_norm": 2.0899763107299805,
|
| 56 |
+
"learning_rate": 1.9968653337272262e-05,
|
| 57 |
+
"loss": 0.8586,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1951219512195122,
|
| 62 |
+
"grad_norm": 1.674285650253296,
|
| 63 |
+
"learning_rate": 1.9944295254705187e-05,
|
| 64 |
+
"loss": 0.8489,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.21951219512195122,
|
| 69 |
+
"grad_norm": 1.6163828372955322,
|
| 70 |
+
"learning_rate": 1.9913006831057967e-05,
|
| 71 |
+
"loss": 0.9198,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.24390243902439024,
|
| 76 |
+
"grad_norm": 1.5275254249572754,
|
| 77 |
+
"learning_rate": 1.9874809871741877e-05,
|
| 78 |
+
"loss": 0.8248,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2682926829268293,
|
| 83 |
+
"grad_norm": 1.4473155736923218,
|
| 84 |
+
"learning_rate": 1.982973099683902e-05,
|
| 85 |
+
"loss": 0.8372,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2926829268292683,
|
| 90 |
+
"grad_norm": 1.5344518423080444,
|
| 91 |
+
"learning_rate": 1.977780162255041e-05,
|
| 92 |
+
"loss": 0.8359,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3170731707317073,
|
| 97 |
+
"grad_norm": 1.608710527420044,
|
| 98 |
+
"learning_rate": 1.9719057939301477e-05,
|
| 99 |
+
"loss": 0.8694,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.34146341463414637,
|
| 104 |
+
"grad_norm": 1.820181131362915,
|
| 105 |
+
"learning_rate": 1.9653540886520387e-05,
|
| 106 |
+
"loss": 0.8495,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.36585365853658536,
|
| 111 |
+
"grad_norm": 1.4957431554794312,
|
| 112 |
+
"learning_rate": 1.9581296124106682e-05,
|
| 113 |
+
"loss": 0.9072,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3902439024390244,
|
| 118 |
+
"grad_norm": 1.5328874588012695,
|
| 119 |
+
"learning_rate": 1.9502374000610152e-05,
|
| 120 |
+
"loss": 0.8843,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4146341463414634,
|
| 125 |
+
"grad_norm": 1.4412930011749268,
|
| 126 |
+
"learning_rate": 1.941682951814212e-05,
|
| 127 |
+
"loss": 0.8477,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.43902439024390244,
|
| 132 |
+
"grad_norm": 1.4690426588058472,
|
| 133 |
+
"learning_rate": 1.932472229404356e-05,
|
| 134 |
+
"loss": 0.9151,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4634146341463415,
|
| 139 |
+
"grad_norm": 1.4018275737762451,
|
| 140 |
+
"learning_rate": 1.922611651933683e-05,
|
| 141 |
+
"loss": 0.8409,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4878048780487805,
|
| 146 |
+
"grad_norm": 1.4043678045272827,
|
| 147 |
+
"learning_rate": 1.912108091398988e-05,
|
| 148 |
+
"loss": 0.8653,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5121951219512195,
|
| 153 |
+
"grad_norm": 1.425691843032837,
|
| 154 |
+
"learning_rate": 1.900968867902419e-05,
|
| 155 |
+
"loss": 0.8408,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5365853658536586,
|
| 160 |
+
"grad_norm": 1.3871020078659058,
|
| 161 |
+
"learning_rate": 1.8892017445499812e-05,
|
| 162 |
+
"loss": 0.895,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5609756097560976,
|
| 167 |
+
"grad_norm": 1.4709112644195557,
|
| 168 |
+
"learning_rate": 1.876814922041299e-05,
|
| 169 |
+
"loss": 0.8058,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5853658536585366,
|
| 174 |
+
"grad_norm": 1.745498538017273,
|
| 175 |
+
"learning_rate": 1.8638170329544164e-05,
|
| 176 |
+
"loss": 0.8957,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6097560975609756,
|
| 181 |
+
"grad_norm": 1.4970194101333618,
|
| 182 |
+
"learning_rate": 1.8502171357296144e-05,
|
| 183 |
+
"loss": 0.8246,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6341463414634146,
|
| 188 |
+
"grad_norm": 1.304500937461853,
|
| 189 |
+
"learning_rate": 1.8360247083564343e-05,
|
| 190 |
+
"loss": 0.8601,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6585365853658537,
|
| 195 |
+
"grad_norm": 1.2432986497879028,
|
| 196 |
+
"learning_rate": 1.8212496417683135e-05,
|
| 197 |
+
"loss": 0.7908,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6829268292682927,
|
| 202 |
+
"grad_norm": 1.2986809015274048,
|
| 203 |
+
"learning_rate": 1.805902232949435e-05,
|
| 204 |
+
"loss": 0.8569,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7073170731707317,
|
| 209 |
+
"grad_norm": 1.4464459419250488,
|
| 210 |
+
"learning_rate": 1.789993177758588e-05,
|
| 211 |
+
"loss": 0.8593,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7317073170731707,
|
| 216 |
+
"grad_norm": 1.3145807981491089,
|
| 217 |
+
"learning_rate": 1.773533563475053e-05,
|
| 218 |
+
"loss": 0.7978,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7560975609756098,
|
| 223 |
+
"grad_norm": 1.3444403409957886,
|
| 224 |
+
"learning_rate": 1.7565348610716963e-05,
|
| 225 |
+
"loss": 0.8448,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7804878048780488,
|
| 230 |
+
"grad_norm": 1.39029061794281,
|
| 231 |
+
"learning_rate": 1.7390089172206594e-05,
|
| 232 |
+
"loss": 0.8832,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8048780487804879,
|
| 237 |
+
"grad_norm": 1.4577229022979736,
|
| 238 |
+
"learning_rate": 1.720967946037225e-05,
|
| 239 |
+
"loss": 0.742,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8292682926829268,
|
| 244 |
+
"grad_norm": 1.2066192626953125,
|
| 245 |
+
"learning_rate": 1.7024245205675986e-05,
|
| 246 |
+
"loss": 0.7554,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8536585365853658,
|
| 251 |
+
"grad_norm": 1.329733967781067,
|
| 252 |
+
"learning_rate": 1.6833915640265485e-05,
|
| 253 |
+
"loss": 0.789,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8780487804878049,
|
| 258 |
+
"grad_norm": 1.2490196228027344,
|
| 259 |
+
"learning_rate": 1.6638823407910085e-05,
|
| 260 |
+
"loss": 0.7634,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9024390243902439,
|
| 265 |
+
"grad_norm": 1.3903274536132812,
|
| 266 |
+
"learning_rate": 1.6439104471559157e-05,
|
| 267 |
+
"loss": 0.809,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.926829268292683,
|
| 272 |
+
"grad_norm": 1.3177831172943115,
|
| 273 |
+
"learning_rate": 1.6234898018587336e-05,
|
| 274 |
+
"loss": 0.807,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.9512195121951219,
|
| 279 |
+
"grad_norm": 1.3945300579071045,
|
| 280 |
+
"learning_rate": 1.6026346363792565e-05,
|
| 281 |
+
"loss": 0.846,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.975609756097561,
|
| 286 |
+
"grad_norm": 1.5534132719039917,
|
| 287 |
+
"learning_rate": 1.58135948502146e-05,
|
| 288 |
+
"loss": 0.7778,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0,
|
| 293 |
+
"grad_norm": 1.185361385345459,
|
| 294 |
+
"learning_rate": 1.5596791747843083e-05,
|
| 295 |
+
"loss": 0.6984,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.024390243902439,
|
| 300 |
+
"grad_norm": 1.4634662866592407,
|
| 301 |
+
"learning_rate": 1.5376088150285777e-05,
|
| 302 |
+
"loss": 0.5167,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.048780487804878,
|
| 307 |
+
"grad_norm": 1.5522443056106567,
|
| 308 |
+
"learning_rate": 1.515163786946896e-05,
|
| 309 |
+
"loss": 0.4939,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.0731707317073171,
|
| 314 |
+
"grad_norm": 1.3210049867630005,
|
| 315 |
+
"learning_rate": 1.4923597328443423e-05,
|
| 316 |
+
"loss": 0.4859,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.0975609756097562,
|
| 321 |
+
"grad_norm": 1.2081218957901,
|
| 322 |
+
"learning_rate": 1.4692125452370664e-05,
|
| 323 |
+
"loss": 0.4547,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.1219512195121952,
|
| 328 |
+
"grad_norm": 1.2935004234313965,
|
| 329 |
+
"learning_rate": 1.4457383557765385e-05,
|
| 330 |
+
"loss": 0.4026,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.146341463414634,
|
| 335 |
+
"grad_norm": 1.385277271270752,
|
| 336 |
+
"learning_rate": 1.4219535240071378e-05,
|
| 337 |
+
"loss": 0.3798,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.170731707317073,
|
| 342 |
+
"grad_norm": 1.3052303791046143,
|
| 343 |
+
"learning_rate": 1.397874625964921e-05,
|
| 344 |
+
"loss": 0.3963,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.1951219512195121,
|
| 349 |
+
"grad_norm": 1.296340823173523,
|
| 350 |
+
"learning_rate": 1.3735184426255117e-05,
|
| 351 |
+
"loss": 0.3943,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.2195121951219512,
|
| 356 |
+
"grad_norm": 1.4150645732879639,
|
| 357 |
+
"learning_rate": 1.348901948209167e-05,
|
| 358 |
+
"loss": 0.4199,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.2439024390243902,
|
| 363 |
+
"grad_norm": 1.5740160942077637,
|
| 364 |
+
"learning_rate": 1.324042298351166e-05,
|
| 365 |
+
"loss": 0.4414,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.2682926829268293,
|
| 370 |
+
"grad_norm": 1.2712550163269043,
|
| 371 |
+
"learning_rate": 1.2989568181457704e-05,
|
| 372 |
+
"loss": 0.3815,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.2926829268292683,
|
| 377 |
+
"grad_norm": 1.4063764810562134,
|
| 378 |
+
"learning_rate": 1.2736629900720832e-05,
|
| 379 |
+
"loss": 0.4403,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.3170731707317074,
|
| 384 |
+
"grad_norm": 1.2893191576004028,
|
| 385 |
+
"learning_rate": 1.248178441810224e-05,
|
| 386 |
+
"loss": 0.4368,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.3414634146341464,
|
| 391 |
+
"grad_norm": 1.3080397844314575,
|
| 392 |
+
"learning_rate": 1.2225209339563144e-05,
|
| 393 |
+
"loss": 0.4048,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.3658536585365852,
|
| 398 |
+
"grad_norm": 1.3232578039169312,
|
| 399 |
+
"learning_rate": 1.1967083476448282e-05,
|
| 400 |
+
"loss": 0.3798,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.3902439024390243,
|
| 405 |
+
"grad_norm": 1.289511799812317,
|
| 406 |
+
"learning_rate": 1.1707586720869375e-05,
|
| 407 |
+
"loss": 0.4385,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.4146341463414633,
|
| 412 |
+
"grad_norm": 1.2744817733764648,
|
| 413 |
+
"learning_rate": 1.1446899920335407e-05,
|
| 414 |
+
"loss": 0.42,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.4390243902439024,
|
| 419 |
+
"grad_norm": 1.2086424827575684,
|
| 420 |
+
"learning_rate": 1.118520475171703e-05,
|
| 421 |
+
"loss": 0.4505,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.4634146341463414,
|
| 426 |
+
"grad_norm": 1.1722941398620605,
|
| 427 |
+
"learning_rate": 1.092268359463302e-05,
|
| 428 |
+
"loss": 0.3888,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.4878048780487805,
|
| 433 |
+
"grad_norm": 1.2801525592803955,
|
| 434 |
+
"learning_rate": 1.0659519404346955e-05,
|
| 435 |
+
"loss": 0.3891,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.5121951219512195,
|
| 440 |
+
"grad_norm": 1.2981622219085693,
|
| 441 |
+
"learning_rate": 1.0395895584262696e-05,
|
| 442 |
+
"loss": 0.4271,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.5365853658536586,
|
| 447 |
+
"grad_norm": 1.2647910118103027,
|
| 448 |
+
"learning_rate": 1.013199585810759e-05,
|
| 449 |
+
"loss": 0.4093,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.5609756097560976,
|
| 454 |
+
"grad_norm": 1.0671759843826294,
|
| 455 |
+
"learning_rate": 9.868004141892412e-06,
|
| 456 |
+
"loss": 0.3422,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.5853658536585367,
|
| 461 |
+
"grad_norm": 1.1954787969589233,
|
| 462 |
+
"learning_rate": 9.604104415737309e-06,
|
| 463 |
+
"loss": 0.4171,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.6097560975609757,
|
| 468 |
+
"grad_norm": 1.2861173152923584,
|
| 469 |
+
"learning_rate": 9.340480595653047e-06,
|
| 470 |
+
"loss": 0.4206,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.6341463414634148,
|
| 475 |
+
"grad_norm": 1.2733056545257568,
|
| 476 |
+
"learning_rate": 9.07731640536698e-06,
|
| 477 |
+
"loss": 0.3932,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.6585365853658538,
|
| 482 |
+
"grad_norm": 1.2633076906204224,
|
| 483 |
+
"learning_rate": 8.814795248282974e-06,
|
| 484 |
+
"loss": 0.4233,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.6829268292682928,
|
| 489 |
+
"grad_norm": 1.2530993223190308,
|
| 490 |
+
"learning_rate": 8.553100079664598e-06,
|
| 491 |
+
"loss": 0.4097,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.7073170731707317,
|
| 496 |
+
"grad_norm": 1.2529405355453491,
|
| 497 |
+
"learning_rate": 8.292413279130625e-06,
|
| 498 |
+
"loss": 0.4349,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.7317073170731707,
|
| 503 |
+
"grad_norm": 1.2502950429916382,
|
| 504 |
+
"learning_rate": 8.03291652355172e-06,
|
| 505 |
+
"loss": 0.4108,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.7560975609756098,
|
| 510 |
+
"grad_norm": 1.3019890785217285,
|
| 511 |
+
"learning_rate": 7.774790660436857e-06,
|
| 512 |
+
"loss": 0.4005,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.7804878048780488,
|
| 517 |
+
"grad_norm": 1.2661802768707275,
|
| 518 |
+
"learning_rate": 7.518215581897763e-06,
|
| 519 |
+
"loss": 0.3406,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.8048780487804879,
|
| 524 |
+
"grad_norm": 1.3292313814163208,
|
| 525 |
+
"learning_rate": 7.263370099279173e-06,
|
| 526 |
+
"loss": 0.3701,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.8292682926829267,
|
| 531 |
+
"grad_norm": 1.1148855686187744,
|
| 532 |
+
"learning_rate": 7.010431818542298e-06,
|
| 533 |
+
"loss": 0.3798,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.8536585365853657,
|
| 538 |
+
"grad_norm": 1.256280779838562,
|
| 539 |
+
"learning_rate": 6.759577016488343e-06,
|
| 540 |
+
"loss": 0.3923,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.8780487804878048,
|
| 545 |
+
"grad_norm": 1.257556438446045,
|
| 546 |
+
"learning_rate": 6.510980517908334e-06,
|
| 547 |
+
"loss": 0.407,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.9024390243902438,
|
| 552 |
+
"grad_norm": 1.188286304473877,
|
| 553 |
+
"learning_rate": 6.264815573744884e-06,
|
| 554 |
+
"loss": 0.3958,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.9268292682926829,
|
| 559 |
+
"grad_norm": 1.2357051372528076,
|
| 560 |
+
"learning_rate": 6.021253740350793e-06,
|
| 561 |
+
"loss": 0.3868,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.951219512195122,
|
| 566 |
+
"grad_norm": 1.297119140625,
|
| 567 |
+
"learning_rate": 5.780464759928623e-06,
|
| 568 |
+
"loss": 0.425,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.975609756097561,
|
| 573 |
+
"grad_norm": 1.2975163459777832,
|
| 574 |
+
"learning_rate": 5.542616442234618e-06,
|
| 575 |
+
"loss": 0.4136,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.0,
|
| 580 |
+
"grad_norm": 1.0819505453109741,
|
| 581 |
+
"learning_rate": 5.307874547629339e-06,
|
| 582 |
+
"loss": 0.3263,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.024390243902439,
|
| 587 |
+
"grad_norm": 1.4304354190826416,
|
| 588 |
+
"learning_rate": 5.076402671556578e-06,
|
| 589 |
+
"loss": 0.2532,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.048780487804878,
|
| 594 |
+
"grad_norm": 1.4104777574539185,
|
| 595 |
+
"learning_rate": 4.848362130531039e-06,
|
| 596 |
+
"loss": 0.235,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.073170731707317,
|
| 601 |
+
"grad_norm": 1.2689759731292725,
|
| 602 |
+
"learning_rate": 4.623911849714226e-06,
|
| 603 |
+
"loss": 0.2078,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.097560975609756,
|
| 608 |
+
"grad_norm": 1.2290229797363281,
|
| 609 |
+
"learning_rate": 4.403208252156921e-06,
|
| 610 |
+
"loss": 0.2265,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.1219512195121952,
|
| 615 |
+
"grad_norm": 1.1399704217910767,
|
| 616 |
+
"learning_rate": 4.186405149785403e-06,
|
| 617 |
+
"loss": 0.2101,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.1463414634146343,
|
| 622 |
+
"grad_norm": 0.9104946851730347,
|
| 623 |
+
"learning_rate": 3.973653636207437e-06,
|
| 624 |
+
"loss": 0.1706,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.1707317073170733,
|
| 629 |
+
"grad_norm": 1.1334648132324219,
|
| 630 |
+
"learning_rate": 3.7651019814126656e-06,
|
| 631 |
+
"loss": 0.1692,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.1951219512195124,
|
| 636 |
+
"grad_norm": 1.098779559135437,
|
| 637 |
+
"learning_rate": 3.560895528440844e-06,
|
| 638 |
+
"loss": 0.1919,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.2195121951219514,
|
| 643 |
+
"grad_norm": 1.5511068105697632,
|
| 644 |
+
"learning_rate": 3.361176592089919e-06,
|
| 645 |
+
"loss": 0.2168,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.2439024390243905,
|
| 650 |
+
"grad_norm": 1.3016074895858765,
|
| 651 |
+
"learning_rate": 3.1660843597345137e-06,
|
| 652 |
+
"loss": 0.1801,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.2682926829268295,
|
| 657 |
+
"grad_norm": 1.7183330059051514,
|
| 658 |
+
"learning_rate": 2.975754794324015e-06,
|
| 659 |
+
"loss": 0.1992,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.292682926829268,
|
| 664 |
+
"grad_norm": 1.4131158590316772,
|
| 665 |
+
"learning_rate": 2.7903205396277546e-06,
|
| 666 |
+
"loss": 0.2049,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.317073170731707,
|
| 671 |
+
"grad_norm": 1.5485832691192627,
|
| 672 |
+
"learning_rate": 2.6099108277934105e-06,
|
| 673 |
+
"loss": 0.207,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.341463414634146,
|
| 678 |
+
"grad_norm": 1.326352834701538,
|
| 679 |
+
"learning_rate": 2.4346513892830427e-06,
|
| 680 |
+
"loss": 0.1809,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.3658536585365852,
|
| 685 |
+
"grad_norm": 1.1115378141403198,
|
| 686 |
+
"learning_rate": 2.2646643652494693e-06,
|
| 687 |
+
"loss": 0.1745,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.3902439024390243,
|
| 692 |
+
"grad_norm": 1.332322120666504,
|
| 693 |
+
"learning_rate": 2.100068222414121e-06,
|
| 694 |
+
"loss": 0.2028,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.4146341463414633,
|
| 699 |
+
"grad_norm": 1.0843946933746338,
|
| 700 |
+
"learning_rate": 1.9409776705056514e-06,
|
| 701 |
+
"loss": 0.1786,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.4390243902439024,
|
| 706 |
+
"grad_norm": 1.0428590774536133,
|
| 707 |
+
"learning_rate": 1.7875035823168641e-06,
|
| 708 |
+
"loss": 0.1782,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.4634146341463414,
|
| 713 |
+
"grad_norm": 1.1619765758514404,
|
| 714 |
+
"learning_rate": 1.6397529164356606e-06,
|
| 715 |
+
"loss": 0.1835,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.4878048780487805,
|
| 720 |
+
"grad_norm": 1.1972867250442505,
|
| 721 |
+
"learning_rate": 1.4978286427038602e-06,
|
| 722 |
+
"loss": 0.2128,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.5121951219512195,
|
| 727 |
+
"grad_norm": 1.0437191724777222,
|
| 728 |
+
"learning_rate": 1.3618296704558364e-06,
|
| 729 |
+
"loss": 0.1746,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.5365853658536586,
|
| 734 |
+
"grad_norm": 1.094221591949463,
|
| 735 |
+
"learning_rate": 1.2318507795870138e-06,
|
| 736 |
+
"loss": 0.2146,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.5609756097560976,
|
| 741 |
+
"grad_norm": 1.0217782258987427,
|
| 742 |
+
"learning_rate": 1.1079825545001887e-06,
|
| 743 |
+
"loss": 0.1896,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.5853658536585367,
|
| 748 |
+
"grad_norm": 1.0593501329421997,
|
| 749 |
+
"learning_rate": 9.903113209758098e-07,
|
| 750 |
+
"loss": 0.1949,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.6097560975609757,
|
| 755 |
+
"grad_norm": 1.020802617073059,
|
| 756 |
+
"learning_rate": 8.789190860101226e-07,
|
| 757 |
+
"loss": 0.2053,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.6341463414634148,
|
| 762 |
+
"grad_norm": 1.0524072647094727,
|
| 763 |
+
"learning_rate": 7.738834806631712e-07,
|
| 764 |
+
"loss": 0.1904,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.658536585365854,
|
| 769 |
+
"grad_norm": 1.0518467426300049,
|
| 770 |
+
"learning_rate": 6.752777059564431e-07,
|
| 771 |
+
"loss": 0.1854,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.682926829268293,
|
| 776 |
+
"grad_norm": 1.0113893747329712,
|
| 777 |
+
"learning_rate": 5.831704818578842e-07,
|
| 778 |
+
"loss": 0.177,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.7073170731707314,
|
| 783 |
+
"grad_norm": 1.290737509727478,
|
| 784 |
+
"learning_rate": 4.976259993898503e-07,
|
| 785 |
+
"loss": 0.2217,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.7317073170731705,
|
| 790 |
+
"grad_norm": 1.0283629894256592,
|
| 791 |
+
"learning_rate": 4.187038758933204e-07,
|
| 792 |
+
"loss": 0.1706,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.7560975609756095,
|
| 797 |
+
"grad_norm": 1.0521726608276367,
|
| 798 |
+
"learning_rate": 3.4645911347961357e-07,
|
| 799 |
+
"loss": 0.1787,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.7804878048780486,
|
| 804 |
+
"grad_norm": 1.1081814765930176,
|
| 805 |
+
"learning_rate": 2.809420606985236e-07,
|
| 806 |
+
"loss": 0.1932,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.8048780487804876,
|
| 811 |
+
"grad_norm": 1.0919123888015747,
|
| 812 |
+
"learning_rate": 2.2219837744959284e-07,
|
| 813 |
+
"loss": 0.191,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.8292682926829267,
|
| 818 |
+
"grad_norm": 1.1104843616485596,
|
| 819 |
+
"learning_rate": 1.7026900316098217e-07,
|
| 820 |
+
"loss": 0.1927,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.8536585365853657,
|
| 825 |
+
"grad_norm": 0.9853484034538269,
|
| 826 |
+
"learning_rate": 1.2519012825812804e-07,
|
| 827 |
+
"loss": 0.1629,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.8780487804878048,
|
| 832 |
+
"grad_norm": 2.889988422393799,
|
| 833 |
+
"learning_rate": 8.699316894203225e-08,
|
| 834 |
+
"loss": 0.2047,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.902439024390244,
|
| 839 |
+
"grad_norm": 0.8976815342903137,
|
| 840 |
+
"learning_rate": 5.5704745294815624e-08,
|
| 841 |
+
"loss": 0.1651,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.926829268292683,
|
| 846 |
+
"grad_norm": 1.0708858966827393,
|
| 847 |
+
"learning_rate": 3.134666272774034e-08,
|
| 848 |
+
"loss": 0.1905,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.951219512195122,
|
| 853 |
+
"grad_norm": 1.0499634742736816,
|
| 854 |
+
"learning_rate": 1.3935896784663671e-08,
|
| 855 |
+
"loss": 0.1764,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.975609756097561,
|
| 860 |
+
"grad_norm": 1.1612313985824585,
|
| 861 |
+
"learning_rate": 3.4845813115114147e-09,
|
| 862 |
+
"loss": 0.1966,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.0,
|
| 867 |
+
"grad_norm": 1.011826753616333,
|
| 868 |
+
"learning_rate": 0.0,
|
| 869 |
+
"loss": 0.1712,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 3.0,
|
| 874 |
+
"step": 123,
|
| 875 |
+
"total_flos": 2.673620881065902e+17,
|
| 876 |
+
"train_loss": 0.48636769339805697,
|
| 877 |
+
"train_runtime": 978.8062,
|
| 878 |
+
"train_samples_per_second": 15.938,
|
| 879 |
+
"train_steps_per_second": 0.126
|
| 880 |
+
}
|
| 881 |
+
],
|
| 882 |
+
"logging_steps": 1,
|
| 883 |
+
"max_steps": 123,
|
| 884 |
+
"num_input_tokens_seen": 0,
|
| 885 |
+
"num_train_epochs": 3,
|
| 886 |
+
"save_steps": 0,
|
| 887 |
+
"stateful_callbacks": {
|
| 888 |
+
"TrainerControl": {
|
| 889 |
+
"args": {
|
| 890 |
+
"should_epoch_stop": false,
|
| 891 |
+
"should_evaluate": false,
|
| 892 |
+
"should_log": false,
|
| 893 |
+
"should_save": true,
|
| 894 |
+
"should_training_stop": true
|
| 895 |
+
},
|
| 896 |
+
"attributes": {}
|
| 897 |
+
}
|
| 898 |
+
},
|
| 899 |
+
"total_flos": 2.673620881065902e+17,
|
| 900 |
+
"train_batch_size": 32,
|
| 901 |
+
"trial_name": null,
|
| 902 |
+
"trial_params": null
|
| 903 |
+
}
|