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": 4.83710241317749,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 1.1925,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04878048780487805,
|
| 20 |
+
"grad_norm": 4.962577819824219,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 1.2116,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07317073170731707,
|
| 27 |
+
"grad_norm": 4.546571254730225,
|
| 28 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 29 |
+
"loss": 1.1354,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0975609756097561,
|
| 34 |
+
"grad_norm": 3.580547332763672,
|
| 35 |
+
"learning_rate": 2e-05,
|
| 36 |
+
"loss": 0.988,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.12195121951219512,
|
| 41 |
+
"grad_norm": 2.2663285732269287,
|
| 42 |
+
"learning_rate": 1.9996515418688493e-05,
|
| 43 |
+
"loss": 1.0428,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14634146341463414,
|
| 48 |
+
"grad_norm": 3.42545485496521,
|
| 49 |
+
"learning_rate": 1.998606410321534e-05,
|
| 50 |
+
"loss": 1.0216,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.17073170731707318,
|
| 55 |
+
"grad_norm": 2.05627703666687,
|
| 56 |
+
"learning_rate": 1.9968653337272262e-05,
|
| 57 |
+
"loss": 1.0113,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1951219512195122,
|
| 62 |
+
"grad_norm": 1.698421597480774,
|
| 63 |
+
"learning_rate": 1.9944295254705187e-05,
|
| 64 |
+
"loss": 0.9708,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.21951219512195122,
|
| 69 |
+
"grad_norm": 1.552731990814209,
|
| 70 |
+
"learning_rate": 1.9913006831057967e-05,
|
| 71 |
+
"loss": 0.9859,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.24390243902439024,
|
| 76 |
+
"grad_norm": 1.6869672536849976,
|
| 77 |
+
"learning_rate": 1.9874809871741877e-05,
|
| 78 |
+
"loss": 1.0283,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2682926829268293,
|
| 83 |
+
"grad_norm": 1.5245341062545776,
|
| 84 |
+
"learning_rate": 1.982973099683902e-05,
|
| 85 |
+
"loss": 0.8768,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.2926829268292683,
|
| 90 |
+
"grad_norm": 1.4050577878952026,
|
| 91 |
+
"learning_rate": 1.977780162255041e-05,
|
| 92 |
+
"loss": 0.9131,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3170731707317073,
|
| 97 |
+
"grad_norm": 1.511888861656189,
|
| 98 |
+
"learning_rate": 1.9719057939301477e-05,
|
| 99 |
+
"loss": 0.8791,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.34146341463414637,
|
| 104 |
+
"grad_norm": 1.5156327486038208,
|
| 105 |
+
"learning_rate": 1.9653540886520387e-05,
|
| 106 |
+
"loss": 0.9465,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.36585365853658536,
|
| 111 |
+
"grad_norm": 1.479278564453125,
|
| 112 |
+
"learning_rate": 1.9581296124106682e-05,
|
| 113 |
+
"loss": 0.9486,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3902439024390244,
|
| 118 |
+
"grad_norm": 1.3854690790176392,
|
| 119 |
+
"learning_rate": 1.9502374000610152e-05,
|
| 120 |
+
"loss": 0.8743,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4146341463414634,
|
| 125 |
+
"grad_norm": 1.401302695274353,
|
| 126 |
+
"learning_rate": 1.941682951814212e-05,
|
| 127 |
+
"loss": 0.939,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.43902439024390244,
|
| 132 |
+
"grad_norm": 1.4647853374481201,
|
| 133 |
+
"learning_rate": 1.932472229404356e-05,
|
| 134 |
+
"loss": 0.9477,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4634146341463415,
|
| 139 |
+
"grad_norm": 1.4645260572433472,
|
| 140 |
+
"learning_rate": 1.922611651933683e-05,
|
| 141 |
+
"loss": 0.9322,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.4878048780487805,
|
| 146 |
+
"grad_norm": 1.453171968460083,
|
| 147 |
+
"learning_rate": 1.912108091398988e-05,
|
| 148 |
+
"loss": 0.9804,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.5121951219512195,
|
| 153 |
+
"grad_norm": 1.3499491214752197,
|
| 154 |
+
"learning_rate": 1.900968867902419e-05,
|
| 155 |
+
"loss": 0.8802,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5365853658536586,
|
| 160 |
+
"grad_norm": 1.2831861972808838,
|
| 161 |
+
"learning_rate": 1.8892017445499812e-05,
|
| 162 |
+
"loss": 0.8841,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.5609756097560976,
|
| 167 |
+
"grad_norm": 1.5657117366790771,
|
| 168 |
+
"learning_rate": 1.876814922041299e-05,
|
| 169 |
+
"loss": 0.9256,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.5853658536585366,
|
| 174 |
+
"grad_norm": 1.4021592140197754,
|
| 175 |
+
"learning_rate": 1.8638170329544164e-05,
|
| 176 |
+
"loss": 0.9621,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.6097560975609756,
|
| 181 |
+
"grad_norm": 1.3671259880065918,
|
| 182 |
+
"learning_rate": 1.8502171357296144e-05,
|
| 183 |
+
"loss": 0.8767,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.6341463414634146,
|
| 188 |
+
"grad_norm": 1.3964266777038574,
|
| 189 |
+
"learning_rate": 1.8360247083564343e-05,
|
| 190 |
+
"loss": 0.9264,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.6585365853658537,
|
| 195 |
+
"grad_norm": 1.2335838079452515,
|
| 196 |
+
"learning_rate": 1.8212496417683135e-05,
|
| 197 |
+
"loss": 0.9141,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.6829268292682927,
|
| 202 |
+
"grad_norm": 1.3922336101531982,
|
| 203 |
+
"learning_rate": 1.805902232949435e-05,
|
| 204 |
+
"loss": 0.9302,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.7073170731707317,
|
| 209 |
+
"grad_norm": 1.3955825567245483,
|
| 210 |
+
"learning_rate": 1.789993177758588e-05,
|
| 211 |
+
"loss": 0.8731,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.7317073170731707,
|
| 216 |
+
"grad_norm": 1.4211026430130005,
|
| 217 |
+
"learning_rate": 1.773533563475053e-05,
|
| 218 |
+
"loss": 0.9748,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.7560975609756098,
|
| 223 |
+
"grad_norm": 1.5928425788879395,
|
| 224 |
+
"learning_rate": 1.7565348610716963e-05,
|
| 225 |
+
"loss": 0.9736,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.7804878048780488,
|
| 230 |
+
"grad_norm": 1.4811115264892578,
|
| 231 |
+
"learning_rate": 1.7390089172206594e-05,
|
| 232 |
+
"loss": 0.929,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.8048780487804879,
|
| 237 |
+
"grad_norm": 1.6273949146270752,
|
| 238 |
+
"learning_rate": 1.720967946037225e-05,
|
| 239 |
+
"loss": 0.952,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.8292682926829268,
|
| 244 |
+
"grad_norm": 1.3416365385055542,
|
| 245 |
+
"learning_rate": 1.7024245205675986e-05,
|
| 246 |
+
"loss": 0.8509,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8536585365853658,
|
| 251 |
+
"grad_norm": 1.4344054460525513,
|
| 252 |
+
"learning_rate": 1.6833915640265485e-05,
|
| 253 |
+
"loss": 0.9221,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.8780487804878049,
|
| 258 |
+
"grad_norm": 1.462561011314392,
|
| 259 |
+
"learning_rate": 1.6638823407910085e-05,
|
| 260 |
+
"loss": 0.9638,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.9024390243902439,
|
| 265 |
+
"grad_norm": 1.4724876880645752,
|
| 266 |
+
"learning_rate": 1.6439104471559157e-05,
|
| 267 |
+
"loss": 1.0171,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.926829268292683,
|
| 272 |
+
"grad_norm": 1.3633418083190918,
|
| 273 |
+
"learning_rate": 1.6234898018587336e-05,
|
| 274 |
+
"loss": 0.9402,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.9512195121951219,
|
| 279 |
+
"grad_norm": 1.357893943786621,
|
| 280 |
+
"learning_rate": 1.6026346363792565e-05,
|
| 281 |
+
"loss": 0.9221,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.975609756097561,
|
| 286 |
+
"grad_norm": 1.3580987453460693,
|
| 287 |
+
"learning_rate": 1.58135948502146e-05,
|
| 288 |
+
"loss": 0.8952,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0,
|
| 293 |
+
"grad_norm": 1.2962157726287842,
|
| 294 |
+
"learning_rate": 1.5596791747843083e-05,
|
| 295 |
+
"loss": 0.8754,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.024390243902439,
|
| 300 |
+
"grad_norm": 1.7266184091567993,
|
| 301 |
+
"learning_rate": 1.5376088150285777e-05,
|
| 302 |
+
"loss": 0.5501,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.048780487804878,
|
| 307 |
+
"grad_norm": 1.4698433876037598,
|
| 308 |
+
"learning_rate": 1.515163786946896e-05,
|
| 309 |
+
"loss": 0.4683,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.0731707317073171,
|
| 314 |
+
"grad_norm": 1.244774580001831,
|
| 315 |
+
"learning_rate": 1.4923597328443423e-05,
|
| 316 |
+
"loss": 0.52,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.0975609756097562,
|
| 321 |
+
"grad_norm": 1.2581526041030884,
|
| 322 |
+
"learning_rate": 1.4692125452370664e-05,
|
| 323 |
+
"loss": 0.4701,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.1219512195121952,
|
| 328 |
+
"grad_norm": 1.4708521366119385,
|
| 329 |
+
"learning_rate": 1.4457383557765385e-05,
|
| 330 |
+
"loss": 0.4902,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.146341463414634,
|
| 335 |
+
"grad_norm": 1.6270897388458252,
|
| 336 |
+
"learning_rate": 1.4219535240071378e-05,
|
| 337 |
+
"loss": 0.5485,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.170731707317073,
|
| 342 |
+
"grad_norm": 1.637870192527771,
|
| 343 |
+
"learning_rate": 1.397874625964921e-05,
|
| 344 |
+
"loss": 0.5086,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.1951219512195121,
|
| 349 |
+
"grad_norm": 1.3685152530670166,
|
| 350 |
+
"learning_rate": 1.3735184426255117e-05,
|
| 351 |
+
"loss": 0.4655,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.2195121951219512,
|
| 356 |
+
"grad_norm": 1.4286671876907349,
|
| 357 |
+
"learning_rate": 1.348901948209167e-05,
|
| 358 |
+
"loss": 0.4786,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.2439024390243902,
|
| 363 |
+
"grad_norm": 1.3230304718017578,
|
| 364 |
+
"learning_rate": 1.324042298351166e-05,
|
| 365 |
+
"loss": 0.4821,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.2682926829268293,
|
| 370 |
+
"grad_norm": 1.3360493183135986,
|
| 371 |
+
"learning_rate": 1.2989568181457704e-05,
|
| 372 |
+
"loss": 0.4505,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.2926829268292683,
|
| 377 |
+
"grad_norm": 1.3025678396224976,
|
| 378 |
+
"learning_rate": 1.2736629900720832e-05,
|
| 379 |
+
"loss": 0.4323,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.3170731707317074,
|
| 384 |
+
"grad_norm": 1.2738451957702637,
|
| 385 |
+
"learning_rate": 1.248178441810224e-05,
|
| 386 |
+
"loss": 0.4608,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.3414634146341464,
|
| 391 |
+
"grad_norm": 1.1641204357147217,
|
| 392 |
+
"learning_rate": 1.2225209339563144e-05,
|
| 393 |
+
"loss": 0.4225,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.3658536585365852,
|
| 398 |
+
"grad_norm": 1.2512654066085815,
|
| 399 |
+
"learning_rate": 1.1967083476448282e-05,
|
| 400 |
+
"loss": 0.4327,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.3902439024390243,
|
| 405 |
+
"grad_norm": 1.4062010049819946,
|
| 406 |
+
"learning_rate": 1.1707586720869375e-05,
|
| 407 |
+
"loss": 0.4687,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.4146341463414633,
|
| 412 |
+
"grad_norm": 1.4430186748504639,
|
| 413 |
+
"learning_rate": 1.1446899920335407e-05,
|
| 414 |
+
"loss": 0.4496,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.4390243902439024,
|
| 419 |
+
"grad_norm": 1.5102589130401611,
|
| 420 |
+
"learning_rate": 1.118520475171703e-05,
|
| 421 |
+
"loss": 0.521,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.4634146341463414,
|
| 426 |
+
"grad_norm": 1.4017813205718994,
|
| 427 |
+
"learning_rate": 1.092268359463302e-05,
|
| 428 |
+
"loss": 0.4825,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.4878048780487805,
|
| 433 |
+
"grad_norm": 1.4781465530395508,
|
| 434 |
+
"learning_rate": 1.0659519404346955e-05,
|
| 435 |
+
"loss": 0.4748,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.5121951219512195,
|
| 440 |
+
"grad_norm": 1.3136210441589355,
|
| 441 |
+
"learning_rate": 1.0395895584262696e-05,
|
| 442 |
+
"loss": 0.4759,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.5365853658536586,
|
| 447 |
+
"grad_norm": 1.2702844142913818,
|
| 448 |
+
"learning_rate": 1.013199585810759e-05,
|
| 449 |
+
"loss": 0.4613,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.5609756097560976,
|
| 454 |
+
"grad_norm": 1.2595293521881104,
|
| 455 |
+
"learning_rate": 9.868004141892412e-06,
|
| 456 |
+
"loss": 0.4684,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.5853658536585367,
|
| 461 |
+
"grad_norm": 1.3314820528030396,
|
| 462 |
+
"learning_rate": 9.604104415737309e-06,
|
| 463 |
+
"loss": 0.4622,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.6097560975609757,
|
| 468 |
+
"grad_norm": 1.2721946239471436,
|
| 469 |
+
"learning_rate": 9.340480595653047e-06,
|
| 470 |
+
"loss": 0.4931,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.6341463414634148,
|
| 475 |
+
"grad_norm": 1.2505251169204712,
|
| 476 |
+
"learning_rate": 9.07731640536698e-06,
|
| 477 |
+
"loss": 0.4154,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.6585365853658538,
|
| 482 |
+
"grad_norm": 1.3923282623291016,
|
| 483 |
+
"learning_rate": 8.814795248282974e-06,
|
| 484 |
+
"loss": 0.5148,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.6829268292682928,
|
| 489 |
+
"grad_norm": 1.2958779335021973,
|
| 490 |
+
"learning_rate": 8.553100079664598e-06,
|
| 491 |
+
"loss": 0.4566,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.7073170731707317,
|
| 496 |
+
"grad_norm": 1.313406229019165,
|
| 497 |
+
"learning_rate": 8.292413279130625e-06,
|
| 498 |
+
"loss": 0.4879,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.7317073170731707,
|
| 503 |
+
"grad_norm": 1.3411192893981934,
|
| 504 |
+
"learning_rate": 8.03291652355172e-06,
|
| 505 |
+
"loss": 0.4896,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.7560975609756098,
|
| 510 |
+
"grad_norm": 1.2127251625061035,
|
| 511 |
+
"learning_rate": 7.774790660436857e-06,
|
| 512 |
+
"loss": 0.4898,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.7804878048780488,
|
| 517 |
+
"grad_norm": 1.2077494859695435,
|
| 518 |
+
"learning_rate": 7.518215581897763e-06,
|
| 519 |
+
"loss": 0.4339,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.8048780487804879,
|
| 524 |
+
"grad_norm": 1.2262243032455444,
|
| 525 |
+
"learning_rate": 7.263370099279173e-06,
|
| 526 |
+
"loss": 0.4448,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.8292682926829267,
|
| 531 |
+
"grad_norm": 1.3548513650894165,
|
| 532 |
+
"learning_rate": 7.010431818542298e-06,
|
| 533 |
+
"loss": 0.4535,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.8536585365853657,
|
| 538 |
+
"grad_norm": 1.3579554557800293,
|
| 539 |
+
"learning_rate": 6.759577016488343e-06,
|
| 540 |
+
"loss": 0.4765,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.8780487804878048,
|
| 545 |
+
"grad_norm": 1.3343220949172974,
|
| 546 |
+
"learning_rate": 6.510980517908334e-06,
|
| 547 |
+
"loss": 0.4947,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.9024390243902438,
|
| 552 |
+
"grad_norm": 1.2846651077270508,
|
| 553 |
+
"learning_rate": 6.264815573744884e-06,
|
| 554 |
+
"loss": 0.4286,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.9268292682926829,
|
| 559 |
+
"grad_norm": 1.4076483249664307,
|
| 560 |
+
"learning_rate": 6.021253740350793e-06,
|
| 561 |
+
"loss": 0.4936,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.951219512195122,
|
| 566 |
+
"grad_norm": 1.6410694122314453,
|
| 567 |
+
"learning_rate": 5.780464759928623e-06,
|
| 568 |
+
"loss": 0.4424,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.975609756097561,
|
| 573 |
+
"grad_norm": 1.2024314403533936,
|
| 574 |
+
"learning_rate": 5.542616442234618e-06,
|
| 575 |
+
"loss": 0.4678,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.0,
|
| 580 |
+
"grad_norm": 1.170007348060608,
|
| 581 |
+
"learning_rate": 5.307874547629339e-06,
|
| 582 |
+
"loss": 0.4246,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.024390243902439,
|
| 587 |
+
"grad_norm": 1.3799552917480469,
|
| 588 |
+
"learning_rate": 5.076402671556578e-06,
|
| 589 |
+
"loss": 0.2486,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 2.048780487804878,
|
| 594 |
+
"grad_norm": 1.5064375400543213,
|
| 595 |
+
"learning_rate": 4.848362130531039e-06,
|
| 596 |
+
"loss": 0.2679,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.073170731707317,
|
| 601 |
+
"grad_norm": 1.1963298320770264,
|
| 602 |
+
"learning_rate": 4.623911849714226e-06,
|
| 603 |
+
"loss": 0.1989,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.097560975609756,
|
| 608 |
+
"grad_norm": 1.2926315069198608,
|
| 609 |
+
"learning_rate": 4.403208252156921e-06,
|
| 610 |
+
"loss": 0.2606,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.1219512195121952,
|
| 615 |
+
"grad_norm": 1.330957055091858,
|
| 616 |
+
"learning_rate": 4.186405149785403e-06,
|
| 617 |
+
"loss": 0.25,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.1463414634146343,
|
| 622 |
+
"grad_norm": 1.0437262058258057,
|
| 623 |
+
"learning_rate": 3.973653636207437e-06,
|
| 624 |
+
"loss": 0.2002,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.1707317073170733,
|
| 629 |
+
"grad_norm": 1.0861252546310425,
|
| 630 |
+
"learning_rate": 3.7651019814126656e-06,
|
| 631 |
+
"loss": 0.2101,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.1951219512195124,
|
| 636 |
+
"grad_norm": 1.7576727867126465,
|
| 637 |
+
"learning_rate": 3.560895528440844e-06,
|
| 638 |
+
"loss": 0.2818,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.2195121951219514,
|
| 643 |
+
"grad_norm": 1.5912959575653076,
|
| 644 |
+
"learning_rate": 3.361176592089919e-06,
|
| 645 |
+
"loss": 0.2277,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.2439024390243905,
|
| 650 |
+
"grad_norm": 1.7516396045684814,
|
| 651 |
+
"learning_rate": 3.1660843597345137e-06,
|
| 652 |
+
"loss": 0.2247,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.2682926829268295,
|
| 657 |
+
"grad_norm": 1.6577694416046143,
|
| 658 |
+
"learning_rate": 2.975754794324015e-06,
|
| 659 |
+
"loss": 0.2423,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.292682926829268,
|
| 664 |
+
"grad_norm": 1.6299339532852173,
|
| 665 |
+
"learning_rate": 2.7903205396277546e-06,
|
| 666 |
+
"loss": 0.242,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.317073170731707,
|
| 671 |
+
"grad_norm": 1.7729440927505493,
|
| 672 |
+
"learning_rate": 2.6099108277934105e-06,
|
| 673 |
+
"loss": 0.2422,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.341463414634146,
|
| 678 |
+
"grad_norm": 1.214379906654358,
|
| 679 |
+
"learning_rate": 2.4346513892830427e-06,
|
| 680 |
+
"loss": 0.2153,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.3658536585365852,
|
| 685 |
+
"grad_norm": 1.140561819076538,
|
| 686 |
+
"learning_rate": 2.2646643652494693e-06,
|
| 687 |
+
"loss": 0.1932,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.3902439024390243,
|
| 692 |
+
"grad_norm": 1.1762300729751587,
|
| 693 |
+
"learning_rate": 2.100068222414121e-06,
|
| 694 |
+
"loss": 0.2227,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.4146341463414633,
|
| 699 |
+
"grad_norm": 1.1965961456298828,
|
| 700 |
+
"learning_rate": 1.9409776705056514e-06,
|
| 701 |
+
"loss": 0.2262,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.4390243902439024,
|
| 706 |
+
"grad_norm": 1.1776394844055176,
|
| 707 |
+
"learning_rate": 1.7875035823168641e-06,
|
| 708 |
+
"loss": 0.2354,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.4634146341463414,
|
| 713 |
+
"grad_norm": 1.0380938053131104,
|
| 714 |
+
"learning_rate": 1.6397529164356606e-06,
|
| 715 |
+
"loss": 0.1996,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.4878048780487805,
|
| 720 |
+
"grad_norm": 1.0618942975997925,
|
| 721 |
+
"learning_rate": 1.4978286427038602e-06,
|
| 722 |
+
"loss": 0.202,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.5121951219512195,
|
| 727 |
+
"grad_norm": 1.095805048942566,
|
| 728 |
+
"learning_rate": 1.3618296704558364e-06,
|
| 729 |
+
"loss": 0.2139,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.5365853658536586,
|
| 734 |
+
"grad_norm": 1.1471381187438965,
|
| 735 |
+
"learning_rate": 1.2318507795870138e-06,
|
| 736 |
+
"loss": 0.2248,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.5609756097560976,
|
| 741 |
+
"grad_norm": 1.2113527059555054,
|
| 742 |
+
"learning_rate": 1.1079825545001887e-06,
|
| 743 |
+
"loss": 0.2299,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.5853658536585367,
|
| 748 |
+
"grad_norm": 1.2136269807815552,
|
| 749 |
+
"learning_rate": 9.903113209758098e-07,
|
| 750 |
+
"loss": 0.219,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.6097560975609757,
|
| 755 |
+
"grad_norm": 1.0971758365631104,
|
| 756 |
+
"learning_rate": 8.789190860101226e-07,
|
| 757 |
+
"loss": 0.225,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.6341463414634148,
|
| 762 |
+
"grad_norm": 1.161306381225586,
|
| 763 |
+
"learning_rate": 7.738834806631712e-07,
|
| 764 |
+
"loss": 0.2399,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.658536585365854,
|
| 769 |
+
"grad_norm": 1.0762730836868286,
|
| 770 |
+
"learning_rate": 6.752777059564431e-07,
|
| 771 |
+
"loss": 0.2153,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.682926829268293,
|
| 776 |
+
"grad_norm": 1.2298052310943604,
|
| 777 |
+
"learning_rate": 5.831704818578842e-07,
|
| 778 |
+
"loss": 0.2234,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.7073170731707314,
|
| 783 |
+
"grad_norm": 1.0872737169265747,
|
| 784 |
+
"learning_rate": 4.976259993898503e-07,
|
| 785 |
+
"loss": 0.1843,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.7317073170731705,
|
| 790 |
+
"grad_norm": 1.0804928541183472,
|
| 791 |
+
"learning_rate": 4.187038758933204e-07,
|
| 792 |
+
"loss": 0.1918,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.7560975609756095,
|
| 797 |
+
"grad_norm": 1.111106514930725,
|
| 798 |
+
"learning_rate": 3.4645911347961357e-07,
|
| 799 |
+
"loss": 0.2149,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.7804878048780486,
|
| 804 |
+
"grad_norm": 1.1574560403823853,
|
| 805 |
+
"learning_rate": 2.809420606985236e-07,
|
| 806 |
+
"loss": 0.225,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.8048780487804876,
|
| 811 |
+
"grad_norm": 1.076986312866211,
|
| 812 |
+
"learning_rate": 2.2219837744959284e-07,
|
| 813 |
+
"loss": 0.2168,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.8292682926829267,
|
| 818 |
+
"grad_norm": 1.1566873788833618,
|
| 819 |
+
"learning_rate": 1.7026900316098217e-07,
|
| 820 |
+
"loss": 0.208,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.8536585365853657,
|
| 825 |
+
"grad_norm": 1.1901310682296753,
|
| 826 |
+
"learning_rate": 1.2519012825812804e-07,
|
| 827 |
+
"loss": 0.2346,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.8780487804878048,
|
| 832 |
+
"grad_norm": 1.1799739599227905,
|
| 833 |
+
"learning_rate": 8.699316894203225e-08,
|
| 834 |
+
"loss": 0.2195,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.902439024390244,
|
| 839 |
+
"grad_norm": 1.168531894683838,
|
| 840 |
+
"learning_rate": 5.5704745294815624e-08,
|
| 841 |
+
"loss": 0.1938,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.926829268292683,
|
| 846 |
+
"grad_norm": 1.195745587348938,
|
| 847 |
+
"learning_rate": 3.134666272774034e-08,
|
| 848 |
+
"loss": 0.2204,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.951219512195122,
|
| 853 |
+
"grad_norm": 1.1482133865356445,
|
| 854 |
+
"learning_rate": 1.3935896784663671e-08,
|
| 855 |
+
"loss": 0.2166,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.975609756097561,
|
| 860 |
+
"grad_norm": 1.3828606605529785,
|
| 861 |
+
"learning_rate": 3.4845813115114147e-09,
|
| 862 |
+
"loss": 0.2406,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.0,
|
| 867 |
+
"grad_norm": 1.0298562049865723,
|
| 868 |
+
"learning_rate": 0.0,
|
| 869 |
+
"loss": 0.1997,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 3.0,
|
| 874 |
+
"step": 123,
|
| 875 |
+
"total_flos": 2.6268345567700582e+17,
|
| 876 |
+
"train_loss": 0.550540007347983,
|
| 877 |
+
"train_runtime": 964.7468,
|
| 878 |
+
"train_samples_per_second": 16.17,
|
| 879 |
+
"train_steps_per_second": 0.127
|
| 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": 2.6268345567700582e+17,
|
| 900 |
+
"train_batch_size": 32,
|
| 901 |
+
"trial_name": null,
|
| 902 |
+
"trial_params": null
|
| 903 |
+
}
|