Commit
•
517ab21
1
Parent(s):
adf4f49
End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +636 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.134615384615385,
|
| 3 |
+
"eval_accuracy": 0.6096256684491979,
|
| 4 |
+
"eval_loss": 1.3609113693237305,
|
| 5 |
+
"eval_runtime": 22.0997,
|
| 6 |
+
"eval_samples_per_second": 8.462,
|
| 7 |
+
"eval_steps_per_second": 1.086
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 6.134615384615385,
|
| 3 |
+
"eval_accuracy": 0.6096256684491979,
|
| 4 |
+
"eval_loss": 1.3609113693237305,
|
| 5 |
+
"eval_runtime": 22.0997,
|
| 6 |
+
"eval_samples_per_second": 8.462,
|
| 7 |
+
"eval_steps_per_second": 1.086
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,636 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6436170212765957,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-conflab-traj-direction-lh-v1/checkpoint-630",
|
| 4 |
+
"epoch": 6.134615384615385,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 728,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.013736263736263736,
|
| 13 |
+
"grad_norm": 7.975502967834473,
|
| 14 |
+
"learning_rate": 6.849315068493151e-06,
|
| 15 |
+
"loss": 2.1354,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.027472527472527472,
|
| 20 |
+
"grad_norm": 6.625336647033691,
|
| 21 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 22 |
+
"loss": 2.0408,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04120879120879121,
|
| 27 |
+
"grad_norm": 6.562021732330322,
|
| 28 |
+
"learning_rate": 2.0547945205479453e-05,
|
| 29 |
+
"loss": 1.9823,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.054945054945054944,
|
| 34 |
+
"grad_norm": 6.1020612716674805,
|
| 35 |
+
"learning_rate": 2.7397260273972603e-05,
|
| 36 |
+
"loss": 1.8703,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06868131868131869,
|
| 41 |
+
"grad_norm": 5.716378211975098,
|
| 42 |
+
"learning_rate": 3.424657534246575e-05,
|
| 43 |
+
"loss": 1.938,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08241758241758242,
|
| 48 |
+
"grad_norm": 10.489607810974121,
|
| 49 |
+
"learning_rate": 4.1095890410958905e-05,
|
| 50 |
+
"loss": 1.9538,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09615384615384616,
|
| 55 |
+
"grad_norm": 7.923999786376953,
|
| 56 |
+
"learning_rate": 4.794520547945205e-05,
|
| 57 |
+
"loss": 1.8729,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10989010989010989,
|
| 62 |
+
"grad_norm": 6.225552082061768,
|
| 63 |
+
"learning_rate": 4.9465648854961834e-05,
|
| 64 |
+
"loss": 1.8932,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.12362637362637363,
|
| 69 |
+
"grad_norm": 5.54323673248291,
|
| 70 |
+
"learning_rate": 4.8702290076335885e-05,
|
| 71 |
+
"loss": 1.785,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.13736263736263737,
|
| 76 |
+
"grad_norm": 8.549125671386719,
|
| 77 |
+
"learning_rate": 4.793893129770993e-05,
|
| 78 |
+
"loss": 1.6395,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14423076923076922,
|
| 83 |
+
"eval_accuracy": 0.3723404255319149,
|
| 84 |
+
"eval_loss": 1.6255276203155518,
|
| 85 |
+
"eval_runtime": 26.9945,
|
| 86 |
+
"eval_samples_per_second": 6.964,
|
| 87 |
+
"eval_steps_per_second": 0.889,
|
| 88 |
+
"step": 105
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.0068681318681318,
|
| 92 |
+
"grad_norm": 5.713430404663086,
|
| 93 |
+
"learning_rate": 4.717557251908397e-05,
|
| 94 |
+
"loss": 1.6571,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.0206043956043955,
|
| 99 |
+
"grad_norm": 10.411270141601562,
|
| 100 |
+
"learning_rate": 4.641221374045801e-05,
|
| 101 |
+
"loss": 1.7831,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.0343406593406594,
|
| 106 |
+
"grad_norm": 7.114817142486572,
|
| 107 |
+
"learning_rate": 4.5648854961832065e-05,
|
| 108 |
+
"loss": 1.4513,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.0480769230769231,
|
| 113 |
+
"grad_norm": 8.18074893951416,
|
| 114 |
+
"learning_rate": 4.488549618320611e-05,
|
| 115 |
+
"loss": 1.6207,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.0618131868131868,
|
| 120 |
+
"grad_norm": 8.822683334350586,
|
| 121 |
+
"learning_rate": 4.4122137404580154e-05,
|
| 122 |
+
"loss": 1.6768,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.0755494505494505,
|
| 127 |
+
"grad_norm": 9.030908584594727,
|
| 128 |
+
"learning_rate": 4.33587786259542e-05,
|
| 129 |
+
"loss": 1.4562,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.0892857142857142,
|
| 134 |
+
"grad_norm": 7.88232421875,
|
| 135 |
+
"learning_rate": 4.2595419847328244e-05,
|
| 136 |
+
"loss": 1.4371,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.103021978021978,
|
| 141 |
+
"grad_norm": 10.967424392700195,
|
| 142 |
+
"learning_rate": 4.1832061068702296e-05,
|
| 143 |
+
"loss": 1.5417,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.1167582417582418,
|
| 148 |
+
"grad_norm": 6.950323104858398,
|
| 149 |
+
"learning_rate": 4.106870229007634e-05,
|
| 150 |
+
"loss": 1.3942,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.1304945054945055,
|
| 155 |
+
"grad_norm": 8.856796264648438,
|
| 156 |
+
"learning_rate": 4.0305343511450385e-05,
|
| 157 |
+
"loss": 1.3338,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.1442307692307692,
|
| 162 |
+
"grad_norm": 8.829967498779297,
|
| 163 |
+
"learning_rate": 3.954198473282443e-05,
|
| 164 |
+
"loss": 1.4009,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.1442307692307692,
|
| 169 |
+
"eval_accuracy": 0.5053191489361702,
|
| 170 |
+
"eval_loss": 1.4288405179977417,
|
| 171 |
+
"eval_runtime": 21.3032,
|
| 172 |
+
"eval_samples_per_second": 8.825,
|
| 173 |
+
"eval_steps_per_second": 1.127,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.0137362637362637,
|
| 178 |
+
"grad_norm": 7.487637519836426,
|
| 179 |
+
"learning_rate": 3.8778625954198475e-05,
|
| 180 |
+
"loss": 1.2009,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.0274725274725274,
|
| 185 |
+
"grad_norm": 6.921274662017822,
|
| 186 |
+
"learning_rate": 3.801526717557252e-05,
|
| 187 |
+
"loss": 0.8766,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.041208791208791,
|
| 192 |
+
"grad_norm": 8.127293586730957,
|
| 193 |
+
"learning_rate": 3.7251908396946565e-05,
|
| 194 |
+
"loss": 1.0099,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.0549450549450547,
|
| 199 |
+
"grad_norm": 13.09719467163086,
|
| 200 |
+
"learning_rate": 3.648854961832061e-05,
|
| 201 |
+
"loss": 1.2435,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.068681318681319,
|
| 206 |
+
"grad_norm": 9.744394302368164,
|
| 207 |
+
"learning_rate": 3.5725190839694654e-05,
|
| 208 |
+
"loss": 0.9877,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.0824175824175826,
|
| 213 |
+
"grad_norm": 10.143108367919922,
|
| 214 |
+
"learning_rate": 3.4961832061068706e-05,
|
| 215 |
+
"loss": 1.1423,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.0961538461538463,
|
| 220 |
+
"grad_norm": 11.305399894714355,
|
| 221 |
+
"learning_rate": 3.419847328244275e-05,
|
| 222 |
+
"loss": 1.0389,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.10989010989011,
|
| 227 |
+
"grad_norm": 11.219425201416016,
|
| 228 |
+
"learning_rate": 3.3435114503816796e-05,
|
| 229 |
+
"loss": 0.8836,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.1236263736263736,
|
| 234 |
+
"grad_norm": 8.758235931396484,
|
| 235 |
+
"learning_rate": 3.267175572519084e-05,
|
| 236 |
+
"loss": 0.8425,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.1373626373626373,
|
| 241 |
+
"grad_norm": 11.10839557647705,
|
| 242 |
+
"learning_rate": 3.190839694656489e-05,
|
| 243 |
+
"loss": 1.0856,
|
| 244 |
+
"step": 310
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.144230769230769,
|
| 248 |
+
"eval_accuracy": 0.574468085106383,
|
| 249 |
+
"eval_loss": 1.2734872102737427,
|
| 250 |
+
"eval_runtime": 25.3393,
|
| 251 |
+
"eval_samples_per_second": 7.419,
|
| 252 |
+
"eval_steps_per_second": 0.947,
|
| 253 |
+
"step": 315
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.006868131868132,
|
| 257 |
+
"grad_norm": 6.112276077270508,
|
| 258 |
+
"learning_rate": 3.114503816793894e-05,
|
| 259 |
+
"loss": 1.0481,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.0206043956043955,
|
| 264 |
+
"grad_norm": 3.976270914077759,
|
| 265 |
+
"learning_rate": 3.038167938931298e-05,
|
| 266 |
+
"loss": 0.6704,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 3.034340659340659,
|
| 271 |
+
"grad_norm": 6.606512069702148,
|
| 272 |
+
"learning_rate": 2.9618320610687023e-05,
|
| 273 |
+
"loss": 0.6002,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.048076923076923,
|
| 278 |
+
"grad_norm": 8.499869346618652,
|
| 279 |
+
"learning_rate": 2.885496183206107e-05,
|
| 280 |
+
"loss": 0.576,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.061813186813187,
|
| 285 |
+
"grad_norm": 4.2253499031066895,
|
| 286 |
+
"learning_rate": 2.8091603053435116e-05,
|
| 287 |
+
"loss": 0.4514,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.0755494505494507,
|
| 292 |
+
"grad_norm": 8.19430923461914,
|
| 293 |
+
"learning_rate": 2.732824427480916e-05,
|
| 294 |
+
"loss": 0.6574,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.0892857142857144,
|
| 299 |
+
"grad_norm": 16.499576568603516,
|
| 300 |
+
"learning_rate": 2.6564885496183206e-05,
|
| 301 |
+
"loss": 0.7088,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.103021978021978,
|
| 306 |
+
"grad_norm": 15.13383960723877,
|
| 307 |
+
"learning_rate": 2.580152671755725e-05,
|
| 308 |
+
"loss": 0.7706,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.116758241758242,
|
| 313 |
+
"grad_norm": 10.932879447937012,
|
| 314 |
+
"learning_rate": 2.50381679389313e-05,
|
| 315 |
+
"loss": 0.3813,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.1304945054945055,
|
| 320 |
+
"grad_norm": 7.318915843963623,
|
| 321 |
+
"learning_rate": 2.4274809160305344e-05,
|
| 322 |
+
"loss": 0.4513,
|
| 323 |
+
"step": 410
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.144230769230769,
|
| 327 |
+
"grad_norm": 11.104874610900879,
|
| 328 |
+
"learning_rate": 2.351145038167939e-05,
|
| 329 |
+
"loss": 0.6975,
|
| 330 |
+
"step": 420
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.144230769230769,
|
| 334 |
+
"eval_accuracy": 0.6063829787234043,
|
| 335 |
+
"eval_loss": 1.2628591060638428,
|
| 336 |
+
"eval_runtime": 22.8443,
|
| 337 |
+
"eval_samples_per_second": 8.23,
|
| 338 |
+
"eval_steps_per_second": 1.051,
|
| 339 |
+
"step": 420
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 4.013736263736264,
|
| 343 |
+
"grad_norm": 5.155519008636475,
|
| 344 |
+
"learning_rate": 2.2748091603053437e-05,
|
| 345 |
+
"loss": 0.2989,
|
| 346 |
+
"step": 430
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 4.027472527472527,
|
| 350 |
+
"grad_norm": 2.615933656692505,
|
| 351 |
+
"learning_rate": 2.198473282442748e-05,
|
| 352 |
+
"loss": 0.3818,
|
| 353 |
+
"step": 440
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.041208791208791,
|
| 357 |
+
"grad_norm": 9.644124031066895,
|
| 358 |
+
"learning_rate": 2.122137404580153e-05,
|
| 359 |
+
"loss": 0.4455,
|
| 360 |
+
"step": 450
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 4.054945054945055,
|
| 364 |
+
"grad_norm": 5.017002105712891,
|
| 365 |
+
"learning_rate": 2.0458015267175575e-05,
|
| 366 |
+
"loss": 0.3644,
|
| 367 |
+
"step": 460
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 4.068681318681318,
|
| 371 |
+
"grad_norm": 10.373658180236816,
|
| 372 |
+
"learning_rate": 1.969465648854962e-05,
|
| 373 |
+
"loss": 0.3132,
|
| 374 |
+
"step": 470
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.082417582417582,
|
| 378 |
+
"grad_norm": 11.175583839416504,
|
| 379 |
+
"learning_rate": 1.8931297709923664e-05,
|
| 380 |
+
"loss": 0.3601,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.096153846153846,
|
| 385 |
+
"grad_norm": 6.8020172119140625,
|
| 386 |
+
"learning_rate": 1.816793893129771e-05,
|
| 387 |
+
"loss": 0.3351,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.1098901098901095,
|
| 392 |
+
"grad_norm": 6.1250810623168945,
|
| 393 |
+
"learning_rate": 1.7404580152671757e-05,
|
| 394 |
+
"loss": 0.3077,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.123626373626374,
|
| 399 |
+
"grad_norm": 20.65755844116211,
|
| 400 |
+
"learning_rate": 1.6641221374045802e-05,
|
| 401 |
+
"loss": 0.3634,
|
| 402 |
+
"step": 510
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.137362637362638,
|
| 406 |
+
"grad_norm": 20.21697235107422,
|
| 407 |
+
"learning_rate": 1.5877862595419847e-05,
|
| 408 |
+
"loss": 0.3574,
|
| 409 |
+
"step": 520
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.144230769230769,
|
| 413 |
+
"eval_accuracy": 0.6170212765957447,
|
| 414 |
+
"eval_loss": 1.2030434608459473,
|
| 415 |
+
"eval_runtime": 24.5243,
|
| 416 |
+
"eval_samples_per_second": 7.666,
|
| 417 |
+
"eval_steps_per_second": 0.979,
|
| 418 |
+
"step": 525
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 5.006868131868132,
|
| 422 |
+
"grad_norm": 10.236026763916016,
|
| 423 |
+
"learning_rate": 1.5114503816793894e-05,
|
| 424 |
+
"loss": 0.2425,
|
| 425 |
+
"step": 530
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 5.020604395604396,
|
| 429 |
+
"grad_norm": 3.8250298500061035,
|
| 430 |
+
"learning_rate": 1.435114503816794e-05,
|
| 431 |
+
"loss": 0.1923,
|
| 432 |
+
"step": 540
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 5.03434065934066,
|
| 436 |
+
"grad_norm": 6.404925346374512,
|
| 437 |
+
"learning_rate": 1.3587786259541985e-05,
|
| 438 |
+
"loss": 0.1586,
|
| 439 |
+
"step": 550
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 5.048076923076923,
|
| 443 |
+
"grad_norm": 4.29252815246582,
|
| 444 |
+
"learning_rate": 1.2824427480916032e-05,
|
| 445 |
+
"loss": 0.1486,
|
| 446 |
+
"step": 560
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 5.061813186813187,
|
| 450 |
+
"grad_norm": 3.185380220413208,
|
| 451 |
+
"learning_rate": 1.2061068702290076e-05,
|
| 452 |
+
"loss": 0.1221,
|
| 453 |
+
"step": 570
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 5.075549450549451,
|
| 457 |
+
"grad_norm": 6.5659918785095215,
|
| 458 |
+
"learning_rate": 1.1297709923664123e-05,
|
| 459 |
+
"loss": 0.2548,
|
| 460 |
+
"step": 580
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 5.089285714285714,
|
| 464 |
+
"grad_norm": 4.746553897857666,
|
| 465 |
+
"learning_rate": 1.0534351145038168e-05,
|
| 466 |
+
"loss": 0.1904,
|
| 467 |
+
"step": 590
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 5.103021978021978,
|
| 471 |
+
"grad_norm": 2.2111618518829346,
|
| 472 |
+
"learning_rate": 9.770992366412214e-06,
|
| 473 |
+
"loss": 0.2278,
|
| 474 |
+
"step": 600
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 5.116758241758242,
|
| 478 |
+
"grad_norm": 0.6979553699493408,
|
| 479 |
+
"learning_rate": 9.00763358778626e-06,
|
| 480 |
+
"loss": 0.071,
|
| 481 |
+
"step": 610
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.1304945054945055,
|
| 485 |
+
"grad_norm": 1.7735334634780884,
|
| 486 |
+
"learning_rate": 8.244274809160306e-06,
|
| 487 |
+
"loss": 0.218,
|
| 488 |
+
"step": 620
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.144230769230769,
|
| 492 |
+
"grad_norm": 29.469202041625977,
|
| 493 |
+
"learning_rate": 7.480916030534352e-06,
|
| 494 |
+
"loss": 0.1654,
|
| 495 |
+
"step": 630
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 5.144230769230769,
|
| 499 |
+
"eval_accuracy": 0.6436170212765957,
|
| 500 |
+
"eval_loss": 1.2649437189102173,
|
| 501 |
+
"eval_runtime": 33.5955,
|
| 502 |
+
"eval_samples_per_second": 5.596,
|
| 503 |
+
"eval_steps_per_second": 0.714,
|
| 504 |
+
"step": 630
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 6.013736263736264,
|
| 508 |
+
"grad_norm": 0.3265649080276489,
|
| 509 |
+
"learning_rate": 6.717557251908398e-06,
|
| 510 |
+
"loss": 0.0701,
|
| 511 |
+
"step": 640
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 6.027472527472527,
|
| 515 |
+
"grad_norm": 4.214682102203369,
|
| 516 |
+
"learning_rate": 5.954198473282443e-06,
|
| 517 |
+
"loss": 0.0474,
|
| 518 |
+
"step": 650
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 6.041208791208791,
|
| 522 |
+
"grad_norm": 4.0008697509765625,
|
| 523 |
+
"learning_rate": 5.190839694656488e-06,
|
| 524 |
+
"loss": 0.1185,
|
| 525 |
+
"step": 660
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 6.054945054945055,
|
| 529 |
+
"grad_norm": 16.712289810180664,
|
| 530 |
+
"learning_rate": 4.427480916030535e-06,
|
| 531 |
+
"loss": 0.0714,
|
| 532 |
+
"step": 670
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 6.068681318681318,
|
| 536 |
+
"grad_norm": 1.800746202468872,
|
| 537 |
+
"learning_rate": 3.6641221374045806e-06,
|
| 538 |
+
"loss": 0.1259,
|
| 539 |
+
"step": 680
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 6.082417582417582,
|
| 543 |
+
"grad_norm": 2.648703098297119,
|
| 544 |
+
"learning_rate": 2.900763358778626e-06,
|
| 545 |
+
"loss": 0.0904,
|
| 546 |
+
"step": 690
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 6.096153846153846,
|
| 550 |
+
"grad_norm": 0.37296178936958313,
|
| 551 |
+
"learning_rate": 2.137404580152672e-06,
|
| 552 |
+
"loss": 0.0844,
|
| 553 |
+
"step": 700
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 6.1098901098901095,
|
| 557 |
+
"grad_norm": 3.3622076511383057,
|
| 558 |
+
"learning_rate": 1.3740458015267176e-06,
|
| 559 |
+
"loss": 0.0761,
|
| 560 |
+
"step": 710
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 6.123626373626374,
|
| 564 |
+
"grad_norm": 17.910383224487305,
|
| 565 |
+
"learning_rate": 6.106870229007634e-07,
|
| 566 |
+
"loss": 0.1179,
|
| 567 |
+
"step": 720
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 6.134615384615385,
|
| 571 |
+
"eval_accuracy": 0.6329787234042553,
|
| 572 |
+
"eval_loss": 1.2938581705093384,
|
| 573 |
+
"eval_runtime": 21.981,
|
| 574 |
+
"eval_samples_per_second": 8.553,
|
| 575 |
+
"eval_steps_per_second": 1.092,
|
| 576 |
+
"step": 728
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 6.134615384615385,
|
| 580 |
+
"step": 728,
|
| 581 |
+
"total_flos": 7.220084549098144e+18,
|
| 582 |
+
"train_loss": 0.8182922416663432,
|
| 583 |
+
"train_runtime": 1426.9497,
|
| 584 |
+
"train_samples_per_second": 4.081,
|
| 585 |
+
"train_steps_per_second": 0.51
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.134615384615385,
|
| 589 |
+
"eval_accuracy": 0.6096256684491979,
|
| 590 |
+
"eval_loss": 1.3609111309051514,
|
| 591 |
+
"eval_runtime": 26.248,
|
| 592 |
+
"eval_samples_per_second": 7.124,
|
| 593 |
+
"eval_steps_per_second": 0.914,
|
| 594 |
+
"step": 728
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 6.134615384615385,
|
| 598 |
+
"eval_accuracy": 0.6096256684491979,
|
| 599 |
+
"eval_loss": 1.3609113693237305,
|
| 600 |
+
"eval_runtime": 22.0997,
|
| 601 |
+
"eval_samples_per_second": 8.462,
|
| 602 |
+
"eval_steps_per_second": 1.086,
|
| 603 |
+
"step": 728
|
| 604 |
+
}
|
| 605 |
+
],
|
| 606 |
+
"logging_steps": 10,
|
| 607 |
+
"max_steps": 728,
|
| 608 |
+
"num_input_tokens_seen": 0,
|
| 609 |
+
"num_train_epochs": 9223372036854775807,
|
| 610 |
+
"save_steps": 500,
|
| 611 |
+
"stateful_callbacks": {
|
| 612 |
+
"EarlyStoppingCallback": {
|
| 613 |
+
"args": {
|
| 614 |
+
"early_stopping_patience": 3,
|
| 615 |
+
"early_stopping_threshold": 0.0
|
| 616 |
+
},
|
| 617 |
+
"attributes": {
|
| 618 |
+
"early_stopping_patience_counter": 0
|
| 619 |
+
}
|
| 620 |
+
},
|
| 621 |
+
"TrainerControl": {
|
| 622 |
+
"args": {
|
| 623 |
+
"should_epoch_stop": false,
|
| 624 |
+
"should_evaluate": false,
|
| 625 |
+
"should_log": false,
|
| 626 |
+
"should_save": true,
|
| 627 |
+
"should_training_stop": true
|
| 628 |
+
},
|
| 629 |
+
"attributes": {}
|
| 630 |
+
}
|
| 631 |
+
},
|
| 632 |
+
"total_flos": 7.220084549098144e+18,
|
| 633 |
+
"train_batch_size": 8,
|
| 634 |
+
"trial_name": null,
|
| 635 |
+
"trial_params": null
|
| 636 |
+
}
|