End of training
Browse files
all_results.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"eval_rouge-1": 1.0,
|
| 7 |
"eval_rouge-2": 1.0,
|
| 8 |
"eval_rouge-l": 1.0,
|
| 9 |
-
"eval_runtime": 0.
|
| 10 |
-
"eval_samples_per_second": 3.
|
| 11 |
-
"eval_steps_per_second": 3.
|
| 12 |
}
|
|
|
|
| 6 |
"eval_rouge-1": 1.0,
|
| 7 |
"eval_rouge-2": 1.0,
|
| 8 |
"eval_rouge-l": 1.0,
|
| 9 |
+
"eval_runtime": 0.3004,
|
| 10 |
+
"eval_samples_per_second": 3.328,
|
| 11 |
+
"eval_steps_per_second": 3.328
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"eval_rouge-1": 1.0,
|
| 7 |
"eval_rouge-2": 1.0,
|
| 8 |
"eval_rouge-l": 1.0,
|
| 9 |
-
"eval_runtime": 0.
|
| 10 |
-
"eval_samples_per_second": 3.
|
| 11 |
-
"eval_steps_per_second": 3.
|
| 12 |
}
|
|
|
|
| 6 |
"eval_rouge-1": 1.0,
|
| 7 |
"eval_rouge-2": 1.0,
|
| 8 |
"eval_rouge-l": 1.0,
|
| 9 |
+
"eval_runtime": 0.3004,
|
| 10 |
+
"eval_samples_per_second": 3.328,
|
| 11 |
+
"eval_steps_per_second": 3.328
|
| 12 |
}
|
runs/May11_15-53-03_642c09b47d0a/events.out.tfevents.1683820671.642c09b47d0a.2440.2
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:987fc48605e430ea40079352654181ea9f2c1902661aece95e75bd77114ff7f7
|
| 3 |
+
size 1148
|
trainer_state.json
ADDED
|
@@ -0,0 +1,513 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.002095520496368408,
|
| 3 |
+
"best_model_checkpoint": "qqq-finetuned-on-calls/checkpoint-500",
|
| 4 |
+
"epoch": 78.125,
|
| 5 |
+
"global_step": 625,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.12,
|
| 12 |
+
"learning_rate": 1e-05,
|
| 13 |
+
"loss": 2.1855,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.12,
|
| 18 |
+
"eval_avg_rouge_f": 0.0,
|
| 19 |
+
"eval_gen_len": 15.0,
|
| 20 |
+
"eval_loss": 1.4282418489456177,
|
| 21 |
+
"eval_rouge-1": 0.0,
|
| 22 |
+
"eval_rouge-2": 0.0,
|
| 23 |
+
"eval_rouge-l": 0.0,
|
| 24 |
+
"eval_runtime": 2.7397,
|
| 25 |
+
"eval_samples_per_second": 0.365,
|
| 26 |
+
"eval_steps_per_second": 0.365,
|
| 27 |
+
"step": 25
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 6.25,
|
| 31 |
+
"learning_rate": 2e-05,
|
| 32 |
+
"loss": 1.5665,
|
| 33 |
+
"step": 50
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 6.25,
|
| 37 |
+
"eval_avg_rouge_f": 0.1212,
|
| 38 |
+
"eval_gen_len": 12.0,
|
| 39 |
+
"eval_loss": 0.642043948173523,
|
| 40 |
+
"eval_rouge-1": 0.1818,
|
| 41 |
+
"eval_rouge-2": 0.0,
|
| 42 |
+
"eval_rouge-l": 0.1818,
|
| 43 |
+
"eval_runtime": 0.8049,
|
| 44 |
+
"eval_samples_per_second": 1.242,
|
| 45 |
+
"eval_steps_per_second": 1.242,
|
| 46 |
+
"step": 50
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 9.38,
|
| 50 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 51 |
+
"loss": 1.1046,
|
| 52 |
+
"step": 75
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 9.38,
|
| 56 |
+
"eval_avg_rouge_f": 1.0,
|
| 57 |
+
"eval_gen_len": 11.0,
|
| 58 |
+
"eval_loss": 0.21844333410263062,
|
| 59 |
+
"eval_rouge-1": 1.0,
|
| 60 |
+
"eval_rouge-2": 1.0,
|
| 61 |
+
"eval_rouge-l": 1.0,
|
| 62 |
+
"eval_runtime": 0.239,
|
| 63 |
+
"eval_samples_per_second": 4.184,
|
| 64 |
+
"eval_steps_per_second": 4.184,
|
| 65 |
+
"step": 75
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 12.5,
|
| 69 |
+
"learning_rate": 1.866666666666667e-05,
|
| 70 |
+
"loss": 0.8218,
|
| 71 |
+
"step": 100
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 12.5,
|
| 75 |
+
"eval_avg_rouge_f": 1.0,
|
| 76 |
+
"eval_gen_len": 11.0,
|
| 77 |
+
"eval_loss": 0.10981562733650208,
|
| 78 |
+
"eval_rouge-1": 1.0,
|
| 79 |
+
"eval_rouge-2": 1.0,
|
| 80 |
+
"eval_rouge-l": 1.0,
|
| 81 |
+
"eval_runtime": 0.2279,
|
| 82 |
+
"eval_samples_per_second": 4.388,
|
| 83 |
+
"eval_steps_per_second": 4.388,
|
| 84 |
+
"step": 100
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 15.62,
|
| 88 |
+
"learning_rate": 1.8e-05,
|
| 89 |
+
"loss": 0.606,
|
| 90 |
+
"step": 125
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 15.62,
|
| 94 |
+
"eval_avg_rouge_f": 1.0,
|
| 95 |
+
"eval_gen_len": 11.0,
|
| 96 |
+
"eval_loss": 0.07491956651210785,
|
| 97 |
+
"eval_rouge-1": 1.0,
|
| 98 |
+
"eval_rouge-2": 1.0,
|
| 99 |
+
"eval_rouge-l": 1.0,
|
| 100 |
+
"eval_runtime": 0.3017,
|
| 101 |
+
"eval_samples_per_second": 3.314,
|
| 102 |
+
"eval_steps_per_second": 3.314,
|
| 103 |
+
"step": 125
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 18.75,
|
| 107 |
+
"learning_rate": 1.7333333333333336e-05,
|
| 108 |
+
"loss": 0.5488,
|
| 109 |
+
"step": 150
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 18.75,
|
| 113 |
+
"eval_avg_rouge_f": 1.0,
|
| 114 |
+
"eval_gen_len": 11.0,
|
| 115 |
+
"eval_loss": 0.057742033153772354,
|
| 116 |
+
"eval_rouge-1": 1.0,
|
| 117 |
+
"eval_rouge-2": 1.0,
|
| 118 |
+
"eval_rouge-l": 1.0,
|
| 119 |
+
"eval_runtime": 0.2335,
|
| 120 |
+
"eval_samples_per_second": 4.282,
|
| 121 |
+
"eval_steps_per_second": 4.282,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 21.88,
|
| 126 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 127 |
+
"loss": 0.4161,
|
| 128 |
+
"step": 175
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 21.88,
|
| 132 |
+
"eval_avg_rouge_f": 1.0,
|
| 133 |
+
"eval_gen_len": 11.0,
|
| 134 |
+
"eval_loss": 0.06839721649885178,
|
| 135 |
+
"eval_rouge-1": 1.0,
|
| 136 |
+
"eval_rouge-2": 1.0,
|
| 137 |
+
"eval_rouge-l": 1.0,
|
| 138 |
+
"eval_runtime": 0.2273,
|
| 139 |
+
"eval_samples_per_second": 4.399,
|
| 140 |
+
"eval_steps_per_second": 4.399,
|
| 141 |
+
"step": 175
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 25.0,
|
| 145 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 146 |
+
"loss": 0.3196,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 25.0,
|
| 151 |
+
"eval_avg_rouge_f": 1.0,
|
| 152 |
+
"eval_gen_len": 11.0,
|
| 153 |
+
"eval_loss": 0.057005975395441055,
|
| 154 |
+
"eval_rouge-1": 1.0,
|
| 155 |
+
"eval_rouge-2": 1.0,
|
| 156 |
+
"eval_rouge-l": 1.0,
|
| 157 |
+
"eval_runtime": 0.2443,
|
| 158 |
+
"eval_samples_per_second": 4.093,
|
| 159 |
+
"eval_steps_per_second": 4.093,
|
| 160 |
+
"step": 200
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 28.12,
|
| 164 |
+
"learning_rate": 1.5333333333333334e-05,
|
| 165 |
+
"loss": 0.2929,
|
| 166 |
+
"step": 225
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 28.12,
|
| 170 |
+
"eval_avg_rouge_f": 1.0,
|
| 171 |
+
"eval_gen_len": 11.0,
|
| 172 |
+
"eval_loss": 0.04155131056904793,
|
| 173 |
+
"eval_rouge-1": 1.0,
|
| 174 |
+
"eval_rouge-2": 1.0,
|
| 175 |
+
"eval_rouge-l": 1.0,
|
| 176 |
+
"eval_runtime": 0.3002,
|
| 177 |
+
"eval_samples_per_second": 3.332,
|
| 178 |
+
"eval_steps_per_second": 3.332,
|
| 179 |
+
"step": 225
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 31.25,
|
| 183 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 184 |
+
"loss": 0.2519,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 31.25,
|
| 189 |
+
"eval_avg_rouge_f": 1.0,
|
| 190 |
+
"eval_gen_len": 11.0,
|
| 191 |
+
"eval_loss": 0.024672139436006546,
|
| 192 |
+
"eval_rouge-1": 1.0,
|
| 193 |
+
"eval_rouge-2": 1.0,
|
| 194 |
+
"eval_rouge-l": 1.0,
|
| 195 |
+
"eval_runtime": 0.2328,
|
| 196 |
+
"eval_samples_per_second": 4.296,
|
| 197 |
+
"eval_steps_per_second": 4.296,
|
| 198 |
+
"step": 250
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 34.38,
|
| 202 |
+
"learning_rate": 1.4e-05,
|
| 203 |
+
"loss": 0.178,
|
| 204 |
+
"step": 275
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 34.38,
|
| 208 |
+
"eval_avg_rouge_f": 1.0,
|
| 209 |
+
"eval_gen_len": 11.0,
|
| 210 |
+
"eval_loss": 0.011837620288133621,
|
| 211 |
+
"eval_rouge-1": 1.0,
|
| 212 |
+
"eval_rouge-2": 1.0,
|
| 213 |
+
"eval_rouge-l": 1.0,
|
| 214 |
+
"eval_runtime": 0.2389,
|
| 215 |
+
"eval_samples_per_second": 4.185,
|
| 216 |
+
"eval_steps_per_second": 4.185,
|
| 217 |
+
"step": 275
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 37.5,
|
| 221 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 222 |
+
"loss": 0.1603,
|
| 223 |
+
"step": 300
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 37.5,
|
| 227 |
+
"eval_avg_rouge_f": 1.0,
|
| 228 |
+
"eval_gen_len": 11.0,
|
| 229 |
+
"eval_loss": 0.00636103143915534,
|
| 230 |
+
"eval_rouge-1": 1.0,
|
| 231 |
+
"eval_rouge-2": 1.0,
|
| 232 |
+
"eval_rouge-l": 1.0,
|
| 233 |
+
"eval_runtime": 0.331,
|
| 234 |
+
"eval_samples_per_second": 3.021,
|
| 235 |
+
"eval_steps_per_second": 3.021,
|
| 236 |
+
"step": 300
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 40.62,
|
| 240 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 241 |
+
"loss": 0.1684,
|
| 242 |
+
"step": 325
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 40.62,
|
| 246 |
+
"eval_avg_rouge_f": 1.0,
|
| 247 |
+
"eval_gen_len": 11.0,
|
| 248 |
+
"eval_loss": 0.005132880061864853,
|
| 249 |
+
"eval_rouge-1": 1.0,
|
| 250 |
+
"eval_rouge-2": 1.0,
|
| 251 |
+
"eval_rouge-l": 1.0,
|
| 252 |
+
"eval_runtime": 0.2317,
|
| 253 |
+
"eval_samples_per_second": 4.316,
|
| 254 |
+
"eval_steps_per_second": 4.316,
|
| 255 |
+
"step": 325
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 43.75,
|
| 259 |
+
"learning_rate": 1.2e-05,
|
| 260 |
+
"loss": 0.1326,
|
| 261 |
+
"step": 350
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 43.75,
|
| 265 |
+
"eval_avg_rouge_f": 1.0,
|
| 266 |
+
"eval_gen_len": 11.0,
|
| 267 |
+
"eval_loss": 0.005127913784235716,
|
| 268 |
+
"eval_rouge-1": 1.0,
|
| 269 |
+
"eval_rouge-2": 1.0,
|
| 270 |
+
"eval_rouge-l": 1.0,
|
| 271 |
+
"eval_runtime": 0.2432,
|
| 272 |
+
"eval_samples_per_second": 4.111,
|
| 273 |
+
"eval_steps_per_second": 4.111,
|
| 274 |
+
"step": 350
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 46.88,
|
| 278 |
+
"learning_rate": 1.1333333333333334e-05,
|
| 279 |
+
"loss": 0.1349,
|
| 280 |
+
"step": 375
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 46.88,
|
| 284 |
+
"eval_avg_rouge_f": 1.0,
|
| 285 |
+
"eval_gen_len": 11.0,
|
| 286 |
+
"eval_loss": 0.006368895061314106,
|
| 287 |
+
"eval_rouge-1": 1.0,
|
| 288 |
+
"eval_rouge-2": 1.0,
|
| 289 |
+
"eval_rouge-l": 1.0,
|
| 290 |
+
"eval_runtime": 0.3345,
|
| 291 |
+
"eval_samples_per_second": 2.99,
|
| 292 |
+
"eval_steps_per_second": 2.99,
|
| 293 |
+
"step": 375
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 50.0,
|
| 297 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 298 |
+
"loss": 0.1105,
|
| 299 |
+
"step": 400
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 50.0,
|
| 303 |
+
"eval_avg_rouge_f": 1.0,
|
| 304 |
+
"eval_gen_len": 11.0,
|
| 305 |
+
"eval_loss": 0.006063114386051893,
|
| 306 |
+
"eval_rouge-1": 1.0,
|
| 307 |
+
"eval_rouge-2": 1.0,
|
| 308 |
+
"eval_rouge-l": 1.0,
|
| 309 |
+
"eval_runtime": 0.2318,
|
| 310 |
+
"eval_samples_per_second": 4.315,
|
| 311 |
+
"eval_steps_per_second": 4.315,
|
| 312 |
+
"step": 400
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 53.12,
|
| 316 |
+
"learning_rate": 1e-05,
|
| 317 |
+
"loss": 0.1026,
|
| 318 |
+
"step": 425
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 53.12,
|
| 322 |
+
"eval_avg_rouge_f": 1.0,
|
| 323 |
+
"eval_gen_len": 11.0,
|
| 324 |
+
"eval_loss": 0.004856177140027285,
|
| 325 |
+
"eval_rouge-1": 1.0,
|
| 326 |
+
"eval_rouge-2": 1.0,
|
| 327 |
+
"eval_rouge-l": 1.0,
|
| 328 |
+
"eval_runtime": 0.2374,
|
| 329 |
+
"eval_samples_per_second": 4.212,
|
| 330 |
+
"eval_steps_per_second": 4.212,
|
| 331 |
+
"step": 425
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 56.25,
|
| 335 |
+
"learning_rate": 9.333333333333334e-06,
|
| 336 |
+
"loss": 0.0936,
|
| 337 |
+
"step": 450
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 56.25,
|
| 341 |
+
"eval_avg_rouge_f": 1.0,
|
| 342 |
+
"eval_gen_len": 11.0,
|
| 343 |
+
"eval_loss": 0.003024860518053174,
|
| 344 |
+
"eval_rouge-1": 1.0,
|
| 345 |
+
"eval_rouge-2": 1.0,
|
| 346 |
+
"eval_rouge-l": 1.0,
|
| 347 |
+
"eval_runtime": 0.3213,
|
| 348 |
+
"eval_samples_per_second": 3.112,
|
| 349 |
+
"eval_steps_per_second": 3.112,
|
| 350 |
+
"step": 450
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 59.38,
|
| 354 |
+
"learning_rate": 8.666666666666668e-06,
|
| 355 |
+
"loss": 0.0704,
|
| 356 |
+
"step": 475
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 59.38,
|
| 360 |
+
"eval_avg_rouge_f": 1.0,
|
| 361 |
+
"eval_gen_len": 11.0,
|
| 362 |
+
"eval_loss": 0.002469500992447138,
|
| 363 |
+
"eval_rouge-1": 1.0,
|
| 364 |
+
"eval_rouge-2": 1.0,
|
| 365 |
+
"eval_rouge-l": 1.0,
|
| 366 |
+
"eval_runtime": 0.2553,
|
| 367 |
+
"eval_samples_per_second": 3.916,
|
| 368 |
+
"eval_steps_per_second": 3.916,
|
| 369 |
+
"step": 475
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 62.5,
|
| 373 |
+
"learning_rate": 8.000000000000001e-06,
|
| 374 |
+
"loss": 0.0699,
|
| 375 |
+
"step": 500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 62.5,
|
| 379 |
+
"eval_avg_rouge_f": 1.0,
|
| 380 |
+
"eval_gen_len": 11.0,
|
| 381 |
+
"eval_loss": 0.002095520496368408,
|
| 382 |
+
"eval_rouge-1": 1.0,
|
| 383 |
+
"eval_rouge-2": 1.0,
|
| 384 |
+
"eval_rouge-l": 1.0,
|
| 385 |
+
"eval_runtime": 0.2503,
|
| 386 |
+
"eval_samples_per_second": 3.996,
|
| 387 |
+
"eval_steps_per_second": 3.996,
|
| 388 |
+
"step": 500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 65.62,
|
| 392 |
+
"learning_rate": 7.333333333333333e-06,
|
| 393 |
+
"loss": 0.0863,
|
| 394 |
+
"step": 525
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 65.62,
|
| 398 |
+
"eval_avg_rouge_f": 1.0,
|
| 399 |
+
"eval_gen_len": 11.0,
|
| 400 |
+
"eval_loss": 0.0020483580883592367,
|
| 401 |
+
"eval_rouge-1": 1.0,
|
| 402 |
+
"eval_rouge-2": 1.0,
|
| 403 |
+
"eval_rouge-l": 1.0,
|
| 404 |
+
"eval_runtime": 0.3232,
|
| 405 |
+
"eval_samples_per_second": 3.094,
|
| 406 |
+
"eval_steps_per_second": 3.094,
|
| 407 |
+
"step": 525
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 68.75,
|
| 411 |
+
"learning_rate": 6.666666666666667e-06,
|
| 412 |
+
"loss": 0.0595,
|
| 413 |
+
"step": 550
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 68.75,
|
| 417 |
+
"eval_avg_rouge_f": 1.0,
|
| 418 |
+
"eval_gen_len": 11.0,
|
| 419 |
+
"eval_loss": 0.002404692117124796,
|
| 420 |
+
"eval_rouge-1": 1.0,
|
| 421 |
+
"eval_rouge-2": 1.0,
|
| 422 |
+
"eval_rouge-l": 1.0,
|
| 423 |
+
"eval_runtime": 0.2836,
|
| 424 |
+
"eval_samples_per_second": 3.526,
|
| 425 |
+
"eval_steps_per_second": 3.526,
|
| 426 |
+
"step": 550
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 71.88,
|
| 430 |
+
"learning_rate": 6e-06,
|
| 431 |
+
"loss": 0.0594,
|
| 432 |
+
"step": 575
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 71.88,
|
| 436 |
+
"eval_avg_rouge_f": 1.0,
|
| 437 |
+
"eval_gen_len": 11.0,
|
| 438 |
+
"eval_loss": 0.002847235416993499,
|
| 439 |
+
"eval_rouge-1": 1.0,
|
| 440 |
+
"eval_rouge-2": 1.0,
|
| 441 |
+
"eval_rouge-l": 1.0,
|
| 442 |
+
"eval_runtime": 0.4294,
|
| 443 |
+
"eval_samples_per_second": 2.329,
|
| 444 |
+
"eval_steps_per_second": 2.329,
|
| 445 |
+
"step": 575
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 75.0,
|
| 449 |
+
"learning_rate": 5.333333333333334e-06,
|
| 450 |
+
"loss": 0.0683,
|
| 451 |
+
"step": 600
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 75.0,
|
| 455 |
+
"eval_avg_rouge_f": 1.0,
|
| 456 |
+
"eval_gen_len": 11.0,
|
| 457 |
+
"eval_loss": 0.0025598241481930017,
|
| 458 |
+
"eval_rouge-1": 1.0,
|
| 459 |
+
"eval_rouge-2": 1.0,
|
| 460 |
+
"eval_rouge-l": 1.0,
|
| 461 |
+
"eval_runtime": 0.2348,
|
| 462 |
+
"eval_samples_per_second": 4.259,
|
| 463 |
+
"eval_steps_per_second": 4.259,
|
| 464 |
+
"step": 600
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 78.12,
|
| 468 |
+
"learning_rate": 4.666666666666667e-06,
|
| 469 |
+
"loss": 0.074,
|
| 470 |
+
"step": 625
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 78.12,
|
| 474 |
+
"eval_avg_rouge_f": 1.0,
|
| 475 |
+
"eval_gen_len": 11.0,
|
| 476 |
+
"eval_loss": 0.0025288190227001905,
|
| 477 |
+
"eval_rouge-1": 1.0,
|
| 478 |
+
"eval_rouge-2": 1.0,
|
| 479 |
+
"eval_rouge-l": 1.0,
|
| 480 |
+
"eval_runtime": 0.245,
|
| 481 |
+
"eval_samples_per_second": 4.081,
|
| 482 |
+
"eval_steps_per_second": 4.081,
|
| 483 |
+
"step": 625
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 78.12,
|
| 487 |
+
"step": 625,
|
| 488 |
+
"total_flos": 840313961717760.0,
|
| 489 |
+
"train_loss": 0.38729172706604004,
|
| 490 |
+
"train_runtime": 274.1527,
|
| 491 |
+
"train_samples_per_second": 8.754,
|
| 492 |
+
"train_steps_per_second": 2.918
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 78.12,
|
| 496 |
+
"eval_avg_rouge_f": 1.0,
|
| 497 |
+
"eval_gen_len": 11.0,
|
| 498 |
+
"eval_loss": 0.002095520496368408,
|
| 499 |
+
"eval_rouge-1": 1.0,
|
| 500 |
+
"eval_rouge-2": 1.0,
|
| 501 |
+
"eval_rouge-l": 1.0,
|
| 502 |
+
"eval_runtime": 0.2617,
|
| 503 |
+
"eval_samples_per_second": 3.82,
|
| 504 |
+
"eval_steps_per_second": 3.82,
|
| 505 |
+
"step": 625
|
| 506 |
+
}
|
| 507 |
+
],
|
| 508 |
+
"max_steps": 800,
|
| 509 |
+
"num_train_epochs": 100,
|
| 510 |
+
"total_flos": 840313961717760.0,
|
| 511 |
+
"trial_name": null,
|
| 512 |
+
"trial_params": null
|
| 513 |
+
}
|