End of training
Browse files- all_results.json +16 -0
- test_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +569 -0
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.999832822513235,
|
| 3 |
+
"test_loss": NaN,
|
| 4 |
+
"test_rouge1": 0.0,
|
| 5 |
+
"test_rouge2": 0.0,
|
| 6 |
+
"test_rougeL": 0.0,
|
| 7 |
+
"test_rougeLsum": 0.0,
|
| 8 |
+
"test_runtime": 3105.513,
|
| 9 |
+
"test_samples_per_second": 3.7,
|
| 10 |
+
"test_steps_per_second": 0.232,
|
| 11 |
+
"total_flos": 1.8665120219529216e+18,
|
| 12 |
+
"train_loss": 0.933749852695655,
|
| 13 |
+
"train_runtime": 62743.6825,
|
| 14 |
+
"train_samples_per_second": 13.728,
|
| 15 |
+
"train_steps_per_second": 0.429
|
| 16 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"test_loss": NaN,
|
| 3 |
+
"test_rouge1": 0.0,
|
| 4 |
+
"test_rouge2": 0.0,
|
| 5 |
+
"test_rougeL": 0.0,
|
| 6 |
+
"test_rougeLsum": 0.0,
|
| 7 |
+
"test_runtime": 3105.513,
|
| 8 |
+
"test_samples_per_second": 3.7,
|
| 9 |
+
"test_steps_per_second": 0.232
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.999832822513235,
|
| 3 |
+
"total_flos": 1.8665120219529216e+18,
|
| 4 |
+
"train_loss": 0.933749852695655,
|
| 5 |
+
"train_runtime": 62743.6825,
|
| 6 |
+
"train_samples_per_second": 13.728,
|
| 7 |
+
"train_steps_per_second": 0.429
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,569 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.0672192719716036,
|
| 3 |
+
"best_model_checkpoint": "/bartabsa-reproduce/outputs/gpt22gpt2_42_std_paper/checkpoint-6000",
|
| 4 |
+
"epoch": 2.999832822513235,
|
| 5 |
+
"eval_steps": 2000,
|
| 6 |
+
"global_step": 26916,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05572582892170521,
|
| 13 |
+
"grad_norm": 4.337312698364258,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 4.5364,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11145165784341042,
|
| 20 |
+
"grad_norm": 2.246065378189087,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 3.095,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.16717748676511562,
|
| 27 |
+
"grad_norm": 1.8885526657104492,
|
| 28 |
+
"learning_rate": 4.9035344960642076e-05,
|
| 29 |
+
"loss": 2.935,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.22290331568682084,
|
| 34 |
+
"grad_norm": 1.8426154851913452,
|
| 35 |
+
"learning_rate": 4.807068992128415e-05,
|
| 36 |
+
"loss": 2.8351,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.22290331568682084,
|
| 41 |
+
"eval_loss": 2.6756136417388916,
|
| 42 |
+
"eval_rouge1": 0.21767546421457468,
|
| 43 |
+
"eval_rouge2": 0.04457308091344295,
|
| 44 |
+
"eval_rougeL": 0.13963266291854382,
|
| 45 |
+
"eval_rougeLsum": 0.2009242837336373,
|
| 46 |
+
"eval_runtime": 4022.6625,
|
| 47 |
+
"eval_samples_per_second": 3.323,
|
| 48 |
+
"eval_steps_per_second": 0.208,
|
| 49 |
+
"step": 2000
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.27862914460852606,
|
| 53 |
+
"grad_norm": 1.7531189918518066,
|
| 54 |
+
"learning_rate": 4.7106034881926225e-05,
|
| 55 |
+
"loss": 2.7805,
|
| 56 |
+
"step": 2500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.33435497353023125,
|
| 60 |
+
"grad_norm": 1.800398349761963,
|
| 61 |
+
"learning_rate": 4.61413798425683e-05,
|
| 62 |
+
"loss": 2.7395,
|
| 63 |
+
"step": 3000
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.3900808024519365,
|
| 67 |
+
"grad_norm": 1.683188796043396,
|
| 68 |
+
"learning_rate": 4.517672480321037e-05,
|
| 69 |
+
"loss": 2.7074,
|
| 70 |
+
"step": 3500
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.4458066313736417,
|
| 74 |
+
"grad_norm": 1.655591607093811,
|
| 75 |
+
"learning_rate": 4.421206976385245e-05,
|
| 76 |
+
"loss": 2.6697,
|
| 77 |
+
"step": 4000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.4458066313736417,
|
| 81 |
+
"eval_loss": 2.5354466438293457,
|
| 82 |
+
"eval_rouge1": 0.2465898813529689,
|
| 83 |
+
"eval_rouge2": 0.059463931913317995,
|
| 84 |
+
"eval_rougeL": 0.15370873221183584,
|
| 85 |
+
"eval_rougeLsum": 0.22609159472872586,
|
| 86 |
+
"eval_runtime": 4025.3864,
|
| 87 |
+
"eval_samples_per_second": 3.321,
|
| 88 |
+
"eval_steps_per_second": 0.208,
|
| 89 |
+
"step": 4000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.5015324602953469,
|
| 93 |
+
"grad_norm": 1.6770449876785278,
|
| 94 |
+
"learning_rate": 4.324741472449452e-05,
|
| 95 |
+
"loss": 2.6358,
|
| 96 |
+
"step": 4500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.5572582892170521,
|
| 100 |
+
"grad_norm": 1.5485845804214478,
|
| 101 |
+
"learning_rate": 4.2282759685136595e-05,
|
| 102 |
+
"loss": 2.6176,
|
| 103 |
+
"step": 5000
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.6129841181387573,
|
| 107 |
+
"grad_norm": 1.5292162895202637,
|
| 108 |
+
"learning_rate": 4.131810464577867e-05,
|
| 109 |
+
"loss": 2.5876,
|
| 110 |
+
"step": 5500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.6687099470604625,
|
| 114 |
+
"grad_norm": 1.5697160959243774,
|
| 115 |
+
"learning_rate": 4.035344960642074e-05,
|
| 116 |
+
"loss": 2.5729,
|
| 117 |
+
"step": 6000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.6687099470604625,
|
| 121 |
+
"eval_loss": 2.4532296657562256,
|
| 122 |
+
"eval_rouge1": 0.2635155122068086,
|
| 123 |
+
"eval_rouge2": 0.0672192719716036,
|
| 124 |
+
"eval_rougeL": 0.16217946801816094,
|
| 125 |
+
"eval_rougeLsum": 0.2442778217221693,
|
| 126 |
+
"eval_runtime": 4039.8611,
|
| 127 |
+
"eval_samples_per_second": 3.309,
|
| 128 |
+
"eval_steps_per_second": 0.207,
|
| 129 |
+
"step": 6000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.7244357759821677,
|
| 133 |
+
"grad_norm": 1.5931309461593628,
|
| 134 |
+
"learning_rate": 3.938879456706282e-05,
|
| 135 |
+
"loss": 2.5551,
|
| 136 |
+
"step": 6500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.780161604903873,
|
| 140 |
+
"grad_norm": 1.5113993883132935,
|
| 141 |
+
"learning_rate": 3.84241395277049e-05,
|
| 142 |
+
"loss": 2.5391,
|
| 143 |
+
"step": 7000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.8358874338255782,
|
| 147 |
+
"grad_norm": 1.518239974975586,
|
| 148 |
+
"learning_rate": 3.745948448834697e-05,
|
| 149 |
+
"loss": 2.5258,
|
| 150 |
+
"step": 7500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.8916132627472834,
|
| 154 |
+
"grad_norm": 1.5953375101089478,
|
| 155 |
+
"learning_rate": 3.6494829448989046e-05,
|
| 156 |
+
"loss": 2.5091,
|
| 157 |
+
"step": 8000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.8916132627472834,
|
| 161 |
+
"eval_loss": 2.390613555908203,
|
| 162 |
+
"eval_rouge1": 0.2551450981643076,
|
| 163 |
+
"eval_rouge2": 0.06574215172816694,
|
| 164 |
+
"eval_rougeL": 0.15758381313108855,
|
| 165 |
+
"eval_rougeLsum": 0.23329141932107159,
|
| 166 |
+
"eval_runtime": 4036.7174,
|
| 167 |
+
"eval_samples_per_second": 3.312,
|
| 168 |
+
"eval_steps_per_second": 0.207,
|
| 169 |
+
"step": 8000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.9473390916689886,
|
| 173 |
+
"grad_norm": 1.3557403087615967,
|
| 174 |
+
"learning_rate": 3.553017440963112e-05,
|
| 175 |
+
"loss": 2.4912,
|
| 176 |
+
"step": 8500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.0030649205906939,
|
| 180 |
+
"grad_norm": NaN,
|
| 181 |
+
"learning_rate": 3.4565519370273194e-05,
|
| 182 |
+
"loss": 2.9327,
|
| 183 |
+
"step": 9000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.058790749512399,
|
| 187 |
+
"grad_norm": NaN,
|
| 188 |
+
"learning_rate": 3.360086433091527e-05,
|
| 189 |
+
"loss": 0.0,
|
| 190 |
+
"step": 9500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.1145165784341042,
|
| 194 |
+
"grad_norm": NaN,
|
| 195 |
+
"learning_rate": 3.263620929155734e-05,
|
| 196 |
+
"loss": 0.0,
|
| 197 |
+
"step": 10000
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.1145165784341042,
|
| 201 |
+
"eval_loss": NaN,
|
| 202 |
+
"eval_rouge1": 0.0,
|
| 203 |
+
"eval_rouge2": 0.0,
|
| 204 |
+
"eval_rougeL": 0.0,
|
| 205 |
+
"eval_rougeLsum": 0.0,
|
| 206 |
+
"eval_runtime": 3619.6605,
|
| 207 |
+
"eval_samples_per_second": 3.693,
|
| 208 |
+
"eval_steps_per_second": 0.231,
|
| 209 |
+
"step": 10000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.1702424073558095,
|
| 213 |
+
"grad_norm": NaN,
|
| 214 |
+
"learning_rate": 3.1671554252199416e-05,
|
| 215 |
+
"loss": 0.0,
|
| 216 |
+
"step": 10500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.2259682362775146,
|
| 220 |
+
"grad_norm": NaN,
|
| 221 |
+
"learning_rate": 3.070689921284149e-05,
|
| 222 |
+
"loss": 0.0,
|
| 223 |
+
"step": 11000
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.28169406519922,
|
| 227 |
+
"grad_norm": NaN,
|
| 228 |
+
"learning_rate": 2.9742244173483564e-05,
|
| 229 |
+
"loss": 0.0,
|
| 230 |
+
"step": 11500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 1.337419894120925,
|
| 234 |
+
"grad_norm": NaN,
|
| 235 |
+
"learning_rate": 2.8777589134125638e-05,
|
| 236 |
+
"loss": 0.0,
|
| 237 |
+
"step": 12000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.337419894120925,
|
| 241 |
+
"eval_loss": NaN,
|
| 242 |
+
"eval_rouge1": 0.0,
|
| 243 |
+
"eval_rouge2": 0.0,
|
| 244 |
+
"eval_rougeL": 0.0,
|
| 245 |
+
"eval_rougeLsum": 0.0,
|
| 246 |
+
"eval_runtime": 3617.2771,
|
| 247 |
+
"eval_samples_per_second": 3.696,
|
| 248 |
+
"eval_steps_per_second": 0.231,
|
| 249 |
+
"step": 12000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.3931457230426303,
|
| 253 |
+
"grad_norm": NaN,
|
| 254 |
+
"learning_rate": 2.7812934094767712e-05,
|
| 255 |
+
"loss": 0.0,
|
| 256 |
+
"step": 12500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.4488715519643356,
|
| 260 |
+
"grad_norm": NaN,
|
| 261 |
+
"learning_rate": 2.6848279055409786e-05,
|
| 262 |
+
"loss": 0.0,
|
| 263 |
+
"step": 13000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.5045973808860407,
|
| 267 |
+
"grad_norm": NaN,
|
| 268 |
+
"learning_rate": 2.588362401605186e-05,
|
| 269 |
+
"loss": 0.0,
|
| 270 |
+
"step": 13500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.5603232098077457,
|
| 274 |
+
"grad_norm": NaN,
|
| 275 |
+
"learning_rate": 2.4918968976693934e-05,
|
| 276 |
+
"loss": 0.0,
|
| 277 |
+
"step": 14000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.5603232098077457,
|
| 281 |
+
"eval_loss": NaN,
|
| 282 |
+
"eval_rouge1": 0.0,
|
| 283 |
+
"eval_rouge2": 0.0,
|
| 284 |
+
"eval_rougeL": 0.0,
|
| 285 |
+
"eval_rougeLsum": 0.0,
|
| 286 |
+
"eval_runtime": 3608.0913,
|
| 287 |
+
"eval_samples_per_second": 3.705,
|
| 288 |
+
"eval_steps_per_second": 0.232,
|
| 289 |
+
"step": 14000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.616049038729451,
|
| 293 |
+
"grad_norm": NaN,
|
| 294 |
+
"learning_rate": 2.3954313937336008e-05,
|
| 295 |
+
"loss": 0.0,
|
| 296 |
+
"step": 14500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.6717748676511563,
|
| 300 |
+
"grad_norm": NaN,
|
| 301 |
+
"learning_rate": 2.2989658897978082e-05,
|
| 302 |
+
"loss": 0.0,
|
| 303 |
+
"step": 15000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.7275006965728616,
|
| 307 |
+
"grad_norm": NaN,
|
| 308 |
+
"learning_rate": 2.2025003858620156e-05,
|
| 309 |
+
"loss": 0.0,
|
| 310 |
+
"step": 15500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.7832265254945667,
|
| 314 |
+
"grad_norm": NaN,
|
| 315 |
+
"learning_rate": 2.1060348819262234e-05,
|
| 316 |
+
"loss": 0.0,
|
| 317 |
+
"step": 16000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.7832265254945667,
|
| 321 |
+
"eval_loss": NaN,
|
| 322 |
+
"eval_rouge1": 0.0,
|
| 323 |
+
"eval_rouge2": 0.0,
|
| 324 |
+
"eval_rougeL": 0.0,
|
| 325 |
+
"eval_rougeLsum": 0.0,
|
| 326 |
+
"eval_runtime": 3610.3297,
|
| 327 |
+
"eval_samples_per_second": 3.703,
|
| 328 |
+
"eval_steps_per_second": 0.232,
|
| 329 |
+
"step": 16000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.8389523544162718,
|
| 333 |
+
"grad_norm": NaN,
|
| 334 |
+
"learning_rate": 2.0095693779904308e-05,
|
| 335 |
+
"loss": 0.0,
|
| 336 |
+
"step": 16500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.894678183337977,
|
| 340 |
+
"grad_norm": NaN,
|
| 341 |
+
"learning_rate": 1.9131038740546382e-05,
|
| 342 |
+
"loss": 0.0,
|
| 343 |
+
"step": 17000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.9504040122596824,
|
| 347 |
+
"grad_norm": NaN,
|
| 348 |
+
"learning_rate": 1.8166383701188456e-05,
|
| 349 |
+
"loss": 0.0,
|
| 350 |
+
"step": 17500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 2.0061298411813877,
|
| 354 |
+
"grad_norm": NaN,
|
| 355 |
+
"learning_rate": 1.720172866183053e-05,
|
| 356 |
+
"loss": 0.0,
|
| 357 |
+
"step": 18000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 2.0061298411813877,
|
| 361 |
+
"eval_loss": NaN,
|
| 362 |
+
"eval_rouge1": 0.0,
|
| 363 |
+
"eval_rouge2": 0.0,
|
| 364 |
+
"eval_rougeL": 0.0,
|
| 365 |
+
"eval_rougeLsum": 0.0,
|
| 366 |
+
"eval_runtime": 3609.3659,
|
| 367 |
+
"eval_samples_per_second": 3.704,
|
| 368 |
+
"eval_steps_per_second": 0.232,
|
| 369 |
+
"step": 18000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 2.0618556701030926,
|
| 373 |
+
"grad_norm": NaN,
|
| 374 |
+
"learning_rate": 1.6237073622472604e-05,
|
| 375 |
+
"loss": 0.0,
|
| 376 |
+
"step": 18500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 2.117581499024798,
|
| 380 |
+
"grad_norm": NaN,
|
| 381 |
+
"learning_rate": 1.5272418583114678e-05,
|
| 382 |
+
"loss": 0.0,
|
| 383 |
+
"step": 19000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.173307327946503,
|
| 387 |
+
"grad_norm": NaN,
|
| 388 |
+
"learning_rate": 1.4307763543756752e-05,
|
| 389 |
+
"loss": 0.0,
|
| 390 |
+
"step": 19500
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 2.2290331568682085,
|
| 394 |
+
"grad_norm": NaN,
|
| 395 |
+
"learning_rate": 1.3343108504398828e-05,
|
| 396 |
+
"loss": 0.0,
|
| 397 |
+
"step": 20000
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 2.2290331568682085,
|
| 401 |
+
"eval_loss": NaN,
|
| 402 |
+
"eval_rouge1": 0.0,
|
| 403 |
+
"eval_rouge2": 0.0,
|
| 404 |
+
"eval_rougeL": 0.0,
|
| 405 |
+
"eval_rougeLsum": 0.0,
|
| 406 |
+
"eval_runtime": 3610.8808,
|
| 407 |
+
"eval_samples_per_second": 3.702,
|
| 408 |
+
"eval_steps_per_second": 0.232,
|
| 409 |
+
"step": 20000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.2847589857899138,
|
| 413 |
+
"grad_norm": NaN,
|
| 414 |
+
"learning_rate": 1.2378453465040902e-05,
|
| 415 |
+
"loss": 0.0,
|
| 416 |
+
"step": 20500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 2.340484814711619,
|
| 420 |
+
"grad_norm": NaN,
|
| 421 |
+
"learning_rate": 1.1413798425682977e-05,
|
| 422 |
+
"loss": 0.0,
|
| 423 |
+
"step": 21000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 2.396210643633324,
|
| 427 |
+
"grad_norm": NaN,
|
| 428 |
+
"learning_rate": 1.0449143386325052e-05,
|
| 429 |
+
"loss": 0.0,
|
| 430 |
+
"step": 21500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.4519364725550292,
|
| 434 |
+
"grad_norm": NaN,
|
| 435 |
+
"learning_rate": 9.484488346967126e-06,
|
| 436 |
+
"loss": 0.0,
|
| 437 |
+
"step": 22000
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 2.4519364725550292,
|
| 441 |
+
"eval_loss": NaN,
|
| 442 |
+
"eval_rouge1": 0.0,
|
| 443 |
+
"eval_rouge2": 0.0,
|
| 444 |
+
"eval_rougeL": 0.0,
|
| 445 |
+
"eval_rougeLsum": 0.0,
|
| 446 |
+
"eval_runtime": 3609.367,
|
| 447 |
+
"eval_samples_per_second": 3.704,
|
| 448 |
+
"eval_steps_per_second": 0.232,
|
| 449 |
+
"step": 22000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.5076623014767345,
|
| 453 |
+
"grad_norm": NaN,
|
| 454 |
+
"learning_rate": 8.5198333076092e-06,
|
| 455 |
+
"loss": 0.0,
|
| 456 |
+
"step": 22500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.56338813039844,
|
| 460 |
+
"grad_norm": NaN,
|
| 461 |
+
"learning_rate": 7.5551782682512745e-06,
|
| 462 |
+
"loss": 0.0,
|
| 463 |
+
"step": 23000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 2.6191139593201447,
|
| 467 |
+
"grad_norm": NaN,
|
| 468 |
+
"learning_rate": 6.5905232288933485e-06,
|
| 469 |
+
"loss": 0.0,
|
| 470 |
+
"step": 23500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 2.67483978824185,
|
| 474 |
+
"grad_norm": NaN,
|
| 475 |
+
"learning_rate": 5.6258681895354226e-06,
|
| 476 |
+
"loss": 0.0,
|
| 477 |
+
"step": 24000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.67483978824185,
|
| 481 |
+
"eval_loss": NaN,
|
| 482 |
+
"eval_rouge1": 0.0,
|
| 483 |
+
"eval_rouge2": 0.0,
|
| 484 |
+
"eval_rougeL": 0.0,
|
| 485 |
+
"eval_rougeLsum": 0.0,
|
| 486 |
+
"eval_runtime": 3608.0619,
|
| 487 |
+
"eval_samples_per_second": 3.705,
|
| 488 |
+
"eval_steps_per_second": 0.232,
|
| 489 |
+
"step": 24000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.7305656171635553,
|
| 493 |
+
"grad_norm": NaN,
|
| 494 |
+
"learning_rate": 4.661213150177497e-06,
|
| 495 |
+
"loss": 0.0,
|
| 496 |
+
"step": 24500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.7862914460852606,
|
| 500 |
+
"grad_norm": NaN,
|
| 501 |
+
"learning_rate": 3.6965581108195706e-06,
|
| 502 |
+
"loss": 0.0,
|
| 503 |
+
"step": 25000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.842017275006966,
|
| 507 |
+
"grad_norm": NaN,
|
| 508 |
+
"learning_rate": 2.7319030714616455e-06,
|
| 509 |
+
"loss": 0.0,
|
| 510 |
+
"step": 25500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.897743103928671,
|
| 514 |
+
"grad_norm": NaN,
|
| 515 |
+
"learning_rate": 1.7672480321037198e-06,
|
| 516 |
+
"loss": 0.0,
|
| 517 |
+
"step": 26000
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.897743103928671,
|
| 521 |
+
"eval_loss": NaN,
|
| 522 |
+
"eval_rouge1": 0.0,
|
| 523 |
+
"eval_rouge2": 0.0,
|
| 524 |
+
"eval_rougeL": 0.0,
|
| 525 |
+
"eval_rougeLsum": 0.0,
|
| 526 |
+
"eval_runtime": 3609.5904,
|
| 527 |
+
"eval_samples_per_second": 3.703,
|
| 528 |
+
"eval_steps_per_second": 0.232,
|
| 529 |
+
"step": 26000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.953468932850376,
|
| 533 |
+
"grad_norm": NaN,
|
| 534 |
+
"learning_rate": 8.025929927457941e-07,
|
| 535 |
+
"loss": 0.0,
|
| 536 |
+
"step": 26500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.999832822513235,
|
| 540 |
+
"step": 26916,
|
| 541 |
+
"total_flos": 1.8665120219529216e+18,
|
| 542 |
+
"train_loss": 0.933749852695655,
|
| 543 |
+
"train_runtime": 62743.6825,
|
| 544 |
+
"train_samples_per_second": 13.728,
|
| 545 |
+
"train_steps_per_second": 0.429
|
| 546 |
+
}
|
| 547 |
+
],
|
| 548 |
+
"logging_steps": 500,
|
| 549 |
+
"max_steps": 26916,
|
| 550 |
+
"num_input_tokens_seen": 0,
|
| 551 |
+
"num_train_epochs": 3,
|
| 552 |
+
"save_steps": 2000,
|
| 553 |
+
"stateful_callbacks": {
|
| 554 |
+
"TrainerControl": {
|
| 555 |
+
"args": {
|
| 556 |
+
"should_epoch_stop": false,
|
| 557 |
+
"should_evaluate": false,
|
| 558 |
+
"should_log": false,
|
| 559 |
+
"should_save": true,
|
| 560 |
+
"should_training_stop": true
|
| 561 |
+
},
|
| 562 |
+
"attributes": {}
|
| 563 |
+
}
|
| 564 |
+
},
|
| 565 |
+
"total_flos": 1.8665120219529216e+18,
|
| 566 |
+
"train_batch_size": 8,
|
| 567 |
+
"trial_name": null,
|
| 568 |
+
"trial_params": null
|
| 569 |
+
}
|