Commit
•
298a64b
1
Parent(s):
3c2ac76
Training in progress, step 2000
Browse files- AppE_dev_matched_results.json +8 -0
- ChcE_dev_matched_results.json +8 -0
- CollSgE_dev_matched_results.json +8 -0
- GLUE_validation_matched_results.json +8 -0
- IndE_dev_matched_results.json +8 -0
- MULTI_dev_matched_results.json +8 -0
- VALUE_validation_matched_results.json +8 -0
- all_results.json +12 -0
- pytorch_model.bin +1 -1
- train_results.json +7 -0
- trainer_state.json +460 -0
- training_args.bin +1 -1
AppE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8524053224155578,
|
| 4 |
+
"eval_loss": 0.4613371789455414,
|
| 5 |
+
"eval_runtime": 16.6422,
|
| 6 |
+
"eval_samples_per_second": 587.062,
|
| 7 |
+
"eval_steps_per_second": 18.387
|
| 8 |
+
}
|
ChcE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.87021494370522,
|
| 4 |
+
"eval_loss": 0.4166066646575928,
|
| 5 |
+
"eval_runtime": 16.6152,
|
| 6 |
+
"eval_samples_per_second": 588.016,
|
| 7 |
+
"eval_steps_per_second": 18.417
|
| 8 |
+
}
|
CollSgE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8193449334698055,
|
| 4 |
+
"eval_loss": 0.5677177309989929,
|
| 5 |
+
"eval_runtime": 16.6512,
|
| 6 |
+
"eval_samples_per_second": 586.745,
|
| 7 |
+
"eval_steps_per_second": 18.377
|
| 8 |
+
}
|
GLUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8717269485481406,
|
| 4 |
+
"eval_loss": 0.4091717600822449,
|
| 5 |
+
"eval_runtime": 16.7705,
|
| 6 |
+
"eval_samples_per_second": 585.254,
|
| 7 |
+
"eval_steps_per_second": 18.306
|
| 8 |
+
}
|
IndE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8244626407369499,
|
| 4 |
+
"eval_loss": 0.5456578135490417,
|
| 5 |
+
"eval_runtime": 16.67,
|
| 6 |
+
"eval_samples_per_second": 586.081,
|
| 7 |
+
"eval_steps_per_second": 18.356
|
| 8 |
+
}
|
MULTI_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.7865834952449126,
|
| 4 |
+
"eval_loss": 0.6499571204185486,
|
| 5 |
+
"eval_runtime": 16.7694,
|
| 6 |
+
"eval_samples_per_second": 583.146,
|
| 7 |
+
"eval_steps_per_second": 18.248
|
| 8 |
+
}
|
VALUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8480896586856852,
|
| 4 |
+
"eval_loss": 0.4839687645435333,
|
| 5 |
+
"eval_runtime": 16.7531,
|
| 6 |
+
"eval_samples_per_second": 585.86,
|
| 7 |
+
"eval_steps_per_second": 18.325
|
| 8 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8717269485481406,
|
| 4 |
+
"eval_loss": 0.4091717600822449,
|
| 5 |
+
"eval_runtime": 16.7705,
|
| 6 |
+
"eval_samples_per_second": 585.254,
|
| 7 |
+
"eval_steps_per_second": 18.306,
|
| 8 |
+
"train_loss": 0.17320066710326415,
|
| 9 |
+
"train_runtime": 11406.3483,
|
| 10 |
+
"train_samples_per_second": 168.211,
|
| 11 |
+
"train_steps_per_second": 5.257
|
| 12 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 498665141
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51b39a5c0b0c0cd84bffc9781c095f21390098cbf77e1f3aac0302717e982e82
|
| 3 |
size 498665141
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.17320066710326415,
|
| 4 |
+
"train_runtime": 11406.3483,
|
| 5 |
+
"train_samples_per_second": 168.211,
|
| 6 |
+
"train_steps_per_second": 5.257
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,460 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.87021494370522,
|
| 3 |
+
"best_model_checkpoint": "./outputs/finetuning/mnli_ChcE/checkpoint-20000",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 59960,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.17,
|
| 12 |
+
"learning_rate": 1.933288859239493e-05,
|
| 13 |
+
"loss": 0.3102,
|
| 14 |
+
"step": 2000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.17,
|
| 18 |
+
"eval_acc": 0.8544524053224155,
|
| 19 |
+
"eval_loss": 0.41351282596588135,
|
| 20 |
+
"eval_runtime": 16.7417,
|
| 21 |
+
"eval_samples_per_second": 583.571,
|
| 22 |
+
"eval_steps_per_second": 18.278,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33,
|
| 27 |
+
"learning_rate": 1.866577718478986e-05,
|
| 28 |
+
"loss": 0.3046,
|
| 29 |
+
"step": 4000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.33,
|
| 33 |
+
"eval_acc": 0.8644831115660184,
|
| 34 |
+
"eval_loss": 0.40244850516319275,
|
| 35 |
+
"eval_runtime": 16.6646,
|
| 36 |
+
"eval_samples_per_second": 586.274,
|
| 37 |
+
"eval_steps_per_second": 18.362,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"learning_rate": 1.7998665777184793e-05,
|
| 43 |
+
"loss": 0.3038,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"eval_acc": 0.8668372569089048,
|
| 49 |
+
"eval_loss": 0.39355912804603577,
|
| 50 |
+
"eval_runtime": 16.6429,
|
| 51 |
+
"eval_samples_per_second": 587.039,
|
| 52 |
+
"eval_steps_per_second": 18.386,
|
| 53 |
+
"step": 6000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.67,
|
| 57 |
+
"learning_rate": 1.7331554369579722e-05,
|
| 58 |
+
"loss": 0.3012,
|
| 59 |
+
"step": 8000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.67,
|
| 63 |
+
"eval_acc": 0.8625383828045036,
|
| 64 |
+
"eval_loss": 0.40068650245666504,
|
| 65 |
+
"eval_runtime": 16.6669,
|
| 66 |
+
"eval_samples_per_second": 586.192,
|
| 67 |
+
"eval_steps_per_second": 18.36,
|
| 68 |
+
"step": 8000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.83,
|
| 72 |
+
"learning_rate": 1.666444296197465e-05,
|
| 73 |
+
"loss": 0.2979,
|
| 74 |
+
"step": 10000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.83,
|
| 78 |
+
"eval_acc": 0.8620266120777892,
|
| 79 |
+
"eval_loss": 0.42349421977996826,
|
| 80 |
+
"eval_runtime": 16.6784,
|
| 81 |
+
"eval_samples_per_second": 585.787,
|
| 82 |
+
"eval_steps_per_second": 18.347,
|
| 83 |
+
"step": 10000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.0,
|
| 87 |
+
"learning_rate": 1.599733155436958e-05,
|
| 88 |
+
"loss": 0.2997,
|
| 89 |
+
"step": 12000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.0,
|
| 93 |
+
"eval_acc": 0.8643807574206755,
|
| 94 |
+
"eval_loss": 0.403084933757782,
|
| 95 |
+
"eval_runtime": 16.6652,
|
| 96 |
+
"eval_samples_per_second": 586.251,
|
| 97 |
+
"eval_steps_per_second": 18.362,
|
| 98 |
+
"step": 12000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.17,
|
| 102 |
+
"learning_rate": 1.533022014676451e-05,
|
| 103 |
+
"loss": 0.2099,
|
| 104 |
+
"step": 14000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.17,
|
| 108 |
+
"eval_acc": 0.8632548618219038,
|
| 109 |
+
"eval_loss": 0.43926700949668884,
|
| 110 |
+
"eval_runtime": 16.6492,
|
| 111 |
+
"eval_samples_per_second": 586.816,
|
| 112 |
+
"eval_steps_per_second": 18.379,
|
| 113 |
+
"step": 14000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.33,
|
| 117 |
+
"learning_rate": 1.4663108739159441e-05,
|
| 118 |
+
"loss": 0.2114,
|
| 119 |
+
"step": 16000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.33,
|
| 123 |
+
"eval_acc": 0.8628454452405322,
|
| 124 |
+
"eval_loss": 0.46620267629623413,
|
| 125 |
+
"eval_runtime": 16.6551,
|
| 126 |
+
"eval_samples_per_second": 586.607,
|
| 127 |
+
"eval_steps_per_second": 18.373,
|
| 128 |
+
"step": 16000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.5,
|
| 132 |
+
"learning_rate": 1.3995997331554372e-05,
|
| 133 |
+
"loss": 0.2147,
|
| 134 |
+
"step": 18000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5,
|
| 138 |
+
"eval_acc": 0.864790174002047,
|
| 139 |
+
"eval_loss": 0.4331408143043518,
|
| 140 |
+
"eval_runtime": 16.6453,
|
| 141 |
+
"eval_samples_per_second": 586.954,
|
| 142 |
+
"eval_steps_per_second": 18.384,
|
| 143 |
+
"step": 18000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.67,
|
| 147 |
+
"learning_rate": 1.33288859239493e-05,
|
| 148 |
+
"loss": 0.2122,
|
| 149 |
+
"step": 20000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.67,
|
| 153 |
+
"eval_acc": 0.87021494370522,
|
| 154 |
+
"eval_loss": 0.4166066646575928,
|
| 155 |
+
"eval_runtime": 16.67,
|
| 156 |
+
"eval_samples_per_second": 586.082,
|
| 157 |
+
"eval_steps_per_second": 18.356,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.83,
|
| 162 |
+
"learning_rate": 1.2661774516344229e-05,
|
| 163 |
+
"loss": 0.2156,
|
| 164 |
+
"step": 22000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.83,
|
| 168 |
+
"eval_acc": 0.8632548618219038,
|
| 169 |
+
"eval_loss": 0.4462782144546509,
|
| 170 |
+
"eval_runtime": 16.6855,
|
| 171 |
+
"eval_samples_per_second": 585.539,
|
| 172 |
+
"eval_steps_per_second": 18.339,
|
| 173 |
+
"step": 22000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.0,
|
| 177 |
+
"learning_rate": 1.199466310873916e-05,
|
| 178 |
+
"loss": 0.2117,
|
| 179 |
+
"step": 24000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_acc": 0.8679631525076765,
|
| 184 |
+
"eval_loss": 0.46366986632347107,
|
| 185 |
+
"eval_runtime": 16.6496,
|
| 186 |
+
"eval_samples_per_second": 586.801,
|
| 187 |
+
"eval_steps_per_second": 18.379,
|
| 188 |
+
"step": 24000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.17,
|
| 192 |
+
"learning_rate": 1.132755170113409e-05,
|
| 193 |
+
"loss": 0.1469,
|
| 194 |
+
"step": 26000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.17,
|
| 198 |
+
"eval_acc": 0.8680655066530194,
|
| 199 |
+
"eval_loss": 0.5210850834846497,
|
| 200 |
+
"eval_runtime": 16.6494,
|
| 201 |
+
"eval_samples_per_second": 586.808,
|
| 202 |
+
"eval_steps_per_second": 18.379,
|
| 203 |
+
"step": 26000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.33,
|
| 207 |
+
"learning_rate": 1.066044029352902e-05,
|
| 208 |
+
"loss": 0.1526,
|
| 209 |
+
"step": 28000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.33,
|
| 213 |
+
"eval_acc": 0.8620266120777892,
|
| 214 |
+
"eval_loss": 0.5206254720687866,
|
| 215 |
+
"eval_runtime": 16.6442,
|
| 216 |
+
"eval_samples_per_second": 586.99,
|
| 217 |
+
"eval_steps_per_second": 18.385,
|
| 218 |
+
"step": 28000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.5,
|
| 222 |
+
"learning_rate": 9.99332888592395e-06,
|
| 223 |
+
"loss": 0.1494,
|
| 224 |
+
"step": 30000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.5,
|
| 228 |
+
"eval_acc": 0.8664278403275333,
|
| 229 |
+
"eval_loss": 0.5167897939682007,
|
| 230 |
+
"eval_runtime": 16.6489,
|
| 231 |
+
"eval_samples_per_second": 586.825,
|
| 232 |
+
"eval_steps_per_second": 18.38,
|
| 233 |
+
"step": 30000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.67,
|
| 237 |
+
"learning_rate": 9.326217478318879e-06,
|
| 238 |
+
"loss": 0.1519,
|
| 239 |
+
"step": 32000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.67,
|
| 243 |
+
"eval_acc": 0.8700102354145343,
|
| 244 |
+
"eval_loss": 0.48301637172698975,
|
| 245 |
+
"eval_runtime": 16.6933,
|
| 246 |
+
"eval_samples_per_second": 585.265,
|
| 247 |
+
"eval_steps_per_second": 18.331,
|
| 248 |
+
"step": 32000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.84,
|
| 252 |
+
"learning_rate": 8.65910607071381e-06,
|
| 253 |
+
"loss": 0.152,
|
| 254 |
+
"step": 34000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.84,
|
| 258 |
+
"eval_acc": 0.8635619242579324,
|
| 259 |
+
"eval_loss": 0.5464909672737122,
|
| 260 |
+
"eval_runtime": 16.6533,
|
| 261 |
+
"eval_samples_per_second": 586.671,
|
| 262 |
+
"eval_steps_per_second": 18.375,
|
| 263 |
+
"step": 34000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.0,
|
| 267 |
+
"learning_rate": 7.99199466310874e-06,
|
| 268 |
+
"loss": 0.1498,
|
| 269 |
+
"step": 36000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.0,
|
| 273 |
+
"eval_acc": 0.8679631525076765,
|
| 274 |
+
"eval_loss": 0.5550450682640076,
|
| 275 |
+
"eval_runtime": 16.6712,
|
| 276 |
+
"eval_samples_per_second": 586.039,
|
| 277 |
+
"eval_steps_per_second": 18.355,
|
| 278 |
+
"step": 36000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.17,
|
| 282 |
+
"learning_rate": 7.324883255503669e-06,
|
| 283 |
+
"loss": 0.1131,
|
| 284 |
+
"step": 38000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.17,
|
| 288 |
+
"eval_acc": 0.8601842374616172,
|
| 289 |
+
"eval_loss": 0.6764107346534729,
|
| 290 |
+
"eval_runtime": 16.6576,
|
| 291 |
+
"eval_samples_per_second": 586.52,
|
| 292 |
+
"eval_steps_per_second": 18.37,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.34,
|
| 297 |
+
"learning_rate": 6.6577718478985995e-06,
|
| 298 |
+
"loss": 0.1135,
|
| 299 |
+
"step": 40000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.34,
|
| 303 |
+
"eval_acc": 0.865711361310133,
|
| 304 |
+
"eval_loss": 0.6199905276298523,
|
| 305 |
+
"eval_runtime": 16.6534,
|
| 306 |
+
"eval_samples_per_second": 586.668,
|
| 307 |
+
"eval_steps_per_second": 18.375,
|
| 308 |
+
"step": 40000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.5,
|
| 312 |
+
"learning_rate": 5.99066044029353e-06,
|
| 313 |
+
"loss": 0.1175,
|
| 314 |
+
"step": 42000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"eval_acc": 0.8671443193449335,
|
| 319 |
+
"eval_loss": 0.588898241519928,
|
| 320 |
+
"eval_runtime": 16.664,
|
| 321 |
+
"eval_samples_per_second": 586.293,
|
| 322 |
+
"eval_steps_per_second": 18.363,
|
| 323 |
+
"step": 42000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.67,
|
| 327 |
+
"learning_rate": 5.32354903268846e-06,
|
| 328 |
+
"loss": 0.1156,
|
| 329 |
+
"step": 44000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.67,
|
| 333 |
+
"eval_acc": 0.8663254861821904,
|
| 334 |
+
"eval_loss": 0.6299933195114136,
|
| 335 |
+
"eval_runtime": 16.6491,
|
| 336 |
+
"eval_samples_per_second": 586.818,
|
| 337 |
+
"eval_steps_per_second": 18.379,
|
| 338 |
+
"step": 44000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.84,
|
| 342 |
+
"learning_rate": 4.656437625083389e-06,
|
| 343 |
+
"loss": 0.1104,
|
| 344 |
+
"step": 46000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.84,
|
| 348 |
+
"eval_acc": 0.8689866939611054,
|
| 349 |
+
"eval_loss": 0.6044776439666748,
|
| 350 |
+
"eval_runtime": 16.6309,
|
| 351 |
+
"eval_samples_per_second": 587.46,
|
| 352 |
+
"eval_steps_per_second": 18.399,
|
| 353 |
+
"step": 46000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.0,
|
| 357 |
+
"learning_rate": 3.989326217478319e-06,
|
| 358 |
+
"loss": 0.1111,
|
| 359 |
+
"step": 48000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.0,
|
| 363 |
+
"eval_acc": 0.869396110542477,
|
| 364 |
+
"eval_loss": 0.6412716507911682,
|
| 365 |
+
"eval_runtime": 16.6717,
|
| 366 |
+
"eval_samples_per_second": 586.023,
|
| 367 |
+
"eval_steps_per_second": 18.354,
|
| 368 |
+
"step": 48000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.17,
|
| 372 |
+
"learning_rate": 3.3222148098732494e-06,
|
| 373 |
+
"loss": 0.086,
|
| 374 |
+
"step": 50000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.17,
|
| 378 |
+
"eval_acc": 0.8658137154554759,
|
| 379 |
+
"eval_loss": 0.7271037697792053,
|
| 380 |
+
"eval_runtime": 16.6567,
|
| 381 |
+
"eval_samples_per_second": 586.55,
|
| 382 |
+
"eval_steps_per_second": 18.371,
|
| 383 |
+
"step": 50000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.34,
|
| 387 |
+
"learning_rate": 2.6551034022681787e-06,
|
| 388 |
+
"loss": 0.0895,
|
| 389 |
+
"step": 52000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.34,
|
| 393 |
+
"eval_acc": 0.8682702149437053,
|
| 394 |
+
"eval_loss": 0.7273563146591187,
|
| 395 |
+
"eval_runtime": 16.678,
|
| 396 |
+
"eval_samples_per_second": 585.801,
|
| 397 |
+
"eval_steps_per_second": 18.348,
|
| 398 |
+
"step": 52000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.5,
|
| 402 |
+
"learning_rate": 1.987991994663109e-06,
|
| 403 |
+
"loss": 0.0867,
|
| 404 |
+
"step": 54000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.5,
|
| 408 |
+
"eval_acc": 0.8658137154554759,
|
| 409 |
+
"eval_loss": 0.7226472496986389,
|
| 410 |
+
"eval_runtime": 16.6651,
|
| 411 |
+
"eval_samples_per_second": 586.253,
|
| 412 |
+
"eval_steps_per_second": 18.362,
|
| 413 |
+
"step": 54000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.67,
|
| 417 |
+
"learning_rate": 1.3208805870580388e-06,
|
| 418 |
+
"loss": 0.0886,
|
| 419 |
+
"step": 56000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.67,
|
| 423 |
+
"eval_acc": 0.8690890481064483,
|
| 424 |
+
"eval_loss": 0.7181665897369385,
|
| 425 |
+
"eval_runtime": 16.6571,
|
| 426 |
+
"eval_samples_per_second": 586.536,
|
| 427 |
+
"eval_steps_per_second": 18.371,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.84,
|
| 432 |
+
"learning_rate": 6.537691794529687e-07,
|
| 433 |
+
"loss": 0.0849,
|
| 434 |
+
"step": 58000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.84,
|
| 438 |
+
"eval_acc": 0.8698055271238485,
|
| 439 |
+
"eval_loss": 0.7094203233718872,
|
| 440 |
+
"eval_runtime": 16.6693,
|
| 441 |
+
"eval_samples_per_second": 586.108,
|
| 442 |
+
"eval_steps_per_second": 18.357,
|
| 443 |
+
"step": 58000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 5.0,
|
| 447 |
+
"step": 59960,
|
| 448 |
+
"total_flos": 1.2620695529804544e+17,
|
| 449 |
+
"train_loss": 0.17320066710326415,
|
| 450 |
+
"train_runtime": 11406.3483,
|
| 451 |
+
"train_samples_per_second": 168.211,
|
| 452 |
+
"train_steps_per_second": 5.257
|
| 453 |
+
}
|
| 454 |
+
],
|
| 455 |
+
"max_steps": 59960,
|
| 456 |
+
"num_train_epochs": 5,
|
| 457 |
+
"total_flos": 1.2620695529804544e+17,
|
| 458 |
+
"trial_name": null,
|
| 459 |
+
"trial_params": null
|
| 460 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3451
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56e79755c71603339bde955dcb2e8a6d1359c72222bc22c5b0a3859cadfb4942
|
| 3 |
size 3451
|