Commit
•
812be88
1
Parent(s):
35d8058
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 +475 -0
- training_args.bin +1 -1
AppE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8634595701125896,
|
| 4 |
+
"eval_loss": 0.5700035095214844,
|
| 5 |
+
"eval_runtime": 16.6459,
|
| 6 |
+
"eval_samples_per_second": 586.932,
|
| 7 |
+
"eval_steps_per_second": 18.383
|
| 8 |
+
}
|
ChcE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8649948822927328,
|
| 4 |
+
"eval_loss": 0.546633780002594,
|
| 5 |
+
"eval_runtime": 16.6322,
|
| 6 |
+
"eval_samples_per_second": 587.416,
|
| 7 |
+
"eval_steps_per_second": 18.398
|
| 8 |
+
}
|
CollSgE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8150460593654043,
|
| 4 |
+
"eval_loss": 0.7232102751731873,
|
| 5 |
+
"eval_runtime": 16.6475,
|
| 6 |
+
"eval_samples_per_second": 586.875,
|
| 7 |
+
"eval_steps_per_second": 18.381
|
| 8 |
+
}
|
GLUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8682628629648497,
|
| 4 |
+
"eval_loss": 0.5342766046524048,
|
| 5 |
+
"eval_runtime": 16.7581,
|
| 6 |
+
"eval_samples_per_second": 585.687,
|
| 7 |
+
"eval_steps_per_second": 18.32
|
| 8 |
+
}
|
IndE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8244626407369499,
|
| 4 |
+
"eval_loss": 0.7053251266479492,
|
| 5 |
+
"eval_runtime": 16.6432,
|
| 6 |
+
"eval_samples_per_second": 587.026,
|
| 7 |
+
"eval_steps_per_second": 18.386
|
| 8 |
+
}
|
MULTI_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.7935371714899274,
|
| 4 |
+
"eval_loss": 0.8028998374938965,
|
| 5 |
+
"eval_runtime": 16.7814,
|
| 6 |
+
"eval_samples_per_second": 582.727,
|
| 7 |
+
"eval_steps_per_second": 18.234
|
| 8 |
+
}
|
VALUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8425878757004585,
|
| 4 |
+
"eval_loss": 0.6238889098167419,
|
| 5 |
+
"eval_runtime": 16.7103,
|
| 6 |
+
"eval_samples_per_second": 587.361,
|
| 7 |
+
"eval_steps_per_second": 18.372
|
| 8 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8682628629648497,
|
| 4 |
+
"eval_loss": 0.5342766046524048,
|
| 5 |
+
"eval_runtime": 16.7581,
|
| 6 |
+
"eval_samples_per_second": 585.687,
|
| 7 |
+
"eval_steps_per_second": 18.32,
|
| 8 |
+
"train_loss": 0.18972002029418944,
|
| 9 |
+
"train_runtime": 11504.4253,
|
| 10 |
+
"train_samples_per_second": 166.881,
|
| 11 |
+
"train_steps_per_second": 5.215
|
| 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:ddb8e6ae561d353a5c667e06d71a6dd252a2a5fdc06e294519c1d48acf1b67c5
|
| 3 |
size 498665141
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.18972002029418944,
|
| 4 |
+
"train_runtime": 11504.4253,
|
| 5 |
+
"train_samples_per_second": 166.881,
|
| 6 |
+
"train_steps_per_second": 5.215
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,475 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8634595701125896,
|
| 3 |
+
"best_model_checkpoint": "./outputs/finetuning/mnli_AppE/checkpoint-26000",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 60000,
|
| 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": 1.9333333333333333e-05,
|
| 13 |
+
"loss": 0.3562,
|
| 14 |
+
"step": 2000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.17,
|
| 18 |
+
"eval_acc": 0.8524053224155578,
|
| 19 |
+
"eval_loss": 0.41664522886276245,
|
| 20 |
+
"eval_runtime": 16.7668,
|
| 21 |
+
"eval_samples_per_second": 582.7,
|
| 22 |
+
"eval_steps_per_second": 18.25,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33,
|
| 27 |
+
"learning_rate": 1.866666666666667e-05,
|
| 28 |
+
"loss": 0.3414,
|
| 29 |
+
"step": 4000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.33,
|
| 33 |
+
"eval_acc": 0.8475946775844422,
|
| 34 |
+
"eval_loss": 0.43986669182777405,
|
| 35 |
+
"eval_runtime": 16.7213,
|
| 36 |
+
"eval_samples_per_second": 584.285,
|
| 37 |
+
"eval_steps_per_second": 18.3,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"learning_rate": 1.8e-05,
|
| 43 |
+
"loss": 0.3308,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"eval_acc": 0.8560900716479017,
|
| 49 |
+
"eval_loss": 0.395113080739975,
|
| 50 |
+
"eval_runtime": 16.7306,
|
| 51 |
+
"eval_samples_per_second": 583.96,
|
| 52 |
+
"eval_steps_per_second": 18.29,
|
| 53 |
+
"step": 6000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.67,
|
| 57 |
+
"learning_rate": 1.7333333333333336e-05,
|
| 58 |
+
"loss": 0.3306,
|
| 59 |
+
"step": 8000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.67,
|
| 63 |
+
"eval_acc": 0.8488229273285568,
|
| 64 |
+
"eval_loss": 0.41524460911750793,
|
| 65 |
+
"eval_runtime": 16.7003,
|
| 66 |
+
"eval_samples_per_second": 585.021,
|
| 67 |
+
"eval_steps_per_second": 18.323,
|
| 68 |
+
"step": 8000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.83,
|
| 72 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 73 |
+
"loss": 0.3295,
|
| 74 |
+
"step": 10000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.83,
|
| 78 |
+
"eval_acc": 0.8575230296827021,
|
| 79 |
+
"eval_loss": 0.39166104793548584,
|
| 80 |
+
"eval_runtime": 16.6831,
|
| 81 |
+
"eval_samples_per_second": 585.622,
|
| 82 |
+
"eval_steps_per_second": 18.342,
|
| 83 |
+
"step": 10000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.0,
|
| 87 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 88 |
+
"loss": 0.3288,
|
| 89 |
+
"step": 12000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.0,
|
| 93 |
+
"eval_acc": 0.8581371545547595,
|
| 94 |
+
"eval_loss": 0.40951114892959595,
|
| 95 |
+
"eval_runtime": 16.6934,
|
| 96 |
+
"eval_samples_per_second": 585.263,
|
| 97 |
+
"eval_steps_per_second": 18.331,
|
| 98 |
+
"step": 12000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.17,
|
| 102 |
+
"learning_rate": 1.5333333333333334e-05,
|
| 103 |
+
"loss": 0.229,
|
| 104 |
+
"step": 14000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.17,
|
| 108 |
+
"eval_acc": 0.8611054247697032,
|
| 109 |
+
"eval_loss": 0.4420296251773834,
|
| 110 |
+
"eval_runtime": 16.7189,
|
| 111 |
+
"eval_samples_per_second": 584.367,
|
| 112 |
+
"eval_steps_per_second": 18.303,
|
| 113 |
+
"step": 14000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.33,
|
| 117 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 118 |
+
"loss": 0.2307,
|
| 119 |
+
"step": 16000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.33,
|
| 123 |
+
"eval_acc": 0.8560900716479017,
|
| 124 |
+
"eval_loss": 0.4876081645488739,
|
| 125 |
+
"eval_runtime": 16.6966,
|
| 126 |
+
"eval_samples_per_second": 585.148,
|
| 127 |
+
"eval_steps_per_second": 18.327,
|
| 128 |
+
"step": 16000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.5,
|
| 132 |
+
"learning_rate": 1.4e-05,
|
| 133 |
+
"loss": 0.2334,
|
| 134 |
+
"step": 18000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5,
|
| 138 |
+
"eval_acc": 0.857727737973388,
|
| 139 |
+
"eval_loss": 0.449561208486557,
|
| 140 |
+
"eval_runtime": 16.7074,
|
| 141 |
+
"eval_samples_per_second": 584.771,
|
| 142 |
+
"eval_steps_per_second": 18.315,
|
| 143 |
+
"step": 18000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.67,
|
| 147 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 148 |
+
"loss": 0.2392,
|
| 149 |
+
"step": 20000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.67,
|
| 153 |
+
"eval_acc": 0.8607983623336745,
|
| 154 |
+
"eval_loss": 0.4188033640384674,
|
| 155 |
+
"eval_runtime": 16.694,
|
| 156 |
+
"eval_samples_per_second": 585.239,
|
| 157 |
+
"eval_steps_per_second": 18.33,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.83,
|
| 162 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 163 |
+
"loss": 0.233,
|
| 164 |
+
"step": 22000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.83,
|
| 168 |
+
"eval_acc": 0.8578300921187308,
|
| 169 |
+
"eval_loss": 0.44934678077697754,
|
| 170 |
+
"eval_runtime": 16.6959,
|
| 171 |
+
"eval_samples_per_second": 585.174,
|
| 172 |
+
"eval_steps_per_second": 18.328,
|
| 173 |
+
"step": 22000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.0,
|
| 177 |
+
"learning_rate": 1.2e-05,
|
| 178 |
+
"loss": 0.2343,
|
| 179 |
+
"step": 24000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_acc": 0.8602865916069601,
|
| 184 |
+
"eval_loss": 0.42783743143081665,
|
| 185 |
+
"eval_runtime": 16.7147,
|
| 186 |
+
"eval_samples_per_second": 584.517,
|
| 187 |
+
"eval_steps_per_second": 18.307,
|
| 188 |
+
"step": 24000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.17,
|
| 192 |
+
"learning_rate": 1.1333333333333334e-05,
|
| 193 |
+
"loss": 0.163,
|
| 194 |
+
"step": 26000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.17,
|
| 198 |
+
"eval_acc": 0.8634595701125896,
|
| 199 |
+
"eval_loss": 0.5700035095214844,
|
| 200 |
+
"eval_runtime": 16.7469,
|
| 201 |
+
"eval_samples_per_second": 583.393,
|
| 202 |
+
"eval_steps_per_second": 18.272,
|
| 203 |
+
"step": 26000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.33,
|
| 207 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 208 |
+
"loss": 0.1657,
|
| 209 |
+
"step": 28000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.33,
|
| 213 |
+
"eval_acc": 0.8560900716479017,
|
| 214 |
+
"eval_loss": 0.5322555303573608,
|
| 215 |
+
"eval_runtime": 16.7051,
|
| 216 |
+
"eval_samples_per_second": 584.851,
|
| 217 |
+
"eval_steps_per_second": 18.318,
|
| 218 |
+
"step": 28000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.5,
|
| 222 |
+
"learning_rate": 1e-05,
|
| 223 |
+
"loss": 0.1652,
|
| 224 |
+
"step": 30000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.5,
|
| 228 |
+
"eval_acc": 0.8588536335721597,
|
| 229 |
+
"eval_loss": 0.5046815872192383,
|
| 230 |
+
"eval_runtime": 16.7454,
|
| 231 |
+
"eval_samples_per_second": 583.445,
|
| 232 |
+
"eval_steps_per_second": 18.274,
|
| 233 |
+
"step": 30000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.67,
|
| 237 |
+
"learning_rate": 9.333333333333334e-06,
|
| 238 |
+
"loss": 0.1654,
|
| 239 |
+
"step": 32000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.67,
|
| 243 |
+
"eval_acc": 0.8589559877175026,
|
| 244 |
+
"eval_loss": 0.5081538558006287,
|
| 245 |
+
"eval_runtime": 16.6963,
|
| 246 |
+
"eval_samples_per_second": 585.158,
|
| 247 |
+
"eval_steps_per_second": 18.327,
|
| 248 |
+
"step": 32000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.83,
|
| 252 |
+
"learning_rate": 8.666666666666668e-06,
|
| 253 |
+
"loss": 0.1659,
|
| 254 |
+
"step": 34000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.83,
|
| 258 |
+
"eval_acc": 0.8630501535312181,
|
| 259 |
+
"eval_loss": 0.5075551271438599,
|
| 260 |
+
"eval_runtime": 16.7114,
|
| 261 |
+
"eval_samples_per_second": 584.632,
|
| 262 |
+
"eval_steps_per_second": 18.311,
|
| 263 |
+
"step": 34000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.0,
|
| 267 |
+
"learning_rate": 8.000000000000001e-06,
|
| 268 |
+
"loss": 0.1654,
|
| 269 |
+
"step": 36000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.0,
|
| 273 |
+
"eval_acc": 0.8612077789150461,
|
| 274 |
+
"eval_loss": 0.5035998821258545,
|
| 275 |
+
"eval_runtime": 16.7468,
|
| 276 |
+
"eval_samples_per_second": 583.394,
|
| 277 |
+
"eval_steps_per_second": 18.272,
|
| 278 |
+
"step": 36000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.17,
|
| 282 |
+
"learning_rate": 7.333333333333333e-06,
|
| 283 |
+
"loss": 0.1195,
|
| 284 |
+
"step": 38000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.17,
|
| 288 |
+
"eval_acc": 0.8599795291709315,
|
| 289 |
+
"eval_loss": 0.6220871806144714,
|
| 290 |
+
"eval_runtime": 16.6963,
|
| 291 |
+
"eval_samples_per_second": 585.161,
|
| 292 |
+
"eval_steps_per_second": 18.327,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.33,
|
| 297 |
+
"learning_rate": 6.666666666666667e-06,
|
| 298 |
+
"loss": 0.1194,
|
| 299 |
+
"step": 40000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.33,
|
| 303 |
+
"eval_acc": 0.8575230296827021,
|
| 304 |
+
"eval_loss": 0.6466721892356873,
|
| 305 |
+
"eval_runtime": 16.6726,
|
| 306 |
+
"eval_samples_per_second": 585.991,
|
| 307 |
+
"eval_steps_per_second": 18.353,
|
| 308 |
+
"step": 40000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.5,
|
| 312 |
+
"learning_rate": 6e-06,
|
| 313 |
+
"loss": 0.1205,
|
| 314 |
+
"step": 42000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"eval_acc": 0.8578300921187308,
|
| 319 |
+
"eval_loss": 0.6742368340492249,
|
| 320 |
+
"eval_runtime": 16.6999,
|
| 321 |
+
"eval_samples_per_second": 585.034,
|
| 322 |
+
"eval_steps_per_second": 18.323,
|
| 323 |
+
"step": 42000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.67,
|
| 327 |
+
"learning_rate": 5.333333333333334e-06,
|
| 328 |
+
"loss": 0.1209,
|
| 329 |
+
"step": 44000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.67,
|
| 333 |
+
"eval_acc": 0.8580348004094166,
|
| 334 |
+
"eval_loss": 0.6536660194396973,
|
| 335 |
+
"eval_runtime": 16.7907,
|
| 336 |
+
"eval_samples_per_second": 581.87,
|
| 337 |
+
"eval_steps_per_second": 18.224,
|
| 338 |
+
"step": 44000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.83,
|
| 342 |
+
"learning_rate": 4.666666666666667e-06,
|
| 343 |
+
"loss": 0.1234,
|
| 344 |
+
"step": 46000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.83,
|
| 348 |
+
"eval_acc": 0.8580348004094166,
|
| 349 |
+
"eval_loss": 0.6385037899017334,
|
| 350 |
+
"eval_runtime": 16.6773,
|
| 351 |
+
"eval_samples_per_second": 585.826,
|
| 352 |
+
"eval_steps_per_second": 18.348,
|
| 353 |
+
"step": 46000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.0,
|
| 357 |
+
"learning_rate": 4.000000000000001e-06,
|
| 358 |
+
"loss": 0.122,
|
| 359 |
+
"step": 48000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.0,
|
| 363 |
+
"eval_acc": 0.8628454452405322,
|
| 364 |
+
"eval_loss": 0.6103753447532654,
|
| 365 |
+
"eval_runtime": 16.681,
|
| 366 |
+
"eval_samples_per_second": 585.697,
|
| 367 |
+
"eval_steps_per_second": 18.344,
|
| 368 |
+
"step": 48000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.17,
|
| 372 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 373 |
+
"loss": 0.0937,
|
| 374 |
+
"step": 50000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.17,
|
| 378 |
+
"eval_acc": 0.8602865916069601,
|
| 379 |
+
"eval_loss": 0.7260550260543823,
|
| 380 |
+
"eval_runtime": 16.6866,
|
| 381 |
+
"eval_samples_per_second": 585.501,
|
| 382 |
+
"eval_steps_per_second": 18.338,
|
| 383 |
+
"step": 50000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.33,
|
| 387 |
+
"learning_rate": 2.666666666666667e-06,
|
| 388 |
+
"loss": 0.0941,
|
| 389 |
+
"step": 52000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.33,
|
| 393 |
+
"eval_acc": 0.8606960081883316,
|
| 394 |
+
"eval_loss": 0.7633941769599915,
|
| 395 |
+
"eval_runtime": 16.696,
|
| 396 |
+
"eval_samples_per_second": 585.17,
|
| 397 |
+
"eval_steps_per_second": 18.328,
|
| 398 |
+
"step": 52000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.5,
|
| 402 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 403 |
+
"loss": 0.0926,
|
| 404 |
+
"step": 54000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.5,
|
| 408 |
+
"eval_acc": 0.8617195496417605,
|
| 409 |
+
"eval_loss": 0.7736791968345642,
|
| 410 |
+
"eval_runtime": 16.6764,
|
| 411 |
+
"eval_samples_per_second": 585.859,
|
| 412 |
+
"eval_steps_per_second": 18.349,
|
| 413 |
+
"step": 54000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.67,
|
| 417 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 418 |
+
"loss": 0.093,
|
| 419 |
+
"step": 56000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.67,
|
| 423 |
+
"eval_acc": 0.8632548618219038,
|
| 424 |
+
"eval_loss": 0.757666289806366,
|
| 425 |
+
"eval_runtime": 16.6803,
|
| 426 |
+
"eval_samples_per_second": 585.722,
|
| 427 |
+
"eval_steps_per_second": 18.345,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.83,
|
| 432 |
+
"learning_rate": 6.666666666666667e-07,
|
| 433 |
+
"loss": 0.0918,
|
| 434 |
+
"step": 58000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.83,
|
| 438 |
+
"eval_acc": 0.8630501535312181,
|
| 439 |
+
"eval_loss": 0.7534608840942383,
|
| 440 |
+
"eval_runtime": 16.6933,
|
| 441 |
+
"eval_samples_per_second": 585.263,
|
| 442 |
+
"eval_steps_per_second": 18.331,
|
| 443 |
+
"step": 58000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 5.0,
|
| 447 |
+
"learning_rate": 0.0,
|
| 448 |
+
"loss": 0.0935,
|
| 449 |
+
"step": 60000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 5.0,
|
| 453 |
+
"eval_acc": 0.8622313203684749,
|
| 454 |
+
"eval_loss": 0.743323802947998,
|
| 455 |
+
"eval_runtime": 16.7307,
|
| 456 |
+
"eval_samples_per_second": 583.956,
|
| 457 |
+
"eval_steps_per_second": 18.29,
|
| 458 |
+
"step": 60000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 5.0,
|
| 462 |
+
"step": 60000,
|
| 463 |
+
"total_flos": 1.2628556043159168e+17,
|
| 464 |
+
"train_loss": 0.18972002029418944,
|
| 465 |
+
"train_runtime": 11504.4253,
|
| 466 |
+
"train_samples_per_second": 166.881,
|
| 467 |
+
"train_steps_per_second": 5.215
|
| 468 |
+
}
|
| 469 |
+
],
|
| 470 |
+
"max_steps": 60000,
|
| 471 |
+
"num_train_epochs": 5,
|
| 472 |
+
"total_flos": 1.2628556043159168e+17,
|
| 473 |
+
"trial_name": null,
|
| 474 |
+
"trial_params": null
|
| 475 |
+
}
|
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 sha256:9da8440371189831d2889e0dbeabc0b6ffa34e68b097c74fe817a356213a5b91
|
| 3 |
size 3451
|