Commit
•
3f4b0bd
1
Parent(s):
5810c46
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": 0.8482088024564994,
|
| 4 |
+
"eval_loss": 0.6413494348526001,
|
| 5 |
+
"eval_runtime": 16.6247,
|
| 6 |
+
"eval_samples_per_second": 587.679,
|
| 7 |
+
"eval_steps_per_second": 18.406
|
| 8 |
+
}
|
ChcE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8598771750255886,
|
| 4 |
+
"eval_loss": 0.6220324039459229,
|
| 5 |
+
"eval_runtime": 16.618,
|
| 6 |
+
"eval_samples_per_second": 587.918,
|
| 7 |
+
"eval_steps_per_second": 18.414
|
| 8 |
+
}
|
CollSgE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8503582395087002,
|
| 4 |
+
"eval_loss": 0.6494254469871521,
|
| 5 |
+
"eval_runtime": 16.6141,
|
| 6 |
+
"eval_samples_per_second": 588.055,
|
| 7 |
+
"eval_steps_per_second": 18.418
|
| 8 |
+
}
|
GLUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8712175241976566,
|
| 4 |
+
"eval_loss": 0.5597977638244629,
|
| 5 |
+
"eval_runtime": 16.7338,
|
| 6 |
+
"eval_samples_per_second": 586.538,
|
| 7 |
+
"eval_steps_per_second": 18.346
|
| 8 |
+
}
|
IndE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8372569089048106,
|
| 4 |
+
"eval_loss": 0.6829524636268616,
|
| 5 |
+
"eval_runtime": 16.6186,
|
| 6 |
+
"eval_samples_per_second": 587.896,
|
| 7 |
+
"eval_steps_per_second": 18.413
|
| 8 |
+
}
|
MULTI_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.7931281317108089,
|
| 4 |
+
"eval_loss": 0.835823118686676,
|
| 5 |
+
"eval_runtime": 16.7576,
|
| 6 |
+
"eval_samples_per_second": 583.557,
|
| 7 |
+
"eval_steps_per_second": 18.26
|
| 8 |
+
}
|
VALUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8440142638818136,
|
| 4 |
+
"eval_loss": 0.6680275797843933,
|
| 5 |
+
"eval_runtime": 16.6874,
|
| 6 |
+
"eval_samples_per_second": 588.167,
|
| 7 |
+
"eval_steps_per_second": 18.397
|
| 8 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8712175241976566,
|
| 4 |
+
"eval_loss": 0.5597977638244629,
|
| 5 |
+
"eval_runtime": 16.7338,
|
| 6 |
+
"eval_samples_per_second": 586.538,
|
| 7 |
+
"eval_steps_per_second": 18.346,
|
| 8 |
+
"train_loss": 0.22490199587200385,
|
| 9 |
+
"train_runtime": 11405.3413,
|
| 10 |
+
"train_samples_per_second": 168.222,
|
| 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:fbe0b95b13a74f127b959ae7b8e756d3050dd7f57e3de647e2487ee55b87761e
|
| 3 |
size 498665141
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.22490199587200385,
|
| 4 |
+
"train_runtime": 11405.3413,
|
| 5 |
+
"train_samples_per_second": 168.222,
|
| 6 |
+
"train_steps_per_second": 5.257
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,460 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8503582395087002,
|
| 3 |
+
"best_model_checkpoint": "./outputs/finetuning/mnli_CollSgE/checkpoint-48000",
|
| 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": 0.4123,
|
| 14 |
+
"step": 2000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.17,
|
| 18 |
+
"eval_acc": 0.8331627430910952,
|
| 19 |
+
"eval_loss": 0.46929866075515747,
|
| 20 |
+
"eval_runtime": 16.6562,
|
| 21 |
+
"eval_samples_per_second": 586.569,
|
| 22 |
+
"eval_steps_per_second": 18.372,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33,
|
| 27 |
+
"learning_rate": 1.866577718478986e-05,
|
| 28 |
+
"loss": 0.4028,
|
| 29 |
+
"step": 4000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.33,
|
| 33 |
+
"eval_acc": 0.8337768679631525,
|
| 34 |
+
"eval_loss": 0.46244016289711,
|
| 35 |
+
"eval_runtime": 16.6313,
|
| 36 |
+
"eval_samples_per_second": 587.445,
|
| 37 |
+
"eval_steps_per_second": 18.399,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"learning_rate": 1.7998665777184793e-05,
|
| 43 |
+
"loss": 0.3888,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"eval_acc": 0.8374616171954964,
|
| 49 |
+
"eval_loss": 0.45001623034477234,
|
| 50 |
+
"eval_runtime": 16.6286,
|
| 51 |
+
"eval_samples_per_second": 587.542,
|
| 52 |
+
"eval_steps_per_second": 18.402,
|
| 53 |
+
"step": 6000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.67,
|
| 57 |
+
"learning_rate": 1.7331554369579722e-05,
|
| 58 |
+
"loss": 0.3841,
|
| 59 |
+
"step": 8000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.67,
|
| 63 |
+
"eval_acc": 0.8415557830092119,
|
| 64 |
+
"eval_loss": 0.4280681312084198,
|
| 65 |
+
"eval_runtime": 16.6312,
|
| 66 |
+
"eval_samples_per_second": 587.451,
|
| 67 |
+
"eval_steps_per_second": 18.399,
|
| 68 |
+
"step": 8000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.83,
|
| 72 |
+
"learning_rate": 1.666444296197465e-05,
|
| 73 |
+
"loss": 0.3783,
|
| 74 |
+
"step": 10000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.83,
|
| 78 |
+
"eval_acc": 0.8365404298874104,
|
| 79 |
+
"eval_loss": 0.4433709681034088,
|
| 80 |
+
"eval_runtime": 16.6487,
|
| 81 |
+
"eval_samples_per_second": 586.832,
|
| 82 |
+
"eval_steps_per_second": 18.38,
|
| 83 |
+
"step": 10000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.0,
|
| 87 |
+
"learning_rate": 1.599733155436958e-05,
|
| 88 |
+
"loss": 0.3759,
|
| 89 |
+
"step": 12000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.0,
|
| 93 |
+
"eval_acc": 0.8417604912998976,
|
| 94 |
+
"eval_loss": 0.4399653375148773,
|
| 95 |
+
"eval_runtime": 16.6479,
|
| 96 |
+
"eval_samples_per_second": 586.862,
|
| 97 |
+
"eval_steps_per_second": 18.381,
|
| 98 |
+
"step": 12000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.17,
|
| 102 |
+
"learning_rate": 1.533022014676451e-05,
|
| 103 |
+
"loss": 0.2721,
|
| 104 |
+
"step": 14000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.17,
|
| 108 |
+
"eval_acc": 0.8426816786079836,
|
| 109 |
+
"eval_loss": 0.5021729469299316,
|
| 110 |
+
"eval_runtime": 16.6186,
|
| 111 |
+
"eval_samples_per_second": 587.897,
|
| 112 |
+
"eval_steps_per_second": 18.413,
|
| 113 |
+
"step": 14000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.33,
|
| 117 |
+
"learning_rate": 1.4663108739159441e-05,
|
| 118 |
+
"loss": 0.2736,
|
| 119 |
+
"step": 16000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.33,
|
| 123 |
+
"eval_acc": 0.8430910951893552,
|
| 124 |
+
"eval_loss": 0.5251772403717041,
|
| 125 |
+
"eval_runtime": 16.6517,
|
| 126 |
+
"eval_samples_per_second": 586.728,
|
| 127 |
+
"eval_steps_per_second": 18.377,
|
| 128 |
+
"step": 16000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.5,
|
| 132 |
+
"learning_rate": 1.3995997331554372e-05,
|
| 133 |
+
"loss": 0.2821,
|
| 134 |
+
"step": 18000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5,
|
| 138 |
+
"eval_acc": 0.8409416581371546,
|
| 139 |
+
"eval_loss": 0.48868492245674133,
|
| 140 |
+
"eval_runtime": 16.6338,
|
| 141 |
+
"eval_samples_per_second": 587.358,
|
| 142 |
+
"eval_steps_per_second": 18.396,
|
| 143 |
+
"step": 18000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.67,
|
| 147 |
+
"learning_rate": 1.33288859239493e-05,
|
| 148 |
+
"loss": 0.2802,
|
| 149 |
+
"step": 20000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.67,
|
| 153 |
+
"eval_acc": 0.8457523029682702,
|
| 154 |
+
"eval_loss": 0.4757882356643677,
|
| 155 |
+
"eval_runtime": 16.6205,
|
| 156 |
+
"eval_samples_per_second": 587.829,
|
| 157 |
+
"eval_steps_per_second": 18.411,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.83,
|
| 162 |
+
"learning_rate": 1.2661774516344229e-05,
|
| 163 |
+
"loss": 0.2794,
|
| 164 |
+
"step": 22000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.83,
|
| 168 |
+
"eval_acc": 0.8457523029682702,
|
| 169 |
+
"eval_loss": 0.4611198604106903,
|
| 170 |
+
"eval_runtime": 16.6377,
|
| 171 |
+
"eval_samples_per_second": 587.22,
|
| 172 |
+
"eval_steps_per_second": 18.392,
|
| 173 |
+
"step": 22000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.0,
|
| 177 |
+
"learning_rate": 1.199466310873916e-05,
|
| 178 |
+
"loss": 0.2797,
|
| 179 |
+
"step": 24000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_acc": 0.8456499488229273,
|
| 184 |
+
"eval_loss": 0.49363580346107483,
|
| 185 |
+
"eval_runtime": 16.6189,
|
| 186 |
+
"eval_samples_per_second": 587.883,
|
| 187 |
+
"eval_steps_per_second": 18.413,
|
| 188 |
+
"step": 24000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.17,
|
| 192 |
+
"learning_rate": 1.132755170113409e-05,
|
| 193 |
+
"loss": 0.1915,
|
| 194 |
+
"step": 26000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.17,
|
| 198 |
+
"eval_acc": 0.8461617195496418,
|
| 199 |
+
"eval_loss": 0.5545384883880615,
|
| 200 |
+
"eval_runtime": 16.6314,
|
| 201 |
+
"eval_samples_per_second": 587.443,
|
| 202 |
+
"eval_steps_per_second": 18.399,
|
| 203 |
+
"step": 26000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.33,
|
| 207 |
+
"learning_rate": 1.066044029352902e-05,
|
| 208 |
+
"loss": 0.1946,
|
| 209 |
+
"step": 28000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.33,
|
| 213 |
+
"eval_acc": 0.8443193449334698,
|
| 214 |
+
"eval_loss": 0.5731399059295654,
|
| 215 |
+
"eval_runtime": 16.6331,
|
| 216 |
+
"eval_samples_per_second": 587.383,
|
| 217 |
+
"eval_steps_per_second": 18.397,
|
| 218 |
+
"step": 28000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.5,
|
| 222 |
+
"learning_rate": 9.99332888592395e-06,
|
| 223 |
+
"loss": 0.2007,
|
| 224 |
+
"step": 30000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.5,
|
| 228 |
+
"eval_acc": 0.8427840327533265,
|
| 229 |
+
"eval_loss": 0.5507048964500427,
|
| 230 |
+
"eval_runtime": 16.6762,
|
| 231 |
+
"eval_samples_per_second": 585.864,
|
| 232 |
+
"eval_steps_per_second": 18.349,
|
| 233 |
+
"step": 30000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.67,
|
| 237 |
+
"learning_rate": 9.326217478318879e-06,
|
| 238 |
+
"loss": 0.2008,
|
| 239 |
+
"step": 32000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.67,
|
| 243 |
+
"eval_acc": 0.8454452405322416,
|
| 244 |
+
"eval_loss": 0.5498657822608948,
|
| 245 |
+
"eval_runtime": 16.6352,
|
| 246 |
+
"eval_samples_per_second": 587.307,
|
| 247 |
+
"eval_steps_per_second": 18.395,
|
| 248 |
+
"step": 32000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.84,
|
| 252 |
+
"learning_rate": 8.65910607071381e-06,
|
| 253 |
+
"loss": 0.1971,
|
| 254 |
+
"step": 34000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.84,
|
| 258 |
+
"eval_acc": 0.8483111566018424,
|
| 259 |
+
"eval_loss": 0.5273796916007996,
|
| 260 |
+
"eval_runtime": 16.6214,
|
| 261 |
+
"eval_samples_per_second": 587.797,
|
| 262 |
+
"eval_steps_per_second": 18.41,
|
| 263 |
+
"step": 34000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.0,
|
| 267 |
+
"learning_rate": 7.99199466310874e-06,
|
| 268 |
+
"loss": 0.2054,
|
| 269 |
+
"step": 36000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.0,
|
| 273 |
+
"eval_acc": 0.8475946775844422,
|
| 274 |
+
"eval_loss": 0.5454184412956238,
|
| 275 |
+
"eval_runtime": 16.6713,
|
| 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": 0.1436,
|
| 284 |
+
"step": 38000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.17,
|
| 288 |
+
"eval_acc": 0.844216990788127,
|
| 289 |
+
"eval_loss": 0.6787278652191162,
|
| 290 |
+
"eval_runtime": 16.636,
|
| 291 |
+
"eval_samples_per_second": 587.281,
|
| 292 |
+
"eval_steps_per_second": 18.394,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.34,
|
| 297 |
+
"learning_rate": 6.6577718478985995e-06,
|
| 298 |
+
"loss": 0.1426,
|
| 299 |
+
"step": 40000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.34,
|
| 303 |
+
"eval_acc": 0.8420675537359263,
|
| 304 |
+
"eval_loss": 0.6932513117790222,
|
| 305 |
+
"eval_runtime": 16.6446,
|
| 306 |
+
"eval_samples_per_second": 586.978,
|
| 307 |
+
"eval_steps_per_second": 18.384,
|
| 308 |
+
"step": 40000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.5,
|
| 312 |
+
"learning_rate": 5.99066044029353e-06,
|
| 313 |
+
"loss": 0.1463,
|
| 314 |
+
"step": 42000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"eval_acc": 0.8455475946775844,
|
| 319 |
+
"eval_loss": 0.65469890832901,
|
| 320 |
+
"eval_runtime": 16.6534,
|
| 321 |
+
"eval_samples_per_second": 586.666,
|
| 322 |
+
"eval_steps_per_second": 18.375,
|
| 323 |
+
"step": 42000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.67,
|
| 327 |
+
"learning_rate": 5.32354903268846e-06,
|
| 328 |
+
"loss": 0.1447,
|
| 329 |
+
"step": 44000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.67,
|
| 333 |
+
"eval_acc": 0.8438075742067553,
|
| 334 |
+
"eval_loss": 0.646902859210968,
|
| 335 |
+
"eval_runtime": 16.6236,
|
| 336 |
+
"eval_samples_per_second": 587.717,
|
| 337 |
+
"eval_steps_per_second": 18.408,
|
| 338 |
+
"step": 44000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.84,
|
| 342 |
+
"learning_rate": 4.656437625083389e-06,
|
| 343 |
+
"loss": 0.1445,
|
| 344 |
+
"step": 46000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.84,
|
| 348 |
+
"eval_acc": 0.8471852610030707,
|
| 349 |
+
"eval_loss": 0.6626366376876831,
|
| 350 |
+
"eval_runtime": 16.6568,
|
| 351 |
+
"eval_samples_per_second": 586.548,
|
| 352 |
+
"eval_steps_per_second": 18.371,
|
| 353 |
+
"step": 46000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.0,
|
| 357 |
+
"learning_rate": 3.989326217478319e-06,
|
| 358 |
+
"loss": 0.1457,
|
| 359 |
+
"step": 48000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.0,
|
| 363 |
+
"eval_acc": 0.8503582395087002,
|
| 364 |
+
"eval_loss": 0.6494254469871521,
|
| 365 |
+
"eval_runtime": 16.658,
|
| 366 |
+
"eval_samples_per_second": 586.503,
|
| 367 |
+
"eval_steps_per_second": 18.37,
|
| 368 |
+
"step": 48000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.17,
|
| 372 |
+
"learning_rate": 3.3222148098732494e-06,
|
| 373 |
+
"loss": 0.1133,
|
| 374 |
+
"step": 50000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.17,
|
| 378 |
+
"eval_acc": 0.8458546571136131,
|
| 379 |
+
"eval_loss": 0.7663929462432861,
|
| 380 |
+
"eval_runtime": 16.6347,
|
| 381 |
+
"eval_samples_per_second": 587.326,
|
| 382 |
+
"eval_steps_per_second": 18.395,
|
| 383 |
+
"step": 50000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.34,
|
| 387 |
+
"learning_rate": 2.6551034022681787e-06,
|
| 388 |
+
"loss": 0.1138,
|
| 389 |
+
"step": 52000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.34,
|
| 393 |
+
"eval_acc": 0.8452405322415558,
|
| 394 |
+
"eval_loss": 0.7856881022453308,
|
| 395 |
+
"eval_runtime": 16.65,
|
| 396 |
+
"eval_samples_per_second": 586.787,
|
| 397 |
+
"eval_steps_per_second": 18.378,
|
| 398 |
+
"step": 52000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.5,
|
| 402 |
+
"learning_rate": 1.987991994663109e-06,
|
| 403 |
+
"loss": 0.1154,
|
| 404 |
+
"step": 54000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.5,
|
| 408 |
+
"eval_acc": 0.848618219037871,
|
| 409 |
+
"eval_loss": 0.7623312473297119,
|
| 410 |
+
"eval_runtime": 16.6261,
|
| 411 |
+
"eval_samples_per_second": 587.63,
|
| 412 |
+
"eval_steps_per_second": 18.405,
|
| 413 |
+
"step": 54000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.67,
|
| 417 |
+
"learning_rate": 1.3208805870580388e-06,
|
| 418 |
+
"loss": 0.1102,
|
| 419 |
+
"step": 56000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.67,
|
| 423 |
+
"eval_acc": 0.845957011258956,
|
| 424 |
+
"eval_loss": 0.7739743590354919,
|
| 425 |
+
"eval_runtime": 16.636,
|
| 426 |
+
"eval_samples_per_second": 587.28,
|
| 427 |
+
"eval_steps_per_second": 18.394,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.84,
|
| 432 |
+
"learning_rate": 6.537691794529687e-07,
|
| 433 |
+
"loss": 0.1143,
|
| 434 |
+
"step": 58000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.84,
|
| 438 |
+
"eval_acc": 0.8445240532241556,
|
| 439 |
+
"eval_loss": 0.7610092759132385,
|
| 440 |
+
"eval_runtime": 16.6493,
|
| 441 |
+
"eval_samples_per_second": 586.812,
|
| 442 |
+
"eval_steps_per_second": 18.379,
|
| 443 |
+
"step": 58000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 5.0,
|
| 447 |
+
"step": 59960,
|
| 448 |
+
"total_flos": 1.2620432416386816e+17,
|
| 449 |
+
"train_loss": 0.22490199587200385,
|
| 450 |
+
"train_runtime": 11405.3413,
|
| 451 |
+
"train_samples_per_second": 168.222,
|
| 452 |
+
"train_steps_per_second": 5.257
|
| 453 |
+
}
|
| 454 |
+
],
|
| 455 |
+
"max_steps": 59960,
|
| 456 |
+
"num_train_epochs": 5,
|
| 457 |
+
"total_flos": 1.2620432416386816e+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 3515
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:808fbd6858969de71318b9d7f71d5dee0e445c84ab244699d6966492ba49186b
|
| 3 |
size 3515
|