End of training
Browse files- README.md +2 -2
- all_results.json +15 -0
- eval_results.json +9 -0
- predict_results.txt +0 -0
- train_results.json +9 -0
- trainer_state.json +584 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- F1: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.2168
|
| 22 |
+
- F1: 0.9326
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9326343966094134,
|
| 4 |
+
"eval_loss": 0.21681125462055206,
|
| 5 |
+
"eval_runtime": 24.9667,
|
| 6 |
+
"eval_samples": 8966,
|
| 7 |
+
"eval_samples_per_second": 359.118,
|
| 8 |
+
"eval_steps_per_second": 5.648,
|
| 9 |
+
"total_flos": 5.5714266203904e+16,
|
| 10 |
+
"train_loss": 0.024559360085297206,
|
| 11 |
+
"train_runtime": 8074.1532,
|
| 12 |
+
"train_samples": 21175,
|
| 13 |
+
"train_samples_per_second": 104.903,
|
| 14 |
+
"train_steps_per_second": 1.64
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_f1": 0.9326343966094134,
|
| 4 |
+
"eval_loss": 0.21681125462055206,
|
| 5 |
+
"eval_runtime": 24.9667,
|
| 6 |
+
"eval_samples": 8966,
|
| 7 |
+
"eval_samples_per_second": 359.118,
|
| 8 |
+
"eval_steps_per_second": 5.648
|
| 9 |
+
}
|
predict_results.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 5.5714266203904e+16,
|
| 4 |
+
"train_loss": 0.024559360085297206,
|
| 5 |
+
"train_runtime": 8074.1532,
|
| 6 |
+
"train_samples": 21175,
|
| 7 |
+
"train_samples_per_second": 104.903,
|
| 8 |
+
"train_steps_per_second": 1.64
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,584 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9326343966094134,
|
| 3 |
+
"best_model_checkpoint": "/tmp/classification_hos_bert/checkpoint-662",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 13240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_f1": 0.9258309167967879,
|
| 14 |
+
"eval_loss": 0.21829327940940857,
|
| 15 |
+
"eval_runtime": 25.1238,
|
| 16 |
+
"eval_samples_per_second": 356.873,
|
| 17 |
+
"eval_steps_per_second": 5.612,
|
| 18 |
+
"step": 331
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.510574018126888,
|
| 22 |
+
"grad_norm": 2.2944066524505615,
|
| 23 |
+
"learning_rate": 2.8867069486404837e-05,
|
| 24 |
+
"loss": 0.239,
|
| 25 |
+
"step": 500
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"eval_f1": 0.9326343966094134,
|
| 30 |
+
"eval_loss": 0.21681125462055206,
|
| 31 |
+
"eval_runtime": 25.1665,
|
| 32 |
+
"eval_samples_per_second": 356.267,
|
| 33 |
+
"eval_steps_per_second": 5.603,
|
| 34 |
+
"step": 662
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 3.0,
|
| 38 |
+
"eval_f1": 0.9281730983716261,
|
| 39 |
+
"eval_loss": 0.25273850560188293,
|
| 40 |
+
"eval_runtime": 25.1148,
|
| 41 |
+
"eval_samples_per_second": 357.001,
|
| 42 |
+
"eval_steps_per_second": 5.614,
|
| 43 |
+
"step": 993
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0211480362537766,
|
| 47 |
+
"grad_norm": 0.6900779604911804,
|
| 48 |
+
"learning_rate": 2.7734138972809666e-05,
|
| 49 |
+
"loss": 0.1255,
|
| 50 |
+
"step": 1000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"eval_f1": 0.9288422931072943,
|
| 55 |
+
"eval_loss": 0.28959983587265015,
|
| 56 |
+
"eval_runtime": 25.1245,
|
| 57 |
+
"eval_samples_per_second": 356.863,
|
| 58 |
+
"eval_steps_per_second": 5.612,
|
| 59 |
+
"step": 1324
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.531722054380665,
|
| 63 |
+
"grad_norm": 1.7124501466751099,
|
| 64 |
+
"learning_rate": 2.66012084592145e-05,
|
| 65 |
+
"loss": 0.0662,
|
| 66 |
+
"step": 1500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 5.0,
|
| 70 |
+
"eval_f1": 0.9266116439884007,
|
| 71 |
+
"eval_loss": 0.33890488743782043,
|
| 72 |
+
"eval_runtime": 25.1434,
|
| 73 |
+
"eval_samples_per_second": 356.595,
|
| 74 |
+
"eval_steps_per_second": 5.608,
|
| 75 |
+
"step": 1655
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 6.0,
|
| 79 |
+
"eval_f1": 0.9293999553870176,
|
| 80 |
+
"eval_loss": 0.3792820870876312,
|
| 81 |
+
"eval_runtime": 25.1391,
|
| 82 |
+
"eval_samples_per_second": 356.656,
|
| 83 |
+
"eval_steps_per_second": 5.609,
|
| 84 |
+
"step": 1986
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.042296072507553,
|
| 88 |
+
"grad_norm": 0.8430729508399963,
|
| 89 |
+
"learning_rate": 2.5468277945619337e-05,
|
| 90 |
+
"loss": 0.0453,
|
| 91 |
+
"step": 2000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 7.0,
|
| 95 |
+
"eval_f1": 0.9251617220611198,
|
| 96 |
+
"eval_loss": 0.41103312373161316,
|
| 97 |
+
"eval_runtime": 25.1418,
|
| 98 |
+
"eval_samples_per_second": 356.617,
|
| 99 |
+
"eval_steps_per_second": 5.608,
|
| 100 |
+
"step": 2317
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 7.552870090634441,
|
| 104 |
+
"grad_norm": 0.8941565752029419,
|
| 105 |
+
"learning_rate": 2.433534743202417e-05,
|
| 106 |
+
"loss": 0.0257,
|
| 107 |
+
"step": 2500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 8.0,
|
| 111 |
+
"eval_f1": 0.9204773589114432,
|
| 112 |
+
"eval_loss": 0.4656200110912323,
|
| 113 |
+
"eval_runtime": 25.1662,
|
| 114 |
+
"eval_samples_per_second": 356.271,
|
| 115 |
+
"eval_steps_per_second": 5.603,
|
| 116 |
+
"step": 2648
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 9.0,
|
| 120 |
+
"eval_f1": 0.9262770466205665,
|
| 121 |
+
"eval_loss": 0.49531668424606323,
|
| 122 |
+
"eval_runtime": 25.1846,
|
| 123 |
+
"eval_samples_per_second": 356.011,
|
| 124 |
+
"eval_steps_per_second": 5.599,
|
| 125 |
+
"step": 2979
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 9.06344410876133,
|
| 129 |
+
"grad_norm": 0.01369735598564148,
|
| 130 |
+
"learning_rate": 2.3202416918429002e-05,
|
| 131 |
+
"loss": 0.0196,
|
| 132 |
+
"step": 3000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 10.0,
|
| 136 |
+
"eval_f1": 0.9265001115324559,
|
| 137 |
+
"eval_loss": 0.5412325263023376,
|
| 138 |
+
"eval_runtime": 25.1393,
|
| 139 |
+
"eval_samples_per_second": 356.653,
|
| 140 |
+
"eval_steps_per_second": 5.609,
|
| 141 |
+
"step": 3310
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.574018126888218,
|
| 145 |
+
"grad_norm": 1.39247727394104,
|
| 146 |
+
"learning_rate": 2.2069486404833838e-05,
|
| 147 |
+
"loss": 0.0125,
|
| 148 |
+
"step": 3500
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 11.0,
|
| 152 |
+
"eval_f1": 0.9244925273254517,
|
| 153 |
+
"eval_loss": 0.5528218150138855,
|
| 154 |
+
"eval_runtime": 25.112,
|
| 155 |
+
"eval_samples_per_second": 357.04,
|
| 156 |
+
"eval_steps_per_second": 5.615,
|
| 157 |
+
"step": 3641
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 12.0,
|
| 161 |
+
"eval_f1": 0.9261655141646219,
|
| 162 |
+
"eval_loss": 0.5526648759841919,
|
| 163 |
+
"eval_runtime": 25.1453,
|
| 164 |
+
"eval_samples_per_second": 356.568,
|
| 165 |
+
"eval_steps_per_second": 5.607,
|
| 166 |
+
"step": 3972
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 12.084592145015106,
|
| 170 |
+
"grad_norm": 0.3373314440250397,
|
| 171 |
+
"learning_rate": 2.093655589123867e-05,
|
| 172 |
+
"loss": 0.0141,
|
| 173 |
+
"step": 4000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 13.0,
|
| 177 |
+
"eval_f1": 0.9276154360919028,
|
| 178 |
+
"eval_loss": 0.5682665705680847,
|
| 179 |
+
"eval_runtime": 25.1511,
|
| 180 |
+
"eval_samples_per_second": 356.486,
|
| 181 |
+
"eval_steps_per_second": 5.606,
|
| 182 |
+
"step": 4303
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.595166163141993,
|
| 186 |
+
"grad_norm": 0.07623090595006943,
|
| 187 |
+
"learning_rate": 1.9803625377643507e-05,
|
| 188 |
+
"loss": 0.0097,
|
| 189 |
+
"step": 4500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 14.0,
|
| 193 |
+
"eval_f1": 0.9239348650457283,
|
| 194 |
+
"eval_loss": 0.5835373997688293,
|
| 195 |
+
"eval_runtime": 25.143,
|
| 196 |
+
"eval_samples_per_second": 356.6,
|
| 197 |
+
"eval_steps_per_second": 5.608,
|
| 198 |
+
"step": 4634
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 15.0,
|
| 202 |
+
"eval_f1": 0.9279500334597368,
|
| 203 |
+
"eval_loss": 0.5905042886734009,
|
| 204 |
+
"eval_runtime": 25.1447,
|
| 205 |
+
"eval_samples_per_second": 356.576,
|
| 206 |
+
"eval_steps_per_second": 5.608,
|
| 207 |
+
"step": 4965
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.105740181268882,
|
| 211 |
+
"grad_norm": 0.01646752655506134,
|
| 212 |
+
"learning_rate": 1.867069486404834e-05,
|
| 213 |
+
"loss": 0.0107,
|
| 214 |
+
"step": 5000
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.0,
|
| 218 |
+
"eval_f1": 0.9298460852107964,
|
| 219 |
+
"eval_loss": 0.5799357295036316,
|
| 220 |
+
"eval_runtime": 25.142,
|
| 221 |
+
"eval_samples_per_second": 356.615,
|
| 222 |
+
"eval_steps_per_second": 5.608,
|
| 223 |
+
"step": 5296
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 16.61631419939577,
|
| 227 |
+
"grad_norm": 0.061389509588479996,
|
| 228 |
+
"learning_rate": 1.753776435045317e-05,
|
| 229 |
+
"loss": 0.009,
|
| 230 |
+
"step": 5500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 17.0,
|
| 234 |
+
"eval_f1": 0.9266116439884007,
|
| 235 |
+
"eval_loss": 0.6126909255981445,
|
| 236 |
+
"eval_runtime": 25.1653,
|
| 237 |
+
"eval_samples_per_second": 356.285,
|
| 238 |
+
"eval_steps_per_second": 5.603,
|
| 239 |
+
"step": 5627
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 18.0,
|
| 243 |
+
"eval_f1": 0.9283961632835155,
|
| 244 |
+
"eval_loss": 0.591077446937561,
|
| 245 |
+
"eval_runtime": 25.1503,
|
| 246 |
+
"eval_samples_per_second": 356.497,
|
| 247 |
+
"eval_steps_per_second": 5.606,
|
| 248 |
+
"step": 5958
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 18.12688821752266,
|
| 252 |
+
"grad_norm": 0.0015490599907934666,
|
| 253 |
+
"learning_rate": 1.6404833836858007e-05,
|
| 254 |
+
"loss": 0.0084,
|
| 255 |
+
"step": 6000
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 19.0,
|
| 259 |
+
"eval_f1": 0.930292215034575,
|
| 260 |
+
"eval_loss": 0.5900245308876038,
|
| 261 |
+
"eval_runtime": 25.1515,
|
| 262 |
+
"eval_samples_per_second": 356.479,
|
| 263 |
+
"eval_steps_per_second": 5.606,
|
| 264 |
+
"step": 6289
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 19.637462235649547,
|
| 268 |
+
"grad_norm": 0.34078794717788696,
|
| 269 |
+
"learning_rate": 1.527190332326284e-05,
|
| 270 |
+
"loss": 0.008,
|
| 271 |
+
"step": 6500
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 20.0,
|
| 275 |
+
"eval_f1": 0.9282846308275708,
|
| 276 |
+
"eval_loss": 0.5922934412956238,
|
| 277 |
+
"eval_runtime": 25.1544,
|
| 278 |
+
"eval_samples_per_second": 356.438,
|
| 279 |
+
"eval_steps_per_second": 5.605,
|
| 280 |
+
"step": 6620
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 21.0,
|
| 284 |
+
"eval_f1": 0.9305152799464644,
|
| 285 |
+
"eval_loss": 0.6186188459396362,
|
| 286 |
+
"eval_runtime": 25.1563,
|
| 287 |
+
"eval_samples_per_second": 356.412,
|
| 288 |
+
"eval_steps_per_second": 5.605,
|
| 289 |
+
"step": 6951
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 21.148036253776436,
|
| 293 |
+
"grad_norm": 0.16627806425094604,
|
| 294 |
+
"learning_rate": 1.4138972809667674e-05,
|
| 295 |
+
"loss": 0.0068,
|
| 296 |
+
"step": 7000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 22.0,
|
| 300 |
+
"eval_f1": 0.9291768904751282,
|
| 301 |
+
"eval_loss": 0.6076038479804993,
|
| 302 |
+
"eval_runtime": 25.1577,
|
| 303 |
+
"eval_samples_per_second": 356.392,
|
| 304 |
+
"eval_steps_per_second": 5.605,
|
| 305 |
+
"step": 7282
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 22.658610271903324,
|
| 309 |
+
"grad_norm": 0.02961309626698494,
|
| 310 |
+
"learning_rate": 1.3006042296072508e-05,
|
| 311 |
+
"loss": 0.0064,
|
| 312 |
+
"step": 7500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 23.0,
|
| 316 |
+
"eval_f1": 0.930292215034575,
|
| 317 |
+
"eval_loss": 0.578154444694519,
|
| 318 |
+
"eval_runtime": 25.1751,
|
| 319 |
+
"eval_samples_per_second": 356.145,
|
| 320 |
+
"eval_steps_per_second": 5.601,
|
| 321 |
+
"step": 7613
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 24.0,
|
| 325 |
+
"eval_f1": 0.9319652018737452,
|
| 326 |
+
"eval_loss": 0.607693076133728,
|
| 327 |
+
"eval_runtime": 25.1532,
|
| 328 |
+
"eval_samples_per_second": 356.455,
|
| 329 |
+
"eval_steps_per_second": 5.606,
|
| 330 |
+
"step": 7944
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.169184290030213,
|
| 334 |
+
"grad_norm": 0.012147185392677784,
|
| 335 |
+
"learning_rate": 1.187311178247734e-05,
|
| 336 |
+
"loss": 0.0048,
|
| 337 |
+
"step": 8000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 25.0,
|
| 341 |
+
"eval_f1": 0.9281730983716261,
|
| 342 |
+
"eval_loss": 0.6445909738540649,
|
| 343 |
+
"eval_runtime": 25.1606,
|
| 344 |
+
"eval_samples_per_second": 356.351,
|
| 345 |
+
"eval_steps_per_second": 5.604,
|
| 346 |
+
"step": 8275
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.6797583081571,
|
| 350 |
+
"grad_norm": 0.0304458886384964,
|
| 351 |
+
"learning_rate": 1.0740181268882177e-05,
|
| 352 |
+
"loss": 0.0046,
|
| 353 |
+
"step": 8500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 26.0,
|
| 357 |
+
"eval_f1": 0.9315190720499665,
|
| 358 |
+
"eval_loss": 0.6416810154914856,
|
| 359 |
+
"eval_runtime": 25.1644,
|
| 360 |
+
"eval_samples_per_second": 356.298,
|
| 361 |
+
"eval_steps_per_second": 5.603,
|
| 362 |
+
"step": 8606
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 27.0,
|
| 366 |
+
"eval_f1": 0.9282846308275708,
|
| 367 |
+
"eval_loss": 0.6655632257461548,
|
| 368 |
+
"eval_runtime": 25.1733,
|
| 369 |
+
"eval_samples_per_second": 356.171,
|
| 370 |
+
"eval_steps_per_second": 5.601,
|
| 371 |
+
"step": 8937
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 27.190332326283986,
|
| 375 |
+
"grad_norm": 0.0014442217070609331,
|
| 376 |
+
"learning_rate": 9.60725075528701e-06,
|
| 377 |
+
"loss": 0.0053,
|
| 378 |
+
"step": 9000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 28.0,
|
| 382 |
+
"eval_f1": 0.9288422931072943,
|
| 383 |
+
"eval_loss": 0.6541187763214111,
|
| 384 |
+
"eval_runtime": 25.1766,
|
| 385 |
+
"eval_samples_per_second": 356.124,
|
| 386 |
+
"eval_steps_per_second": 5.6,
|
| 387 |
+
"step": 9268
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.700906344410875,
|
| 391 |
+
"grad_norm": 0.0008725296938791871,
|
| 392 |
+
"learning_rate": 8.474320241691843e-06,
|
| 393 |
+
"loss": 0.0043,
|
| 394 |
+
"step": 9500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 29.0,
|
| 398 |
+
"eval_f1": 0.9277269685478474,
|
| 399 |
+
"eval_loss": 0.6702625155448914,
|
| 400 |
+
"eval_runtime": 25.1263,
|
| 401 |
+
"eval_samples_per_second": 356.837,
|
| 402 |
+
"eval_steps_per_second": 5.612,
|
| 403 |
+
"step": 9599
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 30.0,
|
| 407 |
+
"eval_f1": 0.9251617220611198,
|
| 408 |
+
"eval_loss": 0.6871447563171387,
|
| 409 |
+
"eval_runtime": 25.1371,
|
| 410 |
+
"eval_samples_per_second": 356.684,
|
| 411 |
+
"eval_steps_per_second": 5.609,
|
| 412 |
+
"step": 9930
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 30.211480362537763,
|
| 416 |
+
"grad_norm": 0.0005139079876244068,
|
| 417 |
+
"learning_rate": 7.341389728096677e-06,
|
| 418 |
+
"loss": 0.0041,
|
| 419 |
+
"step": 10000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 31.0,
|
| 423 |
+
"eval_f1": 0.9286192281954049,
|
| 424 |
+
"eval_loss": 0.6735148429870605,
|
| 425 |
+
"eval_runtime": 25.1585,
|
| 426 |
+
"eval_samples_per_second": 356.38,
|
| 427 |
+
"eval_steps_per_second": 5.604,
|
| 428 |
+
"step": 10261
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 31.72205438066465,
|
| 432 |
+
"grad_norm": 0.002090197755023837,
|
| 433 |
+
"learning_rate": 6.208459214501511e-06,
|
| 434 |
+
"loss": 0.0034,
|
| 435 |
+
"step": 10500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 32.0,
|
| 439 |
+
"eval_f1": 0.9306268124024091,
|
| 440 |
+
"eval_loss": 0.6650559306144714,
|
| 441 |
+
"eval_runtime": 25.1466,
|
| 442 |
+
"eval_samples_per_second": 356.549,
|
| 443 |
+
"eval_steps_per_second": 5.607,
|
| 444 |
+
"step": 10592
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 33.0,
|
| 448 |
+
"eval_f1": 0.9305152799464644,
|
| 449 |
+
"eval_loss": 0.6799349188804626,
|
| 450 |
+
"eval_runtime": 25.1547,
|
| 451 |
+
"eval_samples_per_second": 356.435,
|
| 452 |
+
"eval_steps_per_second": 5.605,
|
| 453 |
+
"step": 10923
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 33.23262839879154,
|
| 457 |
+
"grad_norm": 0.016955886036157608,
|
| 458 |
+
"learning_rate": 5.075528700906345e-06,
|
| 459 |
+
"loss": 0.0032,
|
| 460 |
+
"step": 11000
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 34.0,
|
| 464 |
+
"eval_f1": 0.9297345527548516,
|
| 465 |
+
"eval_loss": 0.6752559542655945,
|
| 466 |
+
"eval_runtime": 25.1385,
|
| 467 |
+
"eval_samples_per_second": 356.664,
|
| 468 |
+
"eval_steps_per_second": 5.609,
|
| 469 |
+
"step": 11254
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 34.74320241691843,
|
| 473 |
+
"grad_norm": 0.0011760705383494496,
|
| 474 |
+
"learning_rate": 3.942598187311178e-06,
|
| 475 |
+
"loss": 0.0031,
|
| 476 |
+
"step": 11500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 35.0,
|
| 480 |
+
"eval_f1": 0.9309614097702431,
|
| 481 |
+
"eval_loss": 0.6854746341705322,
|
| 482 |
+
"eval_runtime": 25.1626,
|
| 483 |
+
"eval_samples_per_second": 356.323,
|
| 484 |
+
"eval_steps_per_second": 5.604,
|
| 485 |
+
"step": 11585
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 36.0,
|
| 489 |
+
"eval_f1": 0.9306268124024091,
|
| 490 |
+
"eval_loss": 0.6885010600090027,
|
| 491 |
+
"eval_runtime": 25.1518,
|
| 492 |
+
"eval_samples_per_second": 356.475,
|
| 493 |
+
"eval_steps_per_second": 5.606,
|
| 494 |
+
"step": 11916
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.25377643504532,
|
| 498 |
+
"grad_norm": 0.000429723208071664,
|
| 499 |
+
"learning_rate": 2.809667673716012e-06,
|
| 500 |
+
"loss": 0.003,
|
| 501 |
+
"step": 12000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 37.0,
|
| 505 |
+
"eval_f1": 0.9292884229310729,
|
| 506 |
+
"eval_loss": 0.6960038542747498,
|
| 507 |
+
"eval_runtime": 25.1589,
|
| 508 |
+
"eval_samples_per_second": 356.375,
|
| 509 |
+
"eval_steps_per_second": 5.604,
|
| 510 |
+
"step": 12247
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 37.764350453172206,
|
| 514 |
+
"grad_norm": 0.0003884187317453325,
|
| 515 |
+
"learning_rate": 1.6767371601208459e-06,
|
| 516 |
+
"loss": 0.0026,
|
| 517 |
+
"step": 12500
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 38.0,
|
| 521 |
+
"eval_f1": 0.9291768904751282,
|
| 522 |
+
"eval_loss": 0.6950347423553467,
|
| 523 |
+
"eval_runtime": 25.1564,
|
| 524 |
+
"eval_samples_per_second": 356.41,
|
| 525 |
+
"eval_steps_per_second": 5.605,
|
| 526 |
+
"step": 12578
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 39.0,
|
| 530 |
+
"eval_f1": 0.9297345527548516,
|
| 531 |
+
"eval_loss": 0.6964432597160339,
|
| 532 |
+
"eval_runtime": 25.1685,
|
| 533 |
+
"eval_samples_per_second": 356.238,
|
| 534 |
+
"eval_steps_per_second": 5.602,
|
| 535 |
+
"step": 12909
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 39.274924471299094,
|
| 539 |
+
"grad_norm": 0.03609294071793556,
|
| 540 |
+
"learning_rate": 5.438066465256798e-07,
|
| 541 |
+
"loss": 0.0033,
|
| 542 |
+
"step": 13000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 40.0,
|
| 546 |
+
"eval_f1": 0.928953825563239,
|
| 547 |
+
"eval_loss": 0.6954053640365601,
|
| 548 |
+
"eval_runtime": 25.1629,
|
| 549 |
+
"eval_samples_per_second": 356.318,
|
| 550 |
+
"eval_steps_per_second": 5.603,
|
| 551 |
+
"step": 13240
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 40.0,
|
| 555 |
+
"step": 13240,
|
| 556 |
+
"total_flos": 5.5714266203904e+16,
|
| 557 |
+
"train_loss": 0.024559360085297206,
|
| 558 |
+
"train_runtime": 8074.1532,
|
| 559 |
+
"train_samples_per_second": 104.903,
|
| 560 |
+
"train_steps_per_second": 1.64
|
| 561 |
+
}
|
| 562 |
+
],
|
| 563 |
+
"logging_steps": 500,
|
| 564 |
+
"max_steps": 13240,
|
| 565 |
+
"num_input_tokens_seen": 0,
|
| 566 |
+
"num_train_epochs": 40,
|
| 567 |
+
"save_steps": 500,
|
| 568 |
+
"stateful_callbacks": {
|
| 569 |
+
"TrainerControl": {
|
| 570 |
+
"args": {
|
| 571 |
+
"should_epoch_stop": false,
|
| 572 |
+
"should_evaluate": false,
|
| 573 |
+
"should_log": false,
|
| 574 |
+
"should_save": true,
|
| 575 |
+
"should_training_stop": true
|
| 576 |
+
},
|
| 577 |
+
"attributes": {}
|
| 578 |
+
}
|
| 579 |
+
},
|
| 580 |
+
"total_flos": 5.5714266203904e+16,
|
| 581 |
+
"train_batch_size": 64,
|
| 582 |
+
"trial_name": null,
|
| 583 |
+
"trial_params": null
|
| 584 |
+
}
|