Training in progress, epoch 1
Browse files- all_results.json +26 -16
- eval_results.json +11 -11
- generated_predictions.txt +0 -0
- predict_results.json +12 -0
- pytorch_model.bin +1 -1
- runs/Mar20_21-00-25_benihi/events.out.tfevents.1710939143.benihi +3 -0
- runs/Mar20_22-04-51_benihi/events.out.tfevents.1710939902.benihi +3 -0
- train_results.json +6 -6
- trainer_state.json +260 -126
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,18 +1,28 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_gen_len":
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_rouge1":
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL":
|
| 8 |
-
"eval_rougeLsum":
|
| 9 |
-
"eval_runtime":
|
| 10 |
-
"eval_samples":
|
| 11 |
-
"eval_samples_per_second":
|
| 12 |
-
"eval_steps_per_second":
|
| 13 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
| 16 |
-
"
|
| 17 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_gen_len": 52.15822784810127,
|
| 4 |
+
"eval_loss": 2.7362189292907715,
|
| 5 |
+
"eval_rouge1": 11.716,
|
| 6 |
+
"eval_rouge2": 1.7738,
|
| 7 |
+
"eval_rougeL": 7.7212,
|
| 8 |
+
"eval_rougeLsum": 10.0982,
|
| 9 |
+
"eval_runtime": 334.5899,
|
| 10 |
+
"eval_samples": 632,
|
| 11 |
+
"eval_samples_per_second": 1.889,
|
| 12 |
+
"eval_steps_per_second": 1.889,
|
| 13 |
+
"predict_gen_len": 56.19065420560748,
|
| 14 |
+
"predict_loss": 2.9006059169769287,
|
| 15 |
+
"predict_rouge1": 15.964,
|
| 16 |
+
"predict_rouge2": 2.625,
|
| 17 |
+
"predict_rougeL": 10.5843,
|
| 18 |
+
"predict_rougeLsum": 13.4933,
|
| 19 |
+
"predict_runtime": 286.0012,
|
| 20 |
+
"predict_samples": 535,
|
| 21 |
+
"predict_samples_per_second": 1.871,
|
| 22 |
+
"predict_steps_per_second": 1.871,
|
| 23 |
+
"train_loss": 0.29059598513036994,
|
| 24 |
+
"train_runtime": 2646.0496,
|
| 25 |
+
"train_samples": 4332,
|
| 26 |
+
"train_samples_per_second": 8.186,
|
| 27 |
+
"train_steps_per_second": 8.186
|
| 28 |
}
|
eval_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_gen_len":
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_rouge1":
|
| 6 |
-
"eval_rouge2":
|
| 7 |
-
"eval_rougeL":
|
| 8 |
-
"eval_rougeLsum":
|
| 9 |
-
"eval_runtime":
|
| 10 |
-
"eval_samples":
|
| 11 |
-
"eval_samples_per_second":
|
| 12 |
-
"eval_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_gen_len": 52.15822784810127,
|
| 4 |
+
"eval_loss": 2.7362189292907715,
|
| 5 |
+
"eval_rouge1": 11.716,
|
| 6 |
+
"eval_rouge2": 1.7738,
|
| 7 |
+
"eval_rougeL": 7.7212,
|
| 8 |
+
"eval_rougeLsum": 10.0982,
|
| 9 |
+
"eval_runtime": 334.5899,
|
| 10 |
+
"eval_samples": 632,
|
| 11 |
+
"eval_samples_per_second": 1.889,
|
| 12 |
+
"eval_steps_per_second": 1.889
|
| 13 |
}
|
generated_predictions.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
predict_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"predict_gen_len": 56.19065420560748,
|
| 3 |
+
"predict_loss": 2.9006059169769287,
|
| 4 |
+
"predict_rouge1": 15.964,
|
| 5 |
+
"predict_rouge2": 2.625,
|
| 6 |
+
"predict_rougeL": 10.5843,
|
| 7 |
+
"predict_rougeLsum": 13.4933,
|
| 8 |
+
"predict_runtime": 286.0012,
|
| 9 |
+
"predict_samples": 535,
|
| 10 |
+
"predict_samples_per_second": 1.871,
|
| 11 |
+
"predict_steps_per_second": 1.871
|
| 12 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 501807853
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c425cf9e4c3e2fd9195d6383648d5a87621a02b533e94839dda64a780b60cd14
|
| 3 |
size 501807853
|
runs/Mar20_21-00-25_benihi/events.out.tfevents.1710939143.benihi
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:684a2b5f534fcaa6b48641a4561685287bb4101d795d7ce8c12a5215c785bb52
|
| 3 |
+
size 575
|
runs/Mar20_22-04-51_benihi/events.out.tfevents.1710939902.benihi
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fb64818bd89d12ad01ab9195f7dda71db9cc3eb353c0275535837a9699a1b4a0
|
| 3 |
+
size 7433
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 0.29059598513036994,
|
| 4 |
+
"train_runtime": 2646.0496,
|
| 5 |
+
"train_samples": 4332,
|
| 6 |
+
"train_samples_per_second": 8.186,
|
| 7 |
+
"train_steps_per_second": 8.186
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,147 +1,281 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "saved/tobyoki-pairwise/bart-base-japanese/BaseModel/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"
|
| 6 |
-
"global_step": 90,
|
| 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_gen_len":
|
| 14 |
-
"eval_loss": 2.
|
| 15 |
-
"eval_rouge1":
|
| 16 |
-
"eval_rouge2": 1.
|
| 17 |
-
"eval_rougeL":
|
| 18 |
-
"eval_rougeLsum":
|
| 19 |
-
"eval_runtime":
|
| 20 |
-
"eval_samples_per_second":
|
| 21 |
-
"eval_steps_per_second":
|
| 22 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 23 |
},
|
| 24 |
{
|
| 25 |
"epoch": 2.0,
|
| 26 |
-
"eval_gen_len":
|
| 27 |
-
"eval_loss":
|
| 28 |
-
"eval_rouge1":
|
| 29 |
-
"eval_rouge2": 1.
|
| 30 |
-
"eval_rougeL":
|
| 31 |
-
"eval_rougeLsum":
|
| 32 |
-
"eval_runtime":
|
| 33 |
-
"eval_samples_per_second":
|
| 34 |
-
"eval_steps_per_second":
|
| 35 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 36 |
},
|
| 37 |
{
|
| 38 |
"epoch": 3.0,
|
| 39 |
-
"eval_gen_len":
|
| 40 |
-
"eval_loss":
|
| 41 |
-
"eval_rouge1":
|
| 42 |
-
"eval_rouge2": 2.
|
| 43 |
-
"eval_rougeL":
|
| 44 |
-
"eval_rougeLsum":
|
| 45 |
-
"eval_runtime":
|
| 46 |
-
"eval_samples_per_second":
|
| 47 |
-
"eval_steps_per_second":
|
| 48 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 4.0,
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"
|
| 58 |
-
"eval_runtime": 24.7517,
|
| 59 |
-
"eval_samples_per_second": 0.404,
|
| 60 |
-
"eval_steps_per_second": 0.404,
|
| 61 |
-
"step": 40
|
| 62 |
-
},
|
| 63 |
-
{
|
| 64 |
-
"epoch": 5.0,
|
| 65 |
-
"eval_gen_len": 300.0,
|
| 66 |
-
"eval_loss": 2.332805871963501,
|
| 67 |
-
"eval_rouge1": 20.5428,
|
| 68 |
-
"eval_rouge2": 3.6354,
|
| 69 |
-
"eval_rougeL": 11.3166,
|
| 70 |
-
"eval_rougeLsum": 16.0252,
|
| 71 |
-
"eval_runtime": 25.4485,
|
| 72 |
-
"eval_samples_per_second": 0.393,
|
| 73 |
-
"eval_steps_per_second": 0.393,
|
| 74 |
-
"step": 50
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 6.0,
|
| 78 |
-
"eval_gen_len": 294.2,
|
| 79 |
-
"eval_loss": 2.3246188163757324,
|
| 80 |
-
"eval_rouge1": 24.0821,
|
| 81 |
-
"eval_rouge2": 4.2604,
|
| 82 |
-
"eval_rougeL": 12.4489,
|
| 83 |
-
"eval_rougeLsum": 19.152,
|
| 84 |
-
"eval_runtime": 25.6727,
|
| 85 |
-
"eval_samples_per_second": 0.39,
|
| 86 |
-
"eval_steps_per_second": 0.39,
|
| 87 |
-
"step": 60
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 7.0,
|
| 91 |
-
"eval_gen_len": 300.0,
|
| 92 |
-
"eval_loss": 2.3076653480529785,
|
| 93 |
-
"eval_rouge1": 20.4387,
|
| 94 |
-
"eval_rouge2": 3.8043,
|
| 95 |
-
"eval_rougeL": 11.4571,
|
| 96 |
-
"eval_rougeLsum": 16.7923,
|
| 97 |
-
"eval_runtime": 24.5822,
|
| 98 |
-
"eval_samples_per_second": 0.407,
|
| 99 |
-
"eval_steps_per_second": 0.407,
|
| 100 |
-
"step": 70
|
| 101 |
-
},
|
| 102 |
-
{
|
| 103 |
-
"epoch": 8.0,
|
| 104 |
-
"eval_gen_len": 300.0,
|
| 105 |
-
"eval_loss": 2.305992603302002,
|
| 106 |
-
"eval_rouge1": 22.0434,
|
| 107 |
-
"eval_rouge2": 4.3648,
|
| 108 |
-
"eval_rougeL": 12.386,
|
| 109 |
-
"eval_rougeLsum": 18.3828,
|
| 110 |
-
"eval_runtime": 25.0442,
|
| 111 |
-
"eval_samples_per_second": 0.399,
|
| 112 |
-
"eval_steps_per_second": 0.399,
|
| 113 |
-
"step": 80
|
| 114 |
-
},
|
| 115 |
-
{
|
| 116 |
-
"epoch": 9.0,
|
| 117 |
-
"eval_gen_len": 289.1,
|
| 118 |
-
"eval_loss": 2.302980661392212,
|
| 119 |
-
"eval_rouge1": 20.3362,
|
| 120 |
-
"eval_rouge2": 3.6623,
|
| 121 |
-
"eval_rougeL": 10.9234,
|
| 122 |
-
"eval_rougeLsum": 16.1471,
|
| 123 |
-
"eval_runtime": 25.3281,
|
| 124 |
-
"eval_samples_per_second": 0.395,
|
| 125 |
-
"eval_steps_per_second": 0.395,
|
| 126 |
-
"step": 90
|
| 127 |
-
},
|
| 128 |
-
{
|
| 129 |
-
"epoch": 9.0,
|
| 130 |
-
"step": 90,
|
| 131 |
-
"total_flos": 18151534964736.0,
|
| 132 |
-
"train_loss": 2.465799797905816,
|
| 133 |
-
"train_runtime": 251.0487,
|
| 134 |
-
"train_samples_per_second": 0.398,
|
| 135 |
-
"train_steps_per_second": 0.398
|
| 136 |
}
|
| 137 |
],
|
| 138 |
-
"
|
| 139 |
-
"
|
| 140 |
-
"
|
| 141 |
-
"num_train_epochs": 10,
|
| 142 |
-
"save_steps": 500,
|
| 143 |
-
"total_flos": 18151534964736.0,
|
| 144 |
-
"train_batch_size": 1,
|
| 145 |
"trial_name": null,
|
| 146 |
"trial_params": null
|
| 147 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 11.716,
|
| 3 |
+
"best_model_checkpoint": "saved/tobyoki-pairwise/bart-base-japanese/BaseModel/checkpoint-4332",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 17328,
|
|
|
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 4.8845798707294554e-05,
|
| 13 |
+
"loss": 2.1617,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.23,
|
| 18 |
+
"learning_rate": 4.7691597414589107e-05,
|
| 19 |
+
"loss": 1.346,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.35,
|
| 24 |
+
"learning_rate": 4.653739612188366e-05,
|
| 25 |
+
"loss": 0.96,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.46,
|
| 30 |
+
"learning_rate": 4.538319482917821e-05,
|
| 31 |
+
"loss": 0.7184,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.58,
|
| 36 |
+
"learning_rate": 4.422899353647276e-05,
|
| 37 |
+
"loss": 0.5936,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.69,
|
| 42 |
+
"learning_rate": 4.3074792243767315e-05,
|
| 43 |
+
"loss": 0.4745,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.81,
|
| 48 |
+
"learning_rate": 4.192059095106187e-05,
|
| 49 |
+
"loss": 0.389,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.92,
|
| 54 |
+
"learning_rate": 4.076638965835642e-05,
|
| 55 |
+
"loss": 0.298,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
{
|
| 59 |
"epoch": 1.0,
|
| 60 |
+
"eval_gen_len": 52.15822784810127,
|
| 61 |
+
"eval_loss": 2.7362189292907715,
|
| 62 |
+
"eval_rouge1": 11.716,
|
| 63 |
+
"eval_rouge2": 1.7738,
|
| 64 |
+
"eval_rougeL": 7.7212,
|
| 65 |
+
"eval_rougeLsum": 10.0982,
|
| 66 |
+
"eval_runtime": 282.218,
|
| 67 |
+
"eval_samples_per_second": 2.239,
|
| 68 |
+
"eval_steps_per_second": 2.239,
|
| 69 |
+
"step": 4332
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.04,
|
| 73 |
+
"learning_rate": 3.961218836565097e-05,
|
| 74 |
+
"loss": 0.2931,
|
| 75 |
+
"step": 4500
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.15,
|
| 79 |
+
"learning_rate": 3.845798707294552e-05,
|
| 80 |
+
"loss": 0.27,
|
| 81 |
+
"step": 5000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.27,
|
| 85 |
+
"learning_rate": 3.7303785780240075e-05,
|
| 86 |
+
"loss": 0.2255,
|
| 87 |
+
"step": 5500
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.39,
|
| 91 |
+
"learning_rate": 3.614958448753463e-05,
|
| 92 |
+
"loss": 0.1818,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.5,
|
| 97 |
+
"learning_rate": 3.499538319482918e-05,
|
| 98 |
+
"loss": 0.1736,
|
| 99 |
+
"step": 6500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.62,
|
| 103 |
+
"learning_rate": 3.384118190212373e-05,
|
| 104 |
+
"loss": 0.1894,
|
| 105 |
+
"step": 7000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.73,
|
| 109 |
+
"learning_rate": 3.2686980609418284e-05,
|
| 110 |
+
"loss": 0.1642,
|
| 111 |
+
"step": 7500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.85,
|
| 115 |
+
"learning_rate": 3.1532779316712836e-05,
|
| 116 |
+
"loss": 0.1262,
|
| 117 |
+
"step": 8000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.96,
|
| 121 |
+
"learning_rate": 3.0378578024007388e-05,
|
| 122 |
+
"loss": 0.1596,
|
| 123 |
+
"step": 8500
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.0,
|
| 127 |
+
"eval_gen_len": 42.75791139240506,
|
| 128 |
+
"eval_loss": 3.1079843044281006,
|
| 129 |
+
"eval_rouge1": 11.6339,
|
| 130 |
+
"eval_rouge2": 1.9589,
|
| 131 |
+
"eval_rougeL": 8.2226,
|
| 132 |
+
"eval_rougeLsum": 9.5881,
|
| 133 |
+
"eval_runtime": 452.5754,
|
| 134 |
+
"eval_samples_per_second": 1.396,
|
| 135 |
+
"eval_steps_per_second": 1.396,
|
| 136 |
+
"step": 8664
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.08,
|
| 140 |
+
"learning_rate": 2.922437673130194e-05,
|
| 141 |
+
"loss": 0.1249,
|
| 142 |
+
"step": 9000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.19,
|
| 146 |
+
"learning_rate": 2.8070175438596492e-05,
|
| 147 |
+
"loss": 0.1056,
|
| 148 |
+
"step": 9500
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 2.31,
|
| 152 |
+
"learning_rate": 2.6915974145891044e-05,
|
| 153 |
+
"loss": 0.0981,
|
| 154 |
+
"step": 10000
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 2.42,
|
| 158 |
+
"learning_rate": 2.5761772853185596e-05,
|
| 159 |
+
"loss": 0.1025,
|
| 160 |
+
"step": 10500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.54,
|
| 164 |
+
"learning_rate": 2.460757156048015e-05,
|
| 165 |
+
"loss": 0.0777,
|
| 166 |
+
"step": 11000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.65,
|
| 170 |
+
"learning_rate": 2.3453370267774704e-05,
|
| 171 |
+
"loss": 0.0849,
|
| 172 |
+
"step": 11500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.77,
|
| 176 |
+
"learning_rate": 2.2299168975069256e-05,
|
| 177 |
+
"loss": 0.095,
|
| 178 |
+
"step": 12000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 2.89,
|
| 182 |
+
"learning_rate": 2.1144967682363804e-05,
|
| 183 |
+
"loss": 0.0805,
|
| 184 |
+
"step": 12500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 3.0,
|
| 188 |
+
"eval_gen_len": 43.24683544303797,
|
| 189 |
+
"eval_loss": 3.2877912521362305,
|
| 190 |
+
"eval_rouge1": 11.3717,
|
| 191 |
+
"eval_rouge2": 2.0083,
|
| 192 |
+
"eval_rougeL": 7.8703,
|
| 193 |
+
"eval_rougeLsum": 9.6383,
|
| 194 |
+
"eval_runtime": 236.6502,
|
| 195 |
+
"eval_samples_per_second": 2.671,
|
| 196 |
+
"eval_steps_per_second": 2.671,
|
| 197 |
+
"step": 12996
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.0,
|
| 201 |
+
"learning_rate": 1.9990766389658356e-05,
|
| 202 |
+
"loss": 0.081,
|
| 203 |
+
"step": 13000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.12,
|
| 207 |
+
"learning_rate": 1.883656509695291e-05,
|
| 208 |
+
"loss": 0.0562,
|
| 209 |
+
"step": 13500
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.23,
|
| 213 |
+
"learning_rate": 1.768236380424746e-05,
|
| 214 |
+
"loss": 0.0659,
|
| 215 |
+
"step": 14000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.35,
|
| 219 |
+
"learning_rate": 1.6528162511542013e-05,
|
| 220 |
+
"loss": 0.0641,
|
| 221 |
+
"step": 14500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 3.46,
|
| 225 |
+
"learning_rate": 1.5373961218836565e-05,
|
| 226 |
+
"loss": 0.0621,
|
| 227 |
+
"step": 15000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 3.58,
|
| 231 |
+
"learning_rate": 1.4219759926131118e-05,
|
| 232 |
+
"loss": 0.0495,
|
| 233 |
+
"step": 15500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 3.69,
|
| 237 |
+
"learning_rate": 1.306555863342567e-05,
|
| 238 |
+
"loss": 0.052,
|
| 239 |
+
"step": 16000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.81,
|
| 243 |
+
"learning_rate": 1.1911357340720223e-05,
|
| 244 |
+
"loss": 0.0624,
|
| 245 |
+
"step": 16500
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 3.92,
|
| 249 |
+
"learning_rate": 1.0757156048014775e-05,
|
| 250 |
+
"loss": 0.0576,
|
| 251 |
+
"step": 17000
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.0,
|
| 255 |
+
"eval_gen_len": 43.36867088607595,
|
| 256 |
+
"eval_loss": 3.3865511417388916,
|
| 257 |
+
"eval_rouge1": 11.5579,
|
| 258 |
+
"eval_rouge2": 2.3617,
|
| 259 |
+
"eval_rougeL": 8.299,
|
| 260 |
+
"eval_rougeLsum": 9.8507,
|
| 261 |
+
"eval_runtime": 456.2669,
|
| 262 |
+
"eval_samples_per_second": 1.385,
|
| 263 |
+
"eval_steps_per_second": 1.385,
|
| 264 |
+
"step": 17328
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 4.0,
|
| 268 |
+
"step": 17328,
|
| 269 |
+
"total_flos": 2855176963080192.0,
|
| 270 |
+
"train_loss": 0.29059598513036994,
|
| 271 |
+
"train_runtime": 2646.0496,
|
| 272 |
+
"train_samples_per_second": 8.186,
|
| 273 |
+
"train_steps_per_second": 8.186
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 274 |
}
|
| 275 |
],
|
| 276 |
+
"max_steps": 21660,
|
| 277 |
+
"num_train_epochs": 5,
|
| 278 |
+
"total_flos": 2855176963080192.0,
|
|
|
|
|
|
|
|
|
|
|
|
|
| 279 |
"trial_name": null,
|
| 280 |
"trial_params": null
|
| 281 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4411
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dacf9007d2268ced50160573cc2957c600341983cef3d21a43c082be0f9423f8
|
| 3 |
size 4411
|