Commit
•
0a93dcc
1
Parent(s):
c4e4950
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": 0.8578300921187308,
|
| 4 |
+
"eval_loss": 0.47337478399276733,
|
| 5 |
+
"eval_runtime": 16.6674,
|
| 6 |
+
"eval_samples_per_second": 586.173,
|
| 7 |
+
"eval_steps_per_second": 18.359
|
| 8 |
+
}
|
ChcE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8649948822927328,
|
| 4 |
+
"eval_loss": 0.4547342360019684,
|
| 5 |
+
"eval_runtime": 16.6923,
|
| 6 |
+
"eval_samples_per_second": 585.299,
|
| 7 |
+
"eval_steps_per_second": 18.332
|
| 8 |
+
}
|
CollSgE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8433981576253838,
|
| 4 |
+
"eval_loss": 0.5234562754631042,
|
| 5 |
+
"eval_runtime": 16.6821,
|
| 6 |
+
"eval_samples_per_second": 585.659,
|
| 7 |
+
"eval_steps_per_second": 18.343
|
| 8 |
+
}
|
GLUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8689760570555273,
|
| 4 |
+
"eval_loss": 0.44227728247642517,
|
| 5 |
+
"eval_runtime": 16.7654,
|
| 6 |
+
"eval_samples_per_second": 585.433,
|
| 7 |
+
"eval_steps_per_second": 18.312
|
| 8 |
+
}
|
IndE_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8496417604912999,
|
| 4 |
+
"eval_loss": 0.5003358721733093,
|
| 5 |
+
"eval_runtime": 16.6915,
|
| 6 |
+
"eval_samples_per_second": 585.326,
|
| 7 |
+
"eval_steps_per_second": 18.333
|
| 8 |
+
}
|
MULTI_dev_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8406790060333368,
|
| 4 |
+
"eval_loss": 0.534744143486023,
|
| 5 |
+
"eval_runtime": 16.8125,
|
| 6 |
+
"eval_samples_per_second": 581.65,
|
| 7 |
+
"eval_steps_per_second": 18.201
|
| 8 |
+
}
|
VALUE_validation_matched_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8410596026490066,
|
| 4 |
+
"eval_loss": 0.5228205323219299,
|
| 5 |
+
"eval_runtime": 16.7619,
|
| 6 |
+
"eval_samples_per_second": 585.554,
|
| 7 |
+
"eval_steps_per_second": 18.315
|
| 8 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_acc": 0.8689760570555273,
|
| 4 |
+
"eval_loss": 0.44227728247642517,
|
| 5 |
+
"eval_runtime": 16.7654,
|
| 6 |
+
"eval_samples_per_second": 585.433,
|
| 7 |
+
"eval_steps_per_second": 18.312,
|
| 8 |
+
"train_loss": 0.2560187041367518,
|
| 9 |
+
"train_runtime": 11508.9997,
|
| 10 |
+
"train_samples_per_second": 166.995,
|
| 11 |
+
"train_steps_per_second": 5.219
|
| 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 sha256:18d990fa4d158bd84b9890e5e5cbb1ff7c406ee2fe4472d0724efdd939f675f8
|
| 3 |
size 498665141
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.2560187041367518,
|
| 4 |
+
"train_runtime": 11508.9997,
|
| 5 |
+
"train_samples_per_second": 166.995,
|
| 6 |
+
"train_steps_per_second": 5.219
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,475 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8406790060333368,
|
| 3 |
+
"best_model_checkpoint": "./outputs/finetuning/mnli_MULTI/checkpoint-26000",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 60065,
|
| 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": 1.933405477399484e-05,
|
| 13 |
+
"loss": 0.4798,
|
| 14 |
+
"step": 2000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.17,
|
| 18 |
+
"eval_acc": 0.8211473565804275,
|
| 19 |
+
"eval_loss": 0.4810026288032532,
|
| 20 |
+
"eval_runtime": 16.7598,
|
| 21 |
+
"eval_samples_per_second": 583.48,
|
| 22 |
+
"eval_steps_per_second": 18.258,
|
| 23 |
+
"step": 2000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33,
|
| 27 |
+
"learning_rate": 1.866810954798968e-05,
|
| 28 |
+
"loss": 0.4494,
|
| 29 |
+
"step": 4000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.33,
|
| 33 |
+
"eval_acc": 0.8267716535433071,
|
| 34 |
+
"eval_loss": 0.46056851744651794,
|
| 35 |
+
"eval_runtime": 16.6747,
|
| 36 |
+
"eval_samples_per_second": 586.457,
|
| 37 |
+
"eval_steps_per_second": 18.351,
|
| 38 |
+
"step": 4000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"learning_rate": 1.8002164321984518e-05,
|
| 43 |
+
"loss": 0.431,
|
| 44 |
+
"step": 6000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"eval_acc": 0.8301462317210349,
|
| 49 |
+
"eval_loss": 0.4622686803340912,
|
| 50 |
+
"eval_runtime": 16.6907,
|
| 51 |
+
"eval_samples_per_second": 585.896,
|
| 52 |
+
"eval_steps_per_second": 18.334,
|
| 53 |
+
"step": 6000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.67,
|
| 57 |
+
"learning_rate": 1.7336219095979357e-05,
|
| 58 |
+
"loss": 0.4371,
|
| 59 |
+
"step": 8000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.67,
|
| 63 |
+
"eval_acc": 0.8297371919419163,
|
| 64 |
+
"eval_loss": 0.44371065497398376,
|
| 65 |
+
"eval_runtime": 16.7034,
|
| 66 |
+
"eval_samples_per_second": 585.451,
|
| 67 |
+
"eval_steps_per_second": 18.32,
|
| 68 |
+
"step": 8000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.83,
|
| 72 |
+
"learning_rate": 1.6670273869974196e-05,
|
| 73 |
+
"loss": 0.4297,
|
| 74 |
+
"step": 10000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.83,
|
| 78 |
+
"eval_acc": 0.8312710911136107,
|
| 79 |
+
"eval_loss": 0.45482972264289856,
|
| 80 |
+
"eval_runtime": 16.6876,
|
| 81 |
+
"eval_samples_per_second": 586.004,
|
| 82 |
+
"eval_steps_per_second": 18.337,
|
| 83 |
+
"step": 10000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.0,
|
| 87 |
+
"learning_rate": 1.6004328643969035e-05,
|
| 88 |
+
"loss": 0.4214,
|
| 89 |
+
"step": 12000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.0,
|
| 93 |
+
"eval_acc": 0.8337253297883219,
|
| 94 |
+
"eval_loss": 0.4565775692462921,
|
| 95 |
+
"eval_runtime": 16.6965,
|
| 96 |
+
"eval_samples_per_second": 585.692,
|
| 97 |
+
"eval_steps_per_second": 18.327,
|
| 98 |
+
"step": 12000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.17,
|
| 102 |
+
"learning_rate": 1.5338383417963873e-05,
|
| 103 |
+
"loss": 0.3123,
|
| 104 |
+
"step": 14000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.17,
|
| 108 |
+
"eval_acc": 0.8322936905614071,
|
| 109 |
+
"eval_loss": 0.48932746052742004,
|
| 110 |
+
"eval_runtime": 16.699,
|
| 111 |
+
"eval_samples_per_second": 585.603,
|
| 112 |
+
"eval_steps_per_second": 18.324,
|
| 113 |
+
"step": 14000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.33,
|
| 117 |
+
"learning_rate": 1.4672438191958714e-05,
|
| 118 |
+
"loss": 0.3158,
|
| 119 |
+
"step": 16000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.33,
|
| 123 |
+
"eval_acc": 0.8342366295122201,
|
| 124 |
+
"eval_loss": 0.4861135184764862,
|
| 125 |
+
"eval_runtime": 16.699,
|
| 126 |
+
"eval_samples_per_second": 585.602,
|
| 127 |
+
"eval_steps_per_second": 18.324,
|
| 128 |
+
"step": 16000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.5,
|
| 132 |
+
"learning_rate": 1.4006492965953551e-05,
|
| 133 |
+
"loss": 0.324,
|
| 134 |
+
"step": 18000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5,
|
| 138 |
+
"eval_acc": 0.8307597913897127,
|
| 139 |
+
"eval_loss": 0.4812241792678833,
|
| 140 |
+
"eval_runtime": 16.6905,
|
| 141 |
+
"eval_samples_per_second": 585.902,
|
| 142 |
+
"eval_steps_per_second": 18.334,
|
| 143 |
+
"step": 18000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.66,
|
| 147 |
+
"learning_rate": 1.3340547739948392e-05,
|
| 148 |
+
"loss": 0.3161,
|
| 149 |
+
"step": 20000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.66,
|
| 153 |
+
"eval_acc": 0.8364863482973719,
|
| 154 |
+
"eval_loss": 0.4630277454853058,
|
| 155 |
+
"eval_runtime": 16.702,
|
| 156 |
+
"eval_samples_per_second": 585.498,
|
| 157 |
+
"eval_steps_per_second": 18.321,
|
| 158 |
+
"step": 20000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.83,
|
| 162 |
+
"learning_rate": 1.2674602513943229e-05,
|
| 163 |
+
"loss": 0.32,
|
| 164 |
+
"step": 22000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.83,
|
| 168 |
+
"eval_acc": 0.8365886082421515,
|
| 169 |
+
"eval_loss": 0.46297991275787354,
|
| 170 |
+
"eval_runtime": 16.6817,
|
| 171 |
+
"eval_samples_per_second": 586.213,
|
| 172 |
+
"eval_steps_per_second": 18.343,
|
| 173 |
+
"step": 22000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.0,
|
| 177 |
+
"learning_rate": 1.2008657287938067e-05,
|
| 178 |
+
"loss": 0.3195,
|
| 179 |
+
"step": 24000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"eval_acc": 0.8354637488495756,
|
| 184 |
+
"eval_loss": 0.4681137800216675,
|
| 185 |
+
"eval_runtime": 16.7123,
|
| 186 |
+
"eval_samples_per_second": 585.138,
|
| 187 |
+
"eval_steps_per_second": 18.31,
|
| 188 |
+
"step": 24000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.16,
|
| 192 |
+
"learning_rate": 1.1342712061932908e-05,
|
| 193 |
+
"loss": 0.2274,
|
| 194 |
+
"step": 26000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.16,
|
| 198 |
+
"eval_acc": 0.8406790060333368,
|
| 199 |
+
"eval_loss": 0.534744143486023,
|
| 200 |
+
"eval_runtime": 16.6972,
|
| 201 |
+
"eval_samples_per_second": 585.668,
|
| 202 |
+
"eval_steps_per_second": 18.326,
|
| 203 |
+
"step": 26000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.33,
|
| 207 |
+
"learning_rate": 1.0676766835927745e-05,
|
| 208 |
+
"loss": 0.2311,
|
| 209 |
+
"step": 28000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.33,
|
| 213 |
+
"eval_acc": 0.830964311279272,
|
| 214 |
+
"eval_loss": 0.5649741291999817,
|
| 215 |
+
"eval_runtime": 16.7014,
|
| 216 |
+
"eval_samples_per_second": 585.518,
|
| 217 |
+
"eval_steps_per_second": 18.322,
|
| 218 |
+
"step": 28000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.5,
|
| 222 |
+
"learning_rate": 1.0010821609922586e-05,
|
| 223 |
+
"loss": 0.2293,
|
| 224 |
+
"step": 30000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.5,
|
| 228 |
+
"eval_acc": 0.8354637488495756,
|
| 229 |
+
"eval_loss": 0.5407743453979492,
|
| 230 |
+
"eval_runtime": 16.7112,
|
| 231 |
+
"eval_samples_per_second": 585.175,
|
| 232 |
+
"eval_steps_per_second": 18.311,
|
| 233 |
+
"step": 30000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.66,
|
| 237 |
+
"learning_rate": 9.344876383917424e-06,
|
| 238 |
+
"loss": 0.2296,
|
| 239 |
+
"step": 32000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.66,
|
| 243 |
+
"eval_acc": 0.8374066878003886,
|
| 244 |
+
"eval_loss": 0.5207422971725464,
|
| 245 |
+
"eval_runtime": 16.6974,
|
| 246 |
+
"eval_samples_per_second": 585.661,
|
| 247 |
+
"eval_steps_per_second": 18.326,
|
| 248 |
+
"step": 32000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.83,
|
| 252 |
+
"learning_rate": 8.678931157912263e-06,
|
| 253 |
+
"loss": 0.2274,
|
| 254 |
+
"step": 34000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.83,
|
| 258 |
+
"eval_acc": 0.8352592289600164,
|
| 259 |
+
"eval_loss": 0.5696293115615845,
|
| 260 |
+
"eval_runtime": 16.683,
|
| 261 |
+
"eval_samples_per_second": 586.165,
|
| 262 |
+
"eval_steps_per_second": 18.342,
|
| 263 |
+
"step": 34000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.0,
|
| 267 |
+
"learning_rate": 8.012985931907102e-06,
|
| 268 |
+
"loss": 0.23,
|
| 269 |
+
"step": 36000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.0,
|
| 273 |
+
"eval_acc": 0.8365886082421515,
|
| 274 |
+
"eval_loss": 0.5331636071205139,
|
| 275 |
+
"eval_runtime": 16.7244,
|
| 276 |
+
"eval_samples_per_second": 584.714,
|
| 277 |
+
"eval_steps_per_second": 18.297,
|
| 278 |
+
"step": 36000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.16,
|
| 282 |
+
"learning_rate": 7.34704070590194e-06,
|
| 283 |
+
"loss": 0.1686,
|
| 284 |
+
"step": 38000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.16,
|
| 288 |
+
"eval_acc": 0.8343388894569997,
|
| 289 |
+
"eval_loss": 0.6275357007980347,
|
| 290 |
+
"eval_runtime": 16.6784,
|
| 291 |
+
"eval_samples_per_second": 586.326,
|
| 292 |
+
"eval_steps_per_second": 18.347,
|
| 293 |
+
"step": 38000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.33,
|
| 297 |
+
"learning_rate": 6.681095479896779e-06,
|
| 298 |
+
"loss": 0.1632,
|
| 299 |
+
"step": 40000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.33,
|
| 303 |
+
"eval_acc": 0.8348501891808978,
|
| 304 |
+
"eval_loss": 0.6457108855247498,
|
| 305 |
+
"eval_runtime": 16.7007,
|
| 306 |
+
"eval_samples_per_second": 585.544,
|
| 307 |
+
"eval_steps_per_second": 18.323,
|
| 308 |
+
"step": 40000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.5,
|
| 312 |
+
"learning_rate": 6.0151502538916185e-06,
|
| 313 |
+
"loss": 0.1686,
|
| 314 |
+
"step": 42000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"eval_acc": 0.8338275897331016,
|
| 319 |
+
"eval_loss": 0.5964699983596802,
|
| 320 |
+
"eval_runtime": 16.6934,
|
| 321 |
+
"eval_samples_per_second": 585.8,
|
| 322 |
+
"eval_steps_per_second": 18.331,
|
| 323 |
+
"step": 42000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.66,
|
| 327 |
+
"learning_rate": 5.349205027886457e-06,
|
| 328 |
+
"loss": 0.1634,
|
| 329 |
+
"step": 44000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.66,
|
| 333 |
+
"eval_acc": 0.8342366295122201,
|
| 334 |
+
"eval_loss": 0.6272006034851074,
|
| 335 |
+
"eval_runtime": 16.6672,
|
| 336 |
+
"eval_samples_per_second": 586.722,
|
| 337 |
+
"eval_steps_per_second": 18.359,
|
| 338 |
+
"step": 44000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.83,
|
| 342 |
+
"learning_rate": 4.683259801881296e-06,
|
| 343 |
+
"loss": 0.1656,
|
| 344 |
+
"step": 46000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.83,
|
| 348 |
+
"eval_acc": 0.8311688311688312,
|
| 349 |
+
"eval_loss": 0.6541053652763367,
|
| 350 |
+
"eval_runtime": 16.6856,
|
| 351 |
+
"eval_samples_per_second": 586.076,
|
| 352 |
+
"eval_steps_per_second": 18.339,
|
| 353 |
+
"step": 46000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.0,
|
| 357 |
+
"learning_rate": 4.017314575876134e-06,
|
| 358 |
+
"loss": 0.162,
|
| 359 |
+
"step": 48000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.0,
|
| 363 |
+
"eval_acc": 0.8316801308927293,
|
| 364 |
+
"eval_loss": 0.6408036947250366,
|
| 365 |
+
"eval_runtime": 16.6809,
|
| 366 |
+
"eval_samples_per_second": 586.239,
|
| 367 |
+
"eval_steps_per_second": 18.344,
|
| 368 |
+
"step": 48000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.16,
|
| 372 |
+
"learning_rate": 3.3513693498709734e-06,
|
| 373 |
+
"loss": 0.1288,
|
| 374 |
+
"step": 50000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.16,
|
| 378 |
+
"eval_acc": 0.8348501891808978,
|
| 379 |
+
"eval_loss": 0.7236860990524292,
|
| 380 |
+
"eval_runtime": 16.7011,
|
| 381 |
+
"eval_samples_per_second": 585.529,
|
| 382 |
+
"eval_steps_per_second": 18.322,
|
| 383 |
+
"step": 50000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.33,
|
| 387 |
+
"learning_rate": 2.6854241238658126e-06,
|
| 388 |
+
"loss": 0.1275,
|
| 389 |
+
"step": 52000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.33,
|
| 393 |
+
"eval_acc": 0.8295326720523571,
|
| 394 |
+
"eval_loss": 0.7558159828186035,
|
| 395 |
+
"eval_runtime": 16.7056,
|
| 396 |
+
"eval_samples_per_second": 585.372,
|
| 397 |
+
"eval_steps_per_second": 18.317,
|
| 398 |
+
"step": 52000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.5,
|
| 402 |
+
"learning_rate": 2.019478897860651e-06,
|
| 403 |
+
"loss": 0.1291,
|
| 404 |
+
"step": 54000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.5,
|
| 408 |
+
"eval_acc": 0.8305552715001534,
|
| 409 |
+
"eval_loss": 0.7729807496070862,
|
| 410 |
+
"eval_runtime": 16.6766,
|
| 411 |
+
"eval_samples_per_second": 586.391,
|
| 412 |
+
"eval_steps_per_second": 18.349,
|
| 413 |
+
"step": 54000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.66,
|
| 417 |
+
"learning_rate": 1.35353367185549e-06,
|
| 418 |
+
"loss": 0.1261,
|
| 419 |
+
"step": 56000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.66,
|
| 423 |
+
"eval_acc": 0.8300439717762552,
|
| 424 |
+
"eval_loss": 0.7523751258850098,
|
| 425 |
+
"eval_runtime": 16.688,
|
| 426 |
+
"eval_samples_per_second": 585.99,
|
| 427 |
+
"eval_steps_per_second": 18.337,
|
| 428 |
+
"step": 56000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.83,
|
| 432 |
+
"learning_rate": 6.875884458503289e-07,
|
| 433 |
+
"loss": 0.1272,
|
| 434 |
+
"step": 58000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.83,
|
| 438 |
+
"eval_acc": 0.8316801308927293,
|
| 439 |
+
"eval_loss": 0.7572413682937622,
|
| 440 |
+
"eval_runtime": 16.7304,
|
| 441 |
+
"eval_samples_per_second": 584.506,
|
| 442 |
+
"eval_steps_per_second": 18.29,
|
| 443 |
+
"step": 58000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 4.99,
|
| 447 |
+
"learning_rate": 2.164321984516774e-08,
|
| 448 |
+
"loss": 0.1242,
|
| 449 |
+
"step": 60000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 4.99,
|
| 453 |
+
"eval_acc": 0.830964311279272,
|
| 454 |
+
"eval_loss": 0.7606698870658875,
|
| 455 |
+
"eval_runtime": 16.6838,
|
| 456 |
+
"eval_samples_per_second": 586.137,
|
| 457 |
+
"eval_steps_per_second": 18.341,
|
| 458 |
+
"step": 60000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 5.0,
|
| 462 |
+
"step": 60065,
|
| 463 |
+
"total_flos": 1.2642205051703808e+17,
|
| 464 |
+
"train_loss": 0.2560187041367518,
|
| 465 |
+
"train_runtime": 11508.9997,
|
| 466 |
+
"train_samples_per_second": 166.995,
|
| 467 |
+
"train_steps_per_second": 5.219
|
| 468 |
+
}
|
| 469 |
+
],
|
| 470 |
+
"max_steps": 60065,
|
| 471 |
+
"num_train_epochs": 5,
|
| 472 |
+
"total_flos": 1.2642205051703808e+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:bd5e8c2d370eadb2eb2873e1b0047069e9bc61b25694664c51be74c67e3d650e
|
| 3 |
size 3451
|