Upload 7 files
Browse files- train_validation_epoc10_late1e-5_/config.json +44 -0
- train_validation_epoc10_late1e-5_/model.safetensors +3 -0
- train_validation_epoc10_late1e-5_/optimizer.pt +3 -0
- train_validation_epoc10_late1e-5_/rng_state.pth +3 -0
- train_validation_epoc10_late1e-5_/scheduler.pt +3 -0
- train_validation_epoc10_late1e-5_/trainer_state.json +540 -0
- train_validation_epoc10_late1e-5_/training_args.bin +3 -0
train_validation_epoc10_late1e-5_/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.1,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"id2label": {
|
| 15 |
+
"0": "\uae30\uc068",
|
| 16 |
+
"1": "\ub2f9\ud669",
|
| 17 |
+
"2": "\ubd84\ub178",
|
| 18 |
+
"3": "\ubd88\uc548",
|
| 19 |
+
"4": "\uc0c1\ucc98",
|
| 20 |
+
"5": "\uc2ac\ud514"
|
| 21 |
+
},
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3072,
|
| 24 |
+
"label2id": {
|
| 25 |
+
"\uae30\uc068": 0,
|
| 26 |
+
"\ub2f9\ud669": 1,
|
| 27 |
+
"\ubd84\ub178": 2,
|
| 28 |
+
"\ubd88\uc548": 3,
|
| 29 |
+
"\uc0c1\ucc98": 4,
|
| 30 |
+
"\uc2ac\ud514": 5
|
| 31 |
+
},
|
| 32 |
+
"layer_norm_eps": 1e-05,
|
| 33 |
+
"max_position_embeddings": 514,
|
| 34 |
+
"model_type": "roberta",
|
| 35 |
+
"num_attention_heads": 12,
|
| 36 |
+
"num_hidden_layers": 12,
|
| 37 |
+
"pad_token_id": 1,
|
| 38 |
+
"position_embedding_type": "absolute",
|
| 39 |
+
"tokenizer_class": "BertTokenizer",
|
| 40 |
+
"transformers_version": "4.57.1",
|
| 41 |
+
"type_vocab_size": 1,
|
| 42 |
+
"use_cache": true,
|
| 43 |
+
"vocab_size": 32000
|
| 44 |
+
}
|
train_validation_epoc10_late1e-5_/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a01f1878186591d161ab202ab802b8c24ae616734040613db0bbf380b34c14a6
|
| 3 |
+
size 442515048
|
train_validation_epoc10_late1e-5_/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f699c7a5bceaaa90efb4fd0cc43499e80254b0e03a0c8c703877d713dfa40e33
|
| 3 |
+
size 885149946
|
train_validation_epoc10_late1e-5_/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f6bb84cf6813f85267f549749f7f4a30acafb0596d05b94015e1ea1106d83a1
|
| 3 |
+
size 14244
|
train_validation_epoc10_late1e-5_/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c82bbd8f423e6c39232b4b8e249b779575daaf3e47bfcf053024fb61ebcd92ee
|
| 3 |
+
size 1064
|
train_validation_epoc10_late1e-5_/trainer_state.json
ADDED
|
@@ -0,0 +1,540 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 14525,
|
| 3 |
+
"best_metric": 0.7187681580476467,
|
| 4 |
+
"best_model_checkpoint": "./results\\emotion_model_v2_manual\\checkpoint-14525",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 29050,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.1721170395869191,
|
| 14 |
+
"grad_norm": 4.724106311798096,
|
| 15 |
+
"learning_rate": 1.7177280550774528e-06,
|
| 16 |
+
"loss": 1.6423,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.3442340791738382,
|
| 21 |
+
"grad_norm": 9.161306381225586,
|
| 22 |
+
"learning_rate": 3.438898450946644e-06,
|
| 23 |
+
"loss": 1.0679,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.5163511187607573,
|
| 28 |
+
"grad_norm": 9.363469123840332,
|
| 29 |
+
"learning_rate": 5.160068846815836e-06,
|
| 30 |
+
"loss": 0.8434,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.6884681583476764,
|
| 35 |
+
"grad_norm": 6.987210273742676,
|
| 36 |
+
"learning_rate": 6.881239242685026e-06,
|
| 37 |
+
"loss": 0.7415,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.8605851979345955,
|
| 42 |
+
"grad_norm": 7.236252784729004,
|
| 43 |
+
"learning_rate": 8.602409638554217e-06,
|
| 44 |
+
"loss": 0.7327,
|
| 45 |
+
"step": 2500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.0,
|
| 49 |
+
"eval_accuracy": 0.6786751888436955,
|
| 50 |
+
"eval_f1": 0.6769930163180903,
|
| 51 |
+
"eval_loss": 0.7002930045127869,
|
| 52 |
+
"eval_runtime": 23.5126,
|
| 53 |
+
"eval_samples_per_second": 219.584,
|
| 54 |
+
"eval_steps_per_second": 3.445,
|
| 55 |
+
"step": 2905
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.0327022375215147,
|
| 59 |
+
"grad_norm": 11.019606590270996,
|
| 60 |
+
"learning_rate": 9.999681056649855e-06,
|
| 61 |
+
"loss": 0.6778,
|
| 62 |
+
"step": 3000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.2048192771084336,
|
| 66 |
+
"grad_norm": 5.872101783752441,
|
| 67 |
+
"learning_rate": 9.987269338508347e-06,
|
| 68 |
+
"loss": 0.6805,
|
| 69 |
+
"step": 3500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.3769363166953528,
|
| 73 |
+
"grad_norm": 13.4342679977417,
|
| 74 |
+
"learning_rate": 9.956860847149606e-06,
|
| 75 |
+
"loss": 0.672,
|
| 76 |
+
"step": 4000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.549053356282272,
|
| 80 |
+
"grad_norm": 6.019625663757324,
|
| 81 |
+
"learning_rate": 9.908565312906149e-06,
|
| 82 |
+
"loss": 0.6501,
|
| 83 |
+
"step": 4500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.721170395869191,
|
| 87 |
+
"grad_norm": 32.90696716308594,
|
| 88 |
+
"learning_rate": 9.842557012264402e-06,
|
| 89 |
+
"loss": 0.6514,
|
| 90 |
+
"step": 5000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.8932874354561102,
|
| 94 |
+
"grad_norm": 9.127164840698242,
|
| 95 |
+
"learning_rate": 9.759074138980602e-06,
|
| 96 |
+
"loss": 0.648,
|
| 97 |
+
"step": 5500
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.0,
|
| 101 |
+
"eval_accuracy": 0.6751888436955259,
|
| 102 |
+
"eval_f1": 0.6765777142972994,
|
| 103 |
+
"eval_loss": 0.6460646390914917,
|
| 104 |
+
"eval_runtime": 23.4762,
|
| 105 |
+
"eval_samples_per_second": 219.925,
|
| 106 |
+
"eval_steps_per_second": 3.45,
|
| 107 |
+
"step": 5810
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.0654044750430294,
|
| 111 |
+
"grad_norm": 3.502110242843628,
|
| 112 |
+
"learning_rate": 9.658417944548512e-06,
|
| 113 |
+
"loss": 0.6257,
|
| 114 |
+
"step": 6000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.2375215146299485,
|
| 118 |
+
"grad_norm": 17.171558380126953,
|
| 119 |
+
"learning_rate": 9.540951651120605e-06,
|
| 120 |
+
"loss": 0.5508,
|
| 121 |
+
"step": 6500
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.4096385542168672,
|
| 125 |
+
"grad_norm": 12.519411087036133,
|
| 126 |
+
"learning_rate": 9.407099140805506e-06,
|
| 127 |
+
"loss": 0.5642,
|
| 128 |
+
"step": 7000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.581755593803787,
|
| 132 |
+
"grad_norm": 4.6616106033325195,
|
| 133 |
+
"learning_rate": 9.2573434260714e-06,
|
| 134 |
+
"loss": 0.5705,
|
| 135 |
+
"step": 7500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.7538726333907055,
|
| 139 |
+
"grad_norm": 10.083399772644043,
|
| 140 |
+
"learning_rate": 9.092224906775031e-06,
|
| 141 |
+
"loss": 0.5429,
|
| 142 |
+
"step": 8000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.9259896729776247,
|
| 146 |
+
"grad_norm": 21.949539184570312,
|
| 147 |
+
"learning_rate": 8.912339420105856e-06,
|
| 148 |
+
"loss": 0.5706,
|
| 149 |
+
"step": 8500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.0,
|
| 153 |
+
"eval_accuracy": 0.7090838659693977,
|
| 154 |
+
"eval_f1": 0.7078776705900655,
|
| 155 |
+
"eval_loss": 0.652846097946167,
|
| 156 |
+
"eval_runtime": 23.4941,
|
| 157 |
+
"eval_samples_per_second": 219.757,
|
| 158 |
+
"eval_steps_per_second": 3.448,
|
| 159 |
+
"step": 8715
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.098106712564544,
|
| 163 |
+
"grad_norm": 5.069673538208008,
|
| 164 |
+
"learning_rate": 8.718336090482256e-06,
|
| 165 |
+
"loss": 0.5018,
|
| 166 |
+
"step": 9000
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 3.270223752151463,
|
| 170 |
+
"grad_norm": 25.420305252075195,
|
| 171 |
+
"learning_rate": 8.510914987158492e-06,
|
| 172 |
+
"loss": 0.4446,
|
| 173 |
+
"step": 9500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.442340791738382,
|
| 177 |
+
"grad_norm": 9.161646842956543,
|
| 178 |
+
"learning_rate": 8.29082459799508e-06,
|
| 179 |
+
"loss": 0.479,
|
| 180 |
+
"step": 10000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.6144578313253013,
|
| 184 |
+
"grad_norm": 5.734455585479736,
|
| 185 |
+
"learning_rate": 8.058859128508552e-06,
|
| 186 |
+
"loss": 0.4664,
|
| 187 |
+
"step": 10500
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 3.7865748709122204,
|
| 191 |
+
"grad_norm": 8.140594482421875,
|
| 192 |
+
"learning_rate": 7.815855635947117e-06,
|
| 193 |
+
"loss": 0.4671,
|
| 194 |
+
"step": 11000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.958691910499139,
|
| 198 |
+
"grad_norm": 12.30313777923584,
|
| 199 |
+
"learning_rate": 7.5626910087340345e-06,
|
| 200 |
+
"loss": 0.4672,
|
| 201 |
+
"step": 11500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 4.0,
|
| 205 |
+
"eval_accuracy": 0.6891342242882045,
|
| 206 |
+
"eval_f1": 0.6904309534188946,
|
| 207 |
+
"eval_loss": 0.6823111772537231,
|
| 208 |
+
"eval_runtime": 23.5245,
|
| 209 |
+
"eval_samples_per_second": 219.473,
|
| 210 |
+
"eval_steps_per_second": 3.443,
|
| 211 |
+
"step": 11620
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 4.130808950086059,
|
| 215 |
+
"grad_norm": 6.850505828857422,
|
| 216 |
+
"learning_rate": 7.300278802178502e-06,
|
| 217 |
+
"loss": 0.41,
|
| 218 |
+
"step": 12000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.3029259896729775,
|
| 222 |
+
"grad_norm": 10.596614837646484,
|
| 223 |
+
"learning_rate": 7.029565941872515e-06,
|
| 224 |
+
"loss": 0.3749,
|
| 225 |
+
"step": 12500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 4.475043029259897,
|
| 229 |
+
"grad_norm": 17.926851272583008,
|
| 230 |
+
"learning_rate": 6.751529306669657e-06,
|
| 231 |
+
"loss": 0.3686,
|
| 232 |
+
"step": 13000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 4.647160068846816,
|
| 236 |
+
"grad_norm": 9.452420234680176,
|
| 237 |
+
"learning_rate": 6.467172203576269e-06,
|
| 238 |
+
"loss": 0.3805,
|
| 239 |
+
"step": 13500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.8192771084337345,
|
| 243 |
+
"grad_norm": 10.189457893371582,
|
| 244 |
+
"learning_rate": 6.177520747275505e-06,
|
| 245 |
+
"loss": 0.3808,
|
| 246 |
+
"step": 14000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 4.991394148020654,
|
| 250 |
+
"grad_norm": 9.415667533874512,
|
| 251 |
+
"learning_rate": 5.883620157348912e-06,
|
| 252 |
+
"loss": 0.3761,
|
| 253 |
+
"step": 14500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 5.0,
|
| 257 |
+
"eval_accuracy": 0.7187681580476467,
|
| 258 |
+
"eval_f1": 0.7183692977167631,
|
| 259 |
+
"eval_loss": 0.8157204985618591,
|
| 260 |
+
"eval_runtime": 23.4773,
|
| 261 |
+
"eval_samples_per_second": 219.915,
|
| 262 |
+
"eval_steps_per_second": 3.45,
|
| 263 |
+
"step": 14525
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 5.163511187607573,
|
| 267 |
+
"grad_norm": 15.120372772216797,
|
| 268 |
+
"learning_rate": 5.586530986557198e-06,
|
| 269 |
+
"loss": 0.3165,
|
| 270 |
+
"step": 15000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 5.335628227194492,
|
| 274 |
+
"grad_norm": 39.6508674621582,
|
| 275 |
+
"learning_rate": 5.287325293790533e-06,
|
| 276 |
+
"loss": 0.2895,
|
| 277 |
+
"step": 15500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 5.507745266781411,
|
| 281 |
+
"grad_norm": 13.855805397033691,
|
| 282 |
+
"learning_rate": 4.9870827754985254e-06,
|
| 283 |
+
"loss": 0.2962,
|
| 284 |
+
"step": 16000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 5.679862306368331,
|
| 288 |
+
"grad_norm": 24.66667366027832,
|
| 289 |
+
"learning_rate": 4.686886869559725e-06,
|
| 290 |
+
"loss": 0.297,
|
| 291 |
+
"step": 16500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 5.851979345955249,
|
| 295 |
+
"grad_norm": 7.78758430480957,
|
| 296 |
+
"learning_rate": 4.387820845650027e-06,
|
| 297 |
+
"loss": 0.3033,
|
| 298 |
+
"step": 17000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 6.0,
|
| 302 |
+
"eval_accuracy": 0.7119891535928724,
|
| 303 |
+
"eval_f1": 0.7124070834016881,
|
| 304 |
+
"eval_loss": 0.8771335482597351,
|
| 305 |
+
"eval_runtime": 23.4829,
|
| 306 |
+
"eval_samples_per_second": 219.862,
|
| 307 |
+
"eval_steps_per_second": 3.449,
|
| 308 |
+
"step": 17430
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.024096385542169,
|
| 312 |
+
"grad_norm": 10.739643096923828,
|
| 313 |
+
"learning_rate": 4.090963896218038e-06,
|
| 314 |
+
"loss": 0.2787,
|
| 315 |
+
"step": 17500
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.196213425129088,
|
| 319 |
+
"grad_norm": 32.562843322753906,
|
| 320 |
+
"learning_rate": 3.7973872421732906e-06,
|
| 321 |
+
"loss": 0.2336,
|
| 322 |
+
"step": 18000
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 6.368330464716007,
|
| 326 |
+
"grad_norm": 24.239948272705078,
|
| 327 |
+
"learning_rate": 3.5081502673400967e-06,
|
| 328 |
+
"loss": 0.2229,
|
| 329 |
+
"step": 18500
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 6.540447504302926,
|
| 333 |
+
"grad_norm": 20.512460708618164,
|
| 334 |
+
"learning_rate": 3.224296695626036e-06,
|
| 335 |
+
"loss": 0.2346,
|
| 336 |
+
"step": 19000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 6.712564543889846,
|
| 340 |
+
"grad_norm": 50.5634651184082,
|
| 341 |
+
"learning_rate": 2.9468508246998915e-06,
|
| 342 |
+
"loss": 0.2213,
|
| 343 |
+
"step": 19500
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 6.884681583476764,
|
| 347 |
+
"grad_norm": 7.660253047943115,
|
| 348 |
+
"learning_rate": 2.6768138297700173e-06,
|
| 349 |
+
"loss": 0.2327,
|
| 350 |
+
"step": 20000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 7.0,
|
| 354 |
+
"eval_accuracy": 0.7129575828006973,
|
| 355 |
+
"eval_f1": 0.7124027051593957,
|
| 356 |
+
"eval_loss": 1.0587843656539917,
|
| 357 |
+
"eval_runtime": 23.5153,
|
| 358 |
+
"eval_samples_per_second": 219.559,
|
| 359 |
+
"eval_steps_per_second": 3.445,
|
| 360 |
+
"step": 20335
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 7.056798623063683,
|
| 364 |
+
"grad_norm": 7.952240467071533,
|
| 365 |
+
"learning_rate": 2.4151601508010914e-06,
|
| 366 |
+
"loss": 0.2173,
|
| 367 |
+
"step": 20500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 7.228915662650603,
|
| 371 |
+
"grad_norm": 37.356449127197266,
|
| 372 |
+
"learning_rate": 2.162833976206092e-06,
|
| 373 |
+
"loss": 0.1777,
|
| 374 |
+
"step": 21000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 7.401032702237521,
|
| 378 |
+
"grad_norm": 12.354819297790527,
|
| 379 |
+
"learning_rate": 1.920745835702335e-06,
|
| 380 |
+
"loss": 0.1909,
|
| 381 |
+
"step": 21500
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 7.573149741824441,
|
| 385 |
+
"grad_norm": 27.18878936767578,
|
| 386 |
+
"learning_rate": 1.6897693146263067e-06,
|
| 387 |
+
"loss": 0.1894,
|
| 388 |
+
"step": 22000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 7.74526678141136,
|
| 392 |
+
"grad_norm": 10.569825172424316,
|
| 393 |
+
"learning_rate": 1.4707379015638784e-06,
|
| 394 |
+
"loss": 0.1775,
|
| 395 |
+
"step": 22500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 7.917383820998279,
|
| 399 |
+
"grad_norm": 15.58621883392334,
|
| 400 |
+
"learning_rate": 1.264441980671345e-06,
|
| 401 |
+
"loss": 0.1751,
|
| 402 |
+
"step": 23000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 8.0,
|
| 406 |
+
"eval_accuracy": 0.7129575828006973,
|
| 407 |
+
"eval_f1": 0.7124591956357624,
|
| 408 |
+
"eval_loss": 1.1589008569717407,
|
| 409 |
+
"eval_runtime": 23.5306,
|
| 410 |
+
"eval_samples_per_second": 219.417,
|
| 411 |
+
"eval_steps_per_second": 3.442,
|
| 412 |
+
"step": 23240
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 8.089500860585199,
|
| 416 |
+
"grad_norm": 18.19190788269043,
|
| 417 |
+
"learning_rate": 1.0716259795406304e-06,
|
| 418 |
+
"loss": 0.17,
|
| 419 |
+
"step": 23500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 8.261617900172118,
|
| 423 |
+
"grad_norm": 1.6293628215789795,
|
| 424 |
+
"learning_rate": 8.929856829007067e-07,
|
| 425 |
+
"loss": 0.1527,
|
| 426 |
+
"step": 24000
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 8.433734939759036,
|
| 430 |
+
"grad_norm": 38.503089904785156,
|
| 431 |
+
"learning_rate": 7.291657218488491e-07,
|
| 432 |
+
"loss": 0.1506,
|
| 433 |
+
"step": 24500
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 8.605851979345955,
|
| 437 |
+
"grad_norm": 6.490494728088379,
|
| 438 |
+
"learning_rate": 5.807572476719431e-07,
|
| 439 |
+
"loss": 0.1615,
|
| 440 |
+
"step": 25000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 8.777969018932874,
|
| 444 |
+
"grad_norm": 15.903548240661621,
|
| 445 |
+
"learning_rate": 4.4829579865196096e-07,
|
| 446 |
+
"loss": 0.1439,
|
| 447 |
+
"step": 25500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 8.950086058519794,
|
| 451 |
+
"grad_norm": 7.3351569175720215,
|
| 452 |
+
"learning_rate": 3.32259367553327e-07,
|
| 453 |
+
"loss": 0.1686,
|
| 454 |
+
"step": 26000
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 9.0,
|
| 458 |
+
"eval_accuracy": 0.7112144102266125,
|
| 459 |
+
"eval_f1": 0.7106582757725869,
|
| 460 |
+
"eval_loss": 1.192394733428955,
|
| 461 |
+
"eval_runtime": 23.4543,
|
| 462 |
+
"eval_samples_per_second": 220.13,
|
| 463 |
+
"eval_steps_per_second": 3.454,
|
| 464 |
+
"step": 26145
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 9.122203098106713,
|
| 468 |
+
"grad_norm": 25.813560485839844,
|
| 469 |
+
"learning_rate": 2.3306667676572413e-07,
|
| 470 |
+
"loss": 0.139,
|
| 471 |
+
"step": 26500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 9.294320137693632,
|
| 475 |
+
"grad_norm": 15.110641479492188,
|
| 476 |
+
"learning_rate": 1.5107566732657176e-07,
|
| 477 |
+
"loss": 0.1417,
|
| 478 |
+
"step": 27000
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 9.46643717728055,
|
| 482 |
+
"grad_norm": 4.259241580963135,
|
| 483 |
+
"learning_rate": 8.65822072755873e-08,
|
| 484 |
+
"loss": 0.1331,
|
| 485 |
+
"step": 27500
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 9.638554216867469,
|
| 489 |
+
"grad_norm": 39.04631042480469,
|
| 490 |
+
"learning_rate": 3.981902400240101e-08,
|
| 491 |
+
"loss": 0.1353,
|
| 492 |
+
"step": 28000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 9.81067125645439,
|
| 496 |
+
"grad_norm": 10.42868709564209,
|
| 497 |
+
"learning_rate": 1.095486443988547e-08,
|
| 498 |
+
"loss": 0.141,
|
| 499 |
+
"step": 28500
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 9.982788296041308,
|
| 503 |
+
"grad_norm": 37.38633346557617,
|
| 504 |
+
"learning_rate": 9.388613368188637e-11,
|
| 505 |
+
"loss": 0.1499,
|
| 506 |
+
"step": 29000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 10.0,
|
| 510 |
+
"eval_accuracy": 0.7100522951772226,
|
| 511 |
+
"eval_f1": 0.7096747565777057,
|
| 512 |
+
"eval_loss": 1.2027859687805176,
|
| 513 |
+
"eval_runtime": 23.841,
|
| 514 |
+
"eval_samples_per_second": 216.56,
|
| 515 |
+
"eval_steps_per_second": 3.398,
|
| 516 |
+
"step": 29050
|
| 517 |
+
}
|
| 518 |
+
],
|
| 519 |
+
"logging_steps": 500,
|
| 520 |
+
"max_steps": 29050,
|
| 521 |
+
"num_input_tokens_seen": 0,
|
| 522 |
+
"num_train_epochs": 10,
|
| 523 |
+
"save_steps": 500,
|
| 524 |
+
"stateful_callbacks": {
|
| 525 |
+
"TrainerControl": {
|
| 526 |
+
"args": {
|
| 527 |
+
"should_epoch_stop": false,
|
| 528 |
+
"should_evaluate": false,
|
| 529 |
+
"should_log": false,
|
| 530 |
+
"should_save": true,
|
| 531 |
+
"should_training_stop": true
|
| 532 |
+
},
|
| 533 |
+
"attributes": {}
|
| 534 |
+
}
|
| 535 |
+
},
|
| 536 |
+
"total_flos": 3.05647356425472e+16,
|
| 537 |
+
"train_batch_size": 16,
|
| 538 |
+
"trial_name": null,
|
| 539 |
+
"trial_params": null
|
| 540 |
+
}
|
train_validation_epoc10_late1e-5_/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90fb1d0a1e74848481641c01cac1349a36e03c77cd2f84aa53f19c3dabde55
|
| 3 |
+
size 5432
|