Upload folder using huggingface_hub
Browse files- config.json +25 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +619 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"layer_norm_eps": 1e-12,
|
| 13 |
+
"max_length": 512,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.35.2",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 5120
|
| 25 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"max_length": 512,
|
| 4 |
+
"pad_token_id": 0,
|
| 5 |
+
"transformers_version": "4.35.2"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6677fb2024ace21bf31d9cbd95a9fa578900f93b8536cc6392accdc9286758ae
|
| 3 |
+
size 359944256
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:769b8cf35fbbb7ee3dedb845b68db80a829abb981fbacdfc00ef5682160dad3c
|
| 3 |
+
size 720010699
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a6929480e8da34ff65d3764cc7933f69943b6db1fa0ac74e54af2b7e389043c
|
| 3 |
+
size 14709
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3018e461ce1631f9bf04479faa74b8f3d9fd86bf68d2cd93702db04c8a02131
|
| 3 |
+
size 1465
|
trainer_state.json
ADDED
|
@@ -0,0 +1,619 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.257186081694403,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.3,
|
| 13 |
+
"learning_rate": 4.000000000000001e-06,
|
| 14 |
+
"loss": 7.7084,
|
| 15 |
+
"step": 1000
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.61,
|
| 19 |
+
"learning_rate": 8.000000000000001e-06,
|
| 20 |
+
"loss": 7.1347,
|
| 21 |
+
"step": 2000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.91,
|
| 25 |
+
"learning_rate": 1.2e-05,
|
| 26 |
+
"loss": 6.9286,
|
| 27 |
+
"step": 3000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 1.21,
|
| 31 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 32 |
+
"loss": 6.8404,
|
| 33 |
+
"step": 4000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.51,
|
| 37 |
+
"learning_rate": 2e-05,
|
| 38 |
+
"loss": 6.811,
|
| 39 |
+
"step": 5000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.82,
|
| 43 |
+
"learning_rate": 2.4e-05,
|
| 44 |
+
"loss": 6.7884,
|
| 45 |
+
"step": 6000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.12,
|
| 49 |
+
"learning_rate": 2.8e-05,
|
| 50 |
+
"loss": 6.7671,
|
| 51 |
+
"step": 7000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 2.42,
|
| 55 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 56 |
+
"loss": 6.754,
|
| 57 |
+
"step": 8000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.72,
|
| 61 |
+
"learning_rate": 3.6e-05,
|
| 62 |
+
"loss": 6.7459,
|
| 63 |
+
"step": 9000
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 3.03,
|
| 67 |
+
"learning_rate": 4e-05,
|
| 68 |
+
"loss": 6.7364,
|
| 69 |
+
"step": 10000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.33,
|
| 73 |
+
"learning_rate": 3.955555555555556e-05,
|
| 74 |
+
"loss": 6.7283,
|
| 75 |
+
"step": 11000
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.63,
|
| 79 |
+
"learning_rate": 3.9111111111111115e-05,
|
| 80 |
+
"loss": 6.7226,
|
| 81 |
+
"step": 12000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 3.93,
|
| 85 |
+
"learning_rate": 3.866666666666667e-05,
|
| 86 |
+
"loss": 6.7186,
|
| 87 |
+
"step": 13000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 4.24,
|
| 91 |
+
"learning_rate": 3.8222222222222226e-05,
|
| 92 |
+
"loss": 6.715,
|
| 93 |
+
"step": 14000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.54,
|
| 97 |
+
"learning_rate": 3.777777777777778e-05,
|
| 98 |
+
"loss": 6.7114,
|
| 99 |
+
"step": 15000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.84,
|
| 103 |
+
"learning_rate": 3.733333333333334e-05,
|
| 104 |
+
"loss": 6.7088,
|
| 105 |
+
"step": 16000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 5.14,
|
| 109 |
+
"learning_rate": 3.6888888888888896e-05,
|
| 110 |
+
"loss": 6.7072,
|
| 111 |
+
"step": 17000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 5.45,
|
| 115 |
+
"learning_rate": 3.644444444444445e-05,
|
| 116 |
+
"loss": 6.7059,
|
| 117 |
+
"step": 18000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 5.75,
|
| 121 |
+
"learning_rate": 3.6e-05,
|
| 122 |
+
"loss": 6.7039,
|
| 123 |
+
"step": 19000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 6.05,
|
| 127 |
+
"learning_rate": 3.555555555555555e-05,
|
| 128 |
+
"loss": 6.7023,
|
| 129 |
+
"step": 20000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 6.35,
|
| 133 |
+
"learning_rate": 3.511111111111111e-05,
|
| 134 |
+
"loss": 6.7002,
|
| 135 |
+
"step": 21000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 6.66,
|
| 139 |
+
"learning_rate": 3.466666666666667e-05,
|
| 140 |
+
"loss": 6.698,
|
| 141 |
+
"step": 22000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 6.96,
|
| 145 |
+
"learning_rate": 3.4222222222222224e-05,
|
| 146 |
+
"loss": 6.6975,
|
| 147 |
+
"step": 23000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 7.26,
|
| 151 |
+
"learning_rate": 3.377777777777778e-05,
|
| 152 |
+
"loss": 6.6959,
|
| 153 |
+
"step": 24000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 7.56,
|
| 157 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 158 |
+
"loss": 6.6929,
|
| 159 |
+
"step": 25000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.87,
|
| 163 |
+
"learning_rate": 3.288888888888889e-05,
|
| 164 |
+
"loss": 6.6936,
|
| 165 |
+
"step": 26000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 8.17,
|
| 169 |
+
"learning_rate": 3.2444444444444446e-05,
|
| 170 |
+
"loss": 6.6917,
|
| 171 |
+
"step": 27000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 8.47,
|
| 175 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 176 |
+
"loss": 6.6915,
|
| 177 |
+
"step": 28000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 8.77,
|
| 181 |
+
"learning_rate": 3.155555555555556e-05,
|
| 182 |
+
"loss": 6.69,
|
| 183 |
+
"step": 29000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 9.08,
|
| 187 |
+
"learning_rate": 3.111111111111112e-05,
|
| 188 |
+
"loss": 6.6884,
|
| 189 |
+
"step": 30000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 9.38,
|
| 193 |
+
"learning_rate": 3.066666666666667e-05,
|
| 194 |
+
"loss": 6.6864,
|
| 195 |
+
"step": 31000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 19.37,
|
| 199 |
+
"learning_rate": 3.0222222222222225e-05,
|
| 200 |
+
"loss": 6.6229,
|
| 201 |
+
"step": 32000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 19.97,
|
| 205 |
+
"learning_rate": 2.977777777777778e-05,
|
| 206 |
+
"loss": 6.5542,
|
| 207 |
+
"step": 33000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 20.58,
|
| 211 |
+
"learning_rate": 2.9333333333333333e-05,
|
| 212 |
+
"loss": 6.5168,
|
| 213 |
+
"step": 34000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 21.19,
|
| 217 |
+
"learning_rate": 2.888888888888889e-05,
|
| 218 |
+
"loss": 6.4593,
|
| 219 |
+
"step": 35000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 21.79,
|
| 223 |
+
"learning_rate": 2.8444444444444447e-05,
|
| 224 |
+
"loss": 6.3693,
|
| 225 |
+
"step": 36000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 22.4,
|
| 229 |
+
"learning_rate": 2.8e-05,
|
| 230 |
+
"loss": 6.2642,
|
| 231 |
+
"step": 37000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 23.0,
|
| 235 |
+
"learning_rate": 2.755555555555556e-05,
|
| 236 |
+
"loss": 6.151,
|
| 237 |
+
"step": 38000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 23.61,
|
| 241 |
+
"learning_rate": 2.7111111111111114e-05,
|
| 242 |
+
"loss": 6.0063,
|
| 243 |
+
"step": 39000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 24.21,
|
| 247 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 248 |
+
"loss": 5.9027,
|
| 249 |
+
"step": 40000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 24.82,
|
| 253 |
+
"learning_rate": 2.6222222222222226e-05,
|
| 254 |
+
"loss": 5.826,
|
| 255 |
+
"step": 41000
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 25.42,
|
| 259 |
+
"learning_rate": 2.577777777777778e-05,
|
| 260 |
+
"loss": 5.7664,
|
| 261 |
+
"step": 42000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 26.03,
|
| 265 |
+
"learning_rate": 2.5333333333333334e-05,
|
| 266 |
+
"loss": 5.7222,
|
| 267 |
+
"step": 43000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 26.63,
|
| 271 |
+
"learning_rate": 2.4888888888888893e-05,
|
| 272 |
+
"loss": 5.6801,
|
| 273 |
+
"step": 44000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 27.24,
|
| 277 |
+
"learning_rate": 2.444444444444445e-05,
|
| 278 |
+
"loss": 5.6484,
|
| 279 |
+
"step": 45000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 27.84,
|
| 283 |
+
"learning_rate": 2.4e-05,
|
| 284 |
+
"loss": 5.6185,
|
| 285 |
+
"step": 46000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 28.45,
|
| 289 |
+
"learning_rate": 2.355555555555556e-05,
|
| 290 |
+
"loss": 5.5924,
|
| 291 |
+
"step": 47000
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 29.05,
|
| 295 |
+
"learning_rate": 2.3111111111111112e-05,
|
| 296 |
+
"loss": 5.5701,
|
| 297 |
+
"step": 48000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 29.66,
|
| 301 |
+
"learning_rate": 2.2666666666666668e-05,
|
| 302 |
+
"loss": 5.5486,
|
| 303 |
+
"step": 49000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 30.26,
|
| 307 |
+
"learning_rate": 2.2222222222222227e-05,
|
| 308 |
+
"loss": 5.5285,
|
| 309 |
+
"step": 50000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 30.87,
|
| 313 |
+
"learning_rate": 2.177777777777778e-05,
|
| 314 |
+
"loss": 5.5128,
|
| 315 |
+
"step": 51000
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 31.47,
|
| 319 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 320 |
+
"loss": 5.4946,
|
| 321 |
+
"step": 52000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 32.08,
|
| 325 |
+
"learning_rate": 2.088888888888889e-05,
|
| 326 |
+
"loss": 5.4802,
|
| 327 |
+
"step": 53000
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 32.68,
|
| 331 |
+
"learning_rate": 2.0444444444444446e-05,
|
| 332 |
+
"loss": 5.4654,
|
| 333 |
+
"step": 54000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 33.29,
|
| 337 |
+
"learning_rate": 2e-05,
|
| 338 |
+
"loss": 5.4553,
|
| 339 |
+
"step": 55000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 16.94,
|
| 343 |
+
"learning_rate": 1.9555555555555557e-05,
|
| 344 |
+
"loss": 5.4459,
|
| 345 |
+
"step": 56000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 17.25,
|
| 349 |
+
"learning_rate": 1.9111111111111113e-05,
|
| 350 |
+
"loss": 5.4354,
|
| 351 |
+
"step": 57000
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 17.55,
|
| 355 |
+
"learning_rate": 1.866666666666667e-05,
|
| 356 |
+
"loss": 5.4323,
|
| 357 |
+
"step": 58000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 17.85,
|
| 361 |
+
"learning_rate": 1.8222222222222224e-05,
|
| 362 |
+
"loss": 5.4245,
|
| 363 |
+
"step": 59000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 18.15,
|
| 367 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 368 |
+
"loss": 5.4162,
|
| 369 |
+
"step": 60000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 18.46,
|
| 373 |
+
"learning_rate": 1.7333333333333336e-05,
|
| 374 |
+
"loss": 5.4092,
|
| 375 |
+
"step": 61000
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 18.76,
|
| 379 |
+
"learning_rate": 1.688888888888889e-05,
|
| 380 |
+
"loss": 5.402,
|
| 381 |
+
"step": 62000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 19.06,
|
| 385 |
+
"learning_rate": 1.6444444444444444e-05,
|
| 386 |
+
"loss": 5.3934,
|
| 387 |
+
"step": 63000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 19.36,
|
| 391 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 392 |
+
"loss": 5.392,
|
| 393 |
+
"step": 64000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 19.67,
|
| 397 |
+
"learning_rate": 1.555555555555556e-05,
|
| 398 |
+
"loss": 5.3835,
|
| 399 |
+
"step": 65000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 19.97,
|
| 403 |
+
"learning_rate": 1.5111111111111112e-05,
|
| 404 |
+
"loss": 5.3828,
|
| 405 |
+
"step": 66000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 20.27,
|
| 409 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 410 |
+
"loss": 5.377,
|
| 411 |
+
"step": 67000
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 20.57,
|
| 415 |
+
"learning_rate": 1.4222222222222224e-05,
|
| 416 |
+
"loss": 5.3695,
|
| 417 |
+
"step": 68000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 20.88,
|
| 421 |
+
"learning_rate": 1.377777777777778e-05,
|
| 422 |
+
"loss": 5.3634,
|
| 423 |
+
"step": 69000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 21.18,
|
| 427 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 428 |
+
"loss": 5.358,
|
| 429 |
+
"step": 70000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 21.48,
|
| 433 |
+
"learning_rate": 1.288888888888889e-05,
|
| 434 |
+
"loss": 5.3554,
|
| 435 |
+
"step": 71000
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 21.79,
|
| 439 |
+
"learning_rate": 1.2444444444444446e-05,
|
| 440 |
+
"loss": 5.3537,
|
| 441 |
+
"step": 72000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 22.09,
|
| 445 |
+
"learning_rate": 1.2e-05,
|
| 446 |
+
"loss": 5.3486,
|
| 447 |
+
"step": 73000
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 22.39,
|
| 451 |
+
"learning_rate": 1.1555555555555556e-05,
|
| 452 |
+
"loss": 5.3431,
|
| 453 |
+
"step": 74000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 22.69,
|
| 457 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 458 |
+
"loss": 5.3404,
|
| 459 |
+
"step": 75000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 23.0,
|
| 463 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 464 |
+
"loss": 5.3377,
|
| 465 |
+
"step": 76000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 23.3,
|
| 469 |
+
"learning_rate": 1.0222222222222223e-05,
|
| 470 |
+
"loss": 5.3317,
|
| 471 |
+
"step": 77000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 23.6,
|
| 475 |
+
"learning_rate": 9.777777777777779e-06,
|
| 476 |
+
"loss": 5.3319,
|
| 477 |
+
"step": 78000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 23.9,
|
| 481 |
+
"learning_rate": 9.333333333333334e-06,
|
| 482 |
+
"loss": 5.3299,
|
| 483 |
+
"step": 79000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 24.21,
|
| 487 |
+
"learning_rate": 8.888888888888888e-06,
|
| 488 |
+
"loss": 5.3232,
|
| 489 |
+
"step": 80000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 24.51,
|
| 493 |
+
"learning_rate": 8.444444444444446e-06,
|
| 494 |
+
"loss": 5.3249,
|
| 495 |
+
"step": 81000
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 24.81,
|
| 499 |
+
"learning_rate": 8.000000000000001e-06,
|
| 500 |
+
"loss": 5.3187,
|
| 501 |
+
"step": 82000
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 25.11,
|
| 505 |
+
"learning_rate": 7.555555555555556e-06,
|
| 506 |
+
"loss": 5.3174,
|
| 507 |
+
"step": 83000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 25.42,
|
| 511 |
+
"learning_rate": 7.111111111111112e-06,
|
| 512 |
+
"loss": 5.3139,
|
| 513 |
+
"step": 84000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 25.72,
|
| 517 |
+
"learning_rate": 6.666666666666667e-06,
|
| 518 |
+
"loss": 5.3119,
|
| 519 |
+
"step": 85000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 26.02,
|
| 523 |
+
"learning_rate": 6.222222222222223e-06,
|
| 524 |
+
"loss": 5.3143,
|
| 525 |
+
"step": 86000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 26.32,
|
| 529 |
+
"learning_rate": 5.777777777777778e-06,
|
| 530 |
+
"loss": 5.3063,
|
| 531 |
+
"step": 87000
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 26.63,
|
| 535 |
+
"learning_rate": 5.333333333333334e-06,
|
| 536 |
+
"loss": 5.3063,
|
| 537 |
+
"step": 88000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 26.93,
|
| 541 |
+
"learning_rate": 4.888888888888889e-06,
|
| 542 |
+
"loss": 5.3096,
|
| 543 |
+
"step": 89000
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 27.23,
|
| 547 |
+
"learning_rate": 4.444444444444444e-06,
|
| 548 |
+
"loss": 5.2996,
|
| 549 |
+
"step": 90000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 27.53,
|
| 553 |
+
"learning_rate": 4.000000000000001e-06,
|
| 554 |
+
"loss": 5.3011,
|
| 555 |
+
"step": 91000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 27.84,
|
| 559 |
+
"learning_rate": 3.555555555555556e-06,
|
| 560 |
+
"loss": 5.3027,
|
| 561 |
+
"step": 92000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 28.14,
|
| 565 |
+
"learning_rate": 3.1111111111111116e-06,
|
| 566 |
+
"loss": 5.3007,
|
| 567 |
+
"step": 93000
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 28.44,
|
| 571 |
+
"learning_rate": 2.666666666666667e-06,
|
| 572 |
+
"loss": 5.297,
|
| 573 |
+
"step": 94000
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 28.74,
|
| 577 |
+
"learning_rate": 2.222222222222222e-06,
|
| 578 |
+
"loss": 5.2979,
|
| 579 |
+
"step": 95000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 29.05,
|
| 583 |
+
"learning_rate": 1.777777777777778e-06,
|
| 584 |
+
"loss": 5.2982,
|
| 585 |
+
"step": 96000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 29.35,
|
| 589 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 590 |
+
"loss": 5.2977,
|
| 591 |
+
"step": 97000
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 29.65,
|
| 595 |
+
"learning_rate": 8.88888888888889e-07,
|
| 596 |
+
"loss": 5.2955,
|
| 597 |
+
"step": 98000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 29.95,
|
| 601 |
+
"learning_rate": 4.444444444444445e-07,
|
| 602 |
+
"loss": 5.2938,
|
| 603 |
+
"step": 99000
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 30.26,
|
| 607 |
+
"learning_rate": 0.0,
|
| 608 |
+
"loss": 5.2937,
|
| 609 |
+
"step": 100000
|
| 610 |
+
}
|
| 611 |
+
],
|
| 612 |
+
"logging_steps": 1000,
|
| 613 |
+
"max_steps": 100000,
|
| 614 |
+
"num_train_epochs": 31,
|
| 615 |
+
"save_steps": 1000,
|
| 616 |
+
"total_flos": 9.396281398647587e+18,
|
| 617 |
+
"trial_name": null,
|
| 618 |
+
"trial_params": null
|
| 619 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fedbea814b8e0698bec087c4313eaab91fc0ee4ef38d1f1581968325185357b
|
| 3 |
+
size 5265
|