Ubuntu
commited on
Commit
•
721cb8e
1
Parent(s):
42e0451
initial upload
Browse files- all_results.json +7 -0
- config.json +27 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +7 -0
- trainer_state.json +810 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"train_runtime": 71825.285,
|
| 4 |
+
"train_samples": 131754,
|
| 5 |
+
"train_samples_per_second": 14.675,
|
| 6 |
+
"train_steps_per_second": 0.917
|
| 7 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/electra-large-discriminator",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ElectraForQuestionAnswering"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"embedding_size": 1024,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 1024,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 4096,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "electra",
|
| 16 |
+
"num_attention_heads": 16,
|
| 17 |
+
"num_hidden_layers": 24,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"summary_activation": "gelu",
|
| 21 |
+
"summary_last_dropout": 0.1,
|
| 22 |
+
"summary_type": "first",
|
| 23 |
+
"summary_use_proj": true,
|
| 24 |
+
"transformers_version": "4.7.0.dev0",
|
| 25 |
+
"type_vocab_size": 2,
|
| 26 |
+
"vocab_size": 30522
|
| 27 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c195349d2519ee6f427667adfe5e20bbf637c3628bb87f922a75eb81492a6d42
|
| 3 |
+
size 1336546901
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "google/electra-large-discriminator"}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"train_runtime": 71825.285,
|
| 4 |
+
"train_samples": 131754,
|
| 5 |
+
"train_samples_per_second": 14.675,
|
| 6 |
+
"train_steps_per_second": 0.917
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,810 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"global_step": 65880,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 2.9772313296903464e-05,
|
| 13 |
+
"loss": 5.8692,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.12,
|
| 18 |
+
"learning_rate": 2.954462659380692e-05,
|
| 19 |
+
"loss": 5.9529,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.18,
|
| 24 |
+
"learning_rate": 2.9316939890710385e-05,
|
| 25 |
+
"loss": 5.9524,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.24,
|
| 30 |
+
"learning_rate": 2.9089253187613845e-05,
|
| 31 |
+
"loss": 5.9525,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.3,
|
| 36 |
+
"learning_rate": 2.8861566484517305e-05,
|
| 37 |
+
"loss": 5.9517,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.36,
|
| 42 |
+
"learning_rate": 2.8633879781420765e-05,
|
| 43 |
+
"loss": 5.9514,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.43,
|
| 48 |
+
"learning_rate": 2.8406193078324225e-05,
|
| 49 |
+
"loss": 5.9514,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.49,
|
| 54 |
+
"learning_rate": 2.817850637522769e-05,
|
| 55 |
+
"loss": 5.9515,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.55,
|
| 60 |
+
"learning_rate": 2.795081967213115e-05,
|
| 61 |
+
"loss": 5.9517,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.61,
|
| 66 |
+
"learning_rate": 2.772313296903461e-05,
|
| 67 |
+
"loss": 5.9513,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.67,
|
| 72 |
+
"learning_rate": 2.749544626593807e-05,
|
| 73 |
+
"loss": 5.9504,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.73,
|
| 78 |
+
"learning_rate": 2.7267759562841533e-05,
|
| 79 |
+
"loss": 5.9516,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.79,
|
| 84 |
+
"learning_rate": 2.704007285974499e-05,
|
| 85 |
+
"loss": 5.951,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.85,
|
| 90 |
+
"learning_rate": 2.6812386156648453e-05,
|
| 91 |
+
"loss": 5.952,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.91,
|
| 96 |
+
"learning_rate": 2.6584699453551913e-05,
|
| 97 |
+
"loss": 5.9522,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.97,
|
| 102 |
+
"learning_rate": 2.6357012750455373e-05,
|
| 103 |
+
"loss": 5.9519,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.03,
|
| 108 |
+
"learning_rate": 2.6129326047358837e-05,
|
| 109 |
+
"loss": 5.9517,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.09,
|
| 114 |
+
"learning_rate": 2.5901639344262294e-05,
|
| 115 |
+
"loss": 5.9518,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.15,
|
| 120 |
+
"learning_rate": 2.5673952641165757e-05,
|
| 121 |
+
"loss": 5.9518,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.21,
|
| 126 |
+
"learning_rate": 2.5446265938069218e-05,
|
| 127 |
+
"loss": 5.9513,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.28,
|
| 132 |
+
"learning_rate": 2.5218579234972678e-05,
|
| 133 |
+
"loss": 5.9512,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.34,
|
| 138 |
+
"learning_rate": 2.4990892531876138e-05,
|
| 139 |
+
"loss": 5.9514,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.4,
|
| 144 |
+
"learning_rate": 2.47632058287796e-05,
|
| 145 |
+
"loss": 5.952,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.46,
|
| 150 |
+
"learning_rate": 2.453551912568306e-05,
|
| 151 |
+
"loss": 5.9509,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.52,
|
| 156 |
+
"learning_rate": 2.4307832422586522e-05,
|
| 157 |
+
"loss": 5.9516,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.58,
|
| 162 |
+
"learning_rate": 2.4080145719489982e-05,
|
| 163 |
+
"loss": 5.9517,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.64,
|
| 168 |
+
"learning_rate": 2.3852459016393442e-05,
|
| 169 |
+
"loss": 5.9501,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.7,
|
| 174 |
+
"learning_rate": 2.3624772313296906e-05,
|
| 175 |
+
"loss": 5.9515,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.76,
|
| 180 |
+
"learning_rate": 2.3397085610200362e-05,
|
| 181 |
+
"loss": 5.9505,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.82,
|
| 186 |
+
"learning_rate": 2.3169398907103826e-05,
|
| 187 |
+
"loss": 5.9507,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.88,
|
| 192 |
+
"learning_rate": 2.294171220400729e-05,
|
| 193 |
+
"loss": 5.9512,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.94,
|
| 198 |
+
"learning_rate": 2.2714025500910746e-05,
|
| 199 |
+
"loss": 5.9511,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.0,
|
| 204 |
+
"learning_rate": 2.248633879781421e-05,
|
| 205 |
+
"loss": 5.9514,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 2.06,
|
| 210 |
+
"learning_rate": 2.2258652094717667e-05,
|
| 211 |
+
"loss": 5.9503,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 2.13,
|
| 216 |
+
"learning_rate": 2.203096539162113e-05,
|
| 217 |
+
"loss": 5.9518,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.19,
|
| 222 |
+
"learning_rate": 2.180327868852459e-05,
|
| 223 |
+
"loss": 5.9514,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.25,
|
| 228 |
+
"learning_rate": 2.157559198542805e-05,
|
| 229 |
+
"loss": 5.9518,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.31,
|
| 234 |
+
"learning_rate": 2.1347905282331514e-05,
|
| 235 |
+
"loss": 5.9517,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.37,
|
| 240 |
+
"learning_rate": 2.1120218579234974e-05,
|
| 241 |
+
"loss": 5.9513,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.43,
|
| 246 |
+
"learning_rate": 2.0892531876138434e-05,
|
| 247 |
+
"loss": 5.9503,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.49,
|
| 252 |
+
"learning_rate": 2.0664845173041894e-05,
|
| 253 |
+
"loss": 5.9513,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.55,
|
| 258 |
+
"learning_rate": 2.0437158469945358e-05,
|
| 259 |
+
"loss": 5.9515,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.61,
|
| 264 |
+
"learning_rate": 2.0209471766848815e-05,
|
| 265 |
+
"loss": 5.9511,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.67,
|
| 270 |
+
"learning_rate": 1.998178506375228e-05,
|
| 271 |
+
"loss": 5.9509,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 2.73,
|
| 276 |
+
"learning_rate": 1.975409836065574e-05,
|
| 277 |
+
"loss": 5.9515,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.79,
|
| 282 |
+
"learning_rate": 1.95264116575592e-05,
|
| 283 |
+
"loss": 5.951,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.85,
|
| 288 |
+
"learning_rate": 1.9298724954462662e-05,
|
| 289 |
+
"loss": 5.9511,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 2.91,
|
| 294 |
+
"learning_rate": 1.907103825136612e-05,
|
| 295 |
+
"loss": 5.9511,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 2.98,
|
| 300 |
+
"learning_rate": 1.8843351548269583e-05,
|
| 301 |
+
"loss": 5.9504,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.04,
|
| 306 |
+
"learning_rate": 1.8615664845173043e-05,
|
| 307 |
+
"loss": 5.9507,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.1,
|
| 312 |
+
"learning_rate": 1.8387978142076503e-05,
|
| 313 |
+
"loss": 5.9506,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.16,
|
| 318 |
+
"learning_rate": 1.8160291438979963e-05,
|
| 319 |
+
"loss": 5.9515,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 3.22,
|
| 324 |
+
"learning_rate": 1.7932604735883423e-05,
|
| 325 |
+
"loss": 5.9509,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 3.28,
|
| 330 |
+
"learning_rate": 1.7704918032786887e-05,
|
| 331 |
+
"loss": 5.9505,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 3.34,
|
| 336 |
+
"learning_rate": 1.7477231329690347e-05,
|
| 337 |
+
"loss": 5.951,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.4,
|
| 342 |
+
"learning_rate": 1.7249544626593807e-05,
|
| 343 |
+
"loss": 5.9512,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.46,
|
| 348 |
+
"learning_rate": 1.7021857923497267e-05,
|
| 349 |
+
"loss": 5.9511,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.52,
|
| 354 |
+
"learning_rate": 1.679417122040073e-05,
|
| 355 |
+
"loss": 5.9507,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.58,
|
| 360 |
+
"learning_rate": 1.6566484517304187e-05,
|
| 361 |
+
"loss": 5.9511,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 3.64,
|
| 366 |
+
"learning_rate": 1.633879781420765e-05,
|
| 367 |
+
"loss": 5.9507,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.7,
|
| 372 |
+
"learning_rate": 1.6111111111111115e-05,
|
| 373 |
+
"loss": 5.9514,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.76,
|
| 378 |
+
"learning_rate": 1.588342440801457e-05,
|
| 379 |
+
"loss": 5.9511,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 3.83,
|
| 384 |
+
"learning_rate": 1.5655737704918035e-05,
|
| 385 |
+
"loss": 5.9509,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 3.89,
|
| 390 |
+
"learning_rate": 1.542805100182149e-05,
|
| 391 |
+
"loss": 5.9508,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 3.95,
|
| 396 |
+
"learning_rate": 1.5200364298724955e-05,
|
| 397 |
+
"loss": 5.9511,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.01,
|
| 402 |
+
"learning_rate": 1.4972677595628415e-05,
|
| 403 |
+
"loss": 5.9518,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.07,
|
| 408 |
+
"learning_rate": 1.4744990892531877e-05,
|
| 409 |
+
"loss": 5.9518,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 4.13,
|
| 414 |
+
"learning_rate": 1.4517304189435337e-05,
|
| 415 |
+
"loss": 5.9512,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.19,
|
| 420 |
+
"learning_rate": 1.4289617486338798e-05,
|
| 421 |
+
"loss": 5.9512,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.25,
|
| 426 |
+
"learning_rate": 1.4061930783242258e-05,
|
| 427 |
+
"loss": 5.9508,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.31,
|
| 432 |
+
"learning_rate": 1.383424408014572e-05,
|
| 433 |
+
"loss": 5.951,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.37,
|
| 438 |
+
"learning_rate": 1.3606557377049181e-05,
|
| 439 |
+
"loss": 5.9514,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 4.43,
|
| 444 |
+
"learning_rate": 1.3378870673952642e-05,
|
| 445 |
+
"loss": 5.9505,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.49,
|
| 450 |
+
"learning_rate": 1.3151183970856102e-05,
|
| 451 |
+
"loss": 5.9508,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 4.55,
|
| 456 |
+
"learning_rate": 1.2923497267759564e-05,
|
| 457 |
+
"loss": 5.9513,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 4.61,
|
| 462 |
+
"learning_rate": 1.2695810564663024e-05,
|
| 463 |
+
"loss": 5.9513,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 4.68,
|
| 468 |
+
"learning_rate": 1.2468123861566484e-05,
|
| 469 |
+
"loss": 5.9503,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 4.74,
|
| 474 |
+
"learning_rate": 1.2240437158469946e-05,
|
| 475 |
+
"loss": 5.9508,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.8,
|
| 480 |
+
"learning_rate": 1.2012750455373408e-05,
|
| 481 |
+
"loss": 5.9504,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.86,
|
| 486 |
+
"learning_rate": 1.1785063752276868e-05,
|
| 487 |
+
"loss": 5.9503,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 4.92,
|
| 492 |
+
"learning_rate": 1.1557377049180328e-05,
|
| 493 |
+
"loss": 5.951,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 4.98,
|
| 498 |
+
"learning_rate": 1.132969034608379e-05,
|
| 499 |
+
"loss": 5.9514,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 5.04,
|
| 504 |
+
"learning_rate": 1.110200364298725e-05,
|
| 505 |
+
"loss": 5.9507,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 5.1,
|
| 510 |
+
"learning_rate": 1.087431693989071e-05,
|
| 511 |
+
"loss": 5.9505,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.16,
|
| 516 |
+
"learning_rate": 1.064663023679417e-05,
|
| 517 |
+
"loss": 5.9509,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 5.22,
|
| 522 |
+
"learning_rate": 1.0418943533697632e-05,
|
| 523 |
+
"loss": 5.9506,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 5.28,
|
| 528 |
+
"learning_rate": 1.0191256830601094e-05,
|
| 529 |
+
"loss": 5.9511,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.34,
|
| 534 |
+
"learning_rate": 9.963570127504554e-06,
|
| 535 |
+
"loss": 5.9505,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 5.4,
|
| 540 |
+
"learning_rate": 9.735883424408014e-06,
|
| 541 |
+
"loss": 5.951,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 5.46,
|
| 546 |
+
"learning_rate": 9.508196721311476e-06,
|
| 547 |
+
"loss": 5.9507,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 5.53,
|
| 552 |
+
"learning_rate": 9.280510018214936e-06,
|
| 553 |
+
"loss": 5.9509,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.59,
|
| 558 |
+
"learning_rate": 9.052823315118397e-06,
|
| 559 |
+
"loss": 5.9509,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 5.65,
|
| 564 |
+
"learning_rate": 8.825136612021857e-06,
|
| 565 |
+
"loss": 5.9514,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 5.71,
|
| 570 |
+
"learning_rate": 8.59744990892532e-06,
|
| 571 |
+
"loss": 5.9506,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 5.77,
|
| 576 |
+
"learning_rate": 8.36976320582878e-06,
|
| 577 |
+
"loss": 5.9511,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 5.83,
|
| 582 |
+
"learning_rate": 8.14207650273224e-06,
|
| 583 |
+
"loss": 5.9514,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 5.89,
|
| 588 |
+
"learning_rate": 7.914389799635702e-06,
|
| 589 |
+
"loss": 5.9506,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 5.95,
|
| 594 |
+
"learning_rate": 7.686703096539163e-06,
|
| 595 |
+
"loss": 5.9503,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 6.01,
|
| 600 |
+
"learning_rate": 7.459016393442623e-06,
|
| 601 |
+
"loss": 5.9505,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 6.07,
|
| 606 |
+
"learning_rate": 7.2313296903460846e-06,
|
| 607 |
+
"loss": 5.9509,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 6.13,
|
| 612 |
+
"learning_rate": 7.003642987249545e-06,
|
| 613 |
+
"loss": 5.9507,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 6.19,
|
| 618 |
+
"learning_rate": 6.775956284153006e-06,
|
| 619 |
+
"loss": 5.9506,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 6.25,
|
| 624 |
+
"learning_rate": 6.548269581056467e-06,
|
| 625 |
+
"loss": 5.9508,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 6.31,
|
| 630 |
+
"learning_rate": 6.320582877959928e-06,
|
| 631 |
+
"loss": 5.9508,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 6.38,
|
| 636 |
+
"learning_rate": 6.092896174863388e-06,
|
| 637 |
+
"loss": 5.9508,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 6.44,
|
| 642 |
+
"learning_rate": 5.865209471766849e-06,
|
| 643 |
+
"loss": 5.9509,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 6.5,
|
| 648 |
+
"learning_rate": 5.63752276867031e-06,
|
| 649 |
+
"loss": 5.951,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 6.56,
|
| 654 |
+
"learning_rate": 5.409836065573771e-06,
|
| 655 |
+
"loss": 5.9509,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 6.62,
|
| 660 |
+
"learning_rate": 5.182149362477231e-06,
|
| 661 |
+
"loss": 5.9505,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 6.68,
|
| 666 |
+
"learning_rate": 4.954462659380692e-06,
|
| 667 |
+
"loss": 5.9511,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 6.74,
|
| 672 |
+
"learning_rate": 4.726775956284153e-06,
|
| 673 |
+
"loss": 5.951,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 6.8,
|
| 678 |
+
"learning_rate": 4.499089253187614e-06,
|
| 679 |
+
"loss": 5.9504,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 6.86,
|
| 684 |
+
"learning_rate": 4.271402550091074e-06,
|
| 685 |
+
"loss": 5.9507,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 6.92,
|
| 690 |
+
"learning_rate": 4.043715846994535e-06,
|
| 691 |
+
"loss": 5.9507,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 6.98,
|
| 696 |
+
"learning_rate": 3.816029143897996e-06,
|
| 697 |
+
"loss": 5.9506,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 7.04,
|
| 702 |
+
"learning_rate": 3.5883424408014573e-06,
|
| 703 |
+
"loss": 5.9504,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 7.1,
|
| 708 |
+
"learning_rate": 3.3606557377049183e-06,
|
| 709 |
+
"loss": 5.9508,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 7.16,
|
| 714 |
+
"learning_rate": 3.132969034608379e-06,
|
| 715 |
+
"loss": 5.9505,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 7.23,
|
| 720 |
+
"learning_rate": 2.90528233151184e-06,
|
| 721 |
+
"loss": 5.9507,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 7.29,
|
| 726 |
+
"learning_rate": 2.6775956284153005e-06,
|
| 727 |
+
"loss": 5.9513,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 7.35,
|
| 732 |
+
"learning_rate": 2.4499089253187615e-06,
|
| 733 |
+
"loss": 5.9508,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 7.41,
|
| 738 |
+
"learning_rate": 2.222222222222222e-06,
|
| 739 |
+
"loss": 5.9508,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 7.47,
|
| 744 |
+
"learning_rate": 1.994535519125683e-06,
|
| 745 |
+
"loss": 5.9507,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 7.53,
|
| 750 |
+
"learning_rate": 1.766848816029144e-06,
|
| 751 |
+
"loss": 5.9506,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 7.59,
|
| 756 |
+
"learning_rate": 1.5391621129326049e-06,
|
| 757 |
+
"loss": 5.951,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 7.65,
|
| 762 |
+
"learning_rate": 1.3114754098360657e-06,
|
| 763 |
+
"loss": 5.9504,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 7.71,
|
| 768 |
+
"learning_rate": 1.0837887067395265e-06,
|
| 769 |
+
"loss": 5.9508,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 7.77,
|
| 774 |
+
"learning_rate": 8.561020036429873e-07,
|
| 775 |
+
"loss": 5.9508,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 7.83,
|
| 780 |
+
"learning_rate": 6.284153005464482e-07,
|
| 781 |
+
"loss": 5.9509,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 7.89,
|
| 786 |
+
"learning_rate": 4.0072859744990897e-07,
|
| 787 |
+
"loss": 5.9506,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 7.95,
|
| 792 |
+
"learning_rate": 1.7304189435336977e-07,
|
| 793 |
+
"loss": 5.9511,
|
| 794 |
+
"step": 65500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 8.0,
|
| 798 |
+
"step": 65880,
|
| 799 |
+
"total_flos": 8.113569971306496e+17,
|
| 800 |
+
"train_runtime": 71825.285,
|
| 801 |
+
"train_samples_per_second": 14.675,
|
| 802 |
+
"train_steps_per_second": 0.917
|
| 803 |
+
}
|
| 804 |
+
],
|
| 805 |
+
"max_steps": 65880,
|
| 806 |
+
"num_train_epochs": 8,
|
| 807 |
+
"total_flos": 8.113569971306496e+17,
|
| 808 |
+
"trial_name": null,
|
| 809 |
+
"trial_params": null
|
| 810 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a857016845059a1795072bdcbe12f678083a94676ae63d913a87f2734a9b698e
|
| 3 |
+
size 2479
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|