mathislucka
commited on
Commit
·
eede0a7
1
Parent(s):
33ff9dc
add model
Browse files- added_tokens.json +3 -0
- config.json +41 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +9 -0
- spm.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +17 -0
- trainer_state.json +901 -0
- training_args.bin +3 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-v3-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"id2label": {
|
| 11 |
+
"0": "LABEL_0"
|
| 12 |
+
},
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"label2id": {
|
| 16 |
+
"LABEL_0": 0
|
| 17 |
+
},
|
| 18 |
+
"layer_norm_eps": 1e-07,
|
| 19 |
+
"max_position_embeddings": 512,
|
| 20 |
+
"max_relative_positions": -1,
|
| 21 |
+
"model_type": "deberta-v2",
|
| 22 |
+
"norm_rel_ebd": "layer_norm",
|
| 23 |
+
"num_attention_heads": 12,
|
| 24 |
+
"num_hidden_layers": 12,
|
| 25 |
+
"pad_token_id": 0,
|
| 26 |
+
"pooler_dropout": 0,
|
| 27 |
+
"pooler_hidden_act": "gelu",
|
| 28 |
+
"pooler_hidden_size": 768,
|
| 29 |
+
"pos_att_type": [
|
| 30 |
+
"p2c",
|
| 31 |
+
"c2p"
|
| 32 |
+
],
|
| 33 |
+
"position_biased_input": false,
|
| 34 |
+
"position_buckets": 256,
|
| 35 |
+
"relative_attention": true,
|
| 36 |
+
"share_att_key": true,
|
| 37 |
+
"torch_dtype": "float32",
|
| 38 |
+
"transformers_version": "4.25.1",
|
| 39 |
+
"type_vocab_size": 0,
|
| 40 |
+
"vocab_size": 128100
|
| 41 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:694b621d30efc5917a05b1591f14c19c5e212323c7bcfebfac30755e7d4429b5
|
| 3 |
+
size 1475500825
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:60e9d844431847fe1c1ea3345312d938ab475d20d6d1642f6f2d8a5e21e5520a
|
| 3 |
+
size 737763883
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:25acd4e9247c1087977f0a821a5bf0f46fe34dfbf5f6088627815f4eaaa090f9
|
| 3 |
+
size 14503
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21f8e3585e2d6dfc0020bade21b082af726625f07cedcda299c048f35ca73a3b
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8afe8badaea6b7702b669b87fdceb0392194d091f33d39689e4b65f02e72d16
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "[SEP]",
|
| 5 |
+
"mask_token": "[MASK]",
|
| 6 |
+
"pad_token": "[PAD]",
|
| 7 |
+
"sep_token": "[SEP]",
|
| 8 |
+
"unk_token": "[UNK]"
|
| 9 |
+
}
|
spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"eos_token": "[SEP]",
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 8 |
+
"name_or_path": "microsoft/deberta-v3-base",
|
| 9 |
+
"pad_token": "[PAD]",
|
| 10 |
+
"sep_token": "[SEP]",
|
| 11 |
+
"sp_model_kwargs": {},
|
| 12 |
+
"special_tokens_map_file": null,
|
| 13 |
+
"split_by_punct": false,
|
| 14 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 15 |
+
"unk_token": "[UNK]",
|
| 16 |
+
"vocab_type": "spm"
|
| 17 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,901 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9943937063217163,
|
| 3 |
+
"best_model_checkpoint": "./models/deberta-v3-base-all-v4/checkpoint-73069",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 73069,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 6.952483269238665e-06,
|
| 13 |
+
"loss": 0.8446,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 6.9045833390357056e-06,
|
| 19 |
+
"loss": 0.4856,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 6.856683408832747e-06,
|
| 25 |
+
"loss": 0.4123,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 6.808783478629788e-06,
|
| 31 |
+
"loss": 0.3721,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.03,
|
| 36 |
+
"learning_rate": 6.760883548426829e-06,
|
| 37 |
+
"loss": 0.3444,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04,
|
| 42 |
+
"learning_rate": 6.713079418084276e-06,
|
| 43 |
+
"loss": 0.3378,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"learning_rate": 6.665179487881318e-06,
|
| 49 |
+
"loss": 0.3262,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.05,
|
| 54 |
+
"learning_rate": 6.617279557678359e-06,
|
| 55 |
+
"loss": 0.3156,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.06,
|
| 60 |
+
"learning_rate": 6.5693796274754e-06,
|
| 61 |
+
"loss": 0.3139,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.07,
|
| 66 |
+
"learning_rate": 6.521479697272441e-06,
|
| 67 |
+
"loss": 0.2942,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.08,
|
| 72 |
+
"learning_rate": 6.473579767069482e-06,
|
| 73 |
+
"loss": 0.3003,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.08,
|
| 78 |
+
"learning_rate": 6.4256798368665235e-06,
|
| 79 |
+
"loss": 0.2993,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09,
|
| 84 |
+
"learning_rate": 6.377779906663564e-06,
|
| 85 |
+
"loss": 0.2774,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1,
|
| 90 |
+
"learning_rate": 6.329975776321012e-06,
|
| 91 |
+
"loss": 0.2698,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.1,
|
| 96 |
+
"learning_rate": 6.282075846118053e-06,
|
| 97 |
+
"loss": 0.2782,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.11,
|
| 102 |
+
"learning_rate": 6.234175915915094e-06,
|
| 103 |
+
"loss": 0.2947,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.12,
|
| 108 |
+
"learning_rate": 6.186275985712135e-06,
|
| 109 |
+
"loss": 0.2804,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.12,
|
| 114 |
+
"learning_rate": 6.138471855369582e-06,
|
| 115 |
+
"loss": 0.2591,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.13,
|
| 120 |
+
"learning_rate": 6.0905719251666235e-06,
|
| 121 |
+
"loss": 0.2836,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14,
|
| 126 |
+
"learning_rate": 6.042671994963664e-06,
|
| 127 |
+
"loss": 0.2657,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.14,
|
| 132 |
+
"learning_rate": 5.994772064760705e-06,
|
| 133 |
+
"loss": 0.2613,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.15,
|
| 138 |
+
"learning_rate": 5.946872134557747e-06,
|
| 139 |
+
"loss": 0.263,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.16,
|
| 144 |
+
"learning_rate": 5.8991638040756e-06,
|
| 145 |
+
"loss": 0.2522,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.16,
|
| 150 |
+
"learning_rate": 5.851263873872641e-06,
|
| 151 |
+
"loss": 0.2656,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.17,
|
| 156 |
+
"learning_rate": 5.803363943669682e-06,
|
| 157 |
+
"loss": 0.2611,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.18,
|
| 162 |
+
"learning_rate": 5.755464013466723e-06,
|
| 163 |
+
"loss": 0.2613,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.18,
|
| 168 |
+
"learning_rate": 5.707564083263764e-06,
|
| 169 |
+
"loss": 0.2592,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.19,
|
| 174 |
+
"learning_rate": 5.659664153060805e-06,
|
| 175 |
+
"loss": 0.2625,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.2,
|
| 180 |
+
"learning_rate": 5.611860022718252e-06,
|
| 181 |
+
"loss": 0.2774,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.21,
|
| 186 |
+
"learning_rate": 5.5639600925152934e-06,
|
| 187 |
+
"loss": 0.2428,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.21,
|
| 192 |
+
"learning_rate": 5.5160601623123355e-06,
|
| 193 |
+
"loss": 0.259,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.22,
|
| 198 |
+
"learning_rate": 5.468160232109376e-06,
|
| 199 |
+
"loss": 0.242,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.23,
|
| 204 |
+
"learning_rate": 5.420260301906417e-06,
|
| 205 |
+
"loss": 0.2703,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.23,
|
| 210 |
+
"learning_rate": 5.372456171563864e-06,
|
| 211 |
+
"loss": 0.2466,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.24,
|
| 216 |
+
"learning_rate": 5.324556241360905e-06,
|
| 217 |
+
"loss": 0.2475,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.25,
|
| 222 |
+
"learning_rate": 5.2766563111579465e-06,
|
| 223 |
+
"loss": 0.2443,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.25,
|
| 228 |
+
"learning_rate": 5.228756380954988e-06,
|
| 229 |
+
"loss": 0.253,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.26,
|
| 234 |
+
"learning_rate": 5.180856450752029e-06,
|
| 235 |
+
"loss": 0.2561,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.27,
|
| 240 |
+
"learning_rate": 5.13295652054907e-06,
|
| 241 |
+
"loss": 0.2422,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.27,
|
| 246 |
+
"learning_rate": 5.085056590346111e-06,
|
| 247 |
+
"loss": 0.2437,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.28,
|
| 252 |
+
"learning_rate": 5.037252460003558e-06,
|
| 253 |
+
"loss": 0.2504,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.29,
|
| 258 |
+
"learning_rate": 4.9893525298005995e-06,
|
| 259 |
+
"loss": 0.2238,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.29,
|
| 264 |
+
"learning_rate": 4.941452599597641e-06,
|
| 265 |
+
"loss": 0.2154,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.3,
|
| 270 |
+
"learning_rate": 4.893552669394681e-06,
|
| 271 |
+
"loss": 0.2391,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.31,
|
| 276 |
+
"learning_rate": 4.845652739191722e-06,
|
| 277 |
+
"loss": 0.2445,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.31,
|
| 282 |
+
"learning_rate": 4.797752808988764e-06,
|
| 283 |
+
"loss": 0.2496,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.32,
|
| 288 |
+
"learning_rate": 4.749948678646211e-06,
|
| 289 |
+
"loss": 0.2444,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.33,
|
| 294 |
+
"learning_rate": 4.702048748443253e-06,
|
| 295 |
+
"loss": 0.2332,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.34,
|
| 300 |
+
"learning_rate": 4.654148818240294e-06,
|
| 301 |
+
"loss": 0.2163,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.34,
|
| 306 |
+
"learning_rate": 4.606248888037334e-06,
|
| 307 |
+
"loss": 0.234,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.35,
|
| 312 |
+
"learning_rate": 4.558444757694781e-06,
|
| 313 |
+
"loss": 0.2451,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.36,
|
| 318 |
+
"learning_rate": 4.510544827491822e-06,
|
| 319 |
+
"loss": 0.2478,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.36,
|
| 324 |
+
"learning_rate": 4.462644897288864e-06,
|
| 325 |
+
"loss": 0.2378,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.37,
|
| 330 |
+
"learning_rate": 4.414744967085906e-06,
|
| 331 |
+
"loss": 0.2292,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.38,
|
| 336 |
+
"learning_rate": 4.366940836743353e-06,
|
| 337 |
+
"loss": 0.2391,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.38,
|
| 342 |
+
"learning_rate": 4.319040906540393e-06,
|
| 343 |
+
"loss": 0.218,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.39,
|
| 348 |
+
"learning_rate": 4.271140976337434e-06,
|
| 349 |
+
"loss": 0.2338,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.4,
|
| 354 |
+
"learning_rate": 4.2232410461344755e-06,
|
| 355 |
+
"loss": 0.23,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.4,
|
| 360 |
+
"learning_rate": 4.175341115931517e-06,
|
| 361 |
+
"loss": 0.253,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.41,
|
| 366 |
+
"learning_rate": 4.127441185728558e-06,
|
| 367 |
+
"loss": 0.2252,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.42,
|
| 372 |
+
"learning_rate": 4.079541255525599e-06,
|
| 373 |
+
"loss": 0.2029,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.42,
|
| 378 |
+
"learning_rate": 4.03164132532264e-06,
|
| 379 |
+
"loss": 0.2322,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.43,
|
| 384 |
+
"learning_rate": 3.9837413951196815e-06,
|
| 385 |
+
"loss": 0.235,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.44,
|
| 390 |
+
"learning_rate": 3.9359372647771285e-06,
|
| 391 |
+
"loss": 0.2329,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.44,
|
| 396 |
+
"learning_rate": 3.88803733457417e-06,
|
| 397 |
+
"loss": 0.2244,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.45,
|
| 402 |
+
"learning_rate": 3.840137404371211e-06,
|
| 403 |
+
"loss": 0.2217,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.46,
|
| 408 |
+
"learning_rate": 3.7922374741682517e-06,
|
| 409 |
+
"loss": 0.2155,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.47,
|
| 414 |
+
"learning_rate": 3.7444333438256988e-06,
|
| 415 |
+
"loss": 0.2253,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.47,
|
| 420 |
+
"learning_rate": 3.69653341362274e-06,
|
| 421 |
+
"loss": 0.2218,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.48,
|
| 426 |
+
"learning_rate": 3.648633483419781e-06,
|
| 427 |
+
"loss": 0.2258,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.49,
|
| 432 |
+
"learning_rate": 3.6007335532168224e-06,
|
| 433 |
+
"loss": 0.2336,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.49,
|
| 438 |
+
"learning_rate": 3.552833623013864e-06,
|
| 439 |
+
"loss": 0.2199,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.5,
|
| 444 |
+
"learning_rate": 3.5049336928109044e-06,
|
| 445 |
+
"loss": 0.2278,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.51,
|
| 450 |
+
"learning_rate": 3.457129562468352e-06,
|
| 451 |
+
"loss": 0.211,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.51,
|
| 456 |
+
"learning_rate": 3.409325432125799e-06,
|
| 457 |
+
"loss": 0.2304,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.52,
|
| 462 |
+
"learning_rate": 3.36142550192284e-06,
|
| 463 |
+
"loss": 0.2323,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.53,
|
| 468 |
+
"learning_rate": 3.3135255717198813e-06,
|
| 469 |
+
"loss": 0.2188,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.53,
|
| 474 |
+
"learning_rate": 3.265625641516922e-06,
|
| 475 |
+
"loss": 0.2224,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.54,
|
| 480 |
+
"learning_rate": 3.2177257113139633e-06,
|
| 481 |
+
"loss": 0.2233,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.55,
|
| 486 |
+
"learning_rate": 3.169825781111005e-06,
|
| 487 |
+
"loss": 0.2133,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.55,
|
| 492 |
+
"learning_rate": 3.1219258509080457e-06,
|
| 493 |
+
"loss": 0.2073,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.56,
|
| 498 |
+
"learning_rate": 3.074025920705087e-06,
|
| 499 |
+
"loss": 0.221,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.57,
|
| 504 |
+
"learning_rate": 3.026125990502128e-06,
|
| 505 |
+
"loss": 0.2268,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.57,
|
| 510 |
+
"learning_rate": 2.9782260602991693e-06,
|
| 511 |
+
"loss": 0.2129,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.58,
|
| 516 |
+
"learning_rate": 2.9304219299566163e-06,
|
| 517 |
+
"loss": 0.2144,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.59,
|
| 522 |
+
"learning_rate": 2.8825219997536575e-06,
|
| 523 |
+
"loss": 0.2192,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.6,
|
| 528 |
+
"learning_rate": 2.8346220695506987e-06,
|
| 529 |
+
"loss": 0.2211,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.6,
|
| 534 |
+
"learning_rate": 2.7868179392081458e-06,
|
| 535 |
+
"loss": 0.2356,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.61,
|
| 540 |
+
"learning_rate": 2.738918009005187e-06,
|
| 541 |
+
"loss": 0.2254,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.62,
|
| 546 |
+
"learning_rate": 2.6910180788022277e-06,
|
| 547 |
+
"loss": 0.2258,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.62,
|
| 552 |
+
"learning_rate": 2.6431181485992694e-06,
|
| 553 |
+
"loss": 0.222,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.63,
|
| 558 |
+
"learning_rate": 2.59521821839631e-06,
|
| 559 |
+
"loss": 0.2096,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.64,
|
| 564 |
+
"learning_rate": 2.5473182881933514e-06,
|
| 565 |
+
"loss": 0.2082,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.64,
|
| 570 |
+
"learning_rate": 2.4994183579903926e-06,
|
| 571 |
+
"loss": 0.2008,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.65,
|
| 576 |
+
"learning_rate": 2.45161422764784e-06,
|
| 577 |
+
"loss": 0.2216,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.66,
|
| 582 |
+
"learning_rate": 2.403714297444881e-06,
|
| 583 |
+
"loss": 0.2275,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.66,
|
| 588 |
+
"learning_rate": 2.355814367241922e-06,
|
| 589 |
+
"loss": 0.2146,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.67,
|
| 594 |
+
"learning_rate": 2.307914437038963e-06,
|
| 595 |
+
"loss": 0.2075,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.68,
|
| 600 |
+
"learning_rate": 2.2600145068360044e-06,
|
| 601 |
+
"loss": 0.2183,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.68,
|
| 606 |
+
"learning_rate": 2.2121145766330456e-06,
|
| 607 |
+
"loss": 0.2179,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.69,
|
| 612 |
+
"learning_rate": 2.1642146464300864e-06,
|
| 613 |
+
"loss": 0.2062,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.7,
|
| 618 |
+
"learning_rate": 2.116314716227128e-06,
|
| 619 |
+
"loss": 0.2115,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.7,
|
| 624 |
+
"learning_rate": 2.068510585884575e-06,
|
| 625 |
+
"loss": 0.2176,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.71,
|
| 630 |
+
"learning_rate": 2.020610655681616e-06,
|
| 631 |
+
"loss": 0.2205,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.72,
|
| 636 |
+
"learning_rate": 1.972710725478657e-06,
|
| 637 |
+
"loss": 0.2213,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.73,
|
| 642 |
+
"learning_rate": 1.9248107952756983e-06,
|
| 643 |
+
"loss": 0.201,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.73,
|
| 648 |
+
"learning_rate": 1.8769108650727395e-06,
|
| 649 |
+
"loss": 0.2026,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.74,
|
| 654 |
+
"learning_rate": 1.8290109348697807e-06,
|
| 655 |
+
"loss": 0.2089,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.75,
|
| 660 |
+
"learning_rate": 1.7811110046668217e-06,
|
| 661 |
+
"loss": 0.2035,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.75,
|
| 666 |
+
"learning_rate": 1.733211074463863e-06,
|
| 667 |
+
"loss": 0.2035,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.76,
|
| 672 |
+
"learning_rate": 1.6853111442609041e-06,
|
| 673 |
+
"loss": 0.2055,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.77,
|
| 678 |
+
"learning_rate": 1.6375070139183511e-06,
|
| 679 |
+
"loss": 0.2278,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.77,
|
| 684 |
+
"learning_rate": 1.5897028835757984e-06,
|
| 685 |
+
"loss": 0.2108,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.78,
|
| 690 |
+
"learning_rate": 1.5418029533728394e-06,
|
| 691 |
+
"loss": 0.2169,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.79,
|
| 696 |
+
"learning_rate": 1.4939030231698806e-06,
|
| 697 |
+
"loss": 0.2157,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.79,
|
| 702 |
+
"learning_rate": 1.4460030929669216e-06,
|
| 703 |
+
"loss": 0.2313,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.8,
|
| 708 |
+
"learning_rate": 1.3981031627639628e-06,
|
| 709 |
+
"loss": 0.2176,
|
| 710 |
+
"step": 58500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.81,
|
| 714 |
+
"learning_rate": 1.3503948322818158e-06,
|
| 715 |
+
"loss": 0.2175,
|
| 716 |
+
"step": 59000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.81,
|
| 720 |
+
"learning_rate": 1.3024949020788568e-06,
|
| 721 |
+
"loss": 0.2199,
|
| 722 |
+
"step": 59500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.82,
|
| 726 |
+
"learning_rate": 1.2545949718758982e-06,
|
| 727 |
+
"loss": 0.2193,
|
| 728 |
+
"step": 60000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.83,
|
| 732 |
+
"learning_rate": 1.2066950416729392e-06,
|
| 733 |
+
"loss": 0.2095,
|
| 734 |
+
"step": 60500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.83,
|
| 738 |
+
"learning_rate": 1.1587951114699804e-06,
|
| 739 |
+
"loss": 0.2035,
|
| 740 |
+
"step": 61000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.84,
|
| 744 |
+
"learning_rate": 1.1108951812670216e-06,
|
| 745 |
+
"loss": 0.2105,
|
| 746 |
+
"step": 61500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.85,
|
| 750 |
+
"learning_rate": 1.0629952510640626e-06,
|
| 751 |
+
"loss": 0.2146,
|
| 752 |
+
"step": 62000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.86,
|
| 756 |
+
"learning_rate": 1.0150953208611039e-06,
|
| 757 |
+
"loss": 0.184,
|
| 758 |
+
"step": 62500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.86,
|
| 762 |
+
"learning_rate": 9.672911905185509e-07,
|
| 763 |
+
"loss": 0.2144,
|
| 764 |
+
"step": 63000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.87,
|
| 768 |
+
"learning_rate": 9.194870601759979e-07,
|
| 769 |
+
"loss": 0.2116,
|
| 770 |
+
"step": 63500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.88,
|
| 774 |
+
"learning_rate": 8.716829298334451e-07,
|
| 775 |
+
"loss": 0.2216,
|
| 776 |
+
"step": 64000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.88,
|
| 780 |
+
"learning_rate": 8.237829996304862e-07,
|
| 781 |
+
"loss": 0.2055,
|
| 782 |
+
"step": 64500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.89,
|
| 786 |
+
"learning_rate": 7.758830694275273e-07,
|
| 787 |
+
"loss": 0.2006,
|
| 788 |
+
"step": 65000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.9,
|
| 792 |
+
"learning_rate": 7.279831392245685e-07,
|
| 793 |
+
"loss": 0.2044,
|
| 794 |
+
"step": 65500
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9,
|
| 798 |
+
"learning_rate": 6.800832090216097e-07,
|
| 799 |
+
"loss": 0.2116,
|
| 800 |
+
"step": 66000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.91,
|
| 804 |
+
"learning_rate": 6.321832788186509e-07,
|
| 805 |
+
"loss": 0.223,
|
| 806 |
+
"step": 66500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.92,
|
| 810 |
+
"learning_rate": 5.842833486156921e-07,
|
| 811 |
+
"loss": 0.2058,
|
| 812 |
+
"step": 67000
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.92,
|
| 816 |
+
"learning_rate": 5.363834184127332e-07,
|
| 817 |
+
"loss": 0.2205,
|
| 818 |
+
"step": 67500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.93,
|
| 822 |
+
"learning_rate": 4.884834882097743e-07,
|
| 823 |
+
"loss": 0.1975,
|
| 824 |
+
"step": 68000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.94,
|
| 828 |
+
"learning_rate": 4.4067935786722134e-07,
|
| 829 |
+
"loss": 0.199,
|
| 830 |
+
"step": 68500
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.94,
|
| 834 |
+
"learning_rate": 3.9277942766426255e-07,
|
| 835 |
+
"loss": 0.1955,
|
| 836 |
+
"step": 69000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.95,
|
| 840 |
+
"learning_rate": 3.448794974613037e-07,
|
| 841 |
+
"loss": 0.2121,
|
| 842 |
+
"step": 69500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.96,
|
| 846 |
+
"learning_rate": 2.9697956725834486e-07,
|
| 847 |
+
"loss": 0.2023,
|
| 848 |
+
"step": 70000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.96,
|
| 852 |
+
"learning_rate": 2.4907963705538596e-07,
|
| 853 |
+
"loss": 0.2043,
|
| 854 |
+
"step": 70500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.97,
|
| 858 |
+
"learning_rate": 2.0117970685242714e-07,
|
| 859 |
+
"loss": 0.2183,
|
| 860 |
+
"step": 71000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.98,
|
| 864 |
+
"learning_rate": 1.5337557650987422e-07,
|
| 865 |
+
"loss": 0.2032,
|
| 866 |
+
"step": 71500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.99,
|
| 870 |
+
"learning_rate": 1.0547564630691538e-07,
|
| 871 |
+
"loss": 0.2127,
|
| 872 |
+
"step": 72000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.99,
|
| 876 |
+
"learning_rate": 5.757571610395654e-08,
|
| 877 |
+
"loss": 0.189,
|
| 878 |
+
"step": 72500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0,
|
| 882 |
+
"learning_rate": 9.675785900997687e-09,
|
| 883 |
+
"loss": 0.2012,
|
| 884 |
+
"step": 73000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.0,
|
| 888 |
+
"eval_loss": 0.9943937063217163,
|
| 889 |
+
"eval_rmse": 0.997192919254303,
|
| 890 |
+
"eval_runtime": 2.2905,
|
| 891 |
+
"eval_samples_per_second": 668.41,
|
| 892 |
+
"eval_steps_per_second": 20.956,
|
| 893 |
+
"step": 73069
|
| 894 |
+
}
|
| 895 |
+
],
|
| 896 |
+
"max_steps": 73069,
|
| 897 |
+
"num_train_epochs": 1,
|
| 898 |
+
"total_flos": 7.295661144915454e+16,
|
| 899 |
+
"trial_name": null,
|
| 900 |
+
"trial_params": null
|
| 901 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81cdea6f5fa2ba3cfabd820b579a3e372b035d31aeeef5380e85af345187a58b
|
| 3 |
+
size 3439
|