traintogpb
commited on
Commit
•
762b8f3
1
Parent(s):
1b0fdc6
Initial commit
Browse files- config.json +29 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +3052 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "klue/roberta-large",
|
3 |
+
"architectures": [
|
4 |
+
"RobertaForMaskedLM"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"gradient_checkpointing": false,
|
11 |
+
"hidden_act": "gelu",
|
12 |
+
"hidden_dropout_prob": 0.1,
|
13 |
+
"hidden_size": 1024,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 4096,
|
16 |
+
"layer_norm_eps": 1e-05,
|
17 |
+
"max_position_embeddings": 514,
|
18 |
+
"model_type": "roberta",
|
19 |
+
"num_attention_heads": 16,
|
20 |
+
"num_hidden_layers": 24,
|
21 |
+
"pad_token_id": 1,
|
22 |
+
"position_embedding_type": "absolute",
|
23 |
+
"tokenizer_class": "BertTokenizer",
|
24 |
+
"torch_dtype": "float32",
|
25 |
+
"transformers_version": "4.29.2",
|
26 |
+
"type_vocab_size": 1,
|
27 |
+
"use_cache": true,
|
28 |
+
"vocab_size": 32000
|
29 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2c223b4f255a90124a53c1d2fcd2d2f38a2768ebaaeb40610294f2965171104f
|
3 |
+
size 2693754949
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1b0f9ea76ddba43a5a03d13263624ba6197e2d9491f4c4b58d41fb72e27666df
|
3 |
+
size 1346902073
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:86c58440463ac6ca155cb312ea88bf17497fc47c044b9dbe95c311e62e199a4b
|
3 |
+
size 14575
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a1ffa872c0fa2fb4515a09b85e729bab07424c753932d18a586d5b0b4d7a8d37
|
3 |
+
size 627
|
trainer_state.json
ADDED
@@ -0,0 +1,3052 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7065624395501147,
|
3 |
+
"best_model_checkpoint": "../pretrained_models/klue_roberta_large/checkpoint-44000",
|
4 |
+
"epoch": 4.975686984055185,
|
5 |
+
"global_step": 44000,
|
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": 1.0000000000000002e-06,
|
13 |
+
"loss": 4.7552,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 2.0000000000000003e-06,
|
19 |
+
"loss": 2.2752,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.03,
|
24 |
+
"learning_rate": 3e-06,
|
25 |
+
"loss": 1.9549,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.05,
|
30 |
+
"learning_rate": 4.000000000000001e-06,
|
31 |
+
"loss": 1.8509,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.06,
|
36 |
+
"learning_rate": 5e-06,
|
37 |
+
"loss": 1.8299,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.07,
|
42 |
+
"learning_rate": 6e-06,
|
43 |
+
"loss": 1.753,
|
44 |
+
"step": 600
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.08,
|
48 |
+
"learning_rate": 7.000000000000001e-06,
|
49 |
+
"loss": 1.7313,
|
50 |
+
"step": 700
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.09,
|
54 |
+
"learning_rate": 8.000000000000001e-06,
|
55 |
+
"loss": 1.6829,
|
56 |
+
"step": 800
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.1,
|
60 |
+
"learning_rate": 9e-06,
|
61 |
+
"loss": 1.6651,
|
62 |
+
"step": 900
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.11,
|
66 |
+
"learning_rate": 1e-05,
|
67 |
+
"loss": 1.655,
|
68 |
+
"step": 1000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.11,
|
72 |
+
"eval_loss": 1.5775465965270996,
|
73 |
+
"eval_masked_accuracy": 0.6741395380950737,
|
74 |
+
"eval_runtime": 331.6052,
|
75 |
+
"eval_samples_per_second": 47.409,
|
76 |
+
"eval_steps_per_second": 2.964,
|
77 |
+
"step": 1000
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 0.12,
|
81 |
+
"learning_rate": 1.1000000000000001e-05,
|
82 |
+
"loss": 1.6543,
|
83 |
+
"step": 1100
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 0.14,
|
87 |
+
"learning_rate": 1.2e-05,
|
88 |
+
"loss": 1.656,
|
89 |
+
"step": 1200
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.15,
|
93 |
+
"learning_rate": 1.3000000000000001e-05,
|
94 |
+
"loss": 1.6726,
|
95 |
+
"step": 1300
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.16,
|
99 |
+
"learning_rate": 1.4000000000000001e-05,
|
100 |
+
"loss": 1.6471,
|
101 |
+
"step": 1400
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 0.17,
|
105 |
+
"learning_rate": 1.5e-05,
|
106 |
+
"loss": 1.6679,
|
107 |
+
"step": 1500
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.18,
|
111 |
+
"learning_rate": 1.6000000000000003e-05,
|
112 |
+
"loss": 1.627,
|
113 |
+
"step": 1600
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 0.19,
|
117 |
+
"learning_rate": 1.7000000000000003e-05,
|
118 |
+
"loss": 1.6763,
|
119 |
+
"step": 1700
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.2,
|
123 |
+
"learning_rate": 1.8e-05,
|
124 |
+
"loss": 1.6433,
|
125 |
+
"step": 1800
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.21,
|
129 |
+
"learning_rate": 1.9e-05,
|
130 |
+
"loss": 1.6442,
|
131 |
+
"step": 1900
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 0.23,
|
135 |
+
"learning_rate": 2e-05,
|
136 |
+
"loss": 1.6203,
|
137 |
+
"step": 2000
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 0.23,
|
141 |
+
"eval_loss": 1.562535047531128,
|
142 |
+
"eval_masked_accuracy": 0.6760570671304924,
|
143 |
+
"eval_runtime": 331.4685,
|
144 |
+
"eval_samples_per_second": 47.428,
|
145 |
+
"eval_steps_per_second": 2.966,
|
146 |
+
"step": 2000
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.24,
|
150 |
+
"learning_rate": 2.1e-05,
|
151 |
+
"loss": 1.6762,
|
152 |
+
"step": 2100
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.25,
|
156 |
+
"learning_rate": 2.2000000000000003e-05,
|
157 |
+
"loss": 1.6907,
|
158 |
+
"step": 2200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.26,
|
162 |
+
"learning_rate": 2.3000000000000003e-05,
|
163 |
+
"loss": 1.6851,
|
164 |
+
"step": 2300
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.27,
|
168 |
+
"learning_rate": 2.4e-05,
|
169 |
+
"loss": 1.6431,
|
170 |
+
"step": 2400
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.28,
|
174 |
+
"learning_rate": 2.5e-05,
|
175 |
+
"loss": 1.6542,
|
176 |
+
"step": 2500
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.29,
|
180 |
+
"learning_rate": 2.6000000000000002e-05,
|
181 |
+
"loss": 1.6946,
|
182 |
+
"step": 2600
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.31,
|
186 |
+
"learning_rate": 2.7000000000000002e-05,
|
187 |
+
"loss": 1.683,
|
188 |
+
"step": 2700
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.32,
|
192 |
+
"learning_rate": 2.8000000000000003e-05,
|
193 |
+
"loss": 1.7044,
|
194 |
+
"step": 2800
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.33,
|
198 |
+
"learning_rate": 2.9e-05,
|
199 |
+
"loss": 1.7157,
|
200 |
+
"step": 2900
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.34,
|
204 |
+
"learning_rate": 3e-05,
|
205 |
+
"loss": 1.7084,
|
206 |
+
"step": 3000
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.34,
|
210 |
+
"eval_loss": 1.6114859580993652,
|
211 |
+
"eval_masked_accuracy": 0.6701131345533646,
|
212 |
+
"eval_runtime": 331.5183,
|
213 |
+
"eval_samples_per_second": 47.421,
|
214 |
+
"eval_steps_per_second": 2.965,
|
215 |
+
"step": 3000
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.35,
|
219 |
+
"learning_rate": 3.1e-05,
|
220 |
+
"loss": 1.7268,
|
221 |
+
"step": 3100
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 0.36,
|
225 |
+
"learning_rate": 3.2000000000000005e-05,
|
226 |
+
"loss": 1.715,
|
227 |
+
"step": 3200
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 0.37,
|
231 |
+
"learning_rate": 3.3e-05,
|
232 |
+
"loss": 1.7501,
|
233 |
+
"step": 3300
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.38,
|
237 |
+
"learning_rate": 3.4000000000000007e-05,
|
238 |
+
"loss": 1.6992,
|
239 |
+
"step": 3400
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 0.4,
|
243 |
+
"learning_rate": 3.5e-05,
|
244 |
+
"loss": 1.7397,
|
245 |
+
"step": 3500
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 0.41,
|
249 |
+
"learning_rate": 3.6e-05,
|
250 |
+
"loss": 1.7577,
|
251 |
+
"step": 3600
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 0.42,
|
255 |
+
"learning_rate": 3.7e-05,
|
256 |
+
"loss": 1.728,
|
257 |
+
"step": 3700
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.43,
|
261 |
+
"learning_rate": 3.8e-05,
|
262 |
+
"loss": 1.7584,
|
263 |
+
"step": 3800
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 0.44,
|
267 |
+
"learning_rate": 3.9000000000000006e-05,
|
268 |
+
"loss": 1.7494,
|
269 |
+
"step": 3900
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 0.45,
|
273 |
+
"learning_rate": 4e-05,
|
274 |
+
"loss": 1.7423,
|
275 |
+
"step": 4000
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.45,
|
279 |
+
"eval_loss": 1.6563646793365479,
|
280 |
+
"eval_masked_accuracy": 0.663584889424318,
|
281 |
+
"eval_runtime": 331.7001,
|
282 |
+
"eval_samples_per_second": 47.395,
|
283 |
+
"eval_steps_per_second": 2.964,
|
284 |
+
"step": 4000
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 0.46,
|
288 |
+
"learning_rate": 4.1e-05,
|
289 |
+
"loss": 1.7778,
|
290 |
+
"step": 4100
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 0.47,
|
294 |
+
"learning_rate": 4.2e-05,
|
295 |
+
"loss": 1.7985,
|
296 |
+
"step": 4200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.49,
|
300 |
+
"learning_rate": 4.3e-05,
|
301 |
+
"loss": 1.8133,
|
302 |
+
"step": 4300
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 0.5,
|
306 |
+
"learning_rate": 4.4000000000000006e-05,
|
307 |
+
"loss": 1.8056,
|
308 |
+
"step": 4400
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.51,
|
312 |
+
"learning_rate": 4.5e-05,
|
313 |
+
"loss": 1.7452,
|
314 |
+
"step": 4500
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.52,
|
318 |
+
"learning_rate": 4.600000000000001e-05,
|
319 |
+
"loss": 1.8009,
|
320 |
+
"step": 4600
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.53,
|
324 |
+
"learning_rate": 4.7e-05,
|
325 |
+
"loss": 1.8066,
|
326 |
+
"step": 4700
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.54,
|
330 |
+
"learning_rate": 4.8e-05,
|
331 |
+
"loss": 1.8127,
|
332 |
+
"step": 4800
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.55,
|
336 |
+
"learning_rate": 4.9e-05,
|
337 |
+
"loss": 1.8454,
|
338 |
+
"step": 4900
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.57,
|
342 |
+
"learning_rate": 5e-05,
|
343 |
+
"loss": 1.8455,
|
344 |
+
"step": 5000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.57,
|
348 |
+
"eval_loss": 1.7312405109405518,
|
349 |
+
"eval_masked_accuracy": 0.6539095517673932,
|
350 |
+
"eval_runtime": 331.4967,
|
351 |
+
"eval_samples_per_second": 47.424,
|
352 |
+
"eval_steps_per_second": 2.965,
|
353 |
+
"step": 5000
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 0.58,
|
357 |
+
"learning_rate": 4.987249776871096e-05,
|
358 |
+
"loss": 1.839,
|
359 |
+
"step": 5100
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.59,
|
363 |
+
"learning_rate": 4.9744995537421905e-05,
|
364 |
+
"loss": 1.8038,
|
365 |
+
"step": 5200
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 0.6,
|
369 |
+
"learning_rate": 4.9617493306132864e-05,
|
370 |
+
"loss": 1.837,
|
371 |
+
"step": 5300
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 0.61,
|
375 |
+
"learning_rate": 4.948999107484381e-05,
|
376 |
+
"loss": 1.8565,
|
377 |
+
"step": 5400
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 0.62,
|
381 |
+
"learning_rate": 4.936248884355477e-05,
|
382 |
+
"loss": 1.8285,
|
383 |
+
"step": 5500
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 0.63,
|
387 |
+
"learning_rate": 4.923498661226572e-05,
|
388 |
+
"loss": 1.856,
|
389 |
+
"step": 5600
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 0.64,
|
393 |
+
"learning_rate": 4.910748438097667e-05,
|
394 |
+
"loss": 1.8494,
|
395 |
+
"step": 5700
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 0.66,
|
399 |
+
"learning_rate": 4.897998214968762e-05,
|
400 |
+
"loss": 1.8239,
|
401 |
+
"step": 5800
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.67,
|
405 |
+
"learning_rate": 4.885247991839857e-05,
|
406 |
+
"loss": 1.8294,
|
407 |
+
"step": 5900
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 0.68,
|
411 |
+
"learning_rate": 4.8724977687109524e-05,
|
412 |
+
"loss": 1.8222,
|
413 |
+
"step": 6000
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 0.68,
|
417 |
+
"eval_loss": 1.7369571924209595,
|
418 |
+
"eval_masked_accuracy": 0.6525969197353448,
|
419 |
+
"eval_runtime": 332.1886,
|
420 |
+
"eval_samples_per_second": 47.326,
|
421 |
+
"eval_steps_per_second": 2.959,
|
422 |
+
"step": 6000
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.69,
|
426 |
+
"learning_rate": 4.859747545582048e-05,
|
427 |
+
"loss": 1.8653,
|
428 |
+
"step": 6100
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.7,
|
432 |
+
"learning_rate": 4.8469973224531434e-05,
|
433 |
+
"loss": 1.8235,
|
434 |
+
"step": 6200
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.71,
|
438 |
+
"learning_rate": 4.8342470993242385e-05,
|
439 |
+
"loss": 1.8045,
|
440 |
+
"step": 6300
|
441 |
+
},
|
442 |
+
{
|
443 |
+
"epoch": 0.72,
|
444 |
+
"learning_rate": 4.821496876195334e-05,
|
445 |
+
"loss": 1.8237,
|
446 |
+
"step": 6400
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 0.74,
|
450 |
+
"learning_rate": 4.808746653066429e-05,
|
451 |
+
"loss": 1.8345,
|
452 |
+
"step": 6500
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 0.75,
|
456 |
+
"learning_rate": 4.795996429937524e-05,
|
457 |
+
"loss": 1.838,
|
458 |
+
"step": 6600
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.76,
|
462 |
+
"learning_rate": 4.783246206808619e-05,
|
463 |
+
"loss": 1.7893,
|
464 |
+
"step": 6700
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.77,
|
468 |
+
"learning_rate": 4.770495983679715e-05,
|
469 |
+
"loss": 1.8514,
|
470 |
+
"step": 6800
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.78,
|
474 |
+
"learning_rate": 4.7577457605508094e-05,
|
475 |
+
"loss": 1.8107,
|
476 |
+
"step": 6900
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.79,
|
480 |
+
"learning_rate": 4.744995537421905e-05,
|
481 |
+
"loss": 1.8154,
|
482 |
+
"step": 7000
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.79,
|
486 |
+
"eval_loss": 1.7349368333816528,
|
487 |
+
"eval_masked_accuracy": 0.6534969877432311,
|
488 |
+
"eval_runtime": 332.4558,
|
489 |
+
"eval_samples_per_second": 47.287,
|
490 |
+
"eval_steps_per_second": 2.957,
|
491 |
+
"step": 7000
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 0.8,
|
495 |
+
"learning_rate": 4.7322453142930004e-05,
|
496 |
+
"loss": 1.8566,
|
497 |
+
"step": 7100
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 0.81,
|
501 |
+
"learning_rate": 4.7194950911640956e-05,
|
502 |
+
"loss": 1.8355,
|
503 |
+
"step": 7200
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 0.83,
|
507 |
+
"learning_rate": 4.7067448680351914e-05,
|
508 |
+
"loss": 1.8517,
|
509 |
+
"step": 7300
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 0.84,
|
513 |
+
"learning_rate": 4.693994644906286e-05,
|
514 |
+
"loss": 1.7926,
|
515 |
+
"step": 7400
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 0.85,
|
519 |
+
"learning_rate": 4.681244421777382e-05,
|
520 |
+
"loss": 1.8277,
|
521 |
+
"step": 7500
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 0.86,
|
525 |
+
"learning_rate": 4.668494198648476e-05,
|
526 |
+
"loss": 1.8247,
|
527 |
+
"step": 7600
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.87,
|
531 |
+
"learning_rate": 4.655743975519572e-05,
|
532 |
+
"loss": 1.8339,
|
533 |
+
"step": 7700
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 0.88,
|
537 |
+
"learning_rate": 4.642993752390667e-05,
|
538 |
+
"loss": 1.8542,
|
539 |
+
"step": 7800
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 0.89,
|
543 |
+
"learning_rate": 4.630243529261762e-05,
|
544 |
+
"loss": 1.8018,
|
545 |
+
"step": 7900
|
546 |
+
},
|
547 |
+
{
|
548 |
+
"epoch": 0.9,
|
549 |
+
"learning_rate": 4.6174933061328574e-05,
|
550 |
+
"loss": 1.8652,
|
551 |
+
"step": 8000
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 0.9,
|
555 |
+
"eval_loss": 1.7150492668151855,
|
556 |
+
"eval_masked_accuracy": 0.6540794101708254,
|
557 |
+
"eval_runtime": 331.8362,
|
558 |
+
"eval_samples_per_second": 47.376,
|
559 |
+
"eval_steps_per_second": 2.962,
|
560 |
+
"step": 8000
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 0.92,
|
564 |
+
"learning_rate": 4.6047430830039526e-05,
|
565 |
+
"loss": 1.8236,
|
566 |
+
"step": 8100
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.93,
|
570 |
+
"learning_rate": 4.591992859875048e-05,
|
571 |
+
"loss": 1.8607,
|
572 |
+
"step": 8200
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 0.94,
|
576 |
+
"learning_rate": 4.5792426367461436e-05,
|
577 |
+
"loss": 1.8329,
|
578 |
+
"step": 8300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 0.95,
|
582 |
+
"learning_rate": 4.566492413617239e-05,
|
583 |
+
"loss": 1.8319,
|
584 |
+
"step": 8400
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 0.96,
|
588 |
+
"learning_rate": 4.553742190488334e-05,
|
589 |
+
"loss": 1.8482,
|
590 |
+
"step": 8500
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.97,
|
594 |
+
"learning_rate": 4.540991967359429e-05,
|
595 |
+
"loss": 1.8508,
|
596 |
+
"step": 8600
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 0.98,
|
600 |
+
"learning_rate": 4.528241744230524e-05,
|
601 |
+
"loss": 1.8216,
|
602 |
+
"step": 8700
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 1.0,
|
606 |
+
"learning_rate": 4.515491521101619e-05,
|
607 |
+
"loss": 1.8211,
|
608 |
+
"step": 8800
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 1.01,
|
612 |
+
"learning_rate": 4.5027412979727145e-05,
|
613 |
+
"loss": 1.7892,
|
614 |
+
"step": 8900
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 1.02,
|
618 |
+
"learning_rate": 4.48999107484381e-05,
|
619 |
+
"loss": 1.7627,
|
620 |
+
"step": 9000
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 1.02,
|
624 |
+
"eval_loss": 1.7102247476577759,
|
625 |
+
"eval_masked_accuracy": 0.6570814695769431,
|
626 |
+
"eval_runtime": 332.0611,
|
627 |
+
"eval_samples_per_second": 47.344,
|
628 |
+
"eval_steps_per_second": 2.96,
|
629 |
+
"step": 9000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 1.03,
|
633 |
+
"learning_rate": 4.477240851714905e-05,
|
634 |
+
"loss": 1.773,
|
635 |
+
"step": 9100
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 1.04,
|
639 |
+
"learning_rate": 4.4644906285860006e-05,
|
640 |
+
"loss": 1.744,
|
641 |
+
"step": 9200
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 1.05,
|
645 |
+
"learning_rate": 4.451740405457096e-05,
|
646 |
+
"loss": 1.7884,
|
647 |
+
"step": 9300
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.06,
|
651 |
+
"learning_rate": 4.438990182328191e-05,
|
652 |
+
"loss": 1.7675,
|
653 |
+
"step": 9400
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 1.07,
|
657 |
+
"learning_rate": 4.426239959199287e-05,
|
658 |
+
"loss": 1.7457,
|
659 |
+
"step": 9500
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 1.09,
|
663 |
+
"learning_rate": 4.413489736070381e-05,
|
664 |
+
"loss": 1.7824,
|
665 |
+
"step": 9600
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 1.1,
|
669 |
+
"learning_rate": 4.400739512941477e-05,
|
670 |
+
"loss": 1.7991,
|
671 |
+
"step": 9700
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 1.11,
|
675 |
+
"learning_rate": 4.387989289812572e-05,
|
676 |
+
"loss": 1.768,
|
677 |
+
"step": 9800
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 1.12,
|
681 |
+
"learning_rate": 4.375239066683667e-05,
|
682 |
+
"loss": 1.7601,
|
683 |
+
"step": 9900
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 1.13,
|
687 |
+
"learning_rate": 4.3624888435547625e-05,
|
688 |
+
"loss": 1.7515,
|
689 |
+
"step": 10000
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.13,
|
693 |
+
"eval_loss": 1.7048571109771729,
|
694 |
+
"eval_masked_accuracy": 0.6552869443926557,
|
695 |
+
"eval_runtime": 331.6676,
|
696 |
+
"eval_samples_per_second": 47.4,
|
697 |
+
"eval_steps_per_second": 2.964,
|
698 |
+
"step": 10000
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 1.14,
|
702 |
+
"learning_rate": 4.3497386204258576e-05,
|
703 |
+
"loss": 1.793,
|
704 |
+
"step": 10100
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 1.15,
|
708 |
+
"learning_rate": 4.336988397296953e-05,
|
709 |
+
"loss": 1.7734,
|
710 |
+
"step": 10200
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.16,
|
714 |
+
"learning_rate": 4.324238174168048e-05,
|
715 |
+
"loss": 1.746,
|
716 |
+
"step": 10300
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 1.18,
|
720 |
+
"learning_rate": 4.311487951039143e-05,
|
721 |
+
"loss": 1.7725,
|
722 |
+
"step": 10400
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 1.19,
|
726 |
+
"learning_rate": 4.298737727910239e-05,
|
727 |
+
"loss": 1.7721,
|
728 |
+
"step": 10500
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 1.2,
|
732 |
+
"learning_rate": 4.285987504781334e-05,
|
733 |
+
"loss": 1.7665,
|
734 |
+
"step": 10600
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 1.21,
|
738 |
+
"learning_rate": 4.273237281652429e-05,
|
739 |
+
"loss": 1.746,
|
740 |
+
"step": 10700
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 1.22,
|
744 |
+
"learning_rate": 4.2604870585235244e-05,
|
745 |
+
"loss": 1.7683,
|
746 |
+
"step": 10800
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 1.23,
|
750 |
+
"learning_rate": 4.2477368353946195e-05,
|
751 |
+
"loss": 1.8083,
|
752 |
+
"step": 10900
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.24,
|
756 |
+
"learning_rate": 4.234986612265715e-05,
|
757 |
+
"loss": 1.7701,
|
758 |
+
"step": 11000
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 1.24,
|
762 |
+
"eval_loss": 1.6938451528549194,
|
763 |
+
"eval_masked_accuracy": 0.660930511429812,
|
764 |
+
"eval_runtime": 331.8416,
|
765 |
+
"eval_samples_per_second": 47.375,
|
766 |
+
"eval_steps_per_second": 2.962,
|
767 |
+
"step": 11000
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 1.26,
|
771 |
+
"learning_rate": 4.22223638913681e-05,
|
772 |
+
"loss": 1.802,
|
773 |
+
"step": 11100
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.27,
|
777 |
+
"learning_rate": 4.2094861660079056e-05,
|
778 |
+
"loss": 1.7632,
|
779 |
+
"step": 11200
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 1.28,
|
783 |
+
"learning_rate": 4.196735942879e-05,
|
784 |
+
"loss": 1.751,
|
785 |
+
"step": 11300
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 1.29,
|
789 |
+
"learning_rate": 4.183985719750096e-05,
|
790 |
+
"loss": 1.7969,
|
791 |
+
"step": 11400
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 1.3,
|
795 |
+
"learning_rate": 4.171235496621191e-05,
|
796 |
+
"loss": 1.7728,
|
797 |
+
"step": 11500
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 1.31,
|
801 |
+
"learning_rate": 4.158485273492286e-05,
|
802 |
+
"loss": 1.7399,
|
803 |
+
"step": 11600
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 1.32,
|
807 |
+
"learning_rate": 4.145735050363382e-05,
|
808 |
+
"loss": 1.7988,
|
809 |
+
"step": 11700
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 1.33,
|
813 |
+
"learning_rate": 4.1329848272344765e-05,
|
814 |
+
"loss": 1.7714,
|
815 |
+
"step": 11800
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.35,
|
819 |
+
"learning_rate": 4.1202346041055724e-05,
|
820 |
+
"loss": 1.7627,
|
821 |
+
"step": 11900
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 1.36,
|
825 |
+
"learning_rate": 4.1074843809766675e-05,
|
826 |
+
"loss": 1.7634,
|
827 |
+
"step": 12000
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 1.36,
|
831 |
+
"eval_loss": 1.6735210418701172,
|
832 |
+
"eval_masked_accuracy": 0.6616568320785755,
|
833 |
+
"eval_runtime": 331.8906,
|
834 |
+
"eval_samples_per_second": 47.368,
|
835 |
+
"eval_steps_per_second": 2.962,
|
836 |
+
"step": 12000
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.37,
|
840 |
+
"learning_rate": 4.0947341578477627e-05,
|
841 |
+
"loss": 1.7647,
|
842 |
+
"step": 12100
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 1.38,
|
846 |
+
"learning_rate": 4.081983934718858e-05,
|
847 |
+
"loss": 1.7316,
|
848 |
+
"step": 12200
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 1.39,
|
852 |
+
"learning_rate": 4.069233711589953e-05,
|
853 |
+
"loss": 1.7505,
|
854 |
+
"step": 12300
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.4,
|
858 |
+
"learning_rate": 4.056483488461048e-05,
|
859 |
+
"loss": 1.7623,
|
860 |
+
"step": 12400
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 1.41,
|
864 |
+
"learning_rate": 4.043733265332143e-05,
|
865 |
+
"loss": 1.7462,
|
866 |
+
"step": 12500
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 1.42,
|
870 |
+
"learning_rate": 4.0309830422032384e-05,
|
871 |
+
"loss": 1.7351,
|
872 |
+
"step": 12600
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 1.44,
|
876 |
+
"learning_rate": 4.018232819074334e-05,
|
877 |
+
"loss": 1.7126,
|
878 |
+
"step": 12700
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.45,
|
882 |
+
"learning_rate": 4.005482595945429e-05,
|
883 |
+
"loss": 1.7453,
|
884 |
+
"step": 12800
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 1.46,
|
888 |
+
"learning_rate": 3.9927323728165245e-05,
|
889 |
+
"loss": 1.7464,
|
890 |
+
"step": 12900
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 1.47,
|
894 |
+
"learning_rate": 3.97998214968762e-05,
|
895 |
+
"loss": 1.7489,
|
896 |
+
"step": 13000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 1.47,
|
900 |
+
"eval_loss": 1.6684247255325317,
|
901 |
+
"eval_masked_accuracy": 0.6630148312748695,
|
902 |
+
"eval_runtime": 332.4376,
|
903 |
+
"eval_samples_per_second": 47.29,
|
904 |
+
"eval_steps_per_second": 2.957,
|
905 |
+
"step": 13000
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 1.48,
|
909 |
+
"learning_rate": 3.967231926558715e-05,
|
910 |
+
"loss": 1.7411,
|
911 |
+
"step": 13100
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 1.49,
|
915 |
+
"learning_rate": 3.954481703429811e-05,
|
916 |
+
"loss": 1.7478,
|
917 |
+
"step": 13200
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 1.5,
|
921 |
+
"learning_rate": 3.941731480300905e-05,
|
922 |
+
"loss": 1.7717,
|
923 |
+
"step": 13300
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 1.52,
|
927 |
+
"learning_rate": 3.928981257172001e-05,
|
928 |
+
"loss": 1.7647,
|
929 |
+
"step": 13400
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 1.53,
|
933 |
+
"learning_rate": 3.9162310340430954e-05,
|
934 |
+
"loss": 1.7796,
|
935 |
+
"step": 13500
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 1.54,
|
939 |
+
"learning_rate": 3.903480810914191e-05,
|
940 |
+
"loss": 1.7425,
|
941 |
+
"step": 13600
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 1.55,
|
945 |
+
"learning_rate": 3.8907305877852864e-05,
|
946 |
+
"loss": 1.7404,
|
947 |
+
"step": 13700
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 1.56,
|
951 |
+
"learning_rate": 3.8779803646563816e-05,
|
952 |
+
"loss": 1.7335,
|
953 |
+
"step": 13800
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 1.57,
|
957 |
+
"learning_rate": 3.865230141527477e-05,
|
958 |
+
"loss": 1.7171,
|
959 |
+
"step": 13900
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 1.58,
|
963 |
+
"learning_rate": 3.852479918398572e-05,
|
964 |
+
"loss": 1.7319,
|
965 |
+
"step": 14000
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 1.58,
|
969 |
+
"eval_loss": 1.6560810804367065,
|
970 |
+
"eval_masked_accuracy": 0.6637841256311287,
|
971 |
+
"eval_runtime": 332.3385,
|
972 |
+
"eval_samples_per_second": 47.304,
|
973 |
+
"eval_steps_per_second": 2.958,
|
974 |
+
"step": 14000
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 1.59,
|
978 |
+
"learning_rate": 3.839729695269668e-05,
|
979 |
+
"loss": 1.7296,
|
980 |
+
"step": 14100
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 1.61,
|
984 |
+
"learning_rate": 3.826979472140763e-05,
|
985 |
+
"loss": 1.7562,
|
986 |
+
"step": 14200
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 1.62,
|
990 |
+
"learning_rate": 3.814229249011858e-05,
|
991 |
+
"loss": 1.7446,
|
992 |
+
"step": 14300
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 1.63,
|
996 |
+
"learning_rate": 3.801479025882953e-05,
|
997 |
+
"loss": 1.7525,
|
998 |
+
"step": 14400
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 1.64,
|
1002 |
+
"learning_rate": 3.788728802754048e-05,
|
1003 |
+
"loss": 1.715,
|
1004 |
+
"step": 14500
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 1.65,
|
1008 |
+
"learning_rate": 3.7759785796251434e-05,
|
1009 |
+
"loss": 1.7302,
|
1010 |
+
"step": 14600
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 1.66,
|
1014 |
+
"learning_rate": 3.763228356496239e-05,
|
1015 |
+
"loss": 1.7145,
|
1016 |
+
"step": 14700
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.67,
|
1020 |
+
"learning_rate": 3.750478133367334e-05,
|
1021 |
+
"loss": 1.7382,
|
1022 |
+
"step": 14800
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 1.68,
|
1026 |
+
"learning_rate": 3.7377279102384296e-05,
|
1027 |
+
"loss": 1.7181,
|
1028 |
+
"step": 14900
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 1.7,
|
1032 |
+
"learning_rate": 3.724977687109524e-05,
|
1033 |
+
"loss": 1.7744,
|
1034 |
+
"step": 15000
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 1.7,
|
1038 |
+
"eval_loss": 1.6506950855255127,
|
1039 |
+
"eval_masked_accuracy": 0.6654687548499909,
|
1040 |
+
"eval_runtime": 332.5023,
|
1041 |
+
"eval_samples_per_second": 47.281,
|
1042 |
+
"eval_steps_per_second": 2.956,
|
1043 |
+
"step": 15000
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 1.71,
|
1047 |
+
"learning_rate": 3.71222746398062e-05,
|
1048 |
+
"loss": 1.7438,
|
1049 |
+
"step": 15100
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 1.72,
|
1053 |
+
"learning_rate": 3.699477240851715e-05,
|
1054 |
+
"loss": 1.6953,
|
1055 |
+
"step": 15200
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 1.73,
|
1059 |
+
"learning_rate": 3.68672701772281e-05,
|
1060 |
+
"loss": 1.7364,
|
1061 |
+
"step": 15300
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 1.74,
|
1065 |
+
"learning_rate": 3.673976794593906e-05,
|
1066 |
+
"loss": 1.7148,
|
1067 |
+
"step": 15400
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 1.75,
|
1071 |
+
"learning_rate": 3.6612265714650005e-05,
|
1072 |
+
"loss": 1.7145,
|
1073 |
+
"step": 15500
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 1.76,
|
1077 |
+
"learning_rate": 3.648476348336096e-05,
|
1078 |
+
"loss": 1.7098,
|
1079 |
+
"step": 15600
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 1.78,
|
1083 |
+
"learning_rate": 3.6357261252071914e-05,
|
1084 |
+
"loss": 1.6816,
|
1085 |
+
"step": 15700
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.79,
|
1089 |
+
"learning_rate": 3.6229759020782866e-05,
|
1090 |
+
"loss": 1.6903,
|
1091 |
+
"step": 15800
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 1.8,
|
1095 |
+
"learning_rate": 3.610225678949382e-05,
|
1096 |
+
"loss": 1.7243,
|
1097 |
+
"step": 15900
|
1098 |
+
},
|
1099 |
+
{
|
1100 |
+
"epoch": 1.81,
|
1101 |
+
"learning_rate": 3.597475455820477e-05,
|
1102 |
+
"loss": 1.6786,
|
1103 |
+
"step": 16000
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 1.81,
|
1107 |
+
"eval_loss": 1.633168339729309,
|
1108 |
+
"eval_masked_accuracy": 0.6682277179202242,
|
1109 |
+
"eval_runtime": 332.1898,
|
1110 |
+
"eval_samples_per_second": 47.325,
|
1111 |
+
"eval_steps_per_second": 2.959,
|
1112 |
+
"step": 16000
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 1.82,
|
1116 |
+
"learning_rate": 3.584725232691572e-05,
|
1117 |
+
"loss": 1.7284,
|
1118 |
+
"step": 16100
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 1.83,
|
1122 |
+
"learning_rate": 3.571975009562667e-05,
|
1123 |
+
"loss": 1.684,
|
1124 |
+
"step": 16200
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 1.84,
|
1128 |
+
"learning_rate": 3.559224786433763e-05,
|
1129 |
+
"loss": 1.7172,
|
1130 |
+
"step": 16300
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 1.85,
|
1134 |
+
"learning_rate": 3.546474563304858e-05,
|
1135 |
+
"loss": 1.7348,
|
1136 |
+
"step": 16400
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.87,
|
1140 |
+
"learning_rate": 3.533724340175953e-05,
|
1141 |
+
"loss": 1.7033,
|
1142 |
+
"step": 16500
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 1.88,
|
1146 |
+
"learning_rate": 3.5209741170470485e-05,
|
1147 |
+
"loss": 1.7349,
|
1148 |
+
"step": 16600
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 1.89,
|
1152 |
+
"learning_rate": 3.5082238939181436e-05,
|
1153 |
+
"loss": 1.6813,
|
1154 |
+
"step": 16700
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 1.9,
|
1158 |
+
"learning_rate": 3.495473670789239e-05,
|
1159 |
+
"loss": 1.7038,
|
1160 |
+
"step": 16800
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 1.91,
|
1164 |
+
"learning_rate": 3.4827234476603346e-05,
|
1165 |
+
"loss": 1.6866,
|
1166 |
+
"step": 16900
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 1.92,
|
1170 |
+
"learning_rate": 3.469973224531429e-05,
|
1171 |
+
"loss": 1.7186,
|
1172 |
+
"step": 17000
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 1.92,
|
1176 |
+
"eval_loss": 1.6309564113616943,
|
1177 |
+
"eval_masked_accuracy": 0.6676820097569746,
|
1178 |
+
"eval_runtime": 332.3036,
|
1179 |
+
"eval_samples_per_second": 47.309,
|
1180 |
+
"eval_steps_per_second": 2.958,
|
1181 |
+
"step": 17000
|
1182 |
+
},
|
1183 |
+
{
|
1184 |
+
"epoch": 1.93,
|
1185 |
+
"learning_rate": 3.457223001402525e-05,
|
1186 |
+
"loss": 1.659,
|
1187 |
+
"step": 17100
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 1.95,
|
1191 |
+
"learning_rate": 3.4444727782736194e-05,
|
1192 |
+
"loss": 1.6826,
|
1193 |
+
"step": 17200
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 1.96,
|
1197 |
+
"learning_rate": 3.431722555144715e-05,
|
1198 |
+
"loss": 1.6723,
|
1199 |
+
"step": 17300
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 1.97,
|
1203 |
+
"learning_rate": 3.418972332015811e-05,
|
1204 |
+
"loss": 1.7089,
|
1205 |
+
"step": 17400
|
1206 |
+
},
|
1207 |
+
{
|
1208 |
+
"epoch": 1.98,
|
1209 |
+
"learning_rate": 3.4062221088869055e-05,
|
1210 |
+
"loss": 1.681,
|
1211 |
+
"step": 17500
|
1212 |
+
},
|
1213 |
+
{
|
1214 |
+
"epoch": 1.99,
|
1215 |
+
"learning_rate": 3.393471885758001e-05,
|
1216 |
+
"loss": 1.6889,
|
1217 |
+
"step": 17600
|
1218 |
+
},
|
1219 |
+
{
|
1220 |
+
"epoch": 2.0,
|
1221 |
+
"learning_rate": 3.380721662629096e-05,
|
1222 |
+
"loss": 1.7217,
|
1223 |
+
"step": 17700
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 2.01,
|
1227 |
+
"learning_rate": 3.3679714395001916e-05,
|
1228 |
+
"loss": 1.637,
|
1229 |
+
"step": 17800
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.02,
|
1233 |
+
"learning_rate": 3.355221216371287e-05,
|
1234 |
+
"loss": 1.6631,
|
1235 |
+
"step": 17900
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 2.04,
|
1239 |
+
"learning_rate": 3.342470993242382e-05,
|
1240 |
+
"loss": 1.6505,
|
1241 |
+
"step": 18000
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 2.04,
|
1245 |
+
"eval_loss": 1.6221129894256592,
|
1246 |
+
"eval_masked_accuracy": 0.6697054625754659,
|
1247 |
+
"eval_runtime": 332.1519,
|
1248 |
+
"eval_samples_per_second": 47.331,
|
1249 |
+
"eval_steps_per_second": 2.959,
|
1250 |
+
"step": 18000
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.05,
|
1254 |
+
"learning_rate": 3.329720770113477e-05,
|
1255 |
+
"loss": 1.6624,
|
1256 |
+
"step": 18100
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 2.06,
|
1260 |
+
"learning_rate": 3.316970546984572e-05,
|
1261 |
+
"loss": 1.6445,
|
1262 |
+
"step": 18200
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 2.07,
|
1266 |
+
"learning_rate": 3.3042203238556674e-05,
|
1267 |
+
"loss": 1.6377,
|
1268 |
+
"step": 18300
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 2.08,
|
1272 |
+
"learning_rate": 3.291470100726763e-05,
|
1273 |
+
"loss": 1.6494,
|
1274 |
+
"step": 18400
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 2.09,
|
1278 |
+
"learning_rate": 3.2787198775978584e-05,
|
1279 |
+
"loss": 1.6291,
|
1280 |
+
"step": 18500
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 2.1,
|
1284 |
+
"learning_rate": 3.2659696544689535e-05,
|
1285 |
+
"loss": 1.6128,
|
1286 |
+
"step": 18600
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 2.11,
|
1290 |
+
"learning_rate": 3.2532194313400487e-05,
|
1291 |
+
"loss": 1.648,
|
1292 |
+
"step": 18700
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.13,
|
1296 |
+
"learning_rate": 3.240469208211144e-05,
|
1297 |
+
"loss": 1.6191,
|
1298 |
+
"step": 18800
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 2.14,
|
1302 |
+
"learning_rate": 3.227718985082239e-05,
|
1303 |
+
"loss": 1.652,
|
1304 |
+
"step": 18900
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 2.15,
|
1308 |
+
"learning_rate": 3.214968761953334e-05,
|
1309 |
+
"loss": 1.6222,
|
1310 |
+
"step": 19000
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 2.15,
|
1314 |
+
"eval_loss": 1.6012800931930542,
|
1315 |
+
"eval_masked_accuracy": 0.6742498032867577,
|
1316 |
+
"eval_runtime": 332.2981,
|
1317 |
+
"eval_samples_per_second": 47.31,
|
1318 |
+
"eval_steps_per_second": 2.958,
|
1319 |
+
"step": 19000
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 2.16,
|
1323 |
+
"learning_rate": 3.20221853882443e-05,
|
1324 |
+
"loss": 1.6378,
|
1325 |
+
"step": 19100
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 2.17,
|
1329 |
+
"learning_rate": 3.1894683156955244e-05,
|
1330 |
+
"loss": 1.6373,
|
1331 |
+
"step": 19200
|
1332 |
+
},
|
1333 |
+
{
|
1334 |
+
"epoch": 2.18,
|
1335 |
+
"learning_rate": 3.17671809256662e-05,
|
1336 |
+
"loss": 1.643,
|
1337 |
+
"step": 19300
|
1338 |
+
},
|
1339 |
+
{
|
1340 |
+
"epoch": 2.19,
|
1341 |
+
"learning_rate": 3.1639678694377154e-05,
|
1342 |
+
"loss": 1.6504,
|
1343 |
+
"step": 19400
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 2.21,
|
1347 |
+
"learning_rate": 3.1512176463088105e-05,
|
1348 |
+
"loss": 1.6707,
|
1349 |
+
"step": 19500
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 2.22,
|
1353 |
+
"learning_rate": 3.1384674231799064e-05,
|
1354 |
+
"loss": 1.6684,
|
1355 |
+
"step": 19600
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 2.23,
|
1359 |
+
"learning_rate": 3.125717200051001e-05,
|
1360 |
+
"loss": 1.6431,
|
1361 |
+
"step": 19700
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 2.24,
|
1365 |
+
"learning_rate": 3.1129669769220967e-05,
|
1366 |
+
"loss": 1.6578,
|
1367 |
+
"step": 19800
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 2.25,
|
1371 |
+
"learning_rate": 3.100216753793191e-05,
|
1372 |
+
"loss": 1.6471,
|
1373 |
+
"step": 19900
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 2.26,
|
1377 |
+
"learning_rate": 3.087466530664287e-05,
|
1378 |
+
"loss": 1.6306,
|
1379 |
+
"step": 20000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 2.26,
|
1383 |
+
"eval_loss": 1.5895774364471436,
|
1384 |
+
"eval_masked_accuracy": 0.6750452664697661,
|
1385 |
+
"eval_runtime": 332.6359,
|
1386 |
+
"eval_samples_per_second": 47.262,
|
1387 |
+
"eval_steps_per_second": 2.955,
|
1388 |
+
"step": 20000
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 2.27,
|
1392 |
+
"learning_rate": 3.074716307535382e-05,
|
1393 |
+
"loss": 1.6205,
|
1394 |
+
"step": 20100
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 2.28,
|
1398 |
+
"learning_rate": 3.061966084406477e-05,
|
1399 |
+
"loss": 1.6493,
|
1400 |
+
"step": 20200
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 2.3,
|
1404 |
+
"learning_rate": 3.0492158612775724e-05,
|
1405 |
+
"loss": 1.6142,
|
1406 |
+
"step": 20300
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 2.31,
|
1410 |
+
"learning_rate": 3.036465638148668e-05,
|
1411 |
+
"loss": 1.6115,
|
1412 |
+
"step": 20400
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 2.32,
|
1416 |
+
"learning_rate": 3.0237154150197627e-05,
|
1417 |
+
"loss": 1.6608,
|
1418 |
+
"step": 20500
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 2.33,
|
1422 |
+
"learning_rate": 3.0109651918908582e-05,
|
1423 |
+
"loss": 1.6156,
|
1424 |
+
"step": 20600
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 2.34,
|
1428 |
+
"learning_rate": 2.9982149687619537e-05,
|
1429 |
+
"loss": 1.6199,
|
1430 |
+
"step": 20700
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 2.35,
|
1434 |
+
"learning_rate": 2.985464745633049e-05,
|
1435 |
+
"loss": 1.6255,
|
1436 |
+
"step": 20800
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 2.36,
|
1440 |
+
"learning_rate": 2.9727145225041443e-05,
|
1441 |
+
"loss": 1.6373,
|
1442 |
+
"step": 20900
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 2.37,
|
1446 |
+
"learning_rate": 2.959964299375239e-05,
|
1447 |
+
"loss": 1.6165,
|
1448 |
+
"step": 21000
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 2.37,
|
1452 |
+
"eval_loss": 1.5816543102264404,
|
1453 |
+
"eval_masked_accuracy": 0.6755305159120167,
|
1454 |
+
"eval_runtime": 332.6858,
|
1455 |
+
"eval_samples_per_second": 47.255,
|
1456 |
+
"eval_steps_per_second": 2.955,
|
1457 |
+
"step": 21000
|
1458 |
+
},
|
1459 |
+
{
|
1460 |
+
"epoch": 2.39,
|
1461 |
+
"learning_rate": 2.9472140762463346e-05,
|
1462 |
+
"loss": 1.6529,
|
1463 |
+
"step": 21100
|
1464 |
+
},
|
1465 |
+
{
|
1466 |
+
"epoch": 2.4,
|
1467 |
+
"learning_rate": 2.9344638531174294e-05,
|
1468 |
+
"loss": 1.5853,
|
1469 |
+
"step": 21200
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 2.41,
|
1473 |
+
"learning_rate": 2.921713629988525e-05,
|
1474 |
+
"loss": 1.6273,
|
1475 |
+
"step": 21300
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 2.42,
|
1479 |
+
"learning_rate": 2.90896340685962e-05,
|
1480 |
+
"loss": 1.6308,
|
1481 |
+
"step": 21400
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 2.43,
|
1485 |
+
"learning_rate": 2.8962131837307156e-05,
|
1486 |
+
"loss": 1.6241,
|
1487 |
+
"step": 21500
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 2.44,
|
1491 |
+
"learning_rate": 2.8834629606018104e-05,
|
1492 |
+
"loss": 1.6267,
|
1493 |
+
"step": 21600
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 2.45,
|
1497 |
+
"learning_rate": 2.870712737472906e-05,
|
1498 |
+
"loss": 1.5814,
|
1499 |
+
"step": 21700
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 2.47,
|
1503 |
+
"learning_rate": 2.8579625143440013e-05,
|
1504 |
+
"loss": 1.6185,
|
1505 |
+
"step": 21800
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 2.48,
|
1509 |
+
"learning_rate": 2.8452122912150965e-05,
|
1510 |
+
"loss": 1.5908,
|
1511 |
+
"step": 21900
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 2.49,
|
1515 |
+
"learning_rate": 2.832462068086192e-05,
|
1516 |
+
"loss": 1.6495,
|
1517 |
+
"step": 22000
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 2.49,
|
1521 |
+
"eval_loss": 1.5690662860870361,
|
1522 |
+
"eval_masked_accuracy": 0.6774331565563334,
|
1523 |
+
"eval_runtime": 332.6387,
|
1524 |
+
"eval_samples_per_second": 47.261,
|
1525 |
+
"eval_steps_per_second": 2.955,
|
1526 |
+
"step": 22000
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 2.5,
|
1530 |
+
"learning_rate": 2.8197118449572868e-05,
|
1531 |
+
"loss": 1.6091,
|
1532 |
+
"step": 22100
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 2.51,
|
1536 |
+
"learning_rate": 2.8069616218283823e-05,
|
1537 |
+
"loss": 1.6058,
|
1538 |
+
"step": 22200
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 2.52,
|
1542 |
+
"learning_rate": 2.794211398699477e-05,
|
1543 |
+
"loss": 1.6358,
|
1544 |
+
"step": 22300
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 2.53,
|
1548 |
+
"learning_rate": 2.7814611755705726e-05,
|
1549 |
+
"loss": 1.6094,
|
1550 |
+
"step": 22400
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 2.54,
|
1554 |
+
"learning_rate": 2.7687109524416677e-05,
|
1555 |
+
"loss": 1.6243,
|
1556 |
+
"step": 22500
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 2.56,
|
1560 |
+
"learning_rate": 2.7559607293127632e-05,
|
1561 |
+
"loss": 1.6078,
|
1562 |
+
"step": 22600
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 2.57,
|
1566 |
+
"learning_rate": 2.743210506183858e-05,
|
1567 |
+
"loss": 1.5921,
|
1568 |
+
"step": 22700
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 2.58,
|
1572 |
+
"learning_rate": 2.7304602830549535e-05,
|
1573 |
+
"loss": 1.5934,
|
1574 |
+
"step": 22800
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 2.59,
|
1578 |
+
"learning_rate": 2.717710059926049e-05,
|
1579 |
+
"loss": 1.6134,
|
1580 |
+
"step": 22900
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 2.6,
|
1584 |
+
"learning_rate": 2.704959836797144e-05,
|
1585 |
+
"loss": 1.5809,
|
1586 |
+
"step": 23000
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 2.6,
|
1590 |
+
"eval_loss": 1.555253267288208,
|
1591 |
+
"eval_masked_accuracy": 0.6777390847722294,
|
1592 |
+
"eval_runtime": 332.6441,
|
1593 |
+
"eval_samples_per_second": 47.261,
|
1594 |
+
"eval_steps_per_second": 2.955,
|
1595 |
+
"step": 23000
|
1596 |
+
},
|
1597 |
+
{
|
1598 |
+
"epoch": 2.61,
|
1599 |
+
"learning_rate": 2.6922096136682397e-05,
|
1600 |
+
"loss": 1.6062,
|
1601 |
+
"step": 23100
|
1602 |
+
},
|
1603 |
+
{
|
1604 |
+
"epoch": 2.62,
|
1605 |
+
"learning_rate": 2.6794593905393345e-05,
|
1606 |
+
"loss": 1.5813,
|
1607 |
+
"step": 23200
|
1608 |
+
},
|
1609 |
+
{
|
1610 |
+
"epoch": 2.63,
|
1611 |
+
"learning_rate": 2.66670916741043e-05,
|
1612 |
+
"loss": 1.6014,
|
1613 |
+
"step": 23300
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 2.65,
|
1617 |
+
"learning_rate": 2.6539589442815248e-05,
|
1618 |
+
"loss": 1.5966,
|
1619 |
+
"step": 23400
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 2.66,
|
1623 |
+
"learning_rate": 2.6412087211526203e-05,
|
1624 |
+
"loss": 1.5901,
|
1625 |
+
"step": 23500
|
1626 |
+
},
|
1627 |
+
{
|
1628 |
+
"epoch": 2.67,
|
1629 |
+
"learning_rate": 2.6284584980237154e-05,
|
1630 |
+
"loss": 1.5913,
|
1631 |
+
"step": 23600
|
1632 |
+
},
|
1633 |
+
{
|
1634 |
+
"epoch": 2.68,
|
1635 |
+
"learning_rate": 2.615708274894811e-05,
|
1636 |
+
"loss": 1.6249,
|
1637 |
+
"step": 23700
|
1638 |
+
},
|
1639 |
+
{
|
1640 |
+
"epoch": 2.69,
|
1641 |
+
"learning_rate": 2.6029580517659057e-05,
|
1642 |
+
"loss": 1.6162,
|
1643 |
+
"step": 23800
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 2.7,
|
1647 |
+
"learning_rate": 2.5902078286370012e-05,
|
1648 |
+
"loss": 1.5866,
|
1649 |
+
"step": 23900
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 2.71,
|
1653 |
+
"learning_rate": 2.5774576055080967e-05,
|
1654 |
+
"loss": 1.6091,
|
1655 |
+
"step": 24000
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 2.71,
|
1659 |
+
"eval_loss": 1.544262170791626,
|
1660 |
+
"eval_masked_accuracy": 0.6812164740363877,
|
1661 |
+
"eval_runtime": 332.6866,
|
1662 |
+
"eval_samples_per_second": 47.255,
|
1663 |
+
"eval_steps_per_second": 2.955,
|
1664 |
+
"step": 24000
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 2.73,
|
1668 |
+
"learning_rate": 2.5647073823791918e-05,
|
1669 |
+
"loss": 1.5909,
|
1670 |
+
"step": 24100
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 2.74,
|
1674 |
+
"learning_rate": 2.5519571592502873e-05,
|
1675 |
+
"loss": 1.5965,
|
1676 |
+
"step": 24200
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 2.75,
|
1680 |
+
"learning_rate": 2.539206936121382e-05,
|
1681 |
+
"loss": 1.5999,
|
1682 |
+
"step": 24300
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 2.76,
|
1686 |
+
"learning_rate": 2.5264567129924776e-05,
|
1687 |
+
"loss": 1.6052,
|
1688 |
+
"step": 24400
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 2.77,
|
1692 |
+
"learning_rate": 2.5137064898635728e-05,
|
1693 |
+
"loss": 1.5817,
|
1694 |
+
"step": 24500
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 2.78,
|
1698 |
+
"learning_rate": 2.5009562667346683e-05,
|
1699 |
+
"loss": 1.6076,
|
1700 |
+
"step": 24600
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 2.79,
|
1704 |
+
"learning_rate": 2.4882060436057634e-05,
|
1705 |
+
"loss": 1.6049,
|
1706 |
+
"step": 24700
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 2.8,
|
1710 |
+
"learning_rate": 2.4754558204768586e-05,
|
1711 |
+
"loss": 1.5737,
|
1712 |
+
"step": 24800
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 2.82,
|
1716 |
+
"learning_rate": 2.4627055973479537e-05,
|
1717 |
+
"loss": 1.5943,
|
1718 |
+
"step": 24900
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 2.83,
|
1722 |
+
"learning_rate": 2.449955374219049e-05,
|
1723 |
+
"loss": 1.5687,
|
1724 |
+
"step": 25000
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 2.83,
|
1728 |
+
"eval_loss": 1.5361570119857788,
|
1729 |
+
"eval_masked_accuracy": 0.6823867711149114,
|
1730 |
+
"eval_runtime": 332.6448,
|
1731 |
+
"eval_samples_per_second": 47.261,
|
1732 |
+
"eval_steps_per_second": 2.955,
|
1733 |
+
"step": 25000
|
1734 |
+
},
|
1735 |
+
{
|
1736 |
+
"epoch": 2.84,
|
1737 |
+
"learning_rate": 2.4372051510901443e-05,
|
1738 |
+
"loss": 1.5948,
|
1739 |
+
"step": 25100
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 2.85,
|
1743 |
+
"learning_rate": 2.4244549279612395e-05,
|
1744 |
+
"loss": 1.5808,
|
1745 |
+
"step": 25200
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 2.86,
|
1749 |
+
"learning_rate": 2.4117047048323346e-05,
|
1750 |
+
"loss": 1.5756,
|
1751 |
+
"step": 25300
|
1752 |
+
},
|
1753 |
+
{
|
1754 |
+
"epoch": 2.87,
|
1755 |
+
"learning_rate": 2.3989544817034298e-05,
|
1756 |
+
"loss": 1.5723,
|
1757 |
+
"step": 25400
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 2.88,
|
1761 |
+
"learning_rate": 2.386204258574525e-05,
|
1762 |
+
"loss": 1.5604,
|
1763 |
+
"step": 25500
|
1764 |
+
},
|
1765 |
+
{
|
1766 |
+
"epoch": 2.89,
|
1767 |
+
"learning_rate": 2.3734540354456204e-05,
|
1768 |
+
"loss": 1.5761,
|
1769 |
+
"step": 25600
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 2.91,
|
1773 |
+
"learning_rate": 2.360703812316716e-05,
|
1774 |
+
"loss": 1.5624,
|
1775 |
+
"step": 25700
|
1776 |
+
},
|
1777 |
+
{
|
1778 |
+
"epoch": 2.92,
|
1779 |
+
"learning_rate": 2.347953589187811e-05,
|
1780 |
+
"loss": 1.5607,
|
1781 |
+
"step": 25800
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 2.93,
|
1785 |
+
"learning_rate": 2.3352033660589062e-05,
|
1786 |
+
"loss": 1.5428,
|
1787 |
+
"step": 25900
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 2.94,
|
1791 |
+
"learning_rate": 2.3224531429300014e-05,
|
1792 |
+
"loss": 1.5725,
|
1793 |
+
"step": 26000
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 2.94,
|
1797 |
+
"eval_loss": 1.537841796875,
|
1798 |
+
"eval_masked_accuracy": 0.6820021651738,
|
1799 |
+
"eval_runtime": 332.7089,
|
1800 |
+
"eval_samples_per_second": 47.252,
|
1801 |
+
"eval_steps_per_second": 2.955,
|
1802 |
+
"step": 26000
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 2.95,
|
1806 |
+
"learning_rate": 2.3097029198010965e-05,
|
1807 |
+
"loss": 1.5788,
|
1808 |
+
"step": 26100
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 2.96,
|
1812 |
+
"learning_rate": 2.296952696672192e-05,
|
1813 |
+
"loss": 1.5641,
|
1814 |
+
"step": 26200
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 2.97,
|
1818 |
+
"learning_rate": 2.284202473543287e-05,
|
1819 |
+
"loss": 1.5451,
|
1820 |
+
"step": 26300
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 2.99,
|
1824 |
+
"learning_rate": 2.2714522504143823e-05,
|
1825 |
+
"loss": 1.6074,
|
1826 |
+
"step": 26400
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 3.0,
|
1830 |
+
"learning_rate": 2.2587020272854775e-05,
|
1831 |
+
"loss": 1.5966,
|
1832 |
+
"step": 26500
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 3.01,
|
1836 |
+
"learning_rate": 2.2459518041565726e-05,
|
1837 |
+
"loss": 1.5088,
|
1838 |
+
"step": 26600
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 3.02,
|
1842 |
+
"learning_rate": 2.233201581027668e-05,
|
1843 |
+
"loss": 1.5411,
|
1844 |
+
"step": 26700
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 3.03,
|
1848 |
+
"learning_rate": 2.2204513578987636e-05,
|
1849 |
+
"loss": 1.4903,
|
1850 |
+
"step": 26800
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 3.04,
|
1854 |
+
"learning_rate": 2.2077011347698587e-05,
|
1855 |
+
"loss": 1.5135,
|
1856 |
+
"step": 26900
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 3.05,
|
1860 |
+
"learning_rate": 2.194950911640954e-05,
|
1861 |
+
"loss": 1.5171,
|
1862 |
+
"step": 27000
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 3.05,
|
1866 |
+
"eval_loss": 1.5317901372909546,
|
1867 |
+
"eval_masked_accuracy": 0.683750220460841,
|
1868 |
+
"eval_runtime": 332.6043,
|
1869 |
+
"eval_samples_per_second": 47.266,
|
1870 |
+
"eval_steps_per_second": 2.955,
|
1871 |
+
"step": 27000
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 3.06,
|
1875 |
+
"learning_rate": 2.182200688512049e-05,
|
1876 |
+
"loss": 1.5076,
|
1877 |
+
"step": 27100
|
1878 |
+
},
|
1879 |
+
{
|
1880 |
+
"epoch": 3.08,
|
1881 |
+
"learning_rate": 2.1694504653831442e-05,
|
1882 |
+
"loss": 1.5401,
|
1883 |
+
"step": 27200
|
1884 |
+
},
|
1885 |
+
{
|
1886 |
+
"epoch": 3.09,
|
1887 |
+
"learning_rate": 2.1567002422542397e-05,
|
1888 |
+
"loss": 1.5463,
|
1889 |
+
"step": 27300
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 3.1,
|
1893 |
+
"learning_rate": 2.1439500191253348e-05,
|
1894 |
+
"loss": 1.4955,
|
1895 |
+
"step": 27400
|
1896 |
+
},
|
1897 |
+
{
|
1898 |
+
"epoch": 3.11,
|
1899 |
+
"learning_rate": 2.13119979599643e-05,
|
1900 |
+
"loss": 1.5169,
|
1901 |
+
"step": 27500
|
1902 |
+
},
|
1903 |
+
{
|
1904 |
+
"epoch": 3.12,
|
1905 |
+
"learning_rate": 2.118449572867525e-05,
|
1906 |
+
"loss": 1.5295,
|
1907 |
+
"step": 27600
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 3.13,
|
1911 |
+
"learning_rate": 2.1056993497386203e-05,
|
1912 |
+
"loss": 1.5197,
|
1913 |
+
"step": 27700
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 3.14,
|
1917 |
+
"learning_rate": 2.0929491266097158e-05,
|
1918 |
+
"loss": 1.5525,
|
1919 |
+
"step": 27800
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 3.16,
|
1923 |
+
"learning_rate": 2.0801989034808113e-05,
|
1924 |
+
"loss": 1.5558,
|
1925 |
+
"step": 27900
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 3.17,
|
1929 |
+
"learning_rate": 2.0674486803519064e-05,
|
1930 |
+
"loss": 1.4809,
|
1931 |
+
"step": 28000
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 3.17,
|
1935 |
+
"eval_loss": 1.5124375820159912,
|
1936 |
+
"eval_masked_accuracy": 0.6863205757142956,
|
1937 |
+
"eval_runtime": 332.3275,
|
1938 |
+
"eval_samples_per_second": 47.306,
|
1939 |
+
"eval_steps_per_second": 2.958,
|
1940 |
+
"step": 28000
|
1941 |
+
},
|
1942 |
+
{
|
1943 |
+
"epoch": 3.18,
|
1944 |
+
"learning_rate": 2.0546984572230016e-05,
|
1945 |
+
"loss": 1.4992,
|
1946 |
+
"step": 28100
|
1947 |
+
},
|
1948 |
+
{
|
1949 |
+
"epoch": 3.19,
|
1950 |
+
"learning_rate": 2.0419482340940967e-05,
|
1951 |
+
"loss": 1.5481,
|
1952 |
+
"step": 28200
|
1953 |
+
},
|
1954 |
+
{
|
1955 |
+
"epoch": 3.2,
|
1956 |
+
"learning_rate": 2.0291980109651922e-05,
|
1957 |
+
"loss": 1.5201,
|
1958 |
+
"step": 28300
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 3.21,
|
1962 |
+
"learning_rate": 2.0164477878362873e-05,
|
1963 |
+
"loss": 1.5024,
|
1964 |
+
"step": 28400
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 3.22,
|
1968 |
+
"learning_rate": 2.0036975647073825e-05,
|
1969 |
+
"loss": 1.56,
|
1970 |
+
"step": 28500
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 3.23,
|
1974 |
+
"learning_rate": 1.9909473415784776e-05,
|
1975 |
+
"loss": 1.5406,
|
1976 |
+
"step": 28600
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 3.25,
|
1980 |
+
"learning_rate": 1.9781971184495728e-05,
|
1981 |
+
"loss": 1.5089,
|
1982 |
+
"step": 28700
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 3.26,
|
1986 |
+
"learning_rate": 1.9654468953206683e-05,
|
1987 |
+
"loss": 1.4574,
|
1988 |
+
"step": 28800
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 3.27,
|
1992 |
+
"learning_rate": 1.9526966721917634e-05,
|
1993 |
+
"loss": 1.4856,
|
1994 |
+
"step": 28900
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 3.28,
|
1998 |
+
"learning_rate": 1.939946449062859e-05,
|
1999 |
+
"loss": 1.5158,
|
2000 |
+
"step": 29000
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 3.28,
|
2004 |
+
"eval_loss": 1.4922306537628174,
|
2005 |
+
"eval_masked_accuracy": 0.690069818885663,
|
2006 |
+
"eval_runtime": 332.2549,
|
2007 |
+
"eval_samples_per_second": 47.316,
|
2008 |
+
"eval_steps_per_second": 2.959,
|
2009 |
+
"step": 29000
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 3.29,
|
2013 |
+
"learning_rate": 1.927196225933954e-05,
|
2014 |
+
"loss": 1.5297,
|
2015 |
+
"step": 29100
|
2016 |
+
},
|
2017 |
+
{
|
2018 |
+
"epoch": 3.3,
|
2019 |
+
"learning_rate": 1.9144460028050492e-05,
|
2020 |
+
"loss": 1.4847,
|
2021 |
+
"step": 29200
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 3.31,
|
2025 |
+
"learning_rate": 1.9016957796761444e-05,
|
2026 |
+
"loss": 1.5051,
|
2027 |
+
"step": 29300
|
2028 |
+
},
|
2029 |
+
{
|
2030 |
+
"epoch": 3.32,
|
2031 |
+
"learning_rate": 1.88894555654724e-05,
|
2032 |
+
"loss": 1.5038,
|
2033 |
+
"step": 29400
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 3.34,
|
2037 |
+
"learning_rate": 1.876195333418335e-05,
|
2038 |
+
"loss": 1.5262,
|
2039 |
+
"step": 29500
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 3.35,
|
2043 |
+
"learning_rate": 1.86344511028943e-05,
|
2044 |
+
"loss": 1.4712,
|
2045 |
+
"step": 29600
|
2046 |
+
},
|
2047 |
+
{
|
2048 |
+
"epoch": 3.36,
|
2049 |
+
"learning_rate": 1.8506948871605253e-05,
|
2050 |
+
"loss": 1.5132,
|
2051 |
+
"step": 29700
|
2052 |
+
},
|
2053 |
+
{
|
2054 |
+
"epoch": 3.37,
|
2055 |
+
"learning_rate": 1.8379446640316205e-05,
|
2056 |
+
"loss": 1.5035,
|
2057 |
+
"step": 29800
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 3.38,
|
2061 |
+
"learning_rate": 1.825194440902716e-05,
|
2062 |
+
"loss": 1.5093,
|
2063 |
+
"step": 29900
|
2064 |
+
},
|
2065 |
+
{
|
2066 |
+
"epoch": 3.39,
|
2067 |
+
"learning_rate": 1.812444217773811e-05,
|
2068 |
+
"loss": 1.4825,
|
2069 |
+
"step": 30000
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 3.39,
|
2073 |
+
"eval_loss": 1.4842137098312378,
|
2074 |
+
"eval_masked_accuracy": 0.6907112508204083,
|
2075 |
+
"eval_runtime": 332.0229,
|
2076 |
+
"eval_samples_per_second": 47.349,
|
2077 |
+
"eval_steps_per_second": 2.961,
|
2078 |
+
"step": 30000
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 3.4,
|
2082 |
+
"learning_rate": 1.7996939946449062e-05,
|
2083 |
+
"loss": 1.5421,
|
2084 |
+
"step": 30100
|
2085 |
+
},
|
2086 |
+
{
|
2087 |
+
"epoch": 3.42,
|
2088 |
+
"learning_rate": 1.7869437715160017e-05,
|
2089 |
+
"loss": 1.5267,
|
2090 |
+
"step": 30200
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 3.43,
|
2094 |
+
"learning_rate": 1.774193548387097e-05,
|
2095 |
+
"loss": 1.4855,
|
2096 |
+
"step": 30300
|
2097 |
+
},
|
2098 |
+
{
|
2099 |
+
"epoch": 3.44,
|
2100 |
+
"learning_rate": 1.761443325258192e-05,
|
2101 |
+
"loss": 1.4843,
|
2102 |
+
"step": 30400
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 3.45,
|
2106 |
+
"learning_rate": 1.7486931021292875e-05,
|
2107 |
+
"loss": 1.52,
|
2108 |
+
"step": 30500
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 3.46,
|
2112 |
+
"learning_rate": 1.7359428790003827e-05,
|
2113 |
+
"loss": 1.5077,
|
2114 |
+
"step": 30600
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 3.47,
|
2118 |
+
"learning_rate": 1.7231926558714778e-05,
|
2119 |
+
"loss": 1.4908,
|
2120 |
+
"step": 30700
|
2121 |
+
},
|
2122 |
+
{
|
2123 |
+
"epoch": 3.48,
|
2124 |
+
"learning_rate": 1.710442432742573e-05,
|
2125 |
+
"loss": 1.5223,
|
2126 |
+
"step": 30800
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 3.49,
|
2130 |
+
"learning_rate": 1.697692209613668e-05,
|
2131 |
+
"loss": 1.5103,
|
2132 |
+
"step": 30900
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 3.51,
|
2136 |
+
"learning_rate": 1.6849419864847636e-05,
|
2137 |
+
"loss": 1.4568,
|
2138 |
+
"step": 31000
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 3.51,
|
2142 |
+
"eval_loss": 1.4819023609161377,
|
2143 |
+
"eval_masked_accuracy": 0.6908566639188486,
|
2144 |
+
"eval_runtime": 332.2827,
|
2145 |
+
"eval_samples_per_second": 47.312,
|
2146 |
+
"eval_steps_per_second": 2.958,
|
2147 |
+
"step": 31000
|
2148 |
+
},
|
2149 |
+
{
|
2150 |
+
"epoch": 3.52,
|
2151 |
+
"learning_rate": 1.6721917633558588e-05,
|
2152 |
+
"loss": 1.5105,
|
2153 |
+
"step": 31100
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 3.53,
|
2157 |
+
"learning_rate": 1.659441540226954e-05,
|
2158 |
+
"loss": 1.4714,
|
2159 |
+
"step": 31200
|
2160 |
+
},
|
2161 |
+
{
|
2162 |
+
"epoch": 3.54,
|
2163 |
+
"learning_rate": 1.6466913170980494e-05,
|
2164 |
+
"loss": 1.5056,
|
2165 |
+
"step": 31300
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 3.55,
|
2169 |
+
"learning_rate": 1.6339410939691445e-05,
|
2170 |
+
"loss": 1.4777,
|
2171 |
+
"step": 31400
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 3.56,
|
2175 |
+
"learning_rate": 1.6211908708402397e-05,
|
2176 |
+
"loss": 1.4735,
|
2177 |
+
"step": 31500
|
2178 |
+
},
|
2179 |
+
{
|
2180 |
+
"epoch": 3.57,
|
2181 |
+
"learning_rate": 1.6084406477113352e-05,
|
2182 |
+
"loss": 1.4784,
|
2183 |
+
"step": 31600
|
2184 |
+
},
|
2185 |
+
{
|
2186 |
+
"epoch": 3.58,
|
2187 |
+
"learning_rate": 1.5956904245824303e-05,
|
2188 |
+
"loss": 1.5037,
|
2189 |
+
"step": 31700
|
2190 |
+
},
|
2191 |
+
{
|
2192 |
+
"epoch": 3.6,
|
2193 |
+
"learning_rate": 1.5829402014535255e-05,
|
2194 |
+
"loss": 1.5295,
|
2195 |
+
"step": 31800
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 3.61,
|
2199 |
+
"learning_rate": 1.5701899783246206e-05,
|
2200 |
+
"loss": 1.4655,
|
2201 |
+
"step": 31900
|
2202 |
+
},
|
2203 |
+
{
|
2204 |
+
"epoch": 3.62,
|
2205 |
+
"learning_rate": 1.5574397551957158e-05,
|
2206 |
+
"loss": 1.4868,
|
2207 |
+
"step": 32000
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 3.62,
|
2211 |
+
"eval_loss": 1.4802453517913818,
|
2212 |
+
"eval_masked_accuracy": 0.691725791186244,
|
2213 |
+
"eval_runtime": 332.2068,
|
2214 |
+
"eval_samples_per_second": 47.323,
|
2215 |
+
"eval_steps_per_second": 2.959,
|
2216 |
+
"step": 32000
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 3.63,
|
2220 |
+
"learning_rate": 1.5446895320668113e-05,
|
2221 |
+
"loss": 1.5108,
|
2222 |
+
"step": 32100
|
2223 |
+
},
|
2224 |
+
{
|
2225 |
+
"epoch": 3.64,
|
2226 |
+
"learning_rate": 1.5319393089379064e-05,
|
2227 |
+
"loss": 1.5234,
|
2228 |
+
"step": 32200
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 3.65,
|
2232 |
+
"learning_rate": 1.5191890858090016e-05,
|
2233 |
+
"loss": 1.4669,
|
2234 |
+
"step": 32300
|
2235 |
+
},
|
2236 |
+
{
|
2237 |
+
"epoch": 3.66,
|
2238 |
+
"learning_rate": 1.506438862680097e-05,
|
2239 |
+
"loss": 1.4555,
|
2240 |
+
"step": 32400
|
2241 |
+
},
|
2242 |
+
{
|
2243 |
+
"epoch": 3.68,
|
2244 |
+
"learning_rate": 1.4936886395511924e-05,
|
2245 |
+
"loss": 1.4447,
|
2246 |
+
"step": 32500
|
2247 |
+
},
|
2248 |
+
{
|
2249 |
+
"epoch": 3.69,
|
2250 |
+
"learning_rate": 1.4809384164222875e-05,
|
2251 |
+
"loss": 1.4671,
|
2252 |
+
"step": 32600
|
2253 |
+
},
|
2254 |
+
{
|
2255 |
+
"epoch": 3.7,
|
2256 |
+
"learning_rate": 1.4681881932933827e-05,
|
2257 |
+
"loss": 1.4739,
|
2258 |
+
"step": 32700
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 3.71,
|
2262 |
+
"learning_rate": 1.455437970164478e-05,
|
2263 |
+
"loss": 1.4673,
|
2264 |
+
"step": 32800
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 3.72,
|
2268 |
+
"learning_rate": 1.4426877470355732e-05,
|
2269 |
+
"loss": 1.4821,
|
2270 |
+
"step": 32900
|
2271 |
+
},
|
2272 |
+
{
|
2273 |
+
"epoch": 3.73,
|
2274 |
+
"learning_rate": 1.4299375239066685e-05,
|
2275 |
+
"loss": 1.4826,
|
2276 |
+
"step": 33000
|
2277 |
+
},
|
2278 |
+
{
|
2279 |
+
"epoch": 3.73,
|
2280 |
+
"eval_loss": 1.466788649559021,
|
2281 |
+
"eval_masked_accuracy": 0.6936922350866194,
|
2282 |
+
"eval_runtime": 332.3241,
|
2283 |
+
"eval_samples_per_second": 47.306,
|
2284 |
+
"eval_steps_per_second": 2.958,
|
2285 |
+
"step": 33000
|
2286 |
+
},
|
2287 |
+
{
|
2288 |
+
"epoch": 3.74,
|
2289 |
+
"learning_rate": 1.4171873007777636e-05,
|
2290 |
+
"loss": 1.4993,
|
2291 |
+
"step": 33100
|
2292 |
+
},
|
2293 |
+
{
|
2294 |
+
"epoch": 3.75,
|
2295 |
+
"learning_rate": 1.4044370776488588e-05,
|
2296 |
+
"loss": 1.498,
|
2297 |
+
"step": 33200
|
2298 |
+
},
|
2299 |
+
{
|
2300 |
+
"epoch": 3.77,
|
2301 |
+
"learning_rate": 1.3916868545199541e-05,
|
2302 |
+
"loss": 1.4549,
|
2303 |
+
"step": 33300
|
2304 |
+
},
|
2305 |
+
{
|
2306 |
+
"epoch": 3.78,
|
2307 |
+
"learning_rate": 1.3789366313910492e-05,
|
2308 |
+
"loss": 1.4276,
|
2309 |
+
"step": 33400
|
2310 |
+
},
|
2311 |
+
{
|
2312 |
+
"epoch": 3.79,
|
2313 |
+
"learning_rate": 1.3661864082621447e-05,
|
2314 |
+
"loss": 1.4775,
|
2315 |
+
"step": 33500
|
2316 |
+
},
|
2317 |
+
{
|
2318 |
+
"epoch": 3.8,
|
2319 |
+
"learning_rate": 1.35343618513324e-05,
|
2320 |
+
"loss": 1.5056,
|
2321 |
+
"step": 33600
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 3.81,
|
2325 |
+
"learning_rate": 1.3406859620043352e-05,
|
2326 |
+
"loss": 1.4488,
|
2327 |
+
"step": 33700
|
2328 |
+
},
|
2329 |
+
{
|
2330 |
+
"epoch": 3.82,
|
2331 |
+
"learning_rate": 1.3279357388754305e-05,
|
2332 |
+
"loss": 1.46,
|
2333 |
+
"step": 33800
|
2334 |
+
},
|
2335 |
+
{
|
2336 |
+
"epoch": 3.83,
|
2337 |
+
"learning_rate": 1.3151855157465257e-05,
|
2338 |
+
"loss": 1.4588,
|
2339 |
+
"step": 33900
|
2340 |
+
},
|
2341 |
+
{
|
2342 |
+
"epoch": 3.84,
|
2343 |
+
"learning_rate": 1.3024352926176208e-05,
|
2344 |
+
"loss": 1.4686,
|
2345 |
+
"step": 34000
|
2346 |
+
},
|
2347 |
+
{
|
2348 |
+
"epoch": 3.84,
|
2349 |
+
"eval_loss": 1.4628050327301025,
|
2350 |
+
"eval_masked_accuracy": 0.695108458894874,
|
2351 |
+
"eval_runtime": 332.2343,
|
2352 |
+
"eval_samples_per_second": 47.319,
|
2353 |
+
"eval_steps_per_second": 2.959,
|
2354 |
+
"step": 34000
|
2355 |
+
},
|
2356 |
+
{
|
2357 |
+
"epoch": 3.86,
|
2358 |
+
"learning_rate": 1.2896850694887161e-05,
|
2359 |
+
"loss": 1.4673,
|
2360 |
+
"step": 34100
|
2361 |
+
},
|
2362 |
+
{
|
2363 |
+
"epoch": 3.87,
|
2364 |
+
"learning_rate": 1.2769348463598113e-05,
|
2365 |
+
"loss": 1.4501,
|
2366 |
+
"step": 34200
|
2367 |
+
},
|
2368 |
+
{
|
2369 |
+
"epoch": 3.88,
|
2370 |
+
"learning_rate": 1.2641846232309066e-05,
|
2371 |
+
"loss": 1.4532,
|
2372 |
+
"step": 34300
|
2373 |
+
},
|
2374 |
+
{
|
2375 |
+
"epoch": 3.89,
|
2376 |
+
"learning_rate": 1.2514344001020018e-05,
|
2377 |
+
"loss": 1.4439,
|
2378 |
+
"step": 34400
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 3.9,
|
2382 |
+
"learning_rate": 1.238684176973097e-05,
|
2383 |
+
"loss": 1.4202,
|
2384 |
+
"step": 34500
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 3.91,
|
2388 |
+
"learning_rate": 1.2259339538441924e-05,
|
2389 |
+
"loss": 1.4436,
|
2390 |
+
"step": 34600
|
2391 |
+
},
|
2392 |
+
{
|
2393 |
+
"epoch": 3.92,
|
2394 |
+
"learning_rate": 1.2131837307152875e-05,
|
2395 |
+
"loss": 1.466,
|
2396 |
+
"step": 34700
|
2397 |
+
},
|
2398 |
+
{
|
2399 |
+
"epoch": 3.94,
|
2400 |
+
"learning_rate": 1.2004335075863827e-05,
|
2401 |
+
"loss": 1.4696,
|
2402 |
+
"step": 34800
|
2403 |
+
},
|
2404 |
+
{
|
2405 |
+
"epoch": 3.95,
|
2406 |
+
"learning_rate": 1.1876832844574782e-05,
|
2407 |
+
"loss": 1.4775,
|
2408 |
+
"step": 34900
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 3.96,
|
2412 |
+
"learning_rate": 1.1749330613285733e-05,
|
2413 |
+
"loss": 1.4343,
|
2414 |
+
"step": 35000
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 3.96,
|
2418 |
+
"eval_loss": 1.4424548149108887,
|
2419 |
+
"eval_masked_accuracy": 0.6971389490482162,
|
2420 |
+
"eval_runtime": 331.9906,
|
2421 |
+
"eval_samples_per_second": 47.354,
|
2422 |
+
"eval_steps_per_second": 2.961,
|
2423 |
+
"step": 35000
|
2424 |
+
},
|
2425 |
+
{
|
2426 |
+
"epoch": 3.97,
|
2427 |
+
"learning_rate": 1.1621828381996685e-05,
|
2428 |
+
"loss": 1.4501,
|
2429 |
+
"step": 35100
|
2430 |
+
},
|
2431 |
+
{
|
2432 |
+
"epoch": 3.98,
|
2433 |
+
"learning_rate": 1.1494326150707638e-05,
|
2434 |
+
"loss": 1.4867,
|
2435 |
+
"step": 35200
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 3.99,
|
2439 |
+
"learning_rate": 1.136682391941859e-05,
|
2440 |
+
"loss": 1.4403,
|
2441 |
+
"step": 35300
|
2442 |
+
},
|
2443 |
+
{
|
2444 |
+
"epoch": 4.0,
|
2445 |
+
"learning_rate": 1.1239321688129543e-05,
|
2446 |
+
"loss": 1.4668,
|
2447 |
+
"step": 35400
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 4.01,
|
2451 |
+
"learning_rate": 1.1111819456840496e-05,
|
2452 |
+
"loss": 1.3951,
|
2453 |
+
"step": 35500
|
2454 |
+
},
|
2455 |
+
{
|
2456 |
+
"epoch": 4.03,
|
2457 |
+
"learning_rate": 1.0984317225551447e-05,
|
2458 |
+
"loss": 1.4265,
|
2459 |
+
"step": 35600
|
2460 |
+
},
|
2461 |
+
{
|
2462 |
+
"epoch": 4.04,
|
2463 |
+
"learning_rate": 1.08568149942624e-05,
|
2464 |
+
"loss": 1.4097,
|
2465 |
+
"step": 35700
|
2466 |
+
},
|
2467 |
+
{
|
2468 |
+
"epoch": 4.05,
|
2469 |
+
"learning_rate": 1.0729312762973352e-05,
|
2470 |
+
"loss": 1.4165,
|
2471 |
+
"step": 35800
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 4.06,
|
2475 |
+
"learning_rate": 1.0601810531684305e-05,
|
2476 |
+
"loss": 1.3991,
|
2477 |
+
"step": 35900
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 4.07,
|
2481 |
+
"learning_rate": 1.0474308300395258e-05,
|
2482 |
+
"loss": 1.4286,
|
2483 |
+
"step": 36000
|
2484 |
+
},
|
2485 |
+
{
|
2486 |
+
"epoch": 4.07,
|
2487 |
+
"eval_loss": 1.4411929845809937,
|
2488 |
+
"eval_masked_accuracy": 0.6979967481246483,
|
2489 |
+
"eval_runtime": 332.6761,
|
2490 |
+
"eval_samples_per_second": 47.256,
|
2491 |
+
"eval_steps_per_second": 2.955,
|
2492 |
+
"step": 36000
|
2493 |
+
},
|
2494 |
+
{
|
2495 |
+
"epoch": 4.08,
|
2496 |
+
"learning_rate": 1.034680606910621e-05,
|
2497 |
+
"loss": 1.3967,
|
2498 |
+
"step": 36100
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 4.09,
|
2502 |
+
"learning_rate": 1.0219303837817163e-05,
|
2503 |
+
"loss": 1.4079,
|
2504 |
+
"step": 36200
|
2505 |
+
},
|
2506 |
+
{
|
2507 |
+
"epoch": 4.1,
|
2508 |
+
"learning_rate": 1.0091801606528115e-05,
|
2509 |
+
"loss": 1.4176,
|
2510 |
+
"step": 36300
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 4.12,
|
2514 |
+
"learning_rate": 9.964299375239066e-06,
|
2515 |
+
"loss": 1.4068,
|
2516 |
+
"step": 36400
|
2517 |
+
},
|
2518 |
+
{
|
2519 |
+
"epoch": 4.13,
|
2520 |
+
"learning_rate": 9.83679714395002e-06,
|
2521 |
+
"loss": 1.3876,
|
2522 |
+
"step": 36500
|
2523 |
+
},
|
2524 |
+
{
|
2525 |
+
"epoch": 4.14,
|
2526 |
+
"learning_rate": 9.709294912660973e-06,
|
2527 |
+
"loss": 1.424,
|
2528 |
+
"step": 36600
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 4.15,
|
2532 |
+
"learning_rate": 9.581792681371924e-06,
|
2533 |
+
"loss": 1.3789,
|
2534 |
+
"step": 36700
|
2535 |
+
},
|
2536 |
+
{
|
2537 |
+
"epoch": 4.16,
|
2538 |
+
"learning_rate": 9.454290450082877e-06,
|
2539 |
+
"loss": 1.4157,
|
2540 |
+
"step": 36800
|
2541 |
+
},
|
2542 |
+
{
|
2543 |
+
"epoch": 4.17,
|
2544 |
+
"learning_rate": 9.326788218793829e-06,
|
2545 |
+
"loss": 1.394,
|
2546 |
+
"step": 36900
|
2547 |
+
},
|
2548 |
+
{
|
2549 |
+
"epoch": 4.18,
|
2550 |
+
"learning_rate": 9.199285987504782e-06,
|
2551 |
+
"loss": 1.439,
|
2552 |
+
"step": 37000
|
2553 |
+
},
|
2554 |
+
{
|
2555 |
+
"epoch": 4.18,
|
2556 |
+
"eval_loss": 1.4216623306274414,
|
2557 |
+
"eval_masked_accuracy": 0.7010983390128082,
|
2558 |
+
"eval_runtime": 332.6207,
|
2559 |
+
"eval_samples_per_second": 47.264,
|
2560 |
+
"eval_steps_per_second": 2.955,
|
2561 |
+
"step": 37000
|
2562 |
+
},
|
2563 |
+
{
|
2564 |
+
"epoch": 4.2,
|
2565 |
+
"learning_rate": 9.071783756215735e-06,
|
2566 |
+
"loss": 1.4001,
|
2567 |
+
"step": 37100
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 4.21,
|
2571 |
+
"learning_rate": 8.944281524926687e-06,
|
2572 |
+
"loss": 1.4253,
|
2573 |
+
"step": 37200
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 4.22,
|
2577 |
+
"learning_rate": 8.81677929363764e-06,
|
2578 |
+
"loss": 1.4032,
|
2579 |
+
"step": 37300
|
2580 |
+
},
|
2581 |
+
{
|
2582 |
+
"epoch": 4.23,
|
2583 |
+
"learning_rate": 8.689277062348591e-06,
|
2584 |
+
"loss": 1.4022,
|
2585 |
+
"step": 37400
|
2586 |
+
},
|
2587 |
+
{
|
2588 |
+
"epoch": 4.24,
|
2589 |
+
"learning_rate": 8.561774831059543e-06,
|
2590 |
+
"loss": 1.3839,
|
2591 |
+
"step": 37500
|
2592 |
+
},
|
2593 |
+
{
|
2594 |
+
"epoch": 4.25,
|
2595 |
+
"learning_rate": 8.434272599770496e-06,
|
2596 |
+
"loss": 1.419,
|
2597 |
+
"step": 37600
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 4.26,
|
2601 |
+
"learning_rate": 8.30677036848145e-06,
|
2602 |
+
"loss": 1.3744,
|
2603 |
+
"step": 37700
|
2604 |
+
},
|
2605 |
+
{
|
2606 |
+
"epoch": 4.27,
|
2607 |
+
"learning_rate": 8.1792681371924e-06,
|
2608 |
+
"loss": 1.438,
|
2609 |
+
"step": 37800
|
2610 |
+
},
|
2611 |
+
{
|
2612 |
+
"epoch": 4.29,
|
2613 |
+
"learning_rate": 8.051765905903354e-06,
|
2614 |
+
"loss": 1.3998,
|
2615 |
+
"step": 37900
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 4.3,
|
2619 |
+
"learning_rate": 7.924263674614305e-06,
|
2620 |
+
"loss": 1.4095,
|
2621 |
+
"step": 38000
|
2622 |
+
},
|
2623 |
+
{
|
2624 |
+
"epoch": 4.3,
|
2625 |
+
"eval_loss": 1.4295740127563477,
|
2626 |
+
"eval_masked_accuracy": 0.7001421356222228,
|
2627 |
+
"eval_runtime": 332.5672,
|
2628 |
+
"eval_samples_per_second": 47.272,
|
2629 |
+
"eval_steps_per_second": 2.956,
|
2630 |
+
"step": 38000
|
2631 |
+
},
|
2632 |
+
{
|
2633 |
+
"epoch": 4.31,
|
2634 |
+
"learning_rate": 7.796761443325259e-06,
|
2635 |
+
"loss": 1.4015,
|
2636 |
+
"step": 38100
|
2637 |
+
},
|
2638 |
+
{
|
2639 |
+
"epoch": 4.32,
|
2640 |
+
"learning_rate": 7.669259212036212e-06,
|
2641 |
+
"loss": 1.4069,
|
2642 |
+
"step": 38200
|
2643 |
+
},
|
2644 |
+
{
|
2645 |
+
"epoch": 4.33,
|
2646 |
+
"learning_rate": 7.541756980747164e-06,
|
2647 |
+
"loss": 1.3908,
|
2648 |
+
"step": 38300
|
2649 |
+
},
|
2650 |
+
{
|
2651 |
+
"epoch": 4.34,
|
2652 |
+
"learning_rate": 7.4142547494581165e-06,
|
2653 |
+
"loss": 1.3688,
|
2654 |
+
"step": 38400
|
2655 |
+
},
|
2656 |
+
{
|
2657 |
+
"epoch": 4.35,
|
2658 |
+
"learning_rate": 7.286752518169068e-06,
|
2659 |
+
"loss": 1.3833,
|
2660 |
+
"step": 38500
|
2661 |
+
},
|
2662 |
+
{
|
2663 |
+
"epoch": 4.37,
|
2664 |
+
"learning_rate": 7.15925028688002e-06,
|
2665 |
+
"loss": 1.367,
|
2666 |
+
"step": 38600
|
2667 |
+
},
|
2668 |
+
{
|
2669 |
+
"epoch": 4.38,
|
2670 |
+
"learning_rate": 7.031748055590973e-06,
|
2671 |
+
"loss": 1.3697,
|
2672 |
+
"step": 38700
|
2673 |
+
},
|
2674 |
+
{
|
2675 |
+
"epoch": 4.39,
|
2676 |
+
"learning_rate": 6.904245824301926e-06,
|
2677 |
+
"loss": 1.3803,
|
2678 |
+
"step": 38800
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 4.4,
|
2682 |
+
"learning_rate": 6.776743593012878e-06,
|
2683 |
+
"loss": 1.4184,
|
2684 |
+
"step": 38900
|
2685 |
+
},
|
2686 |
+
{
|
2687 |
+
"epoch": 4.41,
|
2688 |
+
"learning_rate": 6.6492413617238305e-06,
|
2689 |
+
"loss": 1.3987,
|
2690 |
+
"step": 39000
|
2691 |
+
},
|
2692 |
+
{
|
2693 |
+
"epoch": 4.41,
|
2694 |
+
"eval_loss": 1.4123162031173706,
|
2695 |
+
"eval_masked_accuracy": 0.702893200667927,
|
2696 |
+
"eval_runtime": 332.6371,
|
2697 |
+
"eval_samples_per_second": 47.262,
|
2698 |
+
"eval_steps_per_second": 2.955,
|
2699 |
+
"step": 39000
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 4.42,
|
2703 |
+
"learning_rate": 6.521739130434783e-06,
|
2704 |
+
"loss": 1.3888,
|
2705 |
+
"step": 39100
|
2706 |
+
},
|
2707 |
+
{
|
2708 |
+
"epoch": 4.43,
|
2709 |
+
"learning_rate": 6.394236899145735e-06,
|
2710 |
+
"loss": 1.3931,
|
2711 |
+
"step": 39200
|
2712 |
+
},
|
2713 |
+
{
|
2714 |
+
"epoch": 4.44,
|
2715 |
+
"learning_rate": 6.266734667856687e-06,
|
2716 |
+
"loss": 1.4084,
|
2717 |
+
"step": 39300
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 4.46,
|
2721 |
+
"learning_rate": 6.13923243656764e-06,
|
2722 |
+
"loss": 1.3912,
|
2723 |
+
"step": 39400
|
2724 |
+
},
|
2725 |
+
{
|
2726 |
+
"epoch": 4.47,
|
2727 |
+
"learning_rate": 6.011730205278593e-06,
|
2728 |
+
"loss": 1.3928,
|
2729 |
+
"step": 39500
|
2730 |
+
},
|
2731 |
+
{
|
2732 |
+
"epoch": 4.48,
|
2733 |
+
"learning_rate": 5.8842279739895455e-06,
|
2734 |
+
"loss": 1.3732,
|
2735 |
+
"step": 39600
|
2736 |
+
},
|
2737 |
+
{
|
2738 |
+
"epoch": 4.49,
|
2739 |
+
"learning_rate": 5.756725742700497e-06,
|
2740 |
+
"loss": 1.4292,
|
2741 |
+
"step": 39700
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 4.5,
|
2745 |
+
"learning_rate": 5.62922351141145e-06,
|
2746 |
+
"loss": 1.378,
|
2747 |
+
"step": 39800
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 4.51,
|
2751 |
+
"learning_rate": 5.5017212801224025e-06,
|
2752 |
+
"loss": 1.384,
|
2753 |
+
"step": 39900
|
2754 |
+
},
|
2755 |
+
{
|
2756 |
+
"epoch": 4.52,
|
2757 |
+
"learning_rate": 5.374219048833355e-06,
|
2758 |
+
"loss": 1.3725,
|
2759 |
+
"step": 40000
|
2760 |
+
},
|
2761 |
+
{
|
2762 |
+
"epoch": 4.52,
|
2763 |
+
"eval_loss": 1.402807354927063,
|
2764 |
+
"eval_masked_accuracy": 0.7036279403495821,
|
2765 |
+
"eval_runtime": 332.6549,
|
2766 |
+
"eval_samples_per_second": 47.259,
|
2767 |
+
"eval_steps_per_second": 2.955,
|
2768 |
+
"step": 40000
|
2769 |
+
},
|
2770 |
+
{
|
2771 |
+
"epoch": 4.53,
|
2772 |
+
"learning_rate": 5.246716817544307e-06,
|
2773 |
+
"loss": 1.3307,
|
2774 |
+
"step": 40100
|
2775 |
+
},
|
2776 |
+
{
|
2777 |
+
"epoch": 4.55,
|
2778 |
+
"learning_rate": 5.1192145862552595e-06,
|
2779 |
+
"loss": 1.3666,
|
2780 |
+
"step": 40200
|
2781 |
+
},
|
2782 |
+
{
|
2783 |
+
"epoch": 4.56,
|
2784 |
+
"learning_rate": 4.991712354966213e-06,
|
2785 |
+
"loss": 1.3802,
|
2786 |
+
"step": 40300
|
2787 |
+
},
|
2788 |
+
{
|
2789 |
+
"epoch": 4.57,
|
2790 |
+
"learning_rate": 4.864210123677164e-06,
|
2791 |
+
"loss": 1.3732,
|
2792 |
+
"step": 40400
|
2793 |
+
},
|
2794 |
+
{
|
2795 |
+
"epoch": 4.58,
|
2796 |
+
"learning_rate": 4.7367078923881166e-06,
|
2797 |
+
"loss": 1.3749,
|
2798 |
+
"step": 40500
|
2799 |
+
},
|
2800 |
+
{
|
2801 |
+
"epoch": 4.59,
|
2802 |
+
"learning_rate": 4.60920566109907e-06,
|
2803 |
+
"loss": 1.4082,
|
2804 |
+
"step": 40600
|
2805 |
+
},
|
2806 |
+
{
|
2807 |
+
"epoch": 4.6,
|
2808 |
+
"learning_rate": 4.481703429810022e-06,
|
2809 |
+
"loss": 1.4303,
|
2810 |
+
"step": 40700
|
2811 |
+
},
|
2812 |
+
{
|
2813 |
+
"epoch": 4.61,
|
2814 |
+
"learning_rate": 4.3542011985209745e-06,
|
2815 |
+
"loss": 1.3938,
|
2816 |
+
"step": 40800
|
2817 |
+
},
|
2818 |
+
{
|
2819 |
+
"epoch": 4.63,
|
2820 |
+
"learning_rate": 4.226698967231927e-06,
|
2821 |
+
"loss": 1.3797,
|
2822 |
+
"step": 40900
|
2823 |
+
},
|
2824 |
+
{
|
2825 |
+
"epoch": 4.64,
|
2826 |
+
"learning_rate": 4.099196735942879e-06,
|
2827 |
+
"loss": 1.3663,
|
2828 |
+
"step": 41000
|
2829 |
+
},
|
2830 |
+
{
|
2831 |
+
"epoch": 4.64,
|
2832 |
+
"eval_loss": 1.4048454761505127,
|
2833 |
+
"eval_masked_accuracy": 0.7044154426496801,
|
2834 |
+
"eval_runtime": 332.5757,
|
2835 |
+
"eval_samples_per_second": 47.27,
|
2836 |
+
"eval_steps_per_second": 2.956,
|
2837 |
+
"step": 41000
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 4.65,
|
2841 |
+
"learning_rate": 3.9716945046538315e-06,
|
2842 |
+
"loss": 1.3761,
|
2843 |
+
"step": 41100
|
2844 |
+
},
|
2845 |
+
{
|
2846 |
+
"epoch": 4.66,
|
2847 |
+
"learning_rate": 3.844192273364784e-06,
|
2848 |
+
"loss": 1.365,
|
2849 |
+
"step": 41200
|
2850 |
+
},
|
2851 |
+
{
|
2852 |
+
"epoch": 4.67,
|
2853 |
+
"learning_rate": 3.716690042075736e-06,
|
2854 |
+
"loss": 1.3984,
|
2855 |
+
"step": 41300
|
2856 |
+
},
|
2857 |
+
{
|
2858 |
+
"epoch": 4.68,
|
2859 |
+
"learning_rate": 3.5891878107866894e-06,
|
2860 |
+
"loss": 1.3449,
|
2861 |
+
"step": 41400
|
2862 |
+
},
|
2863 |
+
{
|
2864 |
+
"epoch": 4.69,
|
2865 |
+
"learning_rate": 3.4616855794976413e-06,
|
2866 |
+
"loss": 1.3671,
|
2867 |
+
"step": 41500
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 4.7,
|
2871 |
+
"learning_rate": 3.3341833482085937e-06,
|
2872 |
+
"loss": 1.3752,
|
2873 |
+
"step": 41600
|
2874 |
+
},
|
2875 |
+
{
|
2876 |
+
"epoch": 4.72,
|
2877 |
+
"learning_rate": 3.2066811169195464e-06,
|
2878 |
+
"loss": 1.348,
|
2879 |
+
"step": 41700
|
2880 |
+
},
|
2881 |
+
{
|
2882 |
+
"epoch": 4.73,
|
2883 |
+
"learning_rate": 3.0791788856304988e-06,
|
2884 |
+
"loss": 1.3793,
|
2885 |
+
"step": 41800
|
2886 |
+
},
|
2887 |
+
{
|
2888 |
+
"epoch": 4.74,
|
2889 |
+
"learning_rate": 2.951676654341451e-06,
|
2890 |
+
"loss": 1.405,
|
2891 |
+
"step": 41900
|
2892 |
+
},
|
2893 |
+
{
|
2894 |
+
"epoch": 4.75,
|
2895 |
+
"learning_rate": 2.8241744230524035e-06,
|
2896 |
+
"loss": 1.3837,
|
2897 |
+
"step": 42000
|
2898 |
+
},
|
2899 |
+
{
|
2900 |
+
"epoch": 4.75,
|
2901 |
+
"eval_loss": 1.3951770067214966,
|
2902 |
+
"eval_masked_accuracy": 0.7039473004213623,
|
2903 |
+
"eval_runtime": 332.7076,
|
2904 |
+
"eval_samples_per_second": 47.252,
|
2905 |
+
"eval_steps_per_second": 2.955,
|
2906 |
+
"step": 42000
|
2907 |
+
},
|
2908 |
+
{
|
2909 |
+
"epoch": 4.76,
|
2910 |
+
"learning_rate": 2.696672191763356e-06,
|
2911 |
+
"loss": 1.3924,
|
2912 |
+
"step": 42100
|
2913 |
+
},
|
2914 |
+
{
|
2915 |
+
"epoch": 4.77,
|
2916 |
+
"learning_rate": 2.5691699604743086e-06,
|
2917 |
+
"loss": 1.3664,
|
2918 |
+
"step": 42200
|
2919 |
+
},
|
2920 |
+
{
|
2921 |
+
"epoch": 4.78,
|
2922 |
+
"learning_rate": 2.4416677291852605e-06,
|
2923 |
+
"loss": 1.3525,
|
2924 |
+
"step": 42300
|
2925 |
+
},
|
2926 |
+
{
|
2927 |
+
"epoch": 4.79,
|
2928 |
+
"learning_rate": 2.3141654978962133e-06,
|
2929 |
+
"loss": 1.3783,
|
2930 |
+
"step": 42400
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 4.81,
|
2934 |
+
"learning_rate": 2.1866632666071656e-06,
|
2935 |
+
"loss": 1.4034,
|
2936 |
+
"step": 42500
|
2937 |
+
},
|
2938 |
+
{
|
2939 |
+
"epoch": 4.82,
|
2940 |
+
"learning_rate": 2.0591610353181184e-06,
|
2941 |
+
"loss": 1.3807,
|
2942 |
+
"step": 42600
|
2943 |
+
},
|
2944 |
+
{
|
2945 |
+
"epoch": 4.83,
|
2946 |
+
"learning_rate": 1.9316588040290703e-06,
|
2947 |
+
"loss": 1.3791,
|
2948 |
+
"step": 42700
|
2949 |
+
},
|
2950 |
+
{
|
2951 |
+
"epoch": 4.84,
|
2952 |
+
"learning_rate": 1.804156572740023e-06,
|
2953 |
+
"loss": 1.3564,
|
2954 |
+
"step": 42800
|
2955 |
+
},
|
2956 |
+
{
|
2957 |
+
"epoch": 4.85,
|
2958 |
+
"learning_rate": 1.6766543414509756e-06,
|
2959 |
+
"loss": 1.4057,
|
2960 |
+
"step": 42900
|
2961 |
+
},
|
2962 |
+
{
|
2963 |
+
"epoch": 4.86,
|
2964 |
+
"learning_rate": 1.549152110161928e-06,
|
2965 |
+
"loss": 1.3752,
|
2966 |
+
"step": 43000
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 4.86,
|
2970 |
+
"eval_loss": 1.3881962299346924,
|
2971 |
+
"eval_masked_accuracy": 0.7062402711958213,
|
2972 |
+
"eval_runtime": 332.7234,
|
2973 |
+
"eval_samples_per_second": 47.249,
|
2974 |
+
"eval_steps_per_second": 2.954,
|
2975 |
+
"step": 43000
|
2976 |
+
},
|
2977 |
+
{
|
2978 |
+
"epoch": 4.87,
|
2979 |
+
"learning_rate": 1.4216498788728803e-06,
|
2980 |
+
"loss": 1.3468,
|
2981 |
+
"step": 43100
|
2982 |
+
},
|
2983 |
+
{
|
2984 |
+
"epoch": 4.89,
|
2985 |
+
"learning_rate": 1.2941476475838329e-06,
|
2986 |
+
"loss": 1.3939,
|
2987 |
+
"step": 43200
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 4.9,
|
2991 |
+
"learning_rate": 1.1666454162947852e-06,
|
2992 |
+
"loss": 1.4151,
|
2993 |
+
"step": 43300
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 4.91,
|
2997 |
+
"learning_rate": 1.0391431850057378e-06,
|
2998 |
+
"loss": 1.3698,
|
2999 |
+
"step": 43400
|
3000 |
+
},
|
3001 |
+
{
|
3002 |
+
"epoch": 4.92,
|
3003 |
+
"learning_rate": 9.1164095371669e-07,
|
3004 |
+
"loss": 1.3676,
|
3005 |
+
"step": 43500
|
3006 |
+
},
|
3007 |
+
{
|
3008 |
+
"epoch": 4.93,
|
3009 |
+
"learning_rate": 7.841387224276426e-07,
|
3010 |
+
"loss": 1.3844,
|
3011 |
+
"step": 43600
|
3012 |
+
},
|
3013 |
+
{
|
3014 |
+
"epoch": 4.94,
|
3015 |
+
"learning_rate": 6.566364911385949e-07,
|
3016 |
+
"loss": 1.3604,
|
3017 |
+
"step": 43700
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 4.95,
|
3021 |
+
"learning_rate": 5.291342598495474e-07,
|
3022 |
+
"loss": 1.3661,
|
3023 |
+
"step": 43800
|
3024 |
+
},
|
3025 |
+
{
|
3026 |
+
"epoch": 4.96,
|
3027 |
+
"learning_rate": 4.0163202856049983e-07,
|
3028 |
+
"loss": 1.3635,
|
3029 |
+
"step": 43900
|
3030 |
+
},
|
3031 |
+
{
|
3032 |
+
"epoch": 4.98,
|
3033 |
+
"learning_rate": 2.7412979727145223e-07,
|
3034 |
+
"loss": 1.3629,
|
3035 |
+
"step": 44000
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 4.98,
|
3039 |
+
"eval_loss": 1.388177514076233,
|
3040 |
+
"eval_masked_accuracy": 0.7065624395501147,
|
3041 |
+
"eval_runtime": 332.6649,
|
3042 |
+
"eval_samples_per_second": 47.258,
|
3043 |
+
"eval_steps_per_second": 2.955,
|
3044 |
+
"step": 44000
|
3045 |
+
}
|
3046 |
+
],
|
3047 |
+
"max_steps": 44215,
|
3048 |
+
"num_train_epochs": 5,
|
3049 |
+
"total_flos": 4.073071997007176e+17,
|
3050 |
+
"trial_name": null,
|
3051 |
+
"trial_params": null
|
3052 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a9d358bd14f646e22583059763770c006e35b681cc6f6d1fa517fc70f9053a0d
|
3 |
+
size 3963
|