Upload folder using huggingface_hub
Browse files- config.json +34 -0
- generation_config.json +8 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2209 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/nllb-200-distilled-1.3B",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "relu",
|
5 |
+
"architectures": [
|
6 |
+
"M2M100ForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 0,
|
10 |
+
"d_model": 1024,
|
11 |
+
"decoder_attention_heads": 16,
|
12 |
+
"decoder_ffn_dim": 8192,
|
13 |
+
"decoder_layerdrop": 0,
|
14 |
+
"decoder_layers": 24,
|
15 |
+
"decoder_start_token_id": 2,
|
16 |
+
"dropout": 0.1,
|
17 |
+
"encoder_attention_heads": 16,
|
18 |
+
"encoder_ffn_dim": 8192,
|
19 |
+
"encoder_layerdrop": 0,
|
20 |
+
"encoder_layers": 24,
|
21 |
+
"eos_token_id": 2,
|
22 |
+
"init_std": 0.02,
|
23 |
+
"is_encoder_decoder": true,
|
24 |
+
"max_length": 200,
|
25 |
+
"max_position_embeddings": 1024,
|
26 |
+
"model_type": "m2m_100",
|
27 |
+
"num_hidden_layers": 24,
|
28 |
+
"pad_token_id": 1,
|
29 |
+
"scale_embedding": true,
|
30 |
+
"torch_dtype": "float32",
|
31 |
+
"transformers_version": "4.34.1",
|
32 |
+
"use_cache": true,
|
33 |
+
"vocab_size": 256206
|
34 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token_id": 0,
|
3 |
+
"decoder_start_token_id": 2,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"max_length": 200,
|
6 |
+
"pad_token_id": 1,
|
7 |
+
"transformers_version": "4.34.1"
|
8 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76cc9b002119823e0d794e53fc69eae8e20e6fa7bb8289b9e8a045a218dd1398
|
3 |
+
size 10965980889
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:81c5722ab7ce97eda63d267fcbc4625658fad15f79ecf0fda9e15c1b14245378
|
3 |
+
size 5482903439
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:faad0ca488ea82820b7d56445f3ce8cf04a5a1bc39d795578bf62e462a95cce8
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f3b9f6a911fcef6e74d709be1892540c70769200059ae0a29f79b5dda1e2c384
|
3 |
+
size 1064
|
trainer_state.json
ADDED
@@ -0,0 +1,2209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.9959017484099317,
|
5 |
+
"eval_steps": 5000,
|
6 |
+
"global_step": 150000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"learning_rate": 4.9833674854299174e-05,
|
14 |
+
"loss": 1.2201,
|
15 |
+
"step": 500
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.01,
|
19 |
+
"learning_rate": 4.9667349708598347e-05,
|
20 |
+
"loss": 1.1313,
|
21 |
+
"step": 1000
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.02,
|
25 |
+
"learning_rate": 4.950102456289752e-05,
|
26 |
+
"loss": 1.1352,
|
27 |
+
"step": 1500
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.03,
|
31 |
+
"learning_rate": 4.933469941719669e-05,
|
32 |
+
"loss": 1.0939,
|
33 |
+
"step": 2000
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.03,
|
37 |
+
"learning_rate": 4.916837427149586e-05,
|
38 |
+
"loss": 1.0539,
|
39 |
+
"step": 2500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.04,
|
43 |
+
"learning_rate": 4.9002049125795035e-05,
|
44 |
+
"loss": 1.0473,
|
45 |
+
"step": 3000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.05,
|
49 |
+
"learning_rate": 4.8835723980094214e-05,
|
50 |
+
"loss": 1.0461,
|
51 |
+
"step": 3500
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.05,
|
55 |
+
"learning_rate": 4.866939883439338e-05,
|
56 |
+
"loss": 0.9995,
|
57 |
+
"step": 4000
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.06,
|
61 |
+
"learning_rate": 4.850307368869255e-05,
|
62 |
+
"loss": 1.0156,
|
63 |
+
"step": 4500
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.07,
|
67 |
+
"learning_rate": 4.833674854299173e-05,
|
68 |
+
"loss": 1.0091,
|
69 |
+
"step": 5000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.07,
|
73 |
+
"eval_bleu": 39.1306,
|
74 |
+
"eval_chrf++": 61.4553,
|
75 |
+
"eval_gen_len": 24.5812,
|
76 |
+
"eval_loss": 0.983640730381012,
|
77 |
+
"eval_runtime": 4879.4604,
|
78 |
+
"eval_samples_per_second": 4.77,
|
79 |
+
"eval_spbleu": 46.5203,
|
80 |
+
"eval_steps_per_second": 0.954,
|
81 |
+
"eval_ter": 49.0585,
|
82 |
+
"step": 5000
|
83 |
+
},
|
84 |
+
{
|
85 |
+
"epoch": 0.07,
|
86 |
+
"learning_rate": 4.8170423397290895e-05,
|
87 |
+
"loss": 1.0139,
|
88 |
+
"step": 5500
|
89 |
+
},
|
90 |
+
{
|
91 |
+
"epoch": 0.08,
|
92 |
+
"learning_rate": 4.800409825159007e-05,
|
93 |
+
"loss": 0.9984,
|
94 |
+
"step": 6000
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 0.09,
|
98 |
+
"learning_rate": 4.7837773105889246e-05,
|
99 |
+
"loss": 0.9854,
|
100 |
+
"step": 6500
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.09,
|
104 |
+
"learning_rate": 4.767144796018841e-05,
|
105 |
+
"loss": 1.0016,
|
106 |
+
"step": 7000
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.1,
|
110 |
+
"learning_rate": 4.750512281448759e-05,
|
111 |
+
"loss": 0.9851,
|
112 |
+
"step": 7500
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.11,
|
116 |
+
"learning_rate": 4.733879766878676e-05,
|
117 |
+
"loss": 0.9736,
|
118 |
+
"step": 8000
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 0.11,
|
122 |
+
"learning_rate": 4.717247252308593e-05,
|
123 |
+
"loss": 0.968,
|
124 |
+
"step": 8500
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 0.12,
|
128 |
+
"learning_rate": 4.700614737738511e-05,
|
129 |
+
"loss": 0.9772,
|
130 |
+
"step": 9000
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 0.13,
|
134 |
+
"learning_rate": 4.683982223168428e-05,
|
135 |
+
"loss": 0.9452,
|
136 |
+
"step": 9500
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.13,
|
140 |
+
"learning_rate": 4.6673497085983444e-05,
|
141 |
+
"loss": 0.9593,
|
142 |
+
"step": 10000
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.13,
|
146 |
+
"eval_bleu": 40.7812,
|
147 |
+
"eval_chrf++": 62.5078,
|
148 |
+
"eval_gen_len": 24.4548,
|
149 |
+
"eval_loss": 0.9384894967079163,
|
150 |
+
"eval_runtime": 4848.287,
|
151 |
+
"eval_samples_per_second": 4.8,
|
152 |
+
"eval_spbleu": 48.2302,
|
153 |
+
"eval_steps_per_second": 0.96,
|
154 |
+
"eval_ter": 47.6761,
|
155 |
+
"step": 10000
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 0.14,
|
159 |
+
"learning_rate": 4.650717194028262e-05,
|
160 |
+
"loss": 0.9495,
|
161 |
+
"step": 10500
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 0.15,
|
165 |
+
"learning_rate": 4.6340846794581795e-05,
|
166 |
+
"loss": 0.9684,
|
167 |
+
"step": 11000
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 0.15,
|
171 |
+
"learning_rate": 4.617452164888097e-05,
|
172 |
+
"loss": 0.9496,
|
173 |
+
"step": 11500
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 0.16,
|
177 |
+
"learning_rate": 4.600819650318014e-05,
|
178 |
+
"loss": 0.9417,
|
179 |
+
"step": 12000
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.17,
|
183 |
+
"learning_rate": 4.584187135747931e-05,
|
184 |
+
"loss": 0.9415,
|
185 |
+
"step": 12500
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.17,
|
189 |
+
"learning_rate": 4.5675546211778483e-05,
|
190 |
+
"loss": 0.9325,
|
191 |
+
"step": 13000
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.18,
|
195 |
+
"learning_rate": 4.5509221066077656e-05,
|
196 |
+
"loss": 0.9488,
|
197 |
+
"step": 13500
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 0.19,
|
201 |
+
"learning_rate": 4.534289592037683e-05,
|
202 |
+
"loss": 0.9233,
|
203 |
+
"step": 14000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 0.19,
|
207 |
+
"learning_rate": 4.5176570774676e-05,
|
208 |
+
"loss": 0.9541,
|
209 |
+
"step": 14500
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 0.2,
|
213 |
+
"learning_rate": 4.501024562897517e-05,
|
214 |
+
"loss": 0.9716,
|
215 |
+
"step": 15000
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 0.2,
|
219 |
+
"eval_bleu": 41.9435,
|
220 |
+
"eval_chrf++": 63.2793,
|
221 |
+
"eval_gen_len": 24.3083,
|
222 |
+
"eval_loss": 0.9115902781486511,
|
223 |
+
"eval_runtime": 4836.6682,
|
224 |
+
"eval_samples_per_second": 4.812,
|
225 |
+
"eval_spbleu": 49.3278,
|
226 |
+
"eval_steps_per_second": 0.962,
|
227 |
+
"eval_ter": 46.5599,
|
228 |
+
"step": 15000
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 0.21,
|
232 |
+
"learning_rate": 4.4843920483274344e-05,
|
233 |
+
"loss": 0.9314,
|
234 |
+
"step": 15500
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 0.21,
|
238 |
+
"learning_rate": 4.4677595337573516e-05,
|
239 |
+
"loss": 0.9492,
|
240 |
+
"step": 16000
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.22,
|
244 |
+
"learning_rate": 4.451127019187269e-05,
|
245 |
+
"loss": 0.9195,
|
246 |
+
"step": 16500
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.23,
|
250 |
+
"learning_rate": 4.434494504617187e-05,
|
251 |
+
"loss": 0.924,
|
252 |
+
"step": 17000
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 0.23,
|
256 |
+
"learning_rate": 4.417861990047103e-05,
|
257 |
+
"loss": 0.8895,
|
258 |
+
"step": 17500
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 0.24,
|
262 |
+
"learning_rate": 4.4012294754770204e-05,
|
263 |
+
"loss": 0.9303,
|
264 |
+
"step": 18000
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 0.25,
|
268 |
+
"learning_rate": 4.384596960906938e-05,
|
269 |
+
"loss": 0.9499,
|
270 |
+
"step": 18500
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 0.25,
|
274 |
+
"learning_rate": 4.3679644463368555e-05,
|
275 |
+
"loss": 0.9335,
|
276 |
+
"step": 19000
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 0.26,
|
280 |
+
"learning_rate": 4.351331931766772e-05,
|
281 |
+
"loss": 0.9238,
|
282 |
+
"step": 19500
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.27,
|
286 |
+
"learning_rate": 4.33469941719669e-05,
|
287 |
+
"loss": 0.9204,
|
288 |
+
"step": 20000
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 0.27,
|
292 |
+
"eval_bleu": 42.615,
|
293 |
+
"eval_chrf++": 63.6457,
|
294 |
+
"eval_gen_len": 24.1677,
|
295 |
+
"eval_loss": 0.8911730647087097,
|
296 |
+
"eval_runtime": 4794.0613,
|
297 |
+
"eval_samples_per_second": 4.855,
|
298 |
+
"eval_spbleu": 50.0126,
|
299 |
+
"eval_steps_per_second": 0.971,
|
300 |
+
"eval_ter": 45.765,
|
301 |
+
"step": 20000
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 0.27,
|
305 |
+
"learning_rate": 4.318066902626607e-05,
|
306 |
+
"loss": 0.9066,
|
307 |
+
"step": 20500
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 0.28,
|
311 |
+
"learning_rate": 4.301434388056524e-05,
|
312 |
+
"loss": 0.8974,
|
313 |
+
"step": 21000
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 0.29,
|
317 |
+
"learning_rate": 4.2848018734864416e-05,
|
318 |
+
"loss": 0.9196,
|
319 |
+
"step": 21500
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 0.29,
|
323 |
+
"learning_rate": 4.268169358916359e-05,
|
324 |
+
"loss": 0.9046,
|
325 |
+
"step": 22000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 0.3,
|
329 |
+
"learning_rate": 4.251536844346276e-05,
|
330 |
+
"loss": 0.8821,
|
331 |
+
"step": 22500
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.31,
|
335 |
+
"learning_rate": 4.234904329776193e-05,
|
336 |
+
"loss": 0.9095,
|
337 |
+
"step": 23000
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 0.31,
|
341 |
+
"learning_rate": 4.2182718152061104e-05,
|
342 |
+
"loss": 0.8814,
|
343 |
+
"step": 23500
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 0.32,
|
347 |
+
"learning_rate": 4.2016393006360276e-05,
|
348 |
+
"loss": 0.9074,
|
349 |
+
"step": 24000
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 0.33,
|
353 |
+
"learning_rate": 4.185006786065945e-05,
|
354 |
+
"loss": 0.9086,
|
355 |
+
"step": 24500
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 0.33,
|
359 |
+
"learning_rate": 4.168374271495862e-05,
|
360 |
+
"loss": 0.87,
|
361 |
+
"step": 25000
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 0.33,
|
365 |
+
"eval_bleu": 43.372,
|
366 |
+
"eval_chrf++": 64.3092,
|
367 |
+
"eval_gen_len": 24.2173,
|
368 |
+
"eval_loss": 0.8829550743103027,
|
369 |
+
"eval_runtime": 4787.4055,
|
370 |
+
"eval_samples_per_second": 4.862,
|
371 |
+
"eval_spbleu": 50.8714,
|
372 |
+
"eval_steps_per_second": 0.972,
|
373 |
+
"eval_ter": 45.1953,
|
374 |
+
"step": 25000
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.34,
|
378 |
+
"learning_rate": 4.151741756925779e-05,
|
379 |
+
"loss": 0.8688,
|
380 |
+
"step": 25500
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.35,
|
384 |
+
"learning_rate": 4.1351092423556965e-05,
|
385 |
+
"loss": 0.8861,
|
386 |
+
"step": 26000
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 0.35,
|
390 |
+
"learning_rate": 4.1184767277856137e-05,
|
391 |
+
"loss": 0.854,
|
392 |
+
"step": 26500
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 0.36,
|
396 |
+
"learning_rate": 4.101844213215531e-05,
|
397 |
+
"loss": 0.9359,
|
398 |
+
"step": 27000
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 0.37,
|
402 |
+
"learning_rate": 4.085211698645448e-05,
|
403 |
+
"loss": 0.8724,
|
404 |
+
"step": 27500
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 0.37,
|
408 |
+
"learning_rate": 4.068579184075366e-05,
|
409 |
+
"loss": 0.8954,
|
410 |
+
"step": 28000
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 0.38,
|
414 |
+
"learning_rate": 4.0519466695052825e-05,
|
415 |
+
"loss": 0.9009,
|
416 |
+
"step": 28500
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.39,
|
420 |
+
"learning_rate": 4.0353141549352e-05,
|
421 |
+
"loss": 0.889,
|
422 |
+
"step": 29000
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.39,
|
426 |
+
"learning_rate": 4.0186816403651176e-05,
|
427 |
+
"loss": 0.8782,
|
428 |
+
"step": 29500
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.4,
|
432 |
+
"learning_rate": 4.002049125795034e-05,
|
433 |
+
"loss": 0.8891,
|
434 |
+
"step": 30000
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 0.4,
|
438 |
+
"eval_bleu": 43.9744,
|
439 |
+
"eval_chrf++": 64.7903,
|
440 |
+
"eval_gen_len": 24.3772,
|
441 |
+
"eval_loss": 0.8677447438240051,
|
442 |
+
"eval_runtime": 4855.7738,
|
443 |
+
"eval_samples_per_second": 4.793,
|
444 |
+
"eval_spbleu": 51.5389,
|
445 |
+
"eval_steps_per_second": 0.959,
|
446 |
+
"eval_ter": 44.8206,
|
447 |
+
"step": 30000
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 0.41,
|
451 |
+
"learning_rate": 3.985416611224951e-05,
|
452 |
+
"loss": 0.8785,
|
453 |
+
"step": 30500
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 0.41,
|
457 |
+
"learning_rate": 3.968784096654869e-05,
|
458 |
+
"loss": 0.8603,
|
459 |
+
"step": 31000
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 0.42,
|
463 |
+
"learning_rate": 3.952151582084786e-05,
|
464 |
+
"loss": 0.8513,
|
465 |
+
"step": 31500
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 0.43,
|
469 |
+
"learning_rate": 3.9355190675147036e-05,
|
470 |
+
"loss": 0.8737,
|
471 |
+
"step": 32000
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.43,
|
475 |
+
"learning_rate": 3.918886552944621e-05,
|
476 |
+
"loss": 0.8808,
|
477 |
+
"step": 32500
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 0.44,
|
481 |
+
"learning_rate": 3.9022540383745374e-05,
|
482 |
+
"loss": 0.8564,
|
483 |
+
"step": 33000
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 0.45,
|
487 |
+
"learning_rate": 3.885621523804455e-05,
|
488 |
+
"loss": 0.8816,
|
489 |
+
"step": 33500
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 0.45,
|
493 |
+
"learning_rate": 3.8689890092343725e-05,
|
494 |
+
"loss": 0.8593,
|
495 |
+
"step": 34000
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 0.46,
|
499 |
+
"learning_rate": 3.852356494664289e-05,
|
500 |
+
"loss": 0.8382,
|
501 |
+
"step": 34500
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 0.47,
|
505 |
+
"learning_rate": 3.835723980094207e-05,
|
506 |
+
"loss": 0.876,
|
507 |
+
"step": 35000
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 0.47,
|
511 |
+
"eval_bleu": 44.1888,
|
512 |
+
"eval_chrf++": 64.937,
|
513 |
+
"eval_gen_len": 24.3824,
|
514 |
+
"eval_loss": 0.853570818901062,
|
515 |
+
"eval_runtime": 4869.588,
|
516 |
+
"eval_samples_per_second": 4.779,
|
517 |
+
"eval_spbleu": 51.71,
|
518 |
+
"eval_steps_per_second": 0.956,
|
519 |
+
"eval_ter": 44.5618,
|
520 |
+
"step": 35000
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.47,
|
524 |
+
"learning_rate": 3.819091465524124e-05,
|
525 |
+
"loss": 0.8818,
|
526 |
+
"step": 35500
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 0.48,
|
530 |
+
"learning_rate": 3.802458950954041e-05,
|
531 |
+
"loss": 0.8723,
|
532 |
+
"step": 36000
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 0.49,
|
536 |
+
"learning_rate": 3.7858264363839585e-05,
|
537 |
+
"loss": 0.8363,
|
538 |
+
"step": 36500
|
539 |
+
},
|
540 |
+
{
|
541 |
+
"epoch": 0.49,
|
542 |
+
"learning_rate": 3.769193921813876e-05,
|
543 |
+
"loss": 0.8441,
|
544 |
+
"step": 37000
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 0.5,
|
548 |
+
"learning_rate": 3.752561407243793e-05,
|
549 |
+
"loss": 0.8483,
|
550 |
+
"step": 37500
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 0.51,
|
554 |
+
"learning_rate": 3.73592889267371e-05,
|
555 |
+
"loss": 0.8321,
|
556 |
+
"step": 38000
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 0.51,
|
560 |
+
"learning_rate": 3.7192963781036273e-05,
|
561 |
+
"loss": 0.8286,
|
562 |
+
"step": 38500
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.52,
|
566 |
+
"learning_rate": 3.7026638635335446e-05,
|
567 |
+
"loss": 0.8525,
|
568 |
+
"step": 39000
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 0.53,
|
572 |
+
"learning_rate": 3.686031348963462e-05,
|
573 |
+
"loss": 0.8697,
|
574 |
+
"step": 39500
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 0.53,
|
578 |
+
"learning_rate": 3.669398834393379e-05,
|
579 |
+
"loss": 0.8636,
|
580 |
+
"step": 40000
|
581 |
+
},
|
582 |
+
{
|
583 |
+
"epoch": 0.53,
|
584 |
+
"eval_bleu": 44.7097,
|
585 |
+
"eval_chrf++": 65.2018,
|
586 |
+
"eval_gen_len": 24.2135,
|
587 |
+
"eval_loss": 0.8435378670692444,
|
588 |
+
"eval_runtime": 4812.8015,
|
589 |
+
"eval_samples_per_second": 4.836,
|
590 |
+
"eval_spbleu": 52.2257,
|
591 |
+
"eval_steps_per_second": 0.967,
|
592 |
+
"eval_ter": 43.745,
|
593 |
+
"step": 40000
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 0.54,
|
597 |
+
"learning_rate": 3.652766319823296e-05,
|
598 |
+
"loss": 0.873,
|
599 |
+
"step": 40500
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 0.55,
|
603 |
+
"learning_rate": 3.6361338052532134e-05,
|
604 |
+
"loss": 0.8743,
|
605 |
+
"step": 41000
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 0.55,
|
609 |
+
"learning_rate": 3.619501290683131e-05,
|
610 |
+
"loss": 0.86,
|
611 |
+
"step": 41500
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.56,
|
615 |
+
"learning_rate": 3.602868776113048e-05,
|
616 |
+
"loss": 0.8492,
|
617 |
+
"step": 42000
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 0.57,
|
621 |
+
"learning_rate": 3.586236261542965e-05,
|
622 |
+
"loss": 0.8454,
|
623 |
+
"step": 42500
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 0.57,
|
627 |
+
"learning_rate": 3.569603746972883e-05,
|
628 |
+
"loss": 0.8348,
|
629 |
+
"step": 43000
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 0.58,
|
633 |
+
"learning_rate": 3.5529712324027994e-05,
|
634 |
+
"loss": 0.8528,
|
635 |
+
"step": 43500
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 0.59,
|
639 |
+
"learning_rate": 3.5363387178327166e-05,
|
640 |
+
"loss": 0.8358,
|
641 |
+
"step": 44000
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 0.59,
|
645 |
+
"learning_rate": 3.5197062032626345e-05,
|
646 |
+
"loss": 0.8293,
|
647 |
+
"step": 44500
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 0.6,
|
651 |
+
"learning_rate": 3.503073688692551e-05,
|
652 |
+
"loss": 0.8501,
|
653 |
+
"step": 45000
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.6,
|
657 |
+
"eval_bleu": 45.0613,
|
658 |
+
"eval_chrf++": 65.5511,
|
659 |
+
"eval_gen_len": 24.2012,
|
660 |
+
"eval_loss": 0.8350074887275696,
|
661 |
+
"eval_runtime": 4817.2319,
|
662 |
+
"eval_samples_per_second": 4.831,
|
663 |
+
"eval_spbleu": 52.5625,
|
664 |
+
"eval_steps_per_second": 0.966,
|
665 |
+
"eval_ter": 43.4063,
|
666 |
+
"step": 45000
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 0.61,
|
670 |
+
"learning_rate": 3.486441174122469e-05,
|
671 |
+
"loss": 0.8366,
|
672 |
+
"step": 45500
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 0.61,
|
676 |
+
"learning_rate": 3.469808659552386e-05,
|
677 |
+
"loss": 0.8342,
|
678 |
+
"step": 46000
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 0.62,
|
682 |
+
"learning_rate": 3.453176144982303e-05,
|
683 |
+
"loss": 0.8239,
|
684 |
+
"step": 46500
|
685 |
+
},
|
686 |
+
{
|
687 |
+
"epoch": 0.63,
|
688 |
+
"learning_rate": 3.4365436304122206e-05,
|
689 |
+
"loss": 0.8663,
|
690 |
+
"step": 47000
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 0.63,
|
694 |
+
"learning_rate": 3.419911115842138e-05,
|
695 |
+
"loss": 0.8041,
|
696 |
+
"step": 47500
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 0.64,
|
700 |
+
"learning_rate": 3.403278601272055e-05,
|
701 |
+
"loss": 0.8368,
|
702 |
+
"step": 48000
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.65,
|
706 |
+
"learning_rate": 3.386646086701972e-05,
|
707 |
+
"loss": 0.8106,
|
708 |
+
"step": 48500
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 0.65,
|
712 |
+
"learning_rate": 3.3700135721318894e-05,
|
713 |
+
"loss": 0.836,
|
714 |
+
"step": 49000
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 0.66,
|
718 |
+
"learning_rate": 3.3533810575618066e-05,
|
719 |
+
"loss": 0.8145,
|
720 |
+
"step": 49500
|
721 |
+
},
|
722 |
+
{
|
723 |
+
"epoch": 0.67,
|
724 |
+
"learning_rate": 3.336748542991724e-05,
|
725 |
+
"loss": 0.8417,
|
726 |
+
"step": 50000
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 0.67,
|
730 |
+
"eval_bleu": 45.3651,
|
731 |
+
"eval_chrf++": 65.7396,
|
732 |
+
"eval_gen_len": 24.2033,
|
733 |
+
"eval_loss": 0.8263786435127258,
|
734 |
+
"eval_runtime": 4807.1892,
|
735 |
+
"eval_samples_per_second": 4.841,
|
736 |
+
"eval_spbleu": 52.8998,
|
737 |
+
"eval_steps_per_second": 0.968,
|
738 |
+
"eval_ter": 43.3251,
|
739 |
+
"step": 50000
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 0.67,
|
743 |
+
"learning_rate": 3.320116028421641e-05,
|
744 |
+
"loss": 0.7953,
|
745 |
+
"step": 50500
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 0.68,
|
749 |
+
"learning_rate": 3.303483513851558e-05,
|
750 |
+
"loss": 0.8218,
|
751 |
+
"step": 51000
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.69,
|
755 |
+
"learning_rate": 3.2868509992814755e-05,
|
756 |
+
"loss": 0.81,
|
757 |
+
"step": 51500
|
758 |
+
},
|
759 |
+
{
|
760 |
+
"epoch": 0.69,
|
761 |
+
"learning_rate": 3.270218484711393e-05,
|
762 |
+
"loss": 0.822,
|
763 |
+
"step": 52000
|
764 |
+
},
|
765 |
+
{
|
766 |
+
"epoch": 0.7,
|
767 |
+
"learning_rate": 3.25358597014131e-05,
|
768 |
+
"loss": 0.8174,
|
769 |
+
"step": 52500
|
770 |
+
},
|
771 |
+
{
|
772 |
+
"epoch": 0.71,
|
773 |
+
"learning_rate": 3.236953455571227e-05,
|
774 |
+
"loss": 0.8136,
|
775 |
+
"step": 53000
|
776 |
+
},
|
777 |
+
{
|
778 |
+
"epoch": 0.71,
|
779 |
+
"learning_rate": 3.220320941001144e-05,
|
780 |
+
"loss": 0.8669,
|
781 |
+
"step": 53500
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 0.72,
|
785 |
+
"learning_rate": 3.2036884264310615e-05,
|
786 |
+
"loss": 0.8195,
|
787 |
+
"step": 54000
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 0.73,
|
791 |
+
"learning_rate": 3.187055911860979e-05,
|
792 |
+
"loss": 0.825,
|
793 |
+
"step": 54500
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.73,
|
797 |
+
"learning_rate": 3.1704233972908966e-05,
|
798 |
+
"loss": 0.8083,
|
799 |
+
"step": 55000
|
800 |
+
},
|
801 |
+
{
|
802 |
+
"epoch": 0.73,
|
803 |
+
"eval_bleu": 45.8888,
|
804 |
+
"eval_chrf++": 66.0994,
|
805 |
+
"eval_gen_len": 24.2883,
|
806 |
+
"eval_loss": 0.8205437660217285,
|
807 |
+
"eval_runtime": 4815.6563,
|
808 |
+
"eval_samples_per_second": 4.833,
|
809 |
+
"eval_spbleu": 53.2813,
|
810 |
+
"eval_steps_per_second": 0.967,
|
811 |
+
"eval_ter": 42.9175,
|
812 |
+
"step": 55000
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 0.74,
|
816 |
+
"learning_rate": 3.153790882720814e-05,
|
817 |
+
"loss": 0.8357,
|
818 |
+
"step": 55500
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 0.75,
|
822 |
+
"learning_rate": 3.13715836815073e-05,
|
823 |
+
"loss": 0.8097,
|
824 |
+
"step": 56000
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 0.75,
|
828 |
+
"learning_rate": 3.120525853580648e-05,
|
829 |
+
"loss": 0.7932,
|
830 |
+
"step": 56500
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 0.76,
|
834 |
+
"learning_rate": 3.1038933390105654e-05,
|
835 |
+
"loss": 0.8225,
|
836 |
+
"step": 57000
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 0.77,
|
840 |
+
"learning_rate": 3.087260824440482e-05,
|
841 |
+
"loss": 0.7973,
|
842 |
+
"step": 57500
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.77,
|
846 |
+
"learning_rate": 3.0706283098704e-05,
|
847 |
+
"loss": 0.8379,
|
848 |
+
"step": 58000
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 0.78,
|
852 |
+
"learning_rate": 3.053995795300317e-05,
|
853 |
+
"loss": 0.8085,
|
854 |
+
"step": 58500
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 0.79,
|
858 |
+
"learning_rate": 3.037363280730234e-05,
|
859 |
+
"loss": 0.8528,
|
860 |
+
"step": 59000
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 0.79,
|
864 |
+
"learning_rate": 3.0207307661601515e-05,
|
865 |
+
"loss": 0.8262,
|
866 |
+
"step": 59500
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 0.8,
|
870 |
+
"learning_rate": 3.0040982515900683e-05,
|
871 |
+
"loss": 0.7964,
|
872 |
+
"step": 60000
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 0.8,
|
876 |
+
"eval_bleu": 46.137,
|
877 |
+
"eval_chrf++": 66.311,
|
878 |
+
"eval_gen_len": 24.357,
|
879 |
+
"eval_loss": 0.8131713271141052,
|
880 |
+
"eval_runtime": 4833.9641,
|
881 |
+
"eval_samples_per_second": 4.815,
|
882 |
+
"eval_spbleu": 53.6989,
|
883 |
+
"eval_steps_per_second": 0.963,
|
884 |
+
"eval_ter": 42.7845,
|
885 |
+
"step": 60000
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 0.81,
|
889 |
+
"learning_rate": 2.987465737019986e-05,
|
890 |
+
"loss": 0.8137,
|
891 |
+
"step": 60500
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.81,
|
895 |
+
"learning_rate": 2.970833222449903e-05,
|
896 |
+
"loss": 0.8046,
|
897 |
+
"step": 61000
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 0.82,
|
901 |
+
"learning_rate": 2.95420070787982e-05,
|
902 |
+
"loss": 0.8496,
|
903 |
+
"step": 61500
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 0.82,
|
907 |
+
"learning_rate": 2.9375681933097375e-05,
|
908 |
+
"loss": 0.8115,
|
909 |
+
"step": 62000
|
910 |
+
},
|
911 |
+
{
|
912 |
+
"epoch": 0.83,
|
913 |
+
"learning_rate": 2.9209356787396547e-05,
|
914 |
+
"loss": 0.8222,
|
915 |
+
"step": 62500
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 0.84,
|
919 |
+
"learning_rate": 2.9043031641695716e-05,
|
920 |
+
"loss": 0.8067,
|
921 |
+
"step": 63000
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 0.84,
|
925 |
+
"learning_rate": 2.887670649599489e-05,
|
926 |
+
"loss": 0.7789,
|
927 |
+
"step": 63500
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 0.85,
|
931 |
+
"learning_rate": 2.8710381350294063e-05,
|
932 |
+
"loss": 0.8369,
|
933 |
+
"step": 64000
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.86,
|
937 |
+
"learning_rate": 2.854405620459324e-05,
|
938 |
+
"loss": 0.8139,
|
939 |
+
"step": 64500
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 0.86,
|
943 |
+
"learning_rate": 2.837773105889241e-05,
|
944 |
+
"loss": 0.7947,
|
945 |
+
"step": 65000
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 0.86,
|
949 |
+
"eval_bleu": 46.3,
|
950 |
+
"eval_chrf++": 66.3209,
|
951 |
+
"eval_gen_len": 24.3462,
|
952 |
+
"eval_loss": 0.8046720623970032,
|
953 |
+
"eval_runtime": 4808.297,
|
954 |
+
"eval_samples_per_second": 4.84,
|
955 |
+
"eval_spbleu": 53.7389,
|
956 |
+
"eval_steps_per_second": 0.968,
|
957 |
+
"eval_ter": 42.7587,
|
958 |
+
"step": 65000
|
959 |
+
},
|
960 |
+
{
|
961 |
+
"epoch": 0.87,
|
962 |
+
"learning_rate": 2.821140591319158e-05,
|
963 |
+
"loss": 0.7977,
|
964 |
+
"step": 65500
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 0.88,
|
968 |
+
"learning_rate": 2.8045080767490755e-05,
|
969 |
+
"loss": 0.809,
|
970 |
+
"step": 66000
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 0.88,
|
974 |
+
"learning_rate": 2.7878755621789927e-05,
|
975 |
+
"loss": 0.8344,
|
976 |
+
"step": 66500
|
977 |
+
},
|
978 |
+
{
|
979 |
+
"epoch": 0.89,
|
980 |
+
"learning_rate": 2.7712430476089096e-05,
|
981 |
+
"loss": 0.7839,
|
982 |
+
"step": 67000
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.9,
|
986 |
+
"learning_rate": 2.754610533038827e-05,
|
987 |
+
"loss": 0.7905,
|
988 |
+
"step": 67500
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 0.9,
|
992 |
+
"learning_rate": 2.7379780184687444e-05,
|
993 |
+
"loss": 0.806,
|
994 |
+
"step": 68000
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 0.91,
|
998 |
+
"learning_rate": 2.7213455038986612e-05,
|
999 |
+
"loss": 0.7844,
|
1000 |
+
"step": 68500
|
1001 |
+
},
|
1002 |
+
{
|
1003 |
+
"epoch": 0.92,
|
1004 |
+
"learning_rate": 2.7047129893285788e-05,
|
1005 |
+
"loss": 0.7724,
|
1006 |
+
"step": 69000
|
1007 |
+
},
|
1008 |
+
{
|
1009 |
+
"epoch": 0.92,
|
1010 |
+
"learning_rate": 2.688080474758496e-05,
|
1011 |
+
"loss": 0.8193,
|
1012 |
+
"step": 69500
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 0.93,
|
1016 |
+
"learning_rate": 2.6714479601884135e-05,
|
1017 |
+
"loss": 0.7972,
|
1018 |
+
"step": 70000
|
1019 |
+
},
|
1020 |
+
{
|
1021 |
+
"epoch": 0.93,
|
1022 |
+
"eval_bleu": 46.7802,
|
1023 |
+
"eval_chrf++": 66.6403,
|
1024 |
+
"eval_gen_len": 24.1858,
|
1025 |
+
"eval_loss": 0.7967858910560608,
|
1026 |
+
"eval_runtime": 4793.9096,
|
1027 |
+
"eval_samples_per_second": 4.855,
|
1028 |
+
"eval_spbleu": 54.0456,
|
1029 |
+
"eval_steps_per_second": 0.971,
|
1030 |
+
"eval_ter": 42.2931,
|
1031 |
+
"step": 70000
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.94,
|
1035 |
+
"learning_rate": 2.6548154456183304e-05,
|
1036 |
+
"loss": 0.8105,
|
1037 |
+
"step": 70500
|
1038 |
+
},
|
1039 |
+
{
|
1040 |
+
"epoch": 0.94,
|
1041 |
+
"learning_rate": 2.6381829310482476e-05,
|
1042 |
+
"loss": 0.7997,
|
1043 |
+
"step": 71000
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 0.95,
|
1047 |
+
"learning_rate": 2.621550416478165e-05,
|
1048 |
+
"loss": 0.8022,
|
1049 |
+
"step": 71500
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 0.96,
|
1053 |
+
"learning_rate": 2.604917901908082e-05,
|
1054 |
+
"loss": 0.8455,
|
1055 |
+
"step": 72000
|
1056 |
+
},
|
1057 |
+
{
|
1058 |
+
"epoch": 0.96,
|
1059 |
+
"learning_rate": 2.5882853873379992e-05,
|
1060 |
+
"loss": 0.7892,
|
1061 |
+
"step": 72500
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 0.97,
|
1065 |
+
"learning_rate": 2.5716528727679168e-05,
|
1066 |
+
"loss": 0.764,
|
1067 |
+
"step": 73000
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 0.98,
|
1071 |
+
"learning_rate": 2.5550203581978337e-05,
|
1072 |
+
"loss": 0.7833,
|
1073 |
+
"step": 73500
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.98,
|
1077 |
+
"learning_rate": 2.5383878436277515e-05,
|
1078 |
+
"loss": 0.7769,
|
1079 |
+
"step": 74000
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 0.99,
|
1083 |
+
"learning_rate": 2.5217553290576684e-05,
|
1084 |
+
"loss": 0.7938,
|
1085 |
+
"step": 74500
|
1086 |
+
},
|
1087 |
+
{
|
1088 |
+
"epoch": 1.0,
|
1089 |
+
"learning_rate": 2.5051228144875856e-05,
|
1090 |
+
"loss": 0.8199,
|
1091 |
+
"step": 75000
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 1.0,
|
1095 |
+
"eval_bleu": 47.0207,
|
1096 |
+
"eval_chrf++": 66.8796,
|
1097 |
+
"eval_gen_len": 24.2671,
|
1098 |
+
"eval_loss": 0.792296826839447,
|
1099 |
+
"eval_runtime": 4833.1796,
|
1100 |
+
"eval_samples_per_second": 4.815,
|
1101 |
+
"eval_spbleu": 54.3927,
|
1102 |
+
"eval_steps_per_second": 0.963,
|
1103 |
+
"eval_ter": 42.1693,
|
1104 |
+
"step": 75000
|
1105 |
+
},
|
1106 |
+
{
|
1107 |
+
"epoch": 1.0,
|
1108 |
+
"learning_rate": 2.4884902999175028e-05,
|
1109 |
+
"loss": 0.6904,
|
1110 |
+
"step": 75500
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 1.01,
|
1114 |
+
"learning_rate": 2.47185778534742e-05,
|
1115 |
+
"loss": 0.6305,
|
1116 |
+
"step": 76000
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.02,
|
1120 |
+
"learning_rate": 2.4552252707773372e-05,
|
1121 |
+
"loss": 0.5988,
|
1122 |
+
"step": 76500
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 1.02,
|
1126 |
+
"learning_rate": 2.4385927562072548e-05,
|
1127 |
+
"loss": 0.6285,
|
1128 |
+
"step": 77000
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 1.03,
|
1132 |
+
"learning_rate": 2.4219602416371717e-05,
|
1133 |
+
"loss": 0.6293,
|
1134 |
+
"step": 77500
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 1.04,
|
1138 |
+
"learning_rate": 2.405327727067089e-05,
|
1139 |
+
"loss": 0.6043,
|
1140 |
+
"step": 78000
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 1.04,
|
1144 |
+
"learning_rate": 2.3886952124970064e-05,
|
1145 |
+
"loss": 0.6415,
|
1146 |
+
"step": 78500
|
1147 |
+
},
|
1148 |
+
{
|
1149 |
+
"epoch": 1.05,
|
1150 |
+
"learning_rate": 2.3720626979269233e-05,
|
1151 |
+
"loss": 0.6285,
|
1152 |
+
"step": 79000
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 1.06,
|
1156 |
+
"learning_rate": 2.355430183356841e-05,
|
1157 |
+
"loss": 0.6169,
|
1158 |
+
"step": 79500
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 1.06,
|
1162 |
+
"learning_rate": 2.338797668786758e-05,
|
1163 |
+
"loss": 0.6288,
|
1164 |
+
"step": 80000
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 1.06,
|
1168 |
+
"eval_bleu": 47.092,
|
1169 |
+
"eval_chrf++": 66.9397,
|
1170 |
+
"eval_gen_len": 24.3014,
|
1171 |
+
"eval_loss": 0.8010306358337402,
|
1172 |
+
"eval_runtime": 4827.2842,
|
1173 |
+
"eval_samples_per_second": 4.821,
|
1174 |
+
"eval_spbleu": 54.5163,
|
1175 |
+
"eval_steps_per_second": 0.964,
|
1176 |
+
"eval_ter": 41.7379,
|
1177 |
+
"step": 80000
|
1178 |
+
},
|
1179 |
+
{
|
1180 |
+
"epoch": 1.07,
|
1181 |
+
"learning_rate": 2.3221651542166753e-05,
|
1182 |
+
"loss": 0.6062,
|
1183 |
+
"step": 80500
|
1184 |
+
},
|
1185 |
+
{
|
1186 |
+
"epoch": 1.08,
|
1187 |
+
"learning_rate": 2.3055326396465925e-05,
|
1188 |
+
"loss": 0.6193,
|
1189 |
+
"step": 81000
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 1.08,
|
1193 |
+
"learning_rate": 2.2889001250765097e-05,
|
1194 |
+
"loss": 0.6205,
|
1195 |
+
"step": 81500
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 1.09,
|
1199 |
+
"learning_rate": 2.272267610506427e-05,
|
1200 |
+
"loss": 0.6073,
|
1201 |
+
"step": 82000
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 1.1,
|
1205 |
+
"learning_rate": 2.255635095936344e-05,
|
1206 |
+
"loss": 0.5999,
|
1207 |
+
"step": 82500
|
1208 |
+
},
|
1209 |
+
{
|
1210 |
+
"epoch": 1.1,
|
1211 |
+
"learning_rate": 2.2390025813662613e-05,
|
1212 |
+
"loss": 0.6397,
|
1213 |
+
"step": 83000
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 1.11,
|
1217 |
+
"learning_rate": 2.2223700667961785e-05,
|
1218 |
+
"loss": 0.6124,
|
1219 |
+
"step": 83500
|
1220 |
+
},
|
1221 |
+
{
|
1222 |
+
"epoch": 1.12,
|
1223 |
+
"learning_rate": 2.205737552226096e-05,
|
1224 |
+
"loss": 0.6026,
|
1225 |
+
"step": 84000
|
1226 |
+
},
|
1227 |
+
{
|
1228 |
+
"epoch": 1.12,
|
1229 |
+
"learning_rate": 2.1891050376560133e-05,
|
1230 |
+
"loss": 0.6123,
|
1231 |
+
"step": 84500
|
1232 |
+
},
|
1233 |
+
{
|
1234 |
+
"epoch": 1.13,
|
1235 |
+
"learning_rate": 2.17247252308593e-05,
|
1236 |
+
"loss": 0.6313,
|
1237 |
+
"step": 85000
|
1238 |
+
},
|
1239 |
+
{
|
1240 |
+
"epoch": 1.13,
|
1241 |
+
"eval_bleu": 47.321,
|
1242 |
+
"eval_chrf++": 67.0811,
|
1243 |
+
"eval_gen_len": 24.2745,
|
1244 |
+
"eval_loss": 0.7992239594459534,
|
1245 |
+
"eval_runtime": 4828.2975,
|
1246 |
+
"eval_samples_per_second": 4.82,
|
1247 |
+
"eval_spbleu": 54.7759,
|
1248 |
+
"eval_steps_per_second": 0.964,
|
1249 |
+
"eval_ter": 41.5623,
|
1250 |
+
"step": 85000
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.14,
|
1254 |
+
"learning_rate": 2.1558400085158477e-05,
|
1255 |
+
"loss": 0.5976,
|
1256 |
+
"step": 85500
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.14,
|
1260 |
+
"learning_rate": 2.139207493945765e-05,
|
1261 |
+
"loss": 0.6151,
|
1262 |
+
"step": 86000
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 1.15,
|
1266 |
+
"learning_rate": 2.122574979375682e-05,
|
1267 |
+
"loss": 0.607,
|
1268 |
+
"step": 86500
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 1.16,
|
1272 |
+
"learning_rate": 2.1059424648055993e-05,
|
1273 |
+
"loss": 0.6202,
|
1274 |
+
"step": 87000
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.16,
|
1278 |
+
"learning_rate": 2.0893099502355165e-05,
|
1279 |
+
"loss": 0.6135,
|
1280 |
+
"step": 87500
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.17,
|
1284 |
+
"learning_rate": 2.0726774356654337e-05,
|
1285 |
+
"loss": 0.6156,
|
1286 |
+
"step": 88000
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 1.18,
|
1290 |
+
"learning_rate": 2.056044921095351e-05,
|
1291 |
+
"loss": 0.5956,
|
1292 |
+
"step": 88500
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 1.18,
|
1296 |
+
"learning_rate": 2.039412406525268e-05,
|
1297 |
+
"loss": 0.6202,
|
1298 |
+
"step": 89000
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 1.19,
|
1302 |
+
"learning_rate": 2.0227798919551854e-05,
|
1303 |
+
"loss": 0.6041,
|
1304 |
+
"step": 89500
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 1.2,
|
1308 |
+
"learning_rate": 2.006147377385103e-05,
|
1309 |
+
"loss": 0.5869,
|
1310 |
+
"step": 90000
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 1.2,
|
1314 |
+
"eval_bleu": 47.6385,
|
1315 |
+
"eval_chrf++": 67.2633,
|
1316 |
+
"eval_gen_len": 24.3011,
|
1317 |
+
"eval_loss": 0.7958626747131348,
|
1318 |
+
"eval_runtime": 4842.841,
|
1319 |
+
"eval_samples_per_second": 4.806,
|
1320 |
+
"eval_spbleu": 54.9942,
|
1321 |
+
"eval_steps_per_second": 0.961,
|
1322 |
+
"eval_ter": 41.356,
|
1323 |
+
"step": 90000
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 1.2,
|
1327 |
+
"learning_rate": 1.9895148628150198e-05,
|
1328 |
+
"loss": 0.6183,
|
1329 |
+
"step": 90500
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 1.21,
|
1333 |
+
"learning_rate": 1.972882348244937e-05,
|
1334 |
+
"loss": 0.617,
|
1335 |
+
"step": 91000
|
1336 |
+
},
|
1337 |
+
{
|
1338 |
+
"epoch": 1.22,
|
1339 |
+
"learning_rate": 1.9562498336748545e-05,
|
1340 |
+
"loss": 0.5859,
|
1341 |
+
"step": 91500
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 1.22,
|
1345 |
+
"learning_rate": 1.9396173191047717e-05,
|
1346 |
+
"loss": 0.6061,
|
1347 |
+
"step": 92000
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 1.23,
|
1351 |
+
"learning_rate": 1.9229848045346886e-05,
|
1352 |
+
"loss": 0.5971,
|
1353 |
+
"step": 92500
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 1.24,
|
1357 |
+
"learning_rate": 1.906352289964606e-05,
|
1358 |
+
"loss": 0.6302,
|
1359 |
+
"step": 93000
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 1.24,
|
1363 |
+
"learning_rate": 1.8897197753945234e-05,
|
1364 |
+
"loss": 0.5907,
|
1365 |
+
"step": 93500
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 1.25,
|
1369 |
+
"learning_rate": 1.8730872608244406e-05,
|
1370 |
+
"loss": 0.6167,
|
1371 |
+
"step": 94000
|
1372 |
+
},
|
1373 |
+
{
|
1374 |
+
"epoch": 1.26,
|
1375 |
+
"learning_rate": 1.8564547462543578e-05,
|
1376 |
+
"loss": 0.5929,
|
1377 |
+
"step": 94500
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 1.26,
|
1381 |
+
"learning_rate": 1.839822231684275e-05,
|
1382 |
+
"loss": 0.5778,
|
1383 |
+
"step": 95000
|
1384 |
+
},
|
1385 |
+
{
|
1386 |
+
"epoch": 1.26,
|
1387 |
+
"eval_bleu": 47.8127,
|
1388 |
+
"eval_chrf++": 67.3458,
|
1389 |
+
"eval_gen_len": 24.2553,
|
1390 |
+
"eval_loss": 0.7928242683410645,
|
1391 |
+
"eval_runtime": 4822.252,
|
1392 |
+
"eval_samples_per_second": 4.826,
|
1393 |
+
"eval_spbleu": 55.1003,
|
1394 |
+
"eval_steps_per_second": 0.965,
|
1395 |
+
"eval_ter": 41.2697,
|
1396 |
+
"step": 95000
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.27,
|
1400 |
+
"learning_rate": 1.8231897171141922e-05,
|
1401 |
+
"loss": 0.6058,
|
1402 |
+
"step": 95500
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 1.28,
|
1406 |
+
"learning_rate": 1.8065572025441097e-05,
|
1407 |
+
"loss": 0.6104,
|
1408 |
+
"step": 96000
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 1.28,
|
1412 |
+
"learning_rate": 1.7899246879740266e-05,
|
1413 |
+
"loss": 0.5958,
|
1414 |
+
"step": 96500
|
1415 |
+
},
|
1416 |
+
{
|
1417 |
+
"epoch": 1.29,
|
1418 |
+
"learning_rate": 1.7732921734039438e-05,
|
1419 |
+
"loss": 0.5968,
|
1420 |
+
"step": 97000
|
1421 |
+
},
|
1422 |
+
{
|
1423 |
+
"epoch": 1.3,
|
1424 |
+
"learning_rate": 1.7566596588338614e-05,
|
1425 |
+
"loss": 0.6034,
|
1426 |
+
"step": 97500
|
1427 |
+
},
|
1428 |
+
{
|
1429 |
+
"epoch": 1.3,
|
1430 |
+
"learning_rate": 1.7400271442637782e-05,
|
1431 |
+
"loss": 0.6189,
|
1432 |
+
"step": 98000
|
1433 |
+
},
|
1434 |
+
{
|
1435 |
+
"epoch": 1.31,
|
1436 |
+
"learning_rate": 1.7233946296936958e-05,
|
1437 |
+
"loss": 0.6028,
|
1438 |
+
"step": 98500
|
1439 |
+
},
|
1440 |
+
{
|
1441 |
+
"epoch": 1.32,
|
1442 |
+
"learning_rate": 1.706762115123613e-05,
|
1443 |
+
"loss": 0.6044,
|
1444 |
+
"step": 99000
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 1.32,
|
1448 |
+
"learning_rate": 1.6901296005535302e-05,
|
1449 |
+
"loss": 0.6021,
|
1450 |
+
"step": 99500
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 1.33,
|
1454 |
+
"learning_rate": 1.6734970859834474e-05,
|
1455 |
+
"loss": 0.6055,
|
1456 |
+
"step": 100000
|
1457 |
+
},
|
1458 |
+
{
|
1459 |
+
"epoch": 1.33,
|
1460 |
+
"eval_bleu": 47.9968,
|
1461 |
+
"eval_chrf++": 67.5009,
|
1462 |
+
"eval_gen_len": 24.343,
|
1463 |
+
"eval_loss": 0.7921544313430786,
|
1464 |
+
"eval_runtime": 4846.7187,
|
1465 |
+
"eval_samples_per_second": 4.802,
|
1466 |
+
"eval_spbleu": 55.2772,
|
1467 |
+
"eval_steps_per_second": 0.96,
|
1468 |
+
"eval_ter": 41.328,
|
1469 |
+
"step": 100000
|
1470 |
+
},
|
1471 |
+
{
|
1472 |
+
"epoch": 1.34,
|
1473 |
+
"learning_rate": 1.6568645714133646e-05,
|
1474 |
+
"loss": 0.6328,
|
1475 |
+
"step": 100500
|
1476 |
+
},
|
1477 |
+
{
|
1478 |
+
"epoch": 1.34,
|
1479 |
+
"learning_rate": 1.6402320568432818e-05,
|
1480 |
+
"loss": 0.6082,
|
1481 |
+
"step": 101000
|
1482 |
+
},
|
1483 |
+
{
|
1484 |
+
"epoch": 1.35,
|
1485 |
+
"learning_rate": 1.6235995422731994e-05,
|
1486 |
+
"loss": 0.6217,
|
1487 |
+
"step": 101500
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 1.36,
|
1491 |
+
"learning_rate": 1.6069670277031162e-05,
|
1492 |
+
"loss": 0.6225,
|
1493 |
+
"step": 102000
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 1.36,
|
1497 |
+
"learning_rate": 1.5903345131330335e-05,
|
1498 |
+
"loss": 0.6126,
|
1499 |
+
"step": 102500
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 1.37,
|
1503 |
+
"learning_rate": 1.573701998562951e-05,
|
1504 |
+
"loss": 0.6208,
|
1505 |
+
"step": 103000
|
1506 |
+
},
|
1507 |
+
{
|
1508 |
+
"epoch": 1.38,
|
1509 |
+
"learning_rate": 1.5570694839928682e-05,
|
1510 |
+
"loss": 0.6068,
|
1511 |
+
"step": 103500
|
1512 |
+
},
|
1513 |
+
{
|
1514 |
+
"epoch": 1.38,
|
1515 |
+
"learning_rate": 1.540436969422785e-05,
|
1516 |
+
"loss": 0.6062,
|
1517 |
+
"step": 104000
|
1518 |
+
},
|
1519 |
+
{
|
1520 |
+
"epoch": 1.39,
|
1521 |
+
"learning_rate": 1.5238044548527025e-05,
|
1522 |
+
"loss": 0.6063,
|
1523 |
+
"step": 104500
|
1524 |
+
},
|
1525 |
+
{
|
1526 |
+
"epoch": 1.4,
|
1527 |
+
"learning_rate": 1.5071719402826198e-05,
|
1528 |
+
"loss": 0.6136,
|
1529 |
+
"step": 105000
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.4,
|
1533 |
+
"eval_bleu": 48.1927,
|
1534 |
+
"eval_chrf++": 67.6488,
|
1535 |
+
"eval_gen_len": 24.2325,
|
1536 |
+
"eval_loss": 0.7866764664649963,
|
1537 |
+
"eval_runtime": 4833.6742,
|
1538 |
+
"eval_samples_per_second": 4.815,
|
1539 |
+
"eval_spbleu": 55.5056,
|
1540 |
+
"eval_steps_per_second": 0.963,
|
1541 |
+
"eval_ter": 40.8666,
|
1542 |
+
"step": 105000
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 1.4,
|
1546 |
+
"learning_rate": 1.490539425712537e-05,
|
1547 |
+
"loss": 0.5853,
|
1548 |
+
"step": 105500
|
1549 |
+
},
|
1550 |
+
{
|
1551 |
+
"epoch": 1.41,
|
1552 |
+
"learning_rate": 1.4739069111424541e-05,
|
1553 |
+
"loss": 0.6171,
|
1554 |
+
"step": 106000
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 1.42,
|
1558 |
+
"learning_rate": 1.4572743965723715e-05,
|
1559 |
+
"loss": 0.6224,
|
1560 |
+
"step": 106500
|
1561 |
+
},
|
1562 |
+
{
|
1563 |
+
"epoch": 1.42,
|
1564 |
+
"learning_rate": 1.4406418820022888e-05,
|
1565 |
+
"loss": 0.6133,
|
1566 |
+
"step": 107000
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 1.43,
|
1570 |
+
"learning_rate": 1.4240093674322059e-05,
|
1571 |
+
"loss": 0.6056,
|
1572 |
+
"step": 107500
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 1.44,
|
1576 |
+
"learning_rate": 1.4073768528621231e-05,
|
1577 |
+
"loss": 0.6023,
|
1578 |
+
"step": 108000
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 1.44,
|
1582 |
+
"learning_rate": 1.3907443382920405e-05,
|
1583 |
+
"loss": 0.6124,
|
1584 |
+
"step": 108500
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 1.45,
|
1588 |
+
"learning_rate": 1.3741118237219577e-05,
|
1589 |
+
"loss": 0.5926,
|
1590 |
+
"step": 109000
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 1.46,
|
1594 |
+
"learning_rate": 1.3574793091518747e-05,
|
1595 |
+
"loss": 0.6123,
|
1596 |
+
"step": 109500
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.46,
|
1600 |
+
"learning_rate": 1.3408467945817921e-05,
|
1601 |
+
"loss": 0.5983,
|
1602 |
+
"step": 110000
|
1603 |
+
},
|
1604 |
+
{
|
1605 |
+
"epoch": 1.46,
|
1606 |
+
"eval_bleu": 48.2871,
|
1607 |
+
"eval_chrf++": 67.7183,
|
1608 |
+
"eval_gen_len": 24.2734,
|
1609 |
+
"eval_loss": 0.7818799018859863,
|
1610 |
+
"eval_runtime": 4831.9282,
|
1611 |
+
"eval_samples_per_second": 4.817,
|
1612 |
+
"eval_spbleu": 55.6111,
|
1613 |
+
"eval_steps_per_second": 0.963,
|
1614 |
+
"eval_ter": 40.8669,
|
1615 |
+
"step": 110000
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 1.47,
|
1619 |
+
"learning_rate": 1.3242142800117093e-05,
|
1620 |
+
"loss": 0.6285,
|
1621 |
+
"step": 110500
|
1622 |
+
},
|
1623 |
+
{
|
1624 |
+
"epoch": 1.48,
|
1625 |
+
"learning_rate": 1.3075817654416267e-05,
|
1626 |
+
"loss": 0.5953,
|
1627 |
+
"step": 111000
|
1628 |
+
},
|
1629 |
+
{
|
1630 |
+
"epoch": 1.48,
|
1631 |
+
"learning_rate": 1.2909492508715437e-05,
|
1632 |
+
"loss": 0.6105,
|
1633 |
+
"step": 111500
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 1.49,
|
1637 |
+
"learning_rate": 1.2743167363014611e-05,
|
1638 |
+
"loss": 0.6008,
|
1639 |
+
"step": 112000
|
1640 |
+
},
|
1641 |
+
{
|
1642 |
+
"epoch": 1.5,
|
1643 |
+
"learning_rate": 1.2576842217313783e-05,
|
1644 |
+
"loss": 0.615,
|
1645 |
+
"step": 112500
|
1646 |
+
},
|
1647 |
+
{
|
1648 |
+
"epoch": 1.5,
|
1649 |
+
"learning_rate": 1.2410517071612955e-05,
|
1650 |
+
"loss": 0.6252,
|
1651 |
+
"step": 113000
|
1652 |
+
},
|
1653 |
+
{
|
1654 |
+
"epoch": 1.51,
|
1655 |
+
"learning_rate": 1.2244191925912129e-05,
|
1656 |
+
"loss": 0.5896,
|
1657 |
+
"step": 113500
|
1658 |
+
},
|
1659 |
+
{
|
1660 |
+
"epoch": 1.52,
|
1661 |
+
"learning_rate": 1.20778667802113e-05,
|
1662 |
+
"loss": 0.5892,
|
1663 |
+
"step": 114000
|
1664 |
+
},
|
1665 |
+
{
|
1666 |
+
"epoch": 1.52,
|
1667 |
+
"learning_rate": 1.1911541634510473e-05,
|
1668 |
+
"loss": 0.5956,
|
1669 |
+
"step": 114500
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.53,
|
1673 |
+
"learning_rate": 1.1745216488809645e-05,
|
1674 |
+
"loss": 0.6225,
|
1675 |
+
"step": 115000
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 1.53,
|
1679 |
+
"eval_bleu": 48.489,
|
1680 |
+
"eval_chrf++": 67.8465,
|
1681 |
+
"eval_gen_len": 24.3063,
|
1682 |
+
"eval_loss": 0.7788467407226562,
|
1683 |
+
"eval_runtime": 4828.3962,
|
1684 |
+
"eval_samples_per_second": 4.82,
|
1685 |
+
"eval_spbleu": 55.8079,
|
1686 |
+
"eval_steps_per_second": 0.964,
|
1687 |
+
"eval_ter": 40.8166,
|
1688 |
+
"step": 115000
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 1.54,
|
1692 |
+
"learning_rate": 1.1578891343108817e-05,
|
1693 |
+
"loss": 0.5987,
|
1694 |
+
"step": 115500
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 1.54,
|
1698 |
+
"learning_rate": 1.141256619740799e-05,
|
1699 |
+
"loss": 0.5966,
|
1700 |
+
"step": 116000
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 1.55,
|
1704 |
+
"learning_rate": 1.1246241051707161e-05,
|
1705 |
+
"loss": 0.6044,
|
1706 |
+
"step": 116500
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 1.56,
|
1710 |
+
"learning_rate": 1.1079915906006334e-05,
|
1711 |
+
"loss": 0.6069,
|
1712 |
+
"step": 117000
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 1.56,
|
1716 |
+
"learning_rate": 1.0913590760305506e-05,
|
1717 |
+
"loss": 0.586,
|
1718 |
+
"step": 117500
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 1.57,
|
1722 |
+
"learning_rate": 1.074726561460468e-05,
|
1723 |
+
"loss": 0.6076,
|
1724 |
+
"step": 118000
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 1.58,
|
1728 |
+
"learning_rate": 1.0580940468903852e-05,
|
1729 |
+
"loss": 0.5896,
|
1730 |
+
"step": 118500
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 1.58,
|
1734 |
+
"learning_rate": 1.0414615323203024e-05,
|
1735 |
+
"loss": 0.5918,
|
1736 |
+
"step": 119000
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 1.59,
|
1740 |
+
"learning_rate": 1.0248290177502196e-05,
|
1741 |
+
"loss": 0.5885,
|
1742 |
+
"step": 119500
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 1.6,
|
1746 |
+
"learning_rate": 1.0081965031801368e-05,
|
1747 |
+
"loss": 0.5937,
|
1748 |
+
"step": 120000
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 1.6,
|
1752 |
+
"eval_bleu": 48.6385,
|
1753 |
+
"eval_chrf++": 67.9368,
|
1754 |
+
"eval_gen_len": 24.2573,
|
1755 |
+
"eval_loss": 0.7750729322433472,
|
1756 |
+
"eval_runtime": 4834.5977,
|
1757 |
+
"eval_samples_per_second": 4.814,
|
1758 |
+
"eval_spbleu": 55.9507,
|
1759 |
+
"eval_steps_per_second": 0.963,
|
1760 |
+
"eval_ter": 40.604,
|
1761 |
+
"step": 120000
|
1762 |
+
},
|
1763 |
+
{
|
1764 |
+
"epoch": 1.6,
|
1765 |
+
"learning_rate": 9.91563988610054e-06,
|
1766 |
+
"loss": 0.5941,
|
1767 |
+
"step": 120500
|
1768 |
+
},
|
1769 |
+
{
|
1770 |
+
"epoch": 1.61,
|
1771 |
+
"learning_rate": 9.749314740399714e-06,
|
1772 |
+
"loss": 0.5856,
|
1773 |
+
"step": 121000
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 1.62,
|
1777 |
+
"learning_rate": 9.582989594698886e-06,
|
1778 |
+
"loss": 0.5904,
|
1779 |
+
"step": 121500
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 1.62,
|
1783 |
+
"learning_rate": 9.416664448998058e-06,
|
1784 |
+
"loss": 0.5872,
|
1785 |
+
"step": 122000
|
1786 |
+
},
|
1787 |
+
{
|
1788 |
+
"epoch": 1.63,
|
1789 |
+
"learning_rate": 9.25033930329723e-06,
|
1790 |
+
"loss": 0.6085,
|
1791 |
+
"step": 122500
|
1792 |
+
},
|
1793 |
+
{
|
1794 |
+
"epoch": 1.64,
|
1795 |
+
"learning_rate": 9.084014157596404e-06,
|
1796 |
+
"loss": 0.6025,
|
1797 |
+
"step": 123000
|
1798 |
+
},
|
1799 |
+
{
|
1800 |
+
"epoch": 1.64,
|
1801 |
+
"learning_rate": 8.917689011895574e-06,
|
1802 |
+
"loss": 0.6026,
|
1803 |
+
"step": 123500
|
1804 |
+
},
|
1805 |
+
{
|
1806 |
+
"epoch": 1.65,
|
1807 |
+
"learning_rate": 8.751363866194748e-06,
|
1808 |
+
"loss": 0.5826,
|
1809 |
+
"step": 124000
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 1.66,
|
1813 |
+
"learning_rate": 8.58503872049392e-06,
|
1814 |
+
"loss": 0.5989,
|
1815 |
+
"step": 124500
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 1.66,
|
1819 |
+
"learning_rate": 8.418713574793092e-06,
|
1820 |
+
"loss": 0.5934,
|
1821 |
+
"step": 125000
|
1822 |
+
},
|
1823 |
+
{
|
1824 |
+
"epoch": 1.66,
|
1825 |
+
"eval_bleu": 48.7432,
|
1826 |
+
"eval_chrf++": 67.9418,
|
1827 |
+
"eval_gen_len": 24.2522,
|
1828 |
+
"eval_loss": 0.7731887698173523,
|
1829 |
+
"eval_runtime": 4826.8535,
|
1830 |
+
"eval_samples_per_second": 4.822,
|
1831 |
+
"eval_spbleu": 56.0071,
|
1832 |
+
"eval_steps_per_second": 0.964,
|
1833 |
+
"eval_ter": 40.5337,
|
1834 |
+
"step": 125000
|
1835 |
+
},
|
1836 |
+
{
|
1837 |
+
"epoch": 1.67,
|
1838 |
+
"learning_rate": 8.252388429092264e-06,
|
1839 |
+
"loss": 0.5797,
|
1840 |
+
"step": 125500
|
1841 |
+
},
|
1842 |
+
{
|
1843 |
+
"epoch": 1.68,
|
1844 |
+
"learning_rate": 8.086063283391436e-06,
|
1845 |
+
"loss": 0.5713,
|
1846 |
+
"step": 126000
|
1847 |
+
},
|
1848 |
+
{
|
1849 |
+
"epoch": 1.68,
|
1850 |
+
"learning_rate": 7.919738137690608e-06,
|
1851 |
+
"loss": 0.5993,
|
1852 |
+
"step": 126500
|
1853 |
+
},
|
1854 |
+
{
|
1855 |
+
"epoch": 1.69,
|
1856 |
+
"learning_rate": 7.75341299198978e-06,
|
1857 |
+
"loss": 0.5953,
|
1858 |
+
"step": 127000
|
1859 |
+
},
|
1860 |
+
{
|
1861 |
+
"epoch": 1.7,
|
1862 |
+
"learning_rate": 7.587087846288954e-06,
|
1863 |
+
"loss": 0.5866,
|
1864 |
+
"step": 127500
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 1.7,
|
1868 |
+
"learning_rate": 7.4207627005881254e-06,
|
1869 |
+
"loss": 0.5942,
|
1870 |
+
"step": 128000
|
1871 |
+
},
|
1872 |
+
{
|
1873 |
+
"epoch": 1.71,
|
1874 |
+
"learning_rate": 7.254437554887298e-06,
|
1875 |
+
"loss": 0.6301,
|
1876 |
+
"step": 128500
|
1877 |
+
},
|
1878 |
+
{
|
1879 |
+
"epoch": 1.72,
|
1880 |
+
"learning_rate": 7.0881124091864705e-06,
|
1881 |
+
"loss": 0.6143,
|
1882 |
+
"step": 129000
|
1883 |
+
},
|
1884 |
+
{
|
1885 |
+
"epoch": 1.72,
|
1886 |
+
"learning_rate": 6.921787263485643e-06,
|
1887 |
+
"loss": 0.5774,
|
1888 |
+
"step": 129500
|
1889 |
+
},
|
1890 |
+
{
|
1891 |
+
"epoch": 1.73,
|
1892 |
+
"learning_rate": 6.7554621177848155e-06,
|
1893 |
+
"loss": 0.5974,
|
1894 |
+
"step": 130000
|
1895 |
+
},
|
1896 |
+
{
|
1897 |
+
"epoch": 1.73,
|
1898 |
+
"eval_bleu": 48.9376,
|
1899 |
+
"eval_chrf++": 68.1416,
|
1900 |
+
"eval_gen_len": 24.3125,
|
1901 |
+
"eval_loss": 0.7730206847190857,
|
1902 |
+
"eval_runtime": 4839.9438,
|
1903 |
+
"eval_samples_per_second": 4.809,
|
1904 |
+
"eval_spbleu": 56.1898,
|
1905 |
+
"eval_steps_per_second": 0.962,
|
1906 |
+
"eval_ter": 40.479,
|
1907 |
+
"step": 130000
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 1.74,
|
1911 |
+
"learning_rate": 6.589136972083988e-06,
|
1912 |
+
"loss": 0.5838,
|
1913 |
+
"step": 130500
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 1.74,
|
1917 |
+
"learning_rate": 6.42281182638316e-06,
|
1918 |
+
"loss": 0.5732,
|
1919 |
+
"step": 131000
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 1.75,
|
1923 |
+
"learning_rate": 6.2564866806823334e-06,
|
1924 |
+
"loss": 0.5922,
|
1925 |
+
"step": 131500
|
1926 |
+
},
|
1927 |
+
{
|
1928 |
+
"epoch": 1.76,
|
1929 |
+
"learning_rate": 6.090161534981505e-06,
|
1930 |
+
"loss": 0.5852,
|
1931 |
+
"step": 132000
|
1932 |
+
},
|
1933 |
+
{
|
1934 |
+
"epoch": 1.76,
|
1935 |
+
"learning_rate": 5.923836389280677e-06,
|
1936 |
+
"loss": 0.5853,
|
1937 |
+
"step": 132500
|
1938 |
+
},
|
1939 |
+
{
|
1940 |
+
"epoch": 1.77,
|
1941 |
+
"learning_rate": 5.75751124357985e-06,
|
1942 |
+
"loss": 0.6043,
|
1943 |
+
"step": 133000
|
1944 |
+
},
|
1945 |
+
{
|
1946 |
+
"epoch": 1.78,
|
1947 |
+
"learning_rate": 5.591186097879022e-06,
|
1948 |
+
"loss": 0.6143,
|
1949 |
+
"step": 133500
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 1.78,
|
1953 |
+
"learning_rate": 5.424860952178194e-06,
|
1954 |
+
"loss": 0.5775,
|
1955 |
+
"step": 134000
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 1.79,
|
1959 |
+
"learning_rate": 5.258535806477367e-06,
|
1960 |
+
"loss": 0.5763,
|
1961 |
+
"step": 134500
|
1962 |
+
},
|
1963 |
+
{
|
1964 |
+
"epoch": 1.8,
|
1965 |
+
"learning_rate": 5.092210660776539e-06,
|
1966 |
+
"loss": 0.5944,
|
1967 |
+
"step": 135000
|
1968 |
+
},
|
1969 |
+
{
|
1970 |
+
"epoch": 1.8,
|
1971 |
+
"eval_bleu": 49.0549,
|
1972 |
+
"eval_chrf++": 68.2099,
|
1973 |
+
"eval_gen_len": 24.3198,
|
1974 |
+
"eval_loss": 0.7679416537284851,
|
1975 |
+
"eval_runtime": 4837.1327,
|
1976 |
+
"eval_samples_per_second": 4.812,
|
1977 |
+
"eval_spbleu": 56.3421,
|
1978 |
+
"eval_steps_per_second": 0.962,
|
1979 |
+
"eval_ter": 40.2163,
|
1980 |
+
"step": 135000
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 1.8,
|
1984 |
+
"learning_rate": 4.925885515075712e-06,
|
1985 |
+
"loss": 0.6115,
|
1986 |
+
"step": 135500
|
1987 |
+
},
|
1988 |
+
{
|
1989 |
+
"epoch": 1.81,
|
1990 |
+
"learning_rate": 4.759560369374884e-06,
|
1991 |
+
"loss": 0.5966,
|
1992 |
+
"step": 136000
|
1993 |
+
},
|
1994 |
+
{
|
1995 |
+
"epoch": 1.82,
|
1996 |
+
"learning_rate": 4.593235223674056e-06,
|
1997 |
+
"loss": 0.5748,
|
1998 |
+
"step": 136500
|
1999 |
+
},
|
2000 |
+
{
|
2001 |
+
"epoch": 1.82,
|
2002 |
+
"learning_rate": 4.426910077973229e-06,
|
2003 |
+
"loss": 0.573,
|
2004 |
+
"step": 137000
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 1.83,
|
2008 |
+
"learning_rate": 4.260584932272401e-06,
|
2009 |
+
"loss": 0.5958,
|
2010 |
+
"step": 137500
|
2011 |
+
},
|
2012 |
+
{
|
2013 |
+
"epoch": 1.84,
|
2014 |
+
"learning_rate": 4.094259786571573e-06,
|
2015 |
+
"loss": 0.577,
|
2016 |
+
"step": 138000
|
2017 |
+
},
|
2018 |
+
{
|
2019 |
+
"epoch": 1.84,
|
2020 |
+
"learning_rate": 3.927934640870746e-06,
|
2021 |
+
"loss": 0.5966,
|
2022 |
+
"step": 138500
|
2023 |
+
},
|
2024 |
+
{
|
2025 |
+
"epoch": 1.85,
|
2026 |
+
"learning_rate": 3.761609495169918e-06,
|
2027 |
+
"loss": 0.5884,
|
2028 |
+
"step": 139000
|
2029 |
+
},
|
2030 |
+
{
|
2031 |
+
"epoch": 1.86,
|
2032 |
+
"learning_rate": 3.5952843494690906e-06,
|
2033 |
+
"loss": 0.5924,
|
2034 |
+
"step": 139500
|
2035 |
+
},
|
2036 |
+
{
|
2037 |
+
"epoch": 1.86,
|
2038 |
+
"learning_rate": 3.4289592037682623e-06,
|
2039 |
+
"loss": 0.5753,
|
2040 |
+
"step": 140000
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 1.86,
|
2044 |
+
"eval_bleu": 49.1791,
|
2045 |
+
"eval_chrf++": 68.2787,
|
2046 |
+
"eval_gen_len": 24.3117,
|
2047 |
+
"eval_loss": 0.7665663361549377,
|
2048 |
+
"eval_runtime": 4836.1543,
|
2049 |
+
"eval_samples_per_second": 4.813,
|
2050 |
+
"eval_spbleu": 56.4922,
|
2051 |
+
"eval_steps_per_second": 0.963,
|
2052 |
+
"eval_ter": 40.1748,
|
2053 |
+
"step": 140000
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 1.87,
|
2057 |
+
"learning_rate": 3.262634058067435e-06,
|
2058 |
+
"loss": 0.5892,
|
2059 |
+
"step": 140500
|
2060 |
+
},
|
2061 |
+
{
|
2062 |
+
"epoch": 1.88,
|
2063 |
+
"learning_rate": 3.0963089123666073e-06,
|
2064 |
+
"loss": 0.5891,
|
2065 |
+
"step": 141000
|
2066 |
+
},
|
2067 |
+
{
|
2068 |
+
"epoch": 1.88,
|
2069 |
+
"learning_rate": 2.92998376666578e-06,
|
2070 |
+
"loss": 0.5821,
|
2071 |
+
"step": 141500
|
2072 |
+
},
|
2073 |
+
{
|
2074 |
+
"epoch": 1.89,
|
2075 |
+
"learning_rate": 2.7636586209649524e-06,
|
2076 |
+
"loss": 0.5701,
|
2077 |
+
"step": 142000
|
2078 |
+
},
|
2079 |
+
{
|
2080 |
+
"epoch": 1.9,
|
2081 |
+
"learning_rate": 2.5973334752641244e-06,
|
2082 |
+
"loss": 0.5632,
|
2083 |
+
"step": 142500
|
2084 |
+
},
|
2085 |
+
{
|
2086 |
+
"epoch": 1.9,
|
2087 |
+
"learning_rate": 2.4310083295632965e-06,
|
2088 |
+
"loss": 0.588,
|
2089 |
+
"step": 143000
|
2090 |
+
},
|
2091 |
+
{
|
2092 |
+
"epoch": 1.91,
|
2093 |
+
"learning_rate": 2.264683183862469e-06,
|
2094 |
+
"loss": 0.604,
|
2095 |
+
"step": 143500
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 1.92,
|
2099 |
+
"learning_rate": 2.0983580381616416e-06,
|
2100 |
+
"loss": 0.5766,
|
2101 |
+
"step": 144000
|
2102 |
+
},
|
2103 |
+
{
|
2104 |
+
"epoch": 1.92,
|
2105 |
+
"learning_rate": 1.932032892460814e-06,
|
2106 |
+
"loss": 0.595,
|
2107 |
+
"step": 144500
|
2108 |
+
},
|
2109 |
+
{
|
2110 |
+
"epoch": 1.93,
|
2111 |
+
"learning_rate": 1.7657077467599864e-06,
|
2112 |
+
"loss": 0.5907,
|
2113 |
+
"step": 145000
|
2114 |
+
},
|
2115 |
+
{
|
2116 |
+
"epoch": 1.93,
|
2117 |
+
"eval_bleu": 49.2939,
|
2118 |
+
"eval_chrf++": 68.3612,
|
2119 |
+
"eval_gen_len": 24.3154,
|
2120 |
+
"eval_loss": 0.7650372982025146,
|
2121 |
+
"eval_runtime": 4848.1649,
|
2122 |
+
"eval_samples_per_second": 4.801,
|
2123 |
+
"eval_spbleu": 56.5766,
|
2124 |
+
"eval_steps_per_second": 0.96,
|
2125 |
+
"eval_ter": 40.0916,
|
2126 |
+
"step": 145000
|
2127 |
+
},
|
2128 |
+
{
|
2129 |
+
"epoch": 1.94,
|
2130 |
+
"learning_rate": 1.5993826010591587e-06,
|
2131 |
+
"loss": 0.6039,
|
2132 |
+
"step": 145500
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 1.94,
|
2136 |
+
"learning_rate": 1.433057455358331e-06,
|
2137 |
+
"loss": 0.5995,
|
2138 |
+
"step": 146000
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 1.95,
|
2142 |
+
"learning_rate": 1.2667323096575033e-06,
|
2143 |
+
"loss": 0.5899,
|
2144 |
+
"step": 146500
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 1.96,
|
2148 |
+
"learning_rate": 1.1004071639566756e-06,
|
2149 |
+
"loss": 0.5666,
|
2150 |
+
"step": 147000
|
2151 |
+
},
|
2152 |
+
{
|
2153 |
+
"epoch": 1.96,
|
2154 |
+
"learning_rate": 9.340820182558481e-07,
|
2155 |
+
"loss": 0.5925,
|
2156 |
+
"step": 147500
|
2157 |
+
},
|
2158 |
+
{
|
2159 |
+
"epoch": 1.97,
|
2160 |
+
"learning_rate": 7.677568725550204e-07,
|
2161 |
+
"loss": 0.6101,
|
2162 |
+
"step": 148000
|
2163 |
+
},
|
2164 |
+
{
|
2165 |
+
"epoch": 1.98,
|
2166 |
+
"learning_rate": 6.014317268541928e-07,
|
2167 |
+
"loss": 0.6,
|
2168 |
+
"step": 148500
|
2169 |
+
},
|
2170 |
+
{
|
2171 |
+
"epoch": 1.98,
|
2172 |
+
"learning_rate": 4.351065811533651e-07,
|
2173 |
+
"loss": 0.5961,
|
2174 |
+
"step": 149000
|
2175 |
+
},
|
2176 |
+
{
|
2177 |
+
"epoch": 1.99,
|
2178 |
+
"learning_rate": 2.687814354525375e-07,
|
2179 |
+
"loss": 0.5851,
|
2180 |
+
"step": 149500
|
2181 |
+
},
|
2182 |
+
{
|
2183 |
+
"epoch": 2.0,
|
2184 |
+
"learning_rate": 1.0245628975170984e-07,
|
2185 |
+
"loss": 0.5811,
|
2186 |
+
"step": 150000
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 2.0,
|
2190 |
+
"eval_bleu": 49.3031,
|
2191 |
+
"eval_chrf++": 68.3562,
|
2192 |
+
"eval_gen_len": 24.3172,
|
2193 |
+
"eval_loss": 0.7639886140823364,
|
2194 |
+
"eval_runtime": 4840.3648,
|
2195 |
+
"eval_samples_per_second": 4.808,
|
2196 |
+
"eval_spbleu": 56.6,
|
2197 |
+
"eval_steps_per_second": 0.962,
|
2198 |
+
"eval_ter": 40.1023,
|
2199 |
+
"step": 150000
|
2200 |
+
}
|
2201 |
+
],
|
2202 |
+
"logging_steps": 500,
|
2203 |
+
"max_steps": 150308,
|
2204 |
+
"num_train_epochs": 2,
|
2205 |
+
"save_steps": 5000,
|
2206 |
+
"total_flos": 6.383678253160202e+17,
|
2207 |
+
"trial_name": null,
|
2208 |
+
"trial_params": null
|
2209 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d5fc2b62c793ce00568cc8d0a220a46db3853a8757a1fc69eeb752196dcc764
|
3 |
+
size 4664
|