Upload folder using huggingface_hub
Browse files- config.json +33 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +23 -0
- spiece.model +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +38 -0
- trainer_state.json +1810 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "allegro/plt5-base",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"classifier_dropout": 0.0,
|
7 |
+
"d_ff": 2048,
|
8 |
+
"d_kv": 64,
|
9 |
+
"d_model": 768,
|
10 |
+
"decoder_start_token_id": 0,
|
11 |
+
"dense_act_fn": "gelu_new",
|
12 |
+
"dropout_rate": 0.1,
|
13 |
+
"eos_token_id": 1,
|
14 |
+
"feed_forward_proj": "gated-gelu",
|
15 |
+
"initializer_factor": 1.0,
|
16 |
+
"is_encoder_decoder": true,
|
17 |
+
"is_gated_act": true,
|
18 |
+
"layer_norm_epsilon": 1e-06,
|
19 |
+
"model_type": "t5",
|
20 |
+
"num_decoder_layers": 12,
|
21 |
+
"num_heads": 12,
|
22 |
+
"num_layers": 12,
|
23 |
+
"output_past": true,
|
24 |
+
"pad_token_id": 0,
|
25 |
+
"relative_attention_max_distance": 128,
|
26 |
+
"relative_attention_num_buckets": 32,
|
27 |
+
"tie_word_embeddings": false,
|
28 |
+
"tokenizer_class": "T5Tokenizer",
|
29 |
+
"torch_dtype": "float32",
|
30 |
+
"transformers_version": "4.35.2",
|
31 |
+
"use_cache": true,
|
32 |
+
"vocab_size": 50048
|
33 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"decoder_start_token_id": 0,
|
3 |
+
"eos_token_id": 1,
|
4 |
+
"pad_token_id": 0,
|
5 |
+
"transformers_version": "4.35.2"
|
6 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:da5f842d2d896f517a3cdbc5b4a64e371f5f59f58acbf4246b998551df908238
|
3 |
+
size 1100445544
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8e81e8be5eaa829d349c26023c283d54d8ee74ac2c215112ff9a7bb616bd933c
|
3 |
+
size 2515130
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ca7808dc82b7af7b62ec275ab306c805a2d3ff514737b020dbb6239a0cb0c0ba
|
3 |
+
size 14244
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3d66d028b294211f06a27742e89402436e4914db4f8ac387faede1b2eaab537c
|
3 |
+
size 1064
|
special_tokens_map.json
ADDED
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eos_token": {
|
3 |
+
"content": "</s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"pad_token": {
|
10 |
+
"content": "<pad>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"unk_token": {
|
17 |
+
"content": "<unk>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a200b4b9d7106b32ea2340e5cd27033a5c03ae5ece9241ab6e1f685b938800b9
|
3 |
+
size 1124950
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<pad>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "</s>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "<unk>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
}
|
27 |
+
},
|
28 |
+
"additional_special_tokens": [],
|
29 |
+
"clean_up_tokenization_spaces": true,
|
30 |
+
"eos_token": "</s>",
|
31 |
+
"extra_ids": 0,
|
32 |
+
"legacy": true,
|
33 |
+
"model_max_length": 1000000000000000019884624838656,
|
34 |
+
"pad_token": "<pad>",
|
35 |
+
"sp_model_kwargs": {},
|
36 |
+
"tokenizer_class": "T5Tokenizer",
|
37 |
+
"unk_token": "<unk>"
|
38 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1810 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 50,
|
6 |
+
"global_step": 4254,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"learning_rate": 9.998432847516064e-05,
|
14 |
+
"loss": 1.7366,
|
15 |
+
"step": 1
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.02,
|
19 |
+
"learning_rate": 9.921642375803167e-05,
|
20 |
+
"loss": 1.6411,
|
21 |
+
"step": 50
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.02,
|
25 |
+
"eval_bleu": 0.0,
|
26 |
+
"eval_brevity_penalty": 0.0444548482935275,
|
27 |
+
"eval_exact_match": 0.06640625,
|
28 |
+
"eval_gen_len": 2.0,
|
29 |
+
"eval_length_ratio": 0.24311490978157646,
|
30 |
+
"eval_loss": 2.303663492202759,
|
31 |
+
"eval_reference_length": 2106,
|
32 |
+
"eval_runtime": 21.5251,
|
33 |
+
"eval_samples_per_second": 23.786,
|
34 |
+
"eval_steps_per_second": 1.487,
|
35 |
+
"eval_translation_length": 512,
|
36 |
+
"step": 50
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.05,
|
40 |
+
"learning_rate": 9.843284751606332e-05,
|
41 |
+
"loss": 2.8505,
|
42 |
+
"step": 100
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.05,
|
46 |
+
"eval_bleu": 0.0,
|
47 |
+
"eval_brevity_penalty": 0.0444548482935275,
|
48 |
+
"eval_exact_match": 0.068359375,
|
49 |
+
"eval_gen_len": 2.0,
|
50 |
+
"eval_length_ratio": 0.24311490978157646,
|
51 |
+
"eval_loss": 2.0377542972564697,
|
52 |
+
"eval_reference_length": 2106,
|
53 |
+
"eval_runtime": 21.9349,
|
54 |
+
"eval_samples_per_second": 23.342,
|
55 |
+
"eval_steps_per_second": 1.459,
|
56 |
+
"eval_translation_length": 512,
|
57 |
+
"step": 100
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.07,
|
61 |
+
"learning_rate": 9.764927127409498e-05,
|
62 |
+
"loss": 2.5985,
|
63 |
+
"step": 150
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.07,
|
67 |
+
"eval_bleu": 0.0,
|
68 |
+
"eval_brevity_penalty": 0.05906652611110617,
|
69 |
+
"eval_exact_match": 0.060546875,
|
70 |
+
"eval_gen_len": 2.115234375,
|
71 |
+
"eval_length_ratio": 0.2611585944919278,
|
72 |
+
"eval_loss": 1.8557136058807373,
|
73 |
+
"eval_reference_length": 2106,
|
74 |
+
"eval_runtime": 22.3363,
|
75 |
+
"eval_samples_per_second": 22.922,
|
76 |
+
"eval_steps_per_second": 1.433,
|
77 |
+
"eval_translation_length": 550,
|
78 |
+
"step": 150
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 0.09,
|
82 |
+
"learning_rate": 9.686569503212663e-05,
|
83 |
+
"loss": 2.4089,
|
84 |
+
"step": 200
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.09,
|
88 |
+
"eval_bleu": 0.011576539062082333,
|
89 |
+
"eval_brevity_penalty": 0.0658189125921447,
|
90 |
+
"eval_exact_match": 0.060546875,
|
91 |
+
"eval_gen_len": 2.208984375,
|
92 |
+
"eval_length_ratio": 0.26875593542260207,
|
93 |
+
"eval_loss": 1.8144108057022095,
|
94 |
+
"eval_reference_length": 2106,
|
95 |
+
"eval_runtime": 23.7934,
|
96 |
+
"eval_samples_per_second": 21.519,
|
97 |
+
"eval_steps_per_second": 1.345,
|
98 |
+
"eval_translation_length": 566,
|
99 |
+
"step": 200
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.12,
|
103 |
+
"learning_rate": 9.608211879015829e-05,
|
104 |
+
"loss": 2.3867,
|
105 |
+
"step": 250
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.12,
|
109 |
+
"eval_bleu": 0.06335053804735992,
|
110 |
+
"eval_brevity_penalty": 0.40697312998195423,
|
111 |
+
"eval_exact_match": 0.05859375,
|
112 |
+
"eval_gen_len": 3.8125,
|
113 |
+
"eval_length_ratio": 0.52659069325736,
|
114 |
+
"eval_loss": 1.7400383949279785,
|
115 |
+
"eval_reference_length": 2106,
|
116 |
+
"eval_runtime": 29.4794,
|
117 |
+
"eval_samples_per_second": 17.368,
|
118 |
+
"eval_steps_per_second": 1.086,
|
119 |
+
"eval_translation_length": 1109,
|
120 |
+
"step": 250
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.14,
|
124 |
+
"learning_rate": 9.529854254818994e-05,
|
125 |
+
"loss": 2.199,
|
126 |
+
"step": 300
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 0.14,
|
130 |
+
"eval_bleu": 0.1636518792509081,
|
131 |
+
"eval_brevity_penalty": 0.7065164246788794,
|
132 |
+
"eval_exact_match": 0.103515625,
|
133 |
+
"eval_gen_len": 5.21875,
|
134 |
+
"eval_length_ratio": 0.7421652421652422,
|
135 |
+
"eval_loss": 1.551315188407898,
|
136 |
+
"eval_reference_length": 2106,
|
137 |
+
"eval_runtime": 32.1335,
|
138 |
+
"eval_samples_per_second": 15.934,
|
139 |
+
"eval_steps_per_second": 0.996,
|
140 |
+
"eval_translation_length": 1563,
|
141 |
+
"step": 300
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.16,
|
145 |
+
"learning_rate": 9.45149663062216e-05,
|
146 |
+
"loss": 2.1623,
|
147 |
+
"step": 350
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 0.16,
|
151 |
+
"eval_bleu": 0.2018876786015886,
|
152 |
+
"eval_brevity_penalty": 0.6032877660521909,
|
153 |
+
"eval_exact_match": 0.171875,
|
154 |
+
"eval_gen_len": 4.716796875,
|
155 |
+
"eval_length_ratio": 0.664292497625831,
|
156 |
+
"eval_loss": 1.3646727800369263,
|
157 |
+
"eval_reference_length": 2106,
|
158 |
+
"eval_runtime": 31.8812,
|
159 |
+
"eval_samples_per_second": 16.06,
|
160 |
+
"eval_steps_per_second": 1.004,
|
161 |
+
"eval_translation_length": 1399,
|
162 |
+
"step": 350
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.19,
|
166 |
+
"learning_rate": 9.373139006425325e-05,
|
167 |
+
"loss": 1.9871,
|
168 |
+
"step": 400
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.19,
|
172 |
+
"eval_bleu": 0.3336189639058238,
|
173 |
+
"eval_brevity_penalty": 0.9918951577772216,
|
174 |
+
"eval_exact_match": 0.25,
|
175 |
+
"eval_gen_len": 6.9140625,
|
176 |
+
"eval_length_ratio": 0.9919278252611586,
|
177 |
+
"eval_loss": 1.186793327331543,
|
178 |
+
"eval_reference_length": 2106,
|
179 |
+
"eval_runtime": 35.7547,
|
180 |
+
"eval_samples_per_second": 14.32,
|
181 |
+
"eval_steps_per_second": 0.895,
|
182 |
+
"eval_translation_length": 2089,
|
183 |
+
"step": 400
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 0.21,
|
187 |
+
"learning_rate": 9.294781382228491e-05,
|
188 |
+
"loss": 1.7705,
|
189 |
+
"step": 450
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.21,
|
193 |
+
"eval_bleu": 0.3747254791441505,
|
194 |
+
"eval_brevity_penalty": 0.8750477201167791,
|
195 |
+
"eval_exact_match": 0.298828125,
|
196 |
+
"eval_gen_len": 6.1953125,
|
197 |
+
"eval_length_ratio": 0.8822412155745489,
|
198 |
+
"eval_loss": 1.119579553604126,
|
199 |
+
"eval_reference_length": 2106,
|
200 |
+
"eval_runtime": 34.9183,
|
201 |
+
"eval_samples_per_second": 14.663,
|
202 |
+
"eval_steps_per_second": 0.916,
|
203 |
+
"eval_translation_length": 1858,
|
204 |
+
"step": 450
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.24,
|
208 |
+
"learning_rate": 9.216423758031657e-05,
|
209 |
+
"loss": 1.589,
|
210 |
+
"step": 500
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 0.24,
|
214 |
+
"eval_bleu": 0.39875210184754006,
|
215 |
+
"eval_brevity_penalty": 1.0,
|
216 |
+
"eval_exact_match": 0.328125,
|
217 |
+
"eval_gen_len": 7.37109375,
|
218 |
+
"eval_length_ratio": 1.0802469135802468,
|
219 |
+
"eval_loss": 1.0730499029159546,
|
220 |
+
"eval_reference_length": 2106,
|
221 |
+
"eval_runtime": 37.8703,
|
222 |
+
"eval_samples_per_second": 13.52,
|
223 |
+
"eval_steps_per_second": 0.845,
|
224 |
+
"eval_translation_length": 2275,
|
225 |
+
"step": 500
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.26,
|
229 |
+
"learning_rate": 9.138066133834823e-05,
|
230 |
+
"loss": 1.5929,
|
231 |
+
"step": 550
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 0.26,
|
235 |
+
"eval_bleu": 0.46535583502999767,
|
236 |
+
"eval_brevity_penalty": 0.9715887436284736,
|
237 |
+
"eval_exact_match": 0.357421875,
|
238 |
+
"eval_gen_len": 6.783203125,
|
239 |
+
"eval_length_ratio": 0.9719848053181387,
|
240 |
+
"eval_loss": 0.9941992163658142,
|
241 |
+
"eval_reference_length": 2106,
|
242 |
+
"eval_runtime": 35.7255,
|
243 |
+
"eval_samples_per_second": 14.331,
|
244 |
+
"eval_steps_per_second": 0.896,
|
245 |
+
"eval_translation_length": 2047,
|
246 |
+
"step": 550
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.28,
|
250 |
+
"learning_rate": 9.059708509637988e-05,
|
251 |
+
"loss": 1.5104,
|
252 |
+
"step": 600
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 0.28,
|
256 |
+
"eval_bleu": 0.47603827067185095,
|
257 |
+
"eval_brevity_penalty": 1.0,
|
258 |
+
"eval_exact_match": 0.3828125,
|
259 |
+
"eval_gen_len": 7.125,
|
260 |
+
"eval_length_ratio": 1.0261158594491928,
|
261 |
+
"eval_loss": 0.9567559957504272,
|
262 |
+
"eval_reference_length": 2106,
|
263 |
+
"eval_runtime": 36.3056,
|
264 |
+
"eval_samples_per_second": 14.103,
|
265 |
+
"eval_steps_per_second": 0.881,
|
266 |
+
"eval_translation_length": 2161,
|
267 |
+
"step": 600
|
268 |
+
},
|
269 |
+
{
|
270 |
+
"epoch": 0.31,
|
271 |
+
"learning_rate": 8.981350885441154e-05,
|
272 |
+
"loss": 1.382,
|
273 |
+
"step": 650
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 0.31,
|
277 |
+
"eval_bleu": 0.4441517830638515,
|
278 |
+
"eval_brevity_penalty": 1.0,
|
279 |
+
"eval_exact_match": 0.39453125,
|
280 |
+
"eval_gen_len": 7.638671875,
|
281 |
+
"eval_length_ratio": 1.1277302943969612,
|
282 |
+
"eval_loss": 0.9603753089904785,
|
283 |
+
"eval_reference_length": 2106,
|
284 |
+
"eval_runtime": 37.6569,
|
285 |
+
"eval_samples_per_second": 13.596,
|
286 |
+
"eval_steps_per_second": 0.85,
|
287 |
+
"eval_translation_length": 2375,
|
288 |
+
"step": 650
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 0.33,
|
292 |
+
"learning_rate": 8.90299326124432e-05,
|
293 |
+
"loss": 1.4739,
|
294 |
+
"step": 700
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 0.33,
|
298 |
+
"eval_bleu": 0.5062868407195426,
|
299 |
+
"eval_brevity_penalty": 1.0,
|
300 |
+
"eval_exact_match": 0.416015625,
|
301 |
+
"eval_gen_len": 7.041015625,
|
302 |
+
"eval_length_ratio": 1.0090218423551758,
|
303 |
+
"eval_loss": 0.878071665763855,
|
304 |
+
"eval_reference_length": 2106,
|
305 |
+
"eval_runtime": 35.5071,
|
306 |
+
"eval_samples_per_second": 14.42,
|
307 |
+
"eval_steps_per_second": 0.901,
|
308 |
+
"eval_translation_length": 2125,
|
309 |
+
"step": 700
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 0.35,
|
313 |
+
"learning_rate": 8.824635637047485e-05,
|
314 |
+
"loss": 1.3245,
|
315 |
+
"step": 750
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 0.35,
|
319 |
+
"eval_bleu": 0.5210361604149488,
|
320 |
+
"eval_brevity_penalty": 1.0,
|
321 |
+
"eval_exact_match": 0.419921875,
|
322 |
+
"eval_gen_len": 7.25390625,
|
323 |
+
"eval_length_ratio": 1.0574548907882242,
|
324 |
+
"eval_loss": 0.9063434600830078,
|
325 |
+
"eval_reference_length": 2106,
|
326 |
+
"eval_runtime": 34.9391,
|
327 |
+
"eval_samples_per_second": 14.654,
|
328 |
+
"eval_steps_per_second": 0.916,
|
329 |
+
"eval_translation_length": 2227,
|
330 |
+
"step": 750
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 0.38,
|
334 |
+
"learning_rate": 8.746278012850651e-05,
|
335 |
+
"loss": 1.3303,
|
336 |
+
"step": 800
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 0.38,
|
340 |
+
"eval_bleu": 0.4770015676872273,
|
341 |
+
"eval_brevity_penalty": 1.0,
|
342 |
+
"eval_exact_match": 0.3984375,
|
343 |
+
"eval_gen_len": 7.439453125,
|
344 |
+
"eval_length_ratio": 1.0826210826210827,
|
345 |
+
"eval_loss": 0.9164894819259644,
|
346 |
+
"eval_reference_length": 2106,
|
347 |
+
"eval_runtime": 36.5235,
|
348 |
+
"eval_samples_per_second": 14.018,
|
349 |
+
"eval_steps_per_second": 0.876,
|
350 |
+
"eval_translation_length": 2280,
|
351 |
+
"step": 800
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 0.4,
|
355 |
+
"learning_rate": 8.667920388653816e-05,
|
356 |
+
"loss": 1.2626,
|
357 |
+
"step": 850
|
358 |
+
},
|
359 |
+
{
|
360 |
+
"epoch": 0.4,
|
361 |
+
"eval_bleu": 0.5256219338123865,
|
362 |
+
"eval_brevity_penalty": 1.0,
|
363 |
+
"eval_exact_match": 0.40625,
|
364 |
+
"eval_gen_len": 7.05859375,
|
365 |
+
"eval_length_ratio": 1.0237416904083572,
|
366 |
+
"eval_loss": 0.8638509511947632,
|
367 |
+
"eval_reference_length": 2106,
|
368 |
+
"eval_runtime": 34.7224,
|
369 |
+
"eval_samples_per_second": 14.746,
|
370 |
+
"eval_steps_per_second": 0.922,
|
371 |
+
"eval_translation_length": 2156,
|
372 |
+
"step": 850
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 0.42,
|
376 |
+
"learning_rate": 8.589562764456982e-05,
|
377 |
+
"loss": 1.2751,
|
378 |
+
"step": 900
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 0.42,
|
382 |
+
"eval_bleu": 0.5549630399715574,
|
383 |
+
"eval_brevity_penalty": 0.995717351313512,
|
384 |
+
"eval_exact_match": 0.421875,
|
385 |
+
"eval_gen_len": 6.89453125,
|
386 |
+
"eval_length_ratio": 0.9957264957264957,
|
387 |
+
"eval_loss": 0.8271727561950684,
|
388 |
+
"eval_reference_length": 2106,
|
389 |
+
"eval_runtime": 36.4284,
|
390 |
+
"eval_samples_per_second": 14.055,
|
391 |
+
"eval_steps_per_second": 0.878,
|
392 |
+
"eval_translation_length": 2097,
|
393 |
+
"step": 900
|
394 |
+
},
|
395 |
+
{
|
396 |
+
"epoch": 0.45,
|
397 |
+
"learning_rate": 8.511205140260148e-05,
|
398 |
+
"loss": 1.291,
|
399 |
+
"step": 950
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 0.45,
|
403 |
+
"eval_bleu": 0.5601197353946883,
|
404 |
+
"eval_brevity_penalty": 0.9754879045293495,
|
405 |
+
"eval_exact_match": 0.447265625,
|
406 |
+
"eval_gen_len": 6.837890625,
|
407 |
+
"eval_length_ratio": 0.9757834757834758,
|
408 |
+
"eval_loss": 0.827507734298706,
|
409 |
+
"eval_reference_length": 2106,
|
410 |
+
"eval_runtime": 35.2125,
|
411 |
+
"eval_samples_per_second": 14.54,
|
412 |
+
"eval_steps_per_second": 0.909,
|
413 |
+
"eval_translation_length": 2055,
|
414 |
+
"step": 950
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 0.47,
|
418 |
+
"learning_rate": 8.432847516063314e-05,
|
419 |
+
"loss": 1.0992,
|
420 |
+
"step": 1000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 0.47,
|
424 |
+
"eval_bleu": 0.5469392101262274,
|
425 |
+
"eval_brevity_penalty": 0.9538582733351182,
|
426 |
+
"eval_exact_match": 0.4453125,
|
427 |
+
"eval_gen_len": 6.73046875,
|
428 |
+
"eval_length_ratio": 0.9548907882241215,
|
429 |
+
"eval_loss": 0.8413195610046387,
|
430 |
+
"eval_reference_length": 2106,
|
431 |
+
"eval_runtime": 34.3944,
|
432 |
+
"eval_samples_per_second": 14.886,
|
433 |
+
"eval_steps_per_second": 0.93,
|
434 |
+
"eval_translation_length": 2011,
|
435 |
+
"step": 1000
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 0.49,
|
439 |
+
"learning_rate": 8.354489891866479e-05,
|
440 |
+
"loss": 1.1555,
|
441 |
+
"step": 1050
|
442 |
+
},
|
443 |
+
{
|
444 |
+
"epoch": 0.49,
|
445 |
+
"eval_bleu": 0.5455158722304639,
|
446 |
+
"eval_brevity_penalty": 1.0,
|
447 |
+
"eval_exact_match": 0.466796875,
|
448 |
+
"eval_gen_len": 7.326171875,
|
449 |
+
"eval_length_ratio": 1.0607787274453941,
|
450 |
+
"eval_loss": 0.8606413006782532,
|
451 |
+
"eval_reference_length": 2106,
|
452 |
+
"eval_runtime": 35.8914,
|
453 |
+
"eval_samples_per_second": 14.265,
|
454 |
+
"eval_steps_per_second": 0.892,
|
455 |
+
"eval_translation_length": 2234,
|
456 |
+
"step": 1050
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 0.52,
|
460 |
+
"learning_rate": 8.276132267669645e-05,
|
461 |
+
"loss": 1.2137,
|
462 |
+
"step": 1100
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 0.52,
|
466 |
+
"eval_bleu": 0.5620322190043872,
|
467 |
+
"eval_brevity_penalty": 1.0,
|
468 |
+
"eval_exact_match": 0.447265625,
|
469 |
+
"eval_gen_len": 7.017578125,
|
470 |
+
"eval_length_ratio": 1.0118708452041785,
|
471 |
+
"eval_loss": 0.8165176510810852,
|
472 |
+
"eval_reference_length": 2106,
|
473 |
+
"eval_runtime": 35.4665,
|
474 |
+
"eval_samples_per_second": 14.436,
|
475 |
+
"eval_steps_per_second": 0.902,
|
476 |
+
"eval_translation_length": 2131,
|
477 |
+
"step": 1100
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 0.54,
|
481 |
+
"learning_rate": 8.197774643472809e-05,
|
482 |
+
"loss": 1.2334,
|
483 |
+
"step": 1150
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 0.54,
|
487 |
+
"eval_bleu": 0.561256814480964,
|
488 |
+
"eval_brevity_penalty": 0.9573296047876775,
|
489 |
+
"eval_exact_match": 0.443359375,
|
490 |
+
"eval_gen_len": 6.708984375,
|
491 |
+
"eval_length_ratio": 0.9582146248812915,
|
492 |
+
"eval_loss": 0.8095741271972656,
|
493 |
+
"eval_reference_length": 2106,
|
494 |
+
"eval_runtime": 34.1264,
|
495 |
+
"eval_samples_per_second": 15.003,
|
496 |
+
"eval_steps_per_second": 0.938,
|
497 |
+
"eval_translation_length": 2018,
|
498 |
+
"step": 1150
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 0.56,
|
502 |
+
"learning_rate": 8.119417019275976e-05,
|
503 |
+
"loss": 1.2993,
|
504 |
+
"step": 1200
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 0.56,
|
508 |
+
"eval_bleu": 0.588770590240424,
|
509 |
+
"eval_brevity_penalty": 0.9808251378136823,
|
510 |
+
"eval_exact_match": 0.451171875,
|
511 |
+
"eval_gen_len": 6.87109375,
|
512 |
+
"eval_length_ratio": 0.9810066476733144,
|
513 |
+
"eval_loss": 0.7695479393005371,
|
514 |
+
"eval_reference_length": 2106,
|
515 |
+
"eval_runtime": 34.8276,
|
516 |
+
"eval_samples_per_second": 14.701,
|
517 |
+
"eval_steps_per_second": 0.919,
|
518 |
+
"eval_translation_length": 2066,
|
519 |
+
"step": 1200
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 0.59,
|
523 |
+
"learning_rate": 8.04105939507914e-05,
|
524 |
+
"loss": 1.2508,
|
525 |
+
"step": 1250
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 0.59,
|
529 |
+
"eval_bleu": 0.563750421831778,
|
530 |
+
"eval_brevity_penalty": 1.0,
|
531 |
+
"eval_exact_match": 0.462890625,
|
532 |
+
"eval_gen_len": 7.181640625,
|
533 |
+
"eval_length_ratio": 1.047483380816714,
|
534 |
+
"eval_loss": 0.7718470096588135,
|
535 |
+
"eval_reference_length": 2106,
|
536 |
+
"eval_runtime": 35.174,
|
537 |
+
"eval_samples_per_second": 14.556,
|
538 |
+
"eval_steps_per_second": 0.91,
|
539 |
+
"eval_translation_length": 2206,
|
540 |
+
"step": 1250
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 0.61,
|
544 |
+
"learning_rate": 7.962701770882307e-05,
|
545 |
+
"loss": 1.0999,
|
546 |
+
"step": 1300
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 0.61,
|
550 |
+
"eval_bleu": 0.5781765484721586,
|
551 |
+
"eval_brevity_penalty": 0.9715887436284736,
|
552 |
+
"eval_exact_match": 0.447265625,
|
553 |
+
"eval_gen_len": 6.826171875,
|
554 |
+
"eval_length_ratio": 0.9719848053181387,
|
555 |
+
"eval_loss": 0.8051897883415222,
|
556 |
+
"eval_reference_length": 2106,
|
557 |
+
"eval_runtime": 34.2261,
|
558 |
+
"eval_samples_per_second": 14.959,
|
559 |
+
"eval_steps_per_second": 0.935,
|
560 |
+
"eval_translation_length": 2047,
|
561 |
+
"step": 1300
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 0.63,
|
565 |
+
"learning_rate": 7.884344146685473e-05,
|
566 |
+
"loss": 1.1623,
|
567 |
+
"step": 1350
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 0.63,
|
571 |
+
"eval_bleu": 0.5717608964815384,
|
572 |
+
"eval_brevity_penalty": 1.0,
|
573 |
+
"eval_exact_match": 0.478515625,
|
574 |
+
"eval_gen_len": 7.076171875,
|
575 |
+
"eval_length_ratio": 1.017094017094017,
|
576 |
+
"eval_loss": 0.7817097306251526,
|
577 |
+
"eval_reference_length": 2106,
|
578 |
+
"eval_runtime": 34.9684,
|
579 |
+
"eval_samples_per_second": 14.642,
|
580 |
+
"eval_steps_per_second": 0.915,
|
581 |
+
"eval_translation_length": 2142,
|
582 |
+
"step": 1350
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 0.66,
|
586 |
+
"learning_rate": 7.805986522488639e-05,
|
587 |
+
"loss": 1.1405,
|
588 |
+
"step": 1400
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 0.66,
|
592 |
+
"eval_bleu": 0.5807905073436569,
|
593 |
+
"eval_brevity_penalty": 0.9774319412545756,
|
594 |
+
"eval_exact_match": 0.46484375,
|
595 |
+
"eval_gen_len": 6.876953125,
|
596 |
+
"eval_length_ratio": 0.9776828110161444,
|
597 |
+
"eval_loss": 0.8067466616630554,
|
598 |
+
"eval_reference_length": 2106,
|
599 |
+
"eval_runtime": 34.9137,
|
600 |
+
"eval_samples_per_second": 14.665,
|
601 |
+
"eval_steps_per_second": 0.917,
|
602 |
+
"eval_translation_length": 2059,
|
603 |
+
"step": 1400
|
604 |
+
},
|
605 |
+
{
|
606 |
+
"epoch": 0.68,
|
607 |
+
"learning_rate": 7.727628898291805e-05,
|
608 |
+
"loss": 1.0645,
|
609 |
+
"step": 1450
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 0.68,
|
613 |
+
"eval_bleu": 0.6024093506065553,
|
614 |
+
"eval_brevity_penalty": 1.0,
|
615 |
+
"eval_exact_match": 0.48046875,
|
616 |
+
"eval_gen_len": 6.974609375,
|
617 |
+
"eval_length_ratio": 1.0056980056980056,
|
618 |
+
"eval_loss": 0.7935568690299988,
|
619 |
+
"eval_reference_length": 2106,
|
620 |
+
"eval_runtime": 35.237,
|
621 |
+
"eval_samples_per_second": 14.53,
|
622 |
+
"eval_steps_per_second": 0.908,
|
623 |
+
"eval_translation_length": 2118,
|
624 |
+
"step": 1450
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 0.71,
|
628 |
+
"learning_rate": 7.64927127409497e-05,
|
629 |
+
"loss": 1.114,
|
630 |
+
"step": 1500
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 0.71,
|
634 |
+
"eval_bleu": 0.5585335749210942,
|
635 |
+
"eval_brevity_penalty": 1.0,
|
636 |
+
"eval_exact_match": 0.482421875,
|
637 |
+
"eval_gen_len": 7.26953125,
|
638 |
+
"eval_length_ratio": 1.0555555555555556,
|
639 |
+
"eval_loss": 0.7591338157653809,
|
640 |
+
"eval_reference_length": 2106,
|
641 |
+
"eval_runtime": 35.9474,
|
642 |
+
"eval_samples_per_second": 14.243,
|
643 |
+
"eval_steps_per_second": 0.89,
|
644 |
+
"eval_translation_length": 2223,
|
645 |
+
"step": 1500
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 0.73,
|
649 |
+
"learning_rate": 7.570913649898136e-05,
|
650 |
+
"loss": 0.9622,
|
651 |
+
"step": 1550
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 0.73,
|
655 |
+
"eval_bleu": 0.6037520631327163,
|
656 |
+
"eval_brevity_penalty": 0.996194105459075,
|
657 |
+
"eval_exact_match": 0.517578125,
|
658 |
+
"eval_gen_len": 6.9609375,
|
659 |
+
"eval_length_ratio": 0.9962013295346629,
|
660 |
+
"eval_loss": 0.7658359408378601,
|
661 |
+
"eval_reference_length": 2106,
|
662 |
+
"eval_runtime": 35.7224,
|
663 |
+
"eval_samples_per_second": 14.333,
|
664 |
+
"eval_steps_per_second": 0.896,
|
665 |
+
"eval_translation_length": 2098,
|
666 |
+
"step": 1550
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 0.75,
|
670 |
+
"learning_rate": 7.4925560257013e-05,
|
671 |
+
"loss": 1.0521,
|
672 |
+
"step": 1600
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 0.75,
|
676 |
+
"eval_bleu": 0.5825978988431111,
|
677 |
+
"eval_brevity_penalty": 1.0,
|
678 |
+
"eval_exact_match": 0.509765625,
|
679 |
+
"eval_gen_len": 7.033203125,
|
680 |
+
"eval_length_ratio": 1.000474833808167,
|
681 |
+
"eval_loss": 0.7511448860168457,
|
682 |
+
"eval_reference_length": 2106,
|
683 |
+
"eval_runtime": 35.8085,
|
684 |
+
"eval_samples_per_second": 14.298,
|
685 |
+
"eval_steps_per_second": 0.894,
|
686 |
+
"eval_translation_length": 2107,
|
687 |
+
"step": 1600
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 0.78,
|
691 |
+
"learning_rate": 7.414198401504467e-05,
|
692 |
+
"loss": 1.0841,
|
693 |
+
"step": 1650
|
694 |
+
},
|
695 |
+
{
|
696 |
+
"epoch": 0.78,
|
697 |
+
"eval_bleu": 0.5825352487207182,
|
698 |
+
"eval_brevity_penalty": 0.9759742597265093,
|
699 |
+
"eval_exact_match": 0.486328125,
|
700 |
+
"eval_gen_len": 6.845703125,
|
701 |
+
"eval_length_ratio": 0.976258309591643,
|
702 |
+
"eval_loss": 0.7474812269210815,
|
703 |
+
"eval_reference_length": 2106,
|
704 |
+
"eval_runtime": 35.5909,
|
705 |
+
"eval_samples_per_second": 14.386,
|
706 |
+
"eval_steps_per_second": 0.899,
|
707 |
+
"eval_translation_length": 2056,
|
708 |
+
"step": 1650
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 0.8,
|
712 |
+
"learning_rate": 7.335840777307632e-05,
|
713 |
+
"loss": 1.0913,
|
714 |
+
"step": 1700
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 0.8,
|
718 |
+
"eval_bleu": 0.572534516070901,
|
719 |
+
"eval_brevity_penalty": 1.0,
|
720 |
+
"eval_exact_match": 0.486328125,
|
721 |
+
"eval_gen_len": 7.140625,
|
722 |
+
"eval_length_ratio": 1.0332383665717,
|
723 |
+
"eval_loss": 0.7539807558059692,
|
724 |
+
"eval_reference_length": 2106,
|
725 |
+
"eval_runtime": 36.2066,
|
726 |
+
"eval_samples_per_second": 14.141,
|
727 |
+
"eval_steps_per_second": 0.884,
|
728 |
+
"eval_translation_length": 2176,
|
729 |
+
"step": 1700
|
730 |
+
},
|
731 |
+
{
|
732 |
+
"epoch": 0.82,
|
733 |
+
"learning_rate": 7.257483153110798e-05,
|
734 |
+
"loss": 1.0432,
|
735 |
+
"step": 1750
|
736 |
+
},
|
737 |
+
{
|
738 |
+
"epoch": 0.82,
|
739 |
+
"eval_bleu": 0.5471774953144333,
|
740 |
+
"eval_brevity_penalty": 1.0,
|
741 |
+
"eval_exact_match": 0.494140625,
|
742 |
+
"eval_gen_len": 7.560546875,
|
743 |
+
"eval_length_ratio": 1.1054131054131053,
|
744 |
+
"eval_loss": 0.7697689533233643,
|
745 |
+
"eval_reference_length": 2106,
|
746 |
+
"eval_runtime": 36.5988,
|
747 |
+
"eval_samples_per_second": 13.99,
|
748 |
+
"eval_steps_per_second": 0.874,
|
749 |
+
"eval_translation_length": 2328,
|
750 |
+
"step": 1750
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 0.85,
|
754 |
+
"learning_rate": 7.179125528913963e-05,
|
755 |
+
"loss": 1.0081,
|
756 |
+
"step": 1800
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 0.85,
|
760 |
+
"eval_bleu": 0.5805134856584743,
|
761 |
+
"eval_brevity_penalty": 1.0,
|
762 |
+
"eval_exact_match": 0.4921875,
|
763 |
+
"eval_gen_len": 7.1171875,
|
764 |
+
"eval_length_ratio": 1.0270655270655271,
|
765 |
+
"eval_loss": 0.7475109696388245,
|
766 |
+
"eval_reference_length": 2106,
|
767 |
+
"eval_runtime": 35.6381,
|
768 |
+
"eval_samples_per_second": 14.367,
|
769 |
+
"eval_steps_per_second": 0.898,
|
770 |
+
"eval_translation_length": 2163,
|
771 |
+
"step": 1800
|
772 |
+
},
|
773 |
+
{
|
774 |
+
"epoch": 0.87,
|
775 |
+
"learning_rate": 7.10076790471713e-05,
|
776 |
+
"loss": 1.0837,
|
777 |
+
"step": 1850
|
778 |
+
},
|
779 |
+
{
|
780 |
+
"epoch": 0.87,
|
781 |
+
"eval_bleu": 0.565305888773277,
|
782 |
+
"eval_brevity_penalty": 1.0,
|
783 |
+
"eval_exact_match": 0.498046875,
|
784 |
+
"eval_gen_len": 7.259765625,
|
785 |
+
"eval_length_ratio": 1.0517568850902184,
|
786 |
+
"eval_loss": 0.7250198125839233,
|
787 |
+
"eval_reference_length": 2106,
|
788 |
+
"eval_runtime": 35.8261,
|
789 |
+
"eval_samples_per_second": 14.291,
|
790 |
+
"eval_steps_per_second": 0.893,
|
791 |
+
"eval_translation_length": 2215,
|
792 |
+
"step": 1850
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 0.89,
|
796 |
+
"learning_rate": 7.022410280520295e-05,
|
797 |
+
"loss": 1.0115,
|
798 |
+
"step": 1900
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 0.89,
|
802 |
+
"eval_bleu": 0.6107745307911603,
|
803 |
+
"eval_brevity_penalty": 0.997146934891016,
|
804 |
+
"eval_exact_match": 0.51171875,
|
805 |
+
"eval_gen_len": 6.921875,
|
806 |
+
"eval_length_ratio": 0.9971509971509972,
|
807 |
+
"eval_loss": 0.7463927268981934,
|
808 |
+
"eval_reference_length": 2106,
|
809 |
+
"eval_runtime": 34.3792,
|
810 |
+
"eval_samples_per_second": 14.893,
|
811 |
+
"eval_steps_per_second": 0.931,
|
812 |
+
"eval_translation_length": 2100,
|
813 |
+
"step": 1900
|
814 |
+
},
|
815 |
+
{
|
816 |
+
"epoch": 0.92,
|
817 |
+
"learning_rate": 6.944052656323461e-05,
|
818 |
+
"loss": 0.9624,
|
819 |
+
"step": 1950
|
820 |
+
},
|
821 |
+
{
|
822 |
+
"epoch": 0.92,
|
823 |
+
"eval_bleu": 0.6074013490021136,
|
824 |
+
"eval_brevity_penalty": 1.0,
|
825 |
+
"eval_exact_match": 0.515625,
|
826 |
+
"eval_gen_len": 6.962890625,
|
827 |
+
"eval_length_ratio": 1.002849002849003,
|
828 |
+
"eval_loss": 0.7404871582984924,
|
829 |
+
"eval_reference_length": 2106,
|
830 |
+
"eval_runtime": 35.4273,
|
831 |
+
"eval_samples_per_second": 14.452,
|
832 |
+
"eval_steps_per_second": 0.903,
|
833 |
+
"eval_translation_length": 2112,
|
834 |
+
"step": 1950
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 0.94,
|
838 |
+
"learning_rate": 6.865695032126626e-05,
|
839 |
+
"loss": 0.8905,
|
840 |
+
"step": 2000
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 0.94,
|
844 |
+
"eval_bleu": 0.605227378439144,
|
845 |
+
"eval_brevity_penalty": 1.0,
|
846 |
+
"eval_exact_match": 0.51171875,
|
847 |
+
"eval_gen_len": 6.986328125,
|
848 |
+
"eval_length_ratio": 1.0080721747388415,
|
849 |
+
"eval_loss": 0.7403624653816223,
|
850 |
+
"eval_reference_length": 2106,
|
851 |
+
"eval_runtime": 35.8318,
|
852 |
+
"eval_samples_per_second": 14.289,
|
853 |
+
"eval_steps_per_second": 0.893,
|
854 |
+
"eval_translation_length": 2123,
|
855 |
+
"step": 2000
|
856 |
+
},
|
857 |
+
{
|
858 |
+
"epoch": 0.96,
|
859 |
+
"learning_rate": 6.787337407929793e-05,
|
860 |
+
"loss": 1.0936,
|
861 |
+
"step": 2050
|
862 |
+
},
|
863 |
+
{
|
864 |
+
"epoch": 0.96,
|
865 |
+
"eval_bleu": 0.613080912893354,
|
866 |
+
"eval_brevity_penalty": 1.0,
|
867 |
+
"eval_exact_match": 0.52734375,
|
868 |
+
"eval_gen_len": 7.005859375,
|
869 |
+
"eval_length_ratio": 1.0061728395061729,
|
870 |
+
"eval_loss": 0.7132320404052734,
|
871 |
+
"eval_reference_length": 2106,
|
872 |
+
"eval_runtime": 35.4858,
|
873 |
+
"eval_samples_per_second": 14.428,
|
874 |
+
"eval_steps_per_second": 0.902,
|
875 |
+
"eval_translation_length": 2119,
|
876 |
+
"step": 2050
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 0.99,
|
880 |
+
"learning_rate": 6.708979783732958e-05,
|
881 |
+
"loss": 1.0418,
|
882 |
+
"step": 2100
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 0.99,
|
886 |
+
"eval_bleu": 0.6157679585094484,
|
887 |
+
"eval_brevity_penalty": 0.9909373387239356,
|
888 |
+
"eval_exact_match": 0.5234375,
|
889 |
+
"eval_gen_len": 6.91796875,
|
890 |
+
"eval_length_ratio": 0.9909781576448243,
|
891 |
+
"eval_loss": 0.7031872868537903,
|
892 |
+
"eval_reference_length": 2106,
|
893 |
+
"eval_runtime": 35.4742,
|
894 |
+
"eval_samples_per_second": 14.433,
|
895 |
+
"eval_steps_per_second": 0.902,
|
896 |
+
"eval_translation_length": 2087,
|
897 |
+
"step": 2100
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 1.01,
|
901 |
+
"learning_rate": 6.630622159536123e-05,
|
902 |
+
"loss": 0.9631,
|
903 |
+
"step": 2150
|
904 |
+
},
|
905 |
+
{
|
906 |
+
"epoch": 1.01,
|
907 |
+
"eval_bleu": 0.6228792131756815,
|
908 |
+
"eval_brevity_penalty": 0.946880981878215,
|
909 |
+
"eval_exact_match": 0.525390625,
|
910 |
+
"eval_gen_len": 6.642578125,
|
911 |
+
"eval_length_ratio": 0.9482431149097815,
|
912 |
+
"eval_loss": 0.695751428604126,
|
913 |
+
"eval_reference_length": 2106,
|
914 |
+
"eval_runtime": 34.7409,
|
915 |
+
"eval_samples_per_second": 14.738,
|
916 |
+
"eval_steps_per_second": 0.921,
|
917 |
+
"eval_translation_length": 1997,
|
918 |
+
"step": 2150
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 1.03,
|
922 |
+
"learning_rate": 6.552264535339289e-05,
|
923 |
+
"loss": 0.8085,
|
924 |
+
"step": 2200
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 1.03,
|
928 |
+
"eval_bleu": 0.6272859670438183,
|
929 |
+
"eval_brevity_penalty": 0.946880981878215,
|
930 |
+
"eval_exact_match": 0.53125,
|
931 |
+
"eval_gen_len": 6.650390625,
|
932 |
+
"eval_length_ratio": 0.9482431149097815,
|
933 |
+
"eval_loss": 0.7051365375518799,
|
934 |
+
"eval_reference_length": 2106,
|
935 |
+
"eval_runtime": 34.9034,
|
936 |
+
"eval_samples_per_second": 14.669,
|
937 |
+
"eval_steps_per_second": 0.917,
|
938 |
+
"eval_translation_length": 1997,
|
939 |
+
"step": 2200
|
940 |
+
},
|
941 |
+
{
|
942 |
+
"epoch": 1.06,
|
943 |
+
"learning_rate": 6.473906911142454e-05,
|
944 |
+
"loss": 0.9716,
|
945 |
+
"step": 2250
|
946 |
+
},
|
947 |
+
{
|
948 |
+
"epoch": 1.06,
|
949 |
+
"eval_bleu": 0.6325687827118958,
|
950 |
+
"eval_brevity_penalty": 1.0,
|
951 |
+
"eval_exact_match": 0.533203125,
|
952 |
+
"eval_gen_len": 6.943359375,
|
953 |
+
"eval_length_ratio": 1.000474833808167,
|
954 |
+
"eval_loss": 0.6994906067848206,
|
955 |
+
"eval_reference_length": 2106,
|
956 |
+
"eval_runtime": 35.1655,
|
957 |
+
"eval_samples_per_second": 14.56,
|
958 |
+
"eval_steps_per_second": 0.91,
|
959 |
+
"eval_translation_length": 2107,
|
960 |
+
"step": 2250
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 1.08,
|
964 |
+
"learning_rate": 6.39554928694562e-05,
|
965 |
+
"loss": 0.8824,
|
966 |
+
"step": 2300
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 1.08,
|
970 |
+
"eval_bleu": 0.6202715514980656,
|
971 |
+
"eval_brevity_penalty": 0.9691442418767943,
|
972 |
+
"eval_exact_match": 0.5234375,
|
973 |
+
"eval_gen_len": 6.76953125,
|
974 |
+
"eval_length_ratio": 0.9696106362773029,
|
975 |
+
"eval_loss": 0.7111188173294067,
|
976 |
+
"eval_reference_length": 2106,
|
977 |
+
"eval_runtime": 35.1592,
|
978 |
+
"eval_samples_per_second": 14.562,
|
979 |
+
"eval_steps_per_second": 0.91,
|
980 |
+
"eval_translation_length": 2042,
|
981 |
+
"step": 2300
|
982 |
+
},
|
983 |
+
{
|
984 |
+
"epoch": 1.1,
|
985 |
+
"learning_rate": 6.317191662748786e-05,
|
986 |
+
"loss": 0.7888,
|
987 |
+
"step": 2350
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 1.1,
|
991 |
+
"eval_bleu": 0.6314970589714921,
|
992 |
+
"eval_brevity_penalty": 0.9423710963415396,
|
993 |
+
"eval_exact_match": 0.533203125,
|
994 |
+
"eval_gen_len": 6.609375,
|
995 |
+
"eval_length_ratio": 0.9439696106362773,
|
996 |
+
"eval_loss": 0.6993675827980042,
|
997 |
+
"eval_reference_length": 2106,
|
998 |
+
"eval_runtime": 34.3211,
|
999 |
+
"eval_samples_per_second": 14.918,
|
1000 |
+
"eval_steps_per_second": 0.932,
|
1001 |
+
"eval_translation_length": 1988,
|
1002 |
+
"step": 2350
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 1.13,
|
1006 |
+
"learning_rate": 6.238834038551952e-05,
|
1007 |
+
"loss": 0.8409,
|
1008 |
+
"step": 2400
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 1.13,
|
1012 |
+
"eval_bleu": 0.6283952758196866,
|
1013 |
+
"eval_brevity_penalty": 1.0,
|
1014 |
+
"eval_exact_match": 0.54296875,
|
1015 |
+
"eval_gen_len": 6.984375,
|
1016 |
+
"eval_length_ratio": 1.0151946818613484,
|
1017 |
+
"eval_loss": 0.7133020162582397,
|
1018 |
+
"eval_reference_length": 2106,
|
1019 |
+
"eval_runtime": 35.2074,
|
1020 |
+
"eval_samples_per_second": 14.542,
|
1021 |
+
"eval_steps_per_second": 0.909,
|
1022 |
+
"eval_translation_length": 2138,
|
1023 |
+
"step": 2400
|
1024 |
+
},
|
1025 |
+
{
|
1026 |
+
"epoch": 1.15,
|
1027 |
+
"learning_rate": 6.160476414355117e-05,
|
1028 |
+
"loss": 0.8331,
|
1029 |
+
"step": 2450
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 1.15,
|
1033 |
+
"eval_bleu": 0.6355611685013638,
|
1034 |
+
"eval_brevity_penalty": 0.9725649218577119,
|
1035 |
+
"eval_exact_match": 0.541015625,
|
1036 |
+
"eval_gen_len": 6.76953125,
|
1037 |
+
"eval_length_ratio": 0.9729344729344729,
|
1038 |
+
"eval_loss": 0.7177965044975281,
|
1039 |
+
"eval_reference_length": 2106,
|
1040 |
+
"eval_runtime": 35.1779,
|
1041 |
+
"eval_samples_per_second": 14.555,
|
1042 |
+
"eval_steps_per_second": 0.91,
|
1043 |
+
"eval_translation_length": 2049,
|
1044 |
+
"step": 2450
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 1.18,
|
1048 |
+
"learning_rate": 6.0821187901582835e-05,
|
1049 |
+
"loss": 0.8205,
|
1050 |
+
"step": 2500
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 1.18,
|
1054 |
+
"eval_bleu": 0.6392355323482277,
|
1055 |
+
"eval_brevity_penalty": 0.9740274534203012,
|
1056 |
+
"eval_exact_match": 0.53515625,
|
1057 |
+
"eval_gen_len": 6.80078125,
|
1058 |
+
"eval_length_ratio": 0.9743589743589743,
|
1059 |
+
"eval_loss": 0.7102757096290588,
|
1060 |
+
"eval_reference_length": 2106,
|
1061 |
+
"eval_runtime": 35.5038,
|
1062 |
+
"eval_samples_per_second": 14.421,
|
1063 |
+
"eval_steps_per_second": 0.901,
|
1064 |
+
"eval_translation_length": 2052,
|
1065 |
+
"step": 2500
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 1.2,
|
1069 |
+
"learning_rate": 6.003761165961448e-05,
|
1070 |
+
"loss": 0.8214,
|
1071 |
+
"step": 2550
|
1072 |
+
},
|
1073 |
+
{
|
1074 |
+
"epoch": 1.2,
|
1075 |
+
"eval_bleu": 0.6337232384678544,
|
1076 |
+
"eval_brevity_penalty": 0.9711003070476397,
|
1077 |
+
"eval_exact_match": 0.544921875,
|
1078 |
+
"eval_gen_len": 6.759765625,
|
1079 |
+
"eval_length_ratio": 0.9715099715099715,
|
1080 |
+
"eval_loss": 0.7109658718109131,
|
1081 |
+
"eval_reference_length": 2106,
|
1082 |
+
"eval_runtime": 35.4746,
|
1083 |
+
"eval_samples_per_second": 14.433,
|
1084 |
+
"eval_steps_per_second": 0.902,
|
1085 |
+
"eval_translation_length": 2046,
|
1086 |
+
"step": 2550
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 1.22,
|
1090 |
+
"learning_rate": 5.925403541764614e-05,
|
1091 |
+
"loss": 0.8607,
|
1092 |
+
"step": 2600
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 1.22,
|
1096 |
+
"eval_bleu": 0.6413697910504058,
|
1097 |
+
"eval_brevity_penalty": 0.951371692779408,
|
1098 |
+
"eval_exact_match": 0.54296875,
|
1099 |
+
"eval_gen_len": 6.650390625,
|
1100 |
+
"eval_length_ratio": 0.9525166191832859,
|
1101 |
+
"eval_loss": 0.6937527656555176,
|
1102 |
+
"eval_reference_length": 2106,
|
1103 |
+
"eval_runtime": 35.0711,
|
1104 |
+
"eval_samples_per_second": 14.599,
|
1105 |
+
"eval_steps_per_second": 0.912,
|
1106 |
+
"eval_translation_length": 2006,
|
1107 |
+
"step": 2600
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 1.25,
|
1111 |
+
"learning_rate": 5.8470459175677794e-05,
|
1112 |
+
"loss": 0.8754,
|
1113 |
+
"step": 2650
|
1114 |
+
},
|
1115 |
+
{
|
1116 |
+
"epoch": 1.25,
|
1117 |
+
"eval_bleu": 0.6345724305414163,
|
1118 |
+
"eval_brevity_penalty": 0.9880584183776264,
|
1119 |
+
"eval_exact_match": 0.521484375,
|
1120 |
+
"eval_gen_len": 6.88671875,
|
1121 |
+
"eval_length_ratio": 0.9881291547958214,
|
1122 |
+
"eval_loss": 0.7027400732040405,
|
1123 |
+
"eval_reference_length": 2106,
|
1124 |
+
"eval_runtime": 35.2724,
|
1125 |
+
"eval_samples_per_second": 14.516,
|
1126 |
+
"eval_steps_per_second": 0.907,
|
1127 |
+
"eval_translation_length": 2081,
|
1128 |
+
"step": 2650
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 1.27,
|
1132 |
+
"learning_rate": 5.768688293370945e-05,
|
1133 |
+
"loss": 0.8625,
|
1134 |
+
"step": 2700
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 1.27,
|
1138 |
+
"eval_bleu": 0.641086474299546,
|
1139 |
+
"eval_brevity_penalty": 0.9418688112227576,
|
1140 |
+
"eval_exact_match": 0.529296875,
|
1141 |
+
"eval_gen_len": 6.623046875,
|
1142 |
+
"eval_length_ratio": 0.9434947768281101,
|
1143 |
+
"eval_loss": 0.691420316696167,
|
1144 |
+
"eval_reference_length": 2106,
|
1145 |
+
"eval_runtime": 34.6599,
|
1146 |
+
"eval_samples_per_second": 14.772,
|
1147 |
+
"eval_steps_per_second": 0.923,
|
1148 |
+
"eval_translation_length": 1987,
|
1149 |
+
"step": 2700
|
1150 |
+
},
|
1151 |
+
{
|
1152 |
+
"epoch": 1.29,
|
1153 |
+
"learning_rate": 5.690330669174111e-05,
|
1154 |
+
"loss": 0.8625,
|
1155 |
+
"step": 2750
|
1156 |
+
},
|
1157 |
+
{
|
1158 |
+
"epoch": 1.29,
|
1159 |
+
"eval_bleu": 0.6428983658861092,
|
1160 |
+
"eval_brevity_penalty": 0.9543548830374949,
|
1161 |
+
"eval_exact_match": 0.537109375,
|
1162 |
+
"eval_gen_len": 6.693359375,
|
1163 |
+
"eval_length_ratio": 0.9553656220322887,
|
1164 |
+
"eval_loss": 0.7152715921401978,
|
1165 |
+
"eval_reference_length": 2106,
|
1166 |
+
"eval_runtime": 34.5929,
|
1167 |
+
"eval_samples_per_second": 14.801,
|
1168 |
+
"eval_steps_per_second": 0.925,
|
1169 |
+
"eval_translation_length": 2012,
|
1170 |
+
"step": 2750
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 1.32,
|
1174 |
+
"learning_rate": 5.6119730449772766e-05,
|
1175 |
+
"loss": 0.8007,
|
1176 |
+
"step": 2800
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 1.32,
|
1180 |
+
"eval_bleu": 0.6484293170697683,
|
1181 |
+
"eval_brevity_penalty": 0.9528643478208072,
|
1182 |
+
"eval_exact_match": 0.5546875,
|
1183 |
+
"eval_gen_len": 6.67578125,
|
1184 |
+
"eval_length_ratio": 0.9539411206077872,
|
1185 |
+
"eval_loss": 0.691640317440033,
|
1186 |
+
"eval_reference_length": 2106,
|
1187 |
+
"eval_runtime": 34.5133,
|
1188 |
+
"eval_samples_per_second": 14.835,
|
1189 |
+
"eval_steps_per_second": 0.927,
|
1190 |
+
"eval_translation_length": 2009,
|
1191 |
+
"step": 2800
|
1192 |
+
},
|
1193 |
+
{
|
1194 |
+
"epoch": 1.34,
|
1195 |
+
"learning_rate": 5.5336154207804424e-05,
|
1196 |
+
"loss": 0.8345,
|
1197 |
+
"step": 2850
|
1198 |
+
},
|
1199 |
+
{
|
1200 |
+
"epoch": 1.34,
|
1201 |
+
"eval_bleu": 0.6441657194783443,
|
1202 |
+
"eval_brevity_penalty": 0.9418688112227576,
|
1203 |
+
"eval_exact_match": 0.54296875,
|
1204 |
+
"eval_gen_len": 6.638671875,
|
1205 |
+
"eval_length_ratio": 0.9434947768281101,
|
1206 |
+
"eval_loss": 0.7108384370803833,
|
1207 |
+
"eval_reference_length": 2106,
|
1208 |
+
"eval_runtime": 34.9885,
|
1209 |
+
"eval_samples_per_second": 14.633,
|
1210 |
+
"eval_steps_per_second": 0.915,
|
1211 |
+
"eval_translation_length": 1987,
|
1212 |
+
"step": 2850
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 1.36,
|
1216 |
+
"learning_rate": 5.455257796583608e-05,
|
1217 |
+
"loss": 0.8096,
|
1218 |
+
"step": 2900
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 1.36,
|
1222 |
+
"eval_bleu": 0.628018545615884,
|
1223 |
+
"eval_brevity_penalty": 0.9938080693803775,
|
1224 |
+
"eval_exact_match": 0.537109375,
|
1225 |
+
"eval_gen_len": 6.912109375,
|
1226 |
+
"eval_length_ratio": 0.9938271604938271,
|
1227 |
+
"eval_loss": 0.7028455138206482,
|
1228 |
+
"eval_reference_length": 2106,
|
1229 |
+
"eval_runtime": 35.5719,
|
1230 |
+
"eval_samples_per_second": 14.393,
|
1231 |
+
"eval_steps_per_second": 0.9,
|
1232 |
+
"eval_translation_length": 2093,
|
1233 |
+
"step": 2900
|
1234 |
+
},
|
1235 |
+
{
|
1236 |
+
"epoch": 1.39,
|
1237 |
+
"learning_rate": 5.376900172386774e-05,
|
1238 |
+
"loss": 0.7349,
|
1239 |
+
"step": 2950
|
1240 |
+
},
|
1241 |
+
{
|
1242 |
+
"epoch": 1.39,
|
1243 |
+
"eval_bleu": 0.6373148854351528,
|
1244 |
+
"eval_brevity_penalty": 0.9822759172643585,
|
1245 |
+
"eval_exact_match": 0.541015625,
|
1246 |
+
"eval_gen_len": 6.853515625,
|
1247 |
+
"eval_length_ratio": 0.9824311490978158,
|
1248 |
+
"eval_loss": 0.7058258056640625,
|
1249 |
+
"eval_reference_length": 2106,
|
1250 |
+
"eval_runtime": 35.2471,
|
1251 |
+
"eval_samples_per_second": 14.526,
|
1252 |
+
"eval_steps_per_second": 0.908,
|
1253 |
+
"eval_translation_length": 2069,
|
1254 |
+
"step": 2950
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 1.41,
|
1258 |
+
"learning_rate": 5.298542548189939e-05,
|
1259 |
+
"loss": 0.9361,
|
1260 |
+
"step": 3000
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 1.41,
|
1264 |
+
"eval_bleu": 0.6287603417406736,
|
1265 |
+
"eval_brevity_penalty": 0.9923737263316806,
|
1266 |
+
"eval_exact_match": 0.53515625,
|
1267 |
+
"eval_gen_len": 6.89453125,
|
1268 |
+
"eval_length_ratio": 0.9924026590693258,
|
1269 |
+
"eval_loss": 0.6922191381454468,
|
1270 |
+
"eval_reference_length": 2106,
|
1271 |
+
"eval_runtime": 34.6615,
|
1272 |
+
"eval_samples_per_second": 14.771,
|
1273 |
+
"eval_steps_per_second": 0.923,
|
1274 |
+
"eval_translation_length": 2090,
|
1275 |
+
"step": 3000
|
1276 |
+
},
|
1277 |
+
{
|
1278 |
+
"epoch": 1.43,
|
1279 |
+
"learning_rate": 5.220184923993105e-05,
|
1280 |
+
"loss": 0.8269,
|
1281 |
+
"step": 3050
|
1282 |
+
},
|
1283 |
+
{
|
1284 |
+
"epoch": 1.43,
|
1285 |
+
"eval_bleu": 0.6439396809407835,
|
1286 |
+
"eval_brevity_penalty": 0.9642378110498295,
|
1287 |
+
"eval_exact_match": 0.548828125,
|
1288 |
+
"eval_gen_len": 6.740234375,
|
1289 |
+
"eval_length_ratio": 0.9648622981956315,
|
1290 |
+
"eval_loss": 0.6778197288513184,
|
1291 |
+
"eval_reference_length": 2106,
|
1292 |
+
"eval_runtime": 34.7891,
|
1293 |
+
"eval_samples_per_second": 14.717,
|
1294 |
+
"eval_steps_per_second": 0.92,
|
1295 |
+
"eval_translation_length": 2032,
|
1296 |
+
"step": 3050
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 1.46,
|
1300 |
+
"learning_rate": 5.1418272997962704e-05,
|
1301 |
+
"loss": 0.7282,
|
1302 |
+
"step": 3100
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 1.46,
|
1306 |
+
"eval_bleu": 0.642699995934985,
|
1307 |
+
"eval_brevity_penalty": 0.9904580880129722,
|
1308 |
+
"eval_exact_match": 0.55078125,
|
1309 |
+
"eval_gen_len": 6.9140625,
|
1310 |
+
"eval_length_ratio": 0.9905033238366572,
|
1311 |
+
"eval_loss": 0.685681164264679,
|
1312 |
+
"eval_reference_length": 2106,
|
1313 |
+
"eval_runtime": 35.0377,
|
1314 |
+
"eval_samples_per_second": 14.613,
|
1315 |
+
"eval_steps_per_second": 0.913,
|
1316 |
+
"eval_translation_length": 2086,
|
1317 |
+
"step": 3100
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 1.48,
|
1321 |
+
"learning_rate": 5.063469675599436e-05,
|
1322 |
+
"loss": 0.7244,
|
1323 |
+
"step": 3150
|
1324 |
+
},
|
1325 |
+
{
|
1326 |
+
"epoch": 1.48,
|
1327 |
+
"eval_bleu": 0.6412084160600809,
|
1328 |
+
"eval_brevity_penalty": 0.9832419564425784,
|
1329 |
+
"eval_exact_match": 0.541015625,
|
1330 |
+
"eval_gen_len": 6.880859375,
|
1331 |
+
"eval_length_ratio": 0.98338081671415,
|
1332 |
+
"eval_loss": 0.6973811388015747,
|
1333 |
+
"eval_reference_length": 2106,
|
1334 |
+
"eval_runtime": 35.1389,
|
1335 |
+
"eval_samples_per_second": 14.571,
|
1336 |
+
"eval_steps_per_second": 0.911,
|
1337 |
+
"eval_translation_length": 2071,
|
1338 |
+
"step": 3150
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 1.5,
|
1342 |
+
"learning_rate": 4.985112051402602e-05,
|
1343 |
+
"loss": 0.7876,
|
1344 |
+
"step": 3200
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 1.5,
|
1348 |
+
"eval_bleu": 0.6514314556575306,
|
1349 |
+
"eval_brevity_penalty": 0.9947631636301487,
|
1350 |
+
"eval_exact_match": 0.544921875,
|
1351 |
+
"eval_gen_len": 6.9453125,
|
1352 |
+
"eval_length_ratio": 0.9947768281101614,
|
1353 |
+
"eval_loss": 0.6855226755142212,
|
1354 |
+
"eval_reference_length": 2106,
|
1355 |
+
"eval_runtime": 35.5075,
|
1356 |
+
"eval_samples_per_second": 14.42,
|
1357 |
+
"eval_steps_per_second": 0.901,
|
1358 |
+
"eval_translation_length": 2095,
|
1359 |
+
"step": 3200
|
1360 |
+
},
|
1361 |
+
{
|
1362 |
+
"epoch": 1.53,
|
1363 |
+
"learning_rate": 4.9067544272057677e-05,
|
1364 |
+
"loss": 0.8535,
|
1365 |
+
"step": 3250
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 1.53,
|
1369 |
+
"eval_bleu": 0.6500634066257985,
|
1370 |
+
"eval_brevity_penalty": 1.0,
|
1371 |
+
"eval_exact_match": 0.541015625,
|
1372 |
+
"eval_gen_len": 6.986328125,
|
1373 |
+
"eval_length_ratio": 1.0037986704653372,
|
1374 |
+
"eval_loss": 0.6744265556335449,
|
1375 |
+
"eval_reference_length": 2106,
|
1376 |
+
"eval_runtime": 36.0223,
|
1377 |
+
"eval_samples_per_second": 14.213,
|
1378 |
+
"eval_steps_per_second": 0.888,
|
1379 |
+
"eval_translation_length": 2114,
|
1380 |
+
"step": 3250
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 1.55,
|
1384 |
+
"learning_rate": 4.8283968030089334e-05,
|
1385 |
+
"loss": 0.7917,
|
1386 |
+
"step": 3300
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 1.55,
|
1390 |
+
"eval_bleu": 0.6499057150972741,
|
1391 |
+
"eval_brevity_penalty": 0.9985744834917494,
|
1392 |
+
"eval_exact_match": 0.533203125,
|
1393 |
+
"eval_gen_len": 6.986328125,
|
1394 |
+
"eval_length_ratio": 0.9985754985754985,
|
1395 |
+
"eval_loss": 0.672450065612793,
|
1396 |
+
"eval_reference_length": 2106,
|
1397 |
+
"eval_runtime": 35.9664,
|
1398 |
+
"eval_samples_per_second": 14.236,
|
1399 |
+
"eval_steps_per_second": 0.89,
|
1400 |
+
"eval_translation_length": 2103,
|
1401 |
+
"step": 3300
|
1402 |
+
},
|
1403 |
+
{
|
1404 |
+
"epoch": 1.57,
|
1405 |
+
"learning_rate": 4.7500391788120985e-05,
|
1406 |
+
"loss": 0.7538,
|
1407 |
+
"step": 3350
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 1.57,
|
1411 |
+
"eval_bleu": 0.6310972587707046,
|
1412 |
+
"eval_brevity_penalty": 1.0,
|
1413 |
+
"eval_exact_match": 0.5390625,
|
1414 |
+
"eval_gen_len": 7.095703125,
|
1415 |
+
"eval_length_ratio": 1.019468186134853,
|
1416 |
+
"eval_loss": 0.6634331941604614,
|
1417 |
+
"eval_reference_length": 2106,
|
1418 |
+
"eval_runtime": 35.9969,
|
1419 |
+
"eval_samples_per_second": 14.223,
|
1420 |
+
"eval_steps_per_second": 0.889,
|
1421 |
+
"eval_translation_length": 2147,
|
1422 |
+
"step": 3350
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 1.6,
|
1426 |
+
"learning_rate": 4.671681554615264e-05,
|
1427 |
+
"loss": 0.8731,
|
1428 |
+
"step": 3400
|
1429 |
+
},
|
1430 |
+
{
|
1431 |
+
"epoch": 1.6,
|
1432 |
+
"eval_bleu": 0.6365367381748843,
|
1433 |
+
"eval_brevity_penalty": 1.0,
|
1434 |
+
"eval_exact_match": 0.544921875,
|
1435 |
+
"eval_gen_len": 7.03125,
|
1436 |
+
"eval_length_ratio": 1.0113960113960114,
|
1437 |
+
"eval_loss": 0.695984959602356,
|
1438 |
+
"eval_reference_length": 2106,
|
1439 |
+
"eval_runtime": 35.7076,
|
1440 |
+
"eval_samples_per_second": 14.339,
|
1441 |
+
"eval_steps_per_second": 0.896,
|
1442 |
+
"eval_translation_length": 2130,
|
1443 |
+
"step": 3400
|
1444 |
+
},
|
1445 |
+
{
|
1446 |
+
"epoch": 1.62,
|
1447 |
+
"learning_rate": 4.59332393041843e-05,
|
1448 |
+
"loss": 0.749,
|
1449 |
+
"step": 3450
|
1450 |
+
},
|
1451 |
+
{
|
1452 |
+
"epoch": 1.62,
|
1453 |
+
"eval_bleu": 0.6227209715309775,
|
1454 |
+
"eval_brevity_penalty": 0.944377861736799,
|
1455 |
+
"eval_exact_match": 0.5390625,
|
1456 |
+
"eval_gen_len": 6.71484375,
|
1457 |
+
"eval_length_ratio": 0.9458689458689459,
|
1458 |
+
"eval_loss": 0.6684394478797913,
|
1459 |
+
"eval_reference_length": 2106,
|
1460 |
+
"eval_runtime": 35.0788,
|
1461 |
+
"eval_samples_per_second": 14.596,
|
1462 |
+
"eval_steps_per_second": 0.912,
|
1463 |
+
"eval_translation_length": 1992,
|
1464 |
+
"step": 3450
|
1465 |
+
},
|
1466 |
+
{
|
1467 |
+
"epoch": 1.65,
|
1468 |
+
"learning_rate": 4.514966306221596e-05,
|
1469 |
+
"loss": 0.8981,
|
1470 |
+
"step": 3500
|
1471 |
+
},
|
1472 |
+
{
|
1473 |
+
"epoch": 1.65,
|
1474 |
+
"eval_bleu": 0.6248917016218156,
|
1475 |
+
"eval_brevity_penalty": 0.997146934891016,
|
1476 |
+
"eval_exact_match": 0.5390625,
|
1477 |
+
"eval_gen_len": 6.947265625,
|
1478 |
+
"eval_length_ratio": 0.9971509971509972,
|
1479 |
+
"eval_loss": 0.6610915660858154,
|
1480 |
+
"eval_reference_length": 2106,
|
1481 |
+
"eval_runtime": 35.6448,
|
1482 |
+
"eval_samples_per_second": 14.364,
|
1483 |
+
"eval_steps_per_second": 0.898,
|
1484 |
+
"eval_translation_length": 2100,
|
1485 |
+
"step": 3500
|
1486 |
+
},
|
1487 |
+
{
|
1488 |
+
"epoch": 1.67,
|
1489 |
+
"learning_rate": 4.436608682024761e-05,
|
1490 |
+
"loss": 0.6892,
|
1491 |
+
"step": 3550
|
1492 |
+
},
|
1493 |
+
{
|
1494 |
+
"epoch": 1.67,
|
1495 |
+
"eval_bleu": 0.6171345792845108,
|
1496 |
+
"eval_brevity_penalty": 0.9827590514957918,
|
1497 |
+
"eval_exact_match": 0.5390625,
|
1498 |
+
"eval_gen_len": 6.861328125,
|
1499 |
+
"eval_length_ratio": 0.9829059829059829,
|
1500 |
+
"eval_loss": 0.6674365997314453,
|
1501 |
+
"eval_reference_length": 2106,
|
1502 |
+
"eval_runtime": 35.383,
|
1503 |
+
"eval_samples_per_second": 14.47,
|
1504 |
+
"eval_steps_per_second": 0.904,
|
1505 |
+
"eval_translation_length": 2070,
|
1506 |
+
"step": 3550
|
1507 |
+
},
|
1508 |
+
{
|
1509 |
+
"epoch": 1.69,
|
1510 |
+
"learning_rate": 4.3582510578279265e-05,
|
1511 |
+
"loss": 0.721,
|
1512 |
+
"step": 3600
|
1513 |
+
},
|
1514 |
+
{
|
1515 |
+
"epoch": 1.69,
|
1516 |
+
"eval_bleu": 0.6119684034182921,
|
1517 |
+
"eval_brevity_penalty": 0.9784025765941619,
|
1518 |
+
"eval_exact_match": 0.525390625,
|
1519 |
+
"eval_gen_len": 6.888671875,
|
1520 |
+
"eval_length_ratio": 0.9786324786324786,
|
1521 |
+
"eval_loss": 0.6938477754592896,
|
1522 |
+
"eval_reference_length": 2106,
|
1523 |
+
"eval_runtime": 35.4713,
|
1524 |
+
"eval_samples_per_second": 14.434,
|
1525 |
+
"eval_steps_per_second": 0.902,
|
1526 |
+
"eval_translation_length": 2061,
|
1527 |
+
"step": 3600
|
1528 |
+
},
|
1529 |
+
{
|
1530 |
+
"epoch": 1.72,
|
1531 |
+
"learning_rate": 4.279893433631092e-05,
|
1532 |
+
"loss": 0.8326,
|
1533 |
+
"step": 3650
|
1534 |
+
},
|
1535 |
+
{
|
1536 |
+
"epoch": 1.72,
|
1537 |
+
"eval_bleu": 0.6445182499922312,
|
1538 |
+
"eval_brevity_penalty": 0.9942857298792932,
|
1539 |
+
"eval_exact_match": 0.55078125,
|
1540 |
+
"eval_gen_len": 6.951171875,
|
1541 |
+
"eval_length_ratio": 0.9943019943019943,
|
1542 |
+
"eval_loss": 0.6575934886932373,
|
1543 |
+
"eval_reference_length": 2106,
|
1544 |
+
"eval_runtime": 35.1284,
|
1545 |
+
"eval_samples_per_second": 14.575,
|
1546 |
+
"eval_steps_per_second": 0.911,
|
1547 |
+
"eval_translation_length": 2094,
|
1548 |
+
"step": 3650
|
1549 |
+
},
|
1550 |
+
{
|
1551 |
+
"epoch": 1.74,
|
1552 |
+
"learning_rate": 4.201535809434258e-05,
|
1553 |
+
"loss": 0.7976,
|
1554 |
+
"step": 3700
|
1555 |
+
},
|
1556 |
+
{
|
1557 |
+
"epoch": 1.74,
|
1558 |
+
"eval_bleu": 0.6440453785850522,
|
1559 |
+
"eval_brevity_penalty": 0.9508736695620825,
|
1560 |
+
"eval_exact_match": 0.560546875,
|
1561 |
+
"eval_gen_len": 6.767578125,
|
1562 |
+
"eval_length_ratio": 0.9520417853751187,
|
1563 |
+
"eval_loss": 0.6605907678604126,
|
1564 |
+
"eval_reference_length": 2106,
|
1565 |
+
"eval_runtime": 35.1636,
|
1566 |
+
"eval_samples_per_second": 14.561,
|
1567 |
+
"eval_steps_per_second": 0.91,
|
1568 |
+
"eval_translation_length": 2005,
|
1569 |
+
"step": 3700
|
1570 |
+
},
|
1571 |
+
{
|
1572 |
+
"epoch": 1.76,
|
1573 |
+
"learning_rate": 4.123178185237424e-05,
|
1574 |
+
"loss": 0.7771,
|
1575 |
+
"step": 3750
|
1576 |
+
},
|
1577 |
+
{
|
1578 |
+
"epoch": 1.76,
|
1579 |
+
"eval_bleu": 0.644795546998577,
|
1580 |
+
"eval_brevity_penalty": 0.9523670318092967,
|
1581 |
+
"eval_exact_match": 0.560546875,
|
1582 |
+
"eval_gen_len": 6.724609375,
|
1583 |
+
"eval_length_ratio": 0.9534662867996201,
|
1584 |
+
"eval_loss": 0.6605848073959351,
|
1585 |
+
"eval_reference_length": 2106,
|
1586 |
+
"eval_runtime": 34.7093,
|
1587 |
+
"eval_samples_per_second": 14.751,
|
1588 |
+
"eval_steps_per_second": 0.922,
|
1589 |
+
"eval_translation_length": 2008,
|
1590 |
+
"step": 3750
|
1591 |
+
},
|
1592 |
+
{
|
1593 |
+
"epoch": 1.79,
|
1594 |
+
"learning_rate": 4.0448205610405895e-05,
|
1595 |
+
"loss": 0.856,
|
1596 |
+
"step": 3800
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 1.79,
|
1600 |
+
"eval_bleu": 0.6464921288506784,
|
1601 |
+
"eval_brevity_penalty": 0.9817925536433939,
|
1602 |
+
"eval_exact_match": 0.556640625,
|
1603 |
+
"eval_gen_len": 6.865234375,
|
1604 |
+
"eval_length_ratio": 0.9819563152896487,
|
1605 |
+
"eval_loss": 0.6571842432022095,
|
1606 |
+
"eval_reference_length": 2106,
|
1607 |
+
"eval_runtime": 36.0118,
|
1608 |
+
"eval_samples_per_second": 14.218,
|
1609 |
+
"eval_steps_per_second": 0.889,
|
1610 |
+
"eval_translation_length": 2068,
|
1611 |
+
"step": 3800
|
1612 |
+
},
|
1613 |
+
{
|
1614 |
+
"epoch": 1.81,
|
1615 |
+
"learning_rate": 3.966462936843755e-05,
|
1616 |
+
"loss": 0.7667,
|
1617 |
+
"step": 3850
|
1618 |
+
},
|
1619 |
+
{
|
1620 |
+
"epoch": 1.81,
|
1621 |
+
"eval_bleu": 0.6443589957503941,
|
1622 |
+
"eval_brevity_penalty": 0.97694627798514,
|
1623 |
+
"eval_exact_match": 0.55859375,
|
1624 |
+
"eval_gen_len": 6.857421875,
|
1625 |
+
"eval_length_ratio": 0.9772079772079773,
|
1626 |
+
"eval_loss": 0.6501011252403259,
|
1627 |
+
"eval_reference_length": 2106,
|
1628 |
+
"eval_runtime": 36.8454,
|
1629 |
+
"eval_samples_per_second": 13.896,
|
1630 |
+
"eval_steps_per_second": 0.868,
|
1631 |
+
"eval_translation_length": 2058,
|
1632 |
+
"step": 3850
|
1633 |
+
},
|
1634 |
+
{
|
1635 |
+
"epoch": 1.83,
|
1636 |
+
"learning_rate": 3.888105312646921e-05,
|
1637 |
+
"loss": 0.7609,
|
1638 |
+
"step": 3900
|
1639 |
+
},
|
1640 |
+
{
|
1641 |
+
"epoch": 1.83,
|
1642 |
+
"eval_bleu": 0.6357951146509494,
|
1643 |
+
"eval_brevity_penalty": 0.9754879045293495,
|
1644 |
+
"eval_exact_match": 0.560546875,
|
1645 |
+
"eval_gen_len": 6.853515625,
|
1646 |
+
"eval_length_ratio": 0.9757834757834758,
|
1647 |
+
"eval_loss": 0.67498779296875,
|
1648 |
+
"eval_reference_length": 2106,
|
1649 |
+
"eval_runtime": 36.1747,
|
1650 |
+
"eval_samples_per_second": 14.154,
|
1651 |
+
"eval_steps_per_second": 0.885,
|
1652 |
+
"eval_translation_length": 2055,
|
1653 |
+
"step": 3900
|
1654 |
+
},
|
1655 |
+
{
|
1656 |
+
"epoch": 1.86,
|
1657 |
+
"learning_rate": 3.809747688450087e-05,
|
1658 |
+
"loss": 0.871,
|
1659 |
+
"step": 3950
|
1660 |
+
},
|
1661 |
+
{
|
1662 |
+
"epoch": 1.86,
|
1663 |
+
"eval_bleu": 0.6283900991765164,
|
1664 |
+
"eval_brevity_penalty": 0.9478805721747763,
|
1665 |
+
"eval_exact_match": 0.556640625,
|
1666 |
+
"eval_gen_len": 6.693359375,
|
1667 |
+
"eval_length_ratio": 0.9491927825261158,
|
1668 |
+
"eval_loss": 0.6474365592002869,
|
1669 |
+
"eval_reference_length": 2106,
|
1670 |
+
"eval_runtime": 35.4701,
|
1671 |
+
"eval_samples_per_second": 14.435,
|
1672 |
+
"eval_steps_per_second": 0.902,
|
1673 |
+
"eval_translation_length": 1999,
|
1674 |
+
"step": 3950
|
1675 |
+
},
|
1676 |
+
{
|
1677 |
+
"epoch": 1.88,
|
1678 |
+
"learning_rate": 3.731390064253252e-05,
|
1679 |
+
"loss": 0.8432,
|
1680 |
+
"step": 4000
|
1681 |
+
},
|
1682 |
+
{
|
1683 |
+
"epoch": 1.88,
|
1684 |
+
"eval_bleu": 0.6243444944082281,
|
1685 |
+
"eval_brevity_penalty": 1.0,
|
1686 |
+
"eval_exact_match": 0.55859375,
|
1687 |
+
"eval_gen_len": 7.046875,
|
1688 |
+
"eval_length_ratio": 1.0147198480531814,
|
1689 |
+
"eval_loss": 0.6450670957565308,
|
1690 |
+
"eval_reference_length": 2106,
|
1691 |
+
"eval_runtime": 36.3555,
|
1692 |
+
"eval_samples_per_second": 14.083,
|
1693 |
+
"eval_steps_per_second": 0.88,
|
1694 |
+
"eval_translation_length": 2137,
|
1695 |
+
"step": 4000
|
1696 |
+
},
|
1697 |
+
{
|
1698 |
+
"epoch": 1.9,
|
1699 |
+
"learning_rate": 3.6530324400564175e-05,
|
1700 |
+
"loss": 0.7924,
|
1701 |
+
"step": 4050
|
1702 |
+
},
|
1703 |
+
{
|
1704 |
+
"epoch": 1.9,
|
1705 |
+
"eval_bleu": 0.6406129867109561,
|
1706 |
+
"eval_brevity_penalty": 1.0,
|
1707 |
+
"eval_exact_match": 0.55078125,
|
1708 |
+
"eval_gen_len": 6.98828125,
|
1709 |
+
"eval_length_ratio": 1.000474833808167,
|
1710 |
+
"eval_loss": 0.6552107930183411,
|
1711 |
+
"eval_reference_length": 2106,
|
1712 |
+
"eval_runtime": 35.9782,
|
1713 |
+
"eval_samples_per_second": 14.231,
|
1714 |
+
"eval_steps_per_second": 0.889,
|
1715 |
+
"eval_translation_length": 2107,
|
1716 |
+
"step": 4050
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 1.93,
|
1720 |
+
"learning_rate": 3.574674815859583e-05,
|
1721 |
+
"loss": 0.8058,
|
1722 |
+
"step": 4100
|
1723 |
+
},
|
1724 |
+
{
|
1725 |
+
"epoch": 1.93,
|
1726 |
+
"eval_bleu": 0.6521608841609852,
|
1727 |
+
"eval_brevity_penalty": 0.9745145014799987,
|
1728 |
+
"eval_exact_match": 0.5546875,
|
1729 |
+
"eval_gen_len": 6.8515625,
|
1730 |
+
"eval_length_ratio": 0.9748338081671415,
|
1731 |
+
"eval_loss": 0.6530161499977112,
|
1732 |
+
"eval_reference_length": 2106,
|
1733 |
+
"eval_runtime": 36.0535,
|
1734 |
+
"eval_samples_per_second": 14.201,
|
1735 |
+
"eval_steps_per_second": 0.888,
|
1736 |
+
"eval_translation_length": 2053,
|
1737 |
+
"step": 4100
|
1738 |
+
},
|
1739 |
+
{
|
1740 |
+
"epoch": 1.95,
|
1741 |
+
"learning_rate": 3.496317191662749e-05,
|
1742 |
+
"loss": 0.7426,
|
1743 |
+
"step": 4150
|
1744 |
+
},
|
1745 |
+
{
|
1746 |
+
"epoch": 1.95,
|
1747 |
+
"eval_bleu": 0.6517932308118449,
|
1748 |
+
"eval_brevity_penalty": 0.9837246322096701,
|
1749 |
+
"eval_exact_match": 0.552734375,
|
1750 |
+
"eval_gen_len": 6.90625,
|
1751 |
+
"eval_length_ratio": 0.9838556505223172,
|
1752 |
+
"eval_loss": 0.6467949151992798,
|
1753 |
+
"eval_reference_length": 2106,
|
1754 |
+
"eval_runtime": 36.1369,
|
1755 |
+
"eval_samples_per_second": 14.168,
|
1756 |
+
"eval_steps_per_second": 0.886,
|
1757 |
+
"eval_translation_length": 2072,
|
1758 |
+
"step": 4150
|
1759 |
+
},
|
1760 |
+
{
|
1761 |
+
"epoch": 1.97,
|
1762 |
+
"learning_rate": 3.417959567465915e-05,
|
1763 |
+
"loss": 0.806,
|
1764 |
+
"step": 4200
|
1765 |
+
},
|
1766 |
+
{
|
1767 |
+
"epoch": 1.97,
|
1768 |
+
"eval_bleu": 0.6519640951085052,
|
1769 |
+
"eval_brevity_penalty": 0.9740274534203012,
|
1770 |
+
"eval_exact_match": 0.564453125,
|
1771 |
+
"eval_gen_len": 6.8515625,
|
1772 |
+
"eval_length_ratio": 0.9743589743589743,
|
1773 |
+
"eval_loss": 0.6484197378158569,
|
1774 |
+
"eval_reference_length": 2106,
|
1775 |
+
"eval_runtime": 35.8891,
|
1776 |
+
"eval_samples_per_second": 14.266,
|
1777 |
+
"eval_steps_per_second": 0.892,
|
1778 |
+
"eval_translation_length": 2052,
|
1779 |
+
"step": 4200
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 2.0,
|
1783 |
+
"learning_rate": 3.33960194326908e-05,
|
1784 |
+
"loss": 0.8135,
|
1785 |
+
"step": 4250
|
1786 |
+
},
|
1787 |
+
{
|
1788 |
+
"epoch": 2.0,
|
1789 |
+
"eval_bleu": 0.6506610495036242,
|
1790 |
+
"eval_brevity_penalty": 0.97694627798514,
|
1791 |
+
"eval_exact_match": 0.564453125,
|
1792 |
+
"eval_gen_len": 6.859375,
|
1793 |
+
"eval_length_ratio": 0.9772079772079773,
|
1794 |
+
"eval_loss": 0.6449689865112305,
|
1795 |
+
"eval_reference_length": 2106,
|
1796 |
+
"eval_runtime": 35.381,
|
1797 |
+
"eval_samples_per_second": 14.471,
|
1798 |
+
"eval_steps_per_second": 0.904,
|
1799 |
+
"eval_translation_length": 2058,
|
1800 |
+
"step": 4250
|
1801 |
+
}
|
1802 |
+
],
|
1803 |
+
"logging_steps": 50,
|
1804 |
+
"max_steps": 6381,
|
1805 |
+
"num_train_epochs": 3,
|
1806 |
+
"save_steps": 500,
|
1807 |
+
"total_flos": 2.4719244569985024e+16,
|
1808 |
+
"trial_name": null,
|
1809 |
+
"trial_params": null
|
1810 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a01b18e4103a8c7353116c320d7544f9a143a35e35ca2372ac81f888acc50032
|
3 |
+
size 4728
|