Marvin
commited on
Commit
•
8875396
0
Parent(s):
Initial commit
Browse files- .gitattributes +4 -0
- README.md +128 -0
- added_tokens.json +3 -0
- all_results.json +64 -0
- config.json +36 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- special_tokens_map.json +8 -0
- spiece.model +3 -0
- tokenizer.json +3 -0
- tokenizer_config.json +12 -0
- trainer_state.json +768 -0
- training_args.bin +0 -0
.gitattributes
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
3 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
4 |
+
runs/** filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,128 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- de
|
4 |
+
tags:
|
5 |
+
- question-generation
|
6 |
+
- german
|
7 |
+
- text2text-generation
|
8 |
+
- generated_from_trainer
|
9 |
+
datasets:
|
10 |
+
- lmqg/qg_dequad
|
11 |
+
metrics:
|
12 |
+
- bleu4
|
13 |
+
- f1
|
14 |
+
- rouge
|
15 |
+
- exact_match
|
16 |
+
model-index:
|
17 |
+
- name: german-jeopardy-mt5-large-128
|
18 |
+
results:
|
19 |
+
- task:
|
20 |
+
name: Sequence-to-sequence Language Modeling
|
21 |
+
type: text2text-generation
|
22 |
+
dataset:
|
23 |
+
name: lmqg/qg_dequad
|
24 |
+
type: default
|
25 |
+
args: default
|
26 |
+
metrics:
|
27 |
+
- name: BLEU-4
|
28 |
+
type: bleu4
|
29 |
+
value: 16.06
|
30 |
+
- name: F1
|
31 |
+
type: f1
|
32 |
+
value: 42.29
|
33 |
+
- name: ROUGE-1
|
34 |
+
type: rouge1
|
35 |
+
value: 43.40
|
36 |
+
- name: ROUGE-2
|
37 |
+
type: rouge2
|
38 |
+
value: 23.68
|
39 |
+
- name: ROUGE-L
|
40 |
+
type: rougel
|
41 |
+
value: 41.78
|
42 |
+
- name: ROUGE-Lsum
|
43 |
+
type: rougelsum
|
44 |
+
value: 41.79
|
45 |
+
- name: Exact Match
|
46 |
+
type: exact_match
|
47 |
+
value: 3.18
|
48 |
+
---
|
49 |
+
|
50 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
51 |
+
should probably proofread and complete it, then remove this comment. -->
|
52 |
+
|
53 |
+
# german-jeopardy-mt5-large-128
|
54 |
+
|
55 |
+
This model is a fine-tuned version of [google/mt5-large](https://huggingface.co/google/mt5-large) on the [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad) dataset.
|
56 |
+
It achieves the following results on the evaluation set:
|
57 |
+
- Loss: 1.5487
|
58 |
+
- Brevity Penalty: 0.9115
|
59 |
+
- System Length: 19029
|
60 |
+
- Reference Length: 20793
|
61 |
+
- ROUGE-1: 43.40
|
62 |
+
- ROUGE-2: 23.68
|
63 |
+
- ROUGE-L: 41.78
|
64 |
+
- ROUGE-Lsum: 41.79
|
65 |
+
- Exact Match: 3.18
|
66 |
+
- BLEU: 16.06
|
67 |
+
- F1: 42.29
|
68 |
+
|
69 |
+
## Model description
|
70 |
+
|
71 |
+
See [google/mt5-large](https://huggingface.co/google/mt5-large) for the model architecture.
|
72 |
+
The model was trained on a single NVIDIA RTX 3090 GPU with 24GB of VRAM.
|
73 |
+
|
74 |
+
## Intended uses & limitations
|
75 |
+
|
76 |
+
This model can be used for question generation on German text.
|
77 |
+
|
78 |
+
## Training and evaluation data
|
79 |
+
|
80 |
+
See [lmqg/qg_dequad](https://huggingface.co/datasets/lmqg/qg_dequad).
|
81 |
+
|
82 |
+
## Training procedure
|
83 |
+
|
84 |
+
### Training hyperparameters
|
85 |
+
|
86 |
+
The following hyperparameters were used during training:
|
87 |
+
- learning_rate: 0.0001
|
88 |
+
- train_batch_size: 1
|
89 |
+
- eval_batch_size: 1
|
90 |
+
- seed: 7
|
91 |
+
- gradient_accumulation_steps: 128
|
92 |
+
- total_train_batch_size: 128
|
93 |
+
- optimizer: Adafactor
|
94 |
+
- lr_scheduler_type: constant
|
95 |
+
- num_epochs: 20
|
96 |
+
|
97 |
+
### Training results
|
98 |
+
|
99 |
+
| Training Loss | Epoch | Step | Validation Loss | Counts 1 | Counts 2 | Counts 3 | Counts 4 | Totals 1 | Totals 2 | Totals 3 | Totals 4 | Precisions 1 | Precisions 2 | Precisions 3 | Precisions 4 | Brevity Penalty | System Length | Reference Length | ROUGE-1 | ROUGE-2 | ROUGE-L | ROUGE-Lsum | Exact Match | BLEU | Mean Generated Length | F1 |
|
100 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:--------:|:------------:|:------------:|:------------:|:------------:|:---------------:|:-------------:|:----------------:|:-------:|:-------:|:-------:|:----------:|:-----------:|:-------:|:---------------------:|:------:|
|
101 |
+
| 3.9659 | 0.99 | 72 | 1.4145 | 7244 | 2547 | 1183 | 565 | 16296 | 14092 | 11888 | 9684 | 44.4526 | 18.0741 | 9.9512 | 5.8344 | 0.7379 | 16296 | 21250 | 0.3213 | 0.1608 | 0.3091 | 0.309 | 0.0136 | 10.8438 | 11.7786 | 0.3139 |
|
102 |
+
| 1.7081 | 1.99 | 145 | 1.2632 | 7865 | 3037 | 1498 | 759 | 16841 | 14637 | 12433 | 10229 | 46.7015 | 20.7488 | 12.0486 | 7.4201 | 0.7697 | 16841 | 21250 | 0.3577 | 0.189 | 0.3438 | 0.3439 | 0.0181 | 13.2044 | 12.225 | 0.3481 |
|
103 |
+
| 1.4856 | 3.0 | 218 | 1.1974 | 8608 | 3519 | 1818 | 969 | 17627 | 15423 | 13219 | 11015 | 48.8342 | 22.8166 | 13.7529 | 8.7971 | 0.8142 | 17627 | 21250 | 0.3969 | 0.2181 | 0.381 | 0.3812 | 0.0268 | 15.6014 | 13.0027 | 0.3882 |
|
104 |
+
| 1.3277 | 4.0 | 291 | 1.1394 | 9018 | 3702 | 1907 | 1029 | 17465 | 15261 | 13057 | 10853 | 51.6347 | 24.2579 | 14.6052 | 9.4812 | 0.8052 | 17465 | 21250 | 0.424 | 0.2321 | 0.4087 | 0.4085 | 0.0313 | 16.4313 | 12.8716 | 0.4156 |
|
105 |
+
| 1.2314 | 4.99 | 363 | 1.1193 | 9240 | 3869 | 1994 | 1076 | 17794 | 15590 | 13386 | 11182 | 51.9276 | 24.8172 | 14.8962 | 9.6226 | 0.8235 | 17794 | 21250 | 0.4336 | 0.2413 | 0.4183 | 0.418 | 0.0363 | 17.0718 | 13.2137 | 0.4256 |
|
106 |
+
| 1.1264 | 5.99 | 436 | 1.1086 | 9263 | 3908 | 2055 | 1127 | 17502 | 15298 | 13094 | 10890 | 52.9254 | 25.5458 | 15.6942 | 10.3489 | 0.8072 | 17502 | 21250 | 0.4383 | 0.2452 | 0.4239 | 0.4237 | 0.0372 | 17.4744 | 13.034 | 0.4309 |
|
107 |
+
| 1.0469 | 7.0 | 509 | 1.1038 | 9434 | 4034 | 2146 | 1189 | 18028 | 15824 | 13620 | 11416 | 52.3297 | 25.4929 | 15.7562 | 10.4152 | 0.8363 | 18028 | 21250 | 0.4433 | 0.2505 | 0.4286 | 0.4282 | 0.039 | 18.0906 | 13.422 | 0.4348 |
|
108 |
+
| 0.9874 | 8.0 | 582 | 1.0990 | 9746 | 4265 | 2287 | 1285 | 18351 | 16147 | 13943 | 11739 | 53.1088 | 26.4136 | 16.4025 | 10.9464 | 0.8539 | 18351 | 21250 | 0.457 | 0.2627 | 0.4417 | 0.4416 | 0.0454 | 19.1287 | 13.6466 | 0.4498 |
|
109 |
+
| 0.9488 | 8.99 | 654 | 1.1175 | 9484 | 4062 | 2158 | 1197 | 17831 | 15627 | 13423 | 11219 | 53.1883 | 25.9935 | 16.0769 | 10.6694 | 0.8255 | 17831 | 21250 | 0.4482 | 0.2548 | 0.4338 | 0.4333 | 0.0431 | 18.2172 | 13.2763 | 0.4399 |
|
110 |
+
| 0.8893 | 9.99 | 727 | 1.1222 | 9650 | 4205 | 2289 | 1289 | 18017 | 15813 | 13609 | 11405 | 53.5605 | 26.592 | 16.8198 | 11.3021 | 0.8357 | 18017 | 21250 | 0.4543 | 0.262 | 0.4396 | 0.4394 | 0.0463 | 19.064 | 13.4251 | 0.4472 |
|
111 |
+
| 0.8362 | 10.99 | 800 | 1.1342 | 9706 | 4232 | 2279 | 1281 | 18232 | 16028 | 13824 | 11620 | 53.2361 | 26.4038 | 16.4858 | 11.0241 | 0.8474 | 18232 | 21250 | 0.4551 | 0.2632 | 0.4395 | 0.4393 | 0.0472 | 19.052 | 13.6021 | 0.4473 |
|
112 |
+
| 0.7835 | 12.0 | 873 | 1.1427 | 9802 | 4280 | 2292 | 1285 | 18491 | 16287 | 14083 | 11879 | 53.0096 | 26.2786 | 16.2749 | 10.8174 | 0.8614 | 18491 | 21250 | 0.458 | 0.2634 | 0.4414 | 0.4412 | 0.0472 | 19.169 | 14.0168 | 0.4497 |
|
113 |
+
| 0.7441 | 12.99 | 945 | 1.1669 | 9816 | 4323 | 2334 | 1294 | 18498 | 16294 | 14090 | 11886 | 53.0652 | 26.5312 | 16.5649 | 10.8868 | 0.8618 | 18498 | 21250 | 0.4577 | 0.2659 | 0.4418 | 0.4417 | 0.0463 | 19.3443 | 13.8348 | 0.4493 |
|
114 |
+
| 0.7012 | 13.99 | 1018 | 1.1740 | 9856 | 4364 | 2375 | 1360 | 18537 | 16333 | 14129 | 11925 | 53.1693 | 26.7189 | 16.8094 | 11.4046 | 0.8639 | 18537 | 21250 | 0.4591 | 0.2653 | 0.443 | 0.4428 | 0.0476 | 19.7341 | 13.976 | 0.4514 |
|
115 |
+
| 0.6597 | 14.99 | 1091 | 1.1987 | 9780 | 4292 | 2336 | 1302 | 18468 | 16264 | 14060 | 11856 | 52.9565 | 26.3896 | 16.6145 | 10.9818 | 0.8602 | 18468 | 21250 | 0.457 | 0.2633 | 0.4418 | 0.4416 | 0.0485 | 19.3289 | 13.8802 | 0.4492 |
|
116 |
+
| 0.6236 | 16.0 | 1164 | 1.2135 | 9931 | 4388 | 2390 | 1359 | 18717 | 16513 | 14309 | 12105 | 53.0587 | 26.573 | 16.7028 | 11.2268 | 0.8734 | 18717 | 21250 | 0.4618 | 0.2682 | 0.4452 | 0.445 | 0.0495 | 19.8055 | 14.044 | 0.4538 |
|
117 |
+
| 0.5933 | 17.0 | 1237 | 1.2305 | 9806 | 4316 | 2366 | 1348 | 18566 | 16362 | 14158 | 11954 | 52.817 | 26.3782 | 16.7114 | 11.2766 | 0.8654 | 18566 | 21250 | 0.4571 | 0.2628 | 0.4407 | 0.4409 | 0.049 | 19.5893 | 14.0622 | 0.4485 |
|
118 |
+
| 0.5622 | 17.99 | 1309 | 1.2796 | 9787 | 4306 | 2346 | 1338 | 18559 | 16355 | 14151 | 11947 | 52.7345 | 26.3283 | 16.5783 | 11.1995 | 0.865 | 18559 | 21250 | 0.4549 | 0.2609 | 0.4383 | 0.4382 | 0.0476 | 19.4914 | 13.7763 | 0.447 |
|
119 |
+
| 0.5275 | 18.99 | 1382 | 1.2833 | 9918 | 4363 | 2374 | 1355 | 18950 | 16746 | 14542 | 12338 | 52.3377 | 26.054 | 16.3251 | 10.9823 | 0.8857 | 18950 | 21250 | 0.4573 | 0.2624 | 0.441 | 0.4408 | 0.0508 | 19.6947 | 14.1647 | 0.4499 |
|
120 |
+
| 0.4986 | 19.79 | 1440 | 1.3059 | 9879 | 4315 | 2347 | 1324 | 18931 | 16727 | 14523 | 12319 | 52.1842 | 25.7966 | 16.1606 | 10.7476 | 0.8847 | 18931 | 21250 | 0.4564 | 0.2622 | 0.4407 | 0.4403 | 0.0495 | 19.4544 | 14.2827 | 0.4478 |
|
121 |
+
|
122 |
+
|
123 |
+
### Framework versions
|
124 |
+
|
125 |
+
- Transformers 4.32.1
|
126 |
+
- Pytorch 2.1.0
|
127 |
+
- Datasets 2.12.0
|
128 |
+
- Tokenizers 0.13.3
|
added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<hl>": 250100
|
3 |
+
}
|
all_results.json
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.79,
|
3 |
+
"eval_bleu": 19.4544,
|
4 |
+
"eval_bp": 0.8847,
|
5 |
+
"eval_counts_1": 9879,
|
6 |
+
"eval_counts_2": 4315,
|
7 |
+
"eval_counts_3": 2347,
|
8 |
+
"eval_counts_4": 1324,
|
9 |
+
"eval_exact_match": 0.0495,
|
10 |
+
"eval_f1": 0.4478,
|
11 |
+
"eval_gen_len": 14.2827,
|
12 |
+
"eval_loss": 1.3059108257293701,
|
13 |
+
"eval_precisions_1": 52.1842,
|
14 |
+
"eval_precisions_2": 25.7966,
|
15 |
+
"eval_precisions_3": 16.1606,
|
16 |
+
"eval_precisions_4": 10.7476,
|
17 |
+
"eval_ref_len": 21250,
|
18 |
+
"eval_rouge1": 0.4564,
|
19 |
+
"eval_rouge2": 0.2622,
|
20 |
+
"eval_rougeL": 0.4407,
|
21 |
+
"eval_rougeLsum": 0.4403,
|
22 |
+
"eval_runtime": 3710.7373,
|
23 |
+
"eval_samples": 2204,
|
24 |
+
"eval_samples_per_second": 0.594,
|
25 |
+
"eval_steps_per_second": 0.594,
|
26 |
+
"eval_sys_len": 18931,
|
27 |
+
"eval_totals_1": 18931,
|
28 |
+
"eval_totals_2": 16727,
|
29 |
+
"eval_totals_3": 14523,
|
30 |
+
"eval_totals_4": 12319,
|
31 |
+
"predict_bleu": 15.5923,
|
32 |
+
"predict_bp": 0.9152,
|
33 |
+
"predict_counts_1": 9130,
|
34 |
+
"predict_counts_2": 3573,
|
35 |
+
"predict_counts_3": 1732,
|
36 |
+
"predict_counts_4": 883,
|
37 |
+
"predict_exact_match": 0.0318,
|
38 |
+
"predict_f1": 0.4128,
|
39 |
+
"predict_gen_len": 14.554,
|
40 |
+
"predict_loss": 1.5979496240615845,
|
41 |
+
"predict_precisions_1": 47.801,
|
42 |
+
"predict_precisions_2": 21.147,
|
43 |
+
"predict_precisions_3": 11.7887,
|
44 |
+
"predict_precisions_4": 7.0708,
|
45 |
+
"predict_ref_len": 20793,
|
46 |
+
"predict_rouge1": 0.4233,
|
47 |
+
"predict_rouge2": 0.2296,
|
48 |
+
"predict_rougeL": 0.407,
|
49 |
+
"predict_rougeLsum": 0.4067,
|
50 |
+
"predict_runtime": 3256.8207,
|
51 |
+
"predict_samples": 2204,
|
52 |
+
"predict_samples_per_second": 0.677,
|
53 |
+
"predict_steps_per_second": 0.677,
|
54 |
+
"predict_sys_len": 19100,
|
55 |
+
"predict_totals_1": 19100,
|
56 |
+
"predict_totals_2": 16896,
|
57 |
+
"predict_totals_3": 14692,
|
58 |
+
"predict_totals_4": 12488,
|
59 |
+
"train_loss": 1.0667428798145717,
|
60 |
+
"train_runtime": 140813.6912,
|
61 |
+
"train_samples": 9314,
|
62 |
+
"train_samples_per_second": 1.323,
|
63 |
+
"train_steps_per_second": 0.01
|
64 |
+
}
|
config.json
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/mt5-large",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"classifier_dropout": 0.0,
|
7 |
+
"d_ff": 2816,
|
8 |
+
"d_kv": 64,
|
9 |
+
"d_model": 1024,
|
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 |
+
"length_penalty": 0.0,
|
20 |
+
"max_length": 64,
|
21 |
+
"model_type": "mt5",
|
22 |
+
"num_beams": 4,
|
23 |
+
"num_decoder_layers": 24,
|
24 |
+
"num_heads": 16,
|
25 |
+
"num_layers": 24,
|
26 |
+
"output_past": true,
|
27 |
+
"pad_token_id": 0,
|
28 |
+
"relative_attention_max_distance": 128,
|
29 |
+
"relative_attention_num_buckets": 32,
|
30 |
+
"tie_word_embeddings": false,
|
31 |
+
"tokenizer_class": "T5Tokenizer",
|
32 |
+
"torch_dtype": "float32",
|
33 |
+
"transformers_version": "4.32.1",
|
34 |
+
"use_cache": true,
|
35 |
+
"vocab_size": 250112
|
36 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"decoder_start_token_id": 0,
|
3 |
+
"eos_token_id": 1,
|
4 |
+
"length_penalty": 0.0,
|
5 |
+
"max_length": 64,
|
6 |
+
"num_beams": 4,
|
7 |
+
"pad_token_id": 0,
|
8 |
+
"transformers_version": "4.32.1"
|
9 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d027a7e16cc798792ca2a7dd9102e4806e556b8d7c1fad130064107cab1a5bf5
|
3 |
+
size 4918393736
|
special_tokens_map.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<hl>"
|
4 |
+
],
|
5 |
+
"eos_token": "</s>",
|
6 |
+
"pad_token": "<pad>",
|
7 |
+
"unk_token": "<unk>"
|
8 |
+
}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
|
3 |
+
size 4309802
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4ef4273f90fa3341fdb81c66eb7973b1651a8183e57c8ab9e9144aac61b48f9f
|
3 |
+
size 16330550
|
tokenizer_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": null,
|
3 |
+
"clean_up_tokenization_spaces": true,
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"extra_ids": 0,
|
6 |
+
"legacy": true,
|
7 |
+
"model_max_length": 1000000000000000019884624838656,
|
8 |
+
"pad_token": "<pad>",
|
9 |
+
"sp_model_kwargs": {},
|
10 |
+
"tokenizer_class": "T5Tokenizer",
|
11 |
+
"unk_token": "<unk>"
|
12 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 19.789564097058193,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1440,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.99,
|
13 |
+
"learning_rate": 0.0001,
|
14 |
+
"loss": 3.9659,
|
15 |
+
"step": 72
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.99,
|
19 |
+
"eval_bleu": 10.8438,
|
20 |
+
"eval_bp": 0.7379,
|
21 |
+
"eval_counts_1": 7244,
|
22 |
+
"eval_counts_2": 2547,
|
23 |
+
"eval_counts_3": 1183,
|
24 |
+
"eval_counts_4": 565,
|
25 |
+
"eval_exact_match": 0.0136,
|
26 |
+
"eval_f1": 0.3139,
|
27 |
+
"eval_gen_len": 11.7786,
|
28 |
+
"eval_loss": 1.4144511222839355,
|
29 |
+
"eval_precisions_1": 44.4526,
|
30 |
+
"eval_precisions_2": 18.0741,
|
31 |
+
"eval_precisions_3": 9.9512,
|
32 |
+
"eval_precisions_4": 5.8344,
|
33 |
+
"eval_ref_len": 21250,
|
34 |
+
"eval_rouge1": 0.3213,
|
35 |
+
"eval_rouge2": 0.1608,
|
36 |
+
"eval_rougeL": 0.3091,
|
37 |
+
"eval_rougeLsum": 0.309,
|
38 |
+
"eval_runtime": 2106.9539,
|
39 |
+
"eval_samples_per_second": 1.046,
|
40 |
+
"eval_steps_per_second": 1.046,
|
41 |
+
"eval_sys_len": 16296,
|
42 |
+
"eval_totals_1": 16296,
|
43 |
+
"eval_totals_2": 14092,
|
44 |
+
"eval_totals_3": 11888,
|
45 |
+
"eval_totals_4": 9684,
|
46 |
+
"step": 72
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.99,
|
50 |
+
"learning_rate": 0.0001,
|
51 |
+
"loss": 1.7081,
|
52 |
+
"step": 145
|
53 |
+
},
|
54 |
+
{
|
55 |
+
"epoch": 1.99,
|
56 |
+
"eval_bleu": 13.2044,
|
57 |
+
"eval_bp": 0.7697,
|
58 |
+
"eval_counts_1": 7865,
|
59 |
+
"eval_counts_2": 3037,
|
60 |
+
"eval_counts_3": 1498,
|
61 |
+
"eval_counts_4": 759,
|
62 |
+
"eval_exact_match": 0.0181,
|
63 |
+
"eval_f1": 0.3481,
|
64 |
+
"eval_gen_len": 12.225,
|
65 |
+
"eval_loss": 1.263157844543457,
|
66 |
+
"eval_precisions_1": 46.7015,
|
67 |
+
"eval_precisions_2": 20.7488,
|
68 |
+
"eval_precisions_3": 12.0486,
|
69 |
+
"eval_precisions_4": 7.4201,
|
70 |
+
"eval_ref_len": 21250,
|
71 |
+
"eval_rouge1": 0.3577,
|
72 |
+
"eval_rouge2": 0.189,
|
73 |
+
"eval_rougeL": 0.3438,
|
74 |
+
"eval_rougeLsum": 0.3439,
|
75 |
+
"eval_runtime": 3942.8178,
|
76 |
+
"eval_samples_per_second": 0.559,
|
77 |
+
"eval_steps_per_second": 0.559,
|
78 |
+
"eval_sys_len": 16841,
|
79 |
+
"eval_totals_1": 16841,
|
80 |
+
"eval_totals_2": 14637,
|
81 |
+
"eval_totals_3": 12433,
|
82 |
+
"eval_totals_4": 10229,
|
83 |
+
"step": 145
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 3.0,
|
87 |
+
"learning_rate": 0.0001,
|
88 |
+
"loss": 1.4856,
|
89 |
+
"step": 218
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.0,
|
93 |
+
"eval_bleu": 15.6014,
|
94 |
+
"eval_bp": 0.8142,
|
95 |
+
"eval_counts_1": 8608,
|
96 |
+
"eval_counts_2": 3519,
|
97 |
+
"eval_counts_3": 1818,
|
98 |
+
"eval_counts_4": 969,
|
99 |
+
"eval_exact_match": 0.0268,
|
100 |
+
"eval_f1": 0.3882,
|
101 |
+
"eval_gen_len": 13.0027,
|
102 |
+
"eval_loss": 1.1974213123321533,
|
103 |
+
"eval_precisions_1": 48.8342,
|
104 |
+
"eval_precisions_2": 22.8166,
|
105 |
+
"eval_precisions_3": 13.7529,
|
106 |
+
"eval_precisions_4": 8.7971,
|
107 |
+
"eval_ref_len": 21250,
|
108 |
+
"eval_rouge1": 0.3969,
|
109 |
+
"eval_rouge2": 0.2181,
|
110 |
+
"eval_rougeL": 0.381,
|
111 |
+
"eval_rougeLsum": 0.3812,
|
112 |
+
"eval_runtime": 4069.754,
|
113 |
+
"eval_samples_per_second": 0.542,
|
114 |
+
"eval_steps_per_second": 0.542,
|
115 |
+
"eval_sys_len": 17627,
|
116 |
+
"eval_totals_1": 17627,
|
117 |
+
"eval_totals_2": 15423,
|
118 |
+
"eval_totals_3": 13219,
|
119 |
+
"eval_totals_4": 11015,
|
120 |
+
"step": 218
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 4.0,
|
124 |
+
"learning_rate": 0.0001,
|
125 |
+
"loss": 1.3277,
|
126 |
+
"step": 291
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 4.0,
|
130 |
+
"eval_bleu": 16.4313,
|
131 |
+
"eval_bp": 0.8052,
|
132 |
+
"eval_counts_1": 9018,
|
133 |
+
"eval_counts_2": 3702,
|
134 |
+
"eval_counts_3": 1907,
|
135 |
+
"eval_counts_4": 1029,
|
136 |
+
"eval_exact_match": 0.0313,
|
137 |
+
"eval_f1": 0.4156,
|
138 |
+
"eval_gen_len": 12.8716,
|
139 |
+
"eval_loss": 1.1393847465515137,
|
140 |
+
"eval_precisions_1": 51.6347,
|
141 |
+
"eval_precisions_2": 24.2579,
|
142 |
+
"eval_precisions_3": 14.6052,
|
143 |
+
"eval_precisions_4": 9.4812,
|
144 |
+
"eval_ref_len": 21250,
|
145 |
+
"eval_rouge1": 0.424,
|
146 |
+
"eval_rouge2": 0.2321,
|
147 |
+
"eval_rougeL": 0.4087,
|
148 |
+
"eval_rougeLsum": 0.4085,
|
149 |
+
"eval_runtime": 4037.7601,
|
150 |
+
"eval_samples_per_second": 0.546,
|
151 |
+
"eval_steps_per_second": 0.546,
|
152 |
+
"eval_sys_len": 17465,
|
153 |
+
"eval_totals_1": 17465,
|
154 |
+
"eval_totals_2": 15261,
|
155 |
+
"eval_totals_3": 13057,
|
156 |
+
"eval_totals_4": 10853,
|
157 |
+
"step": 291
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 4.99,
|
161 |
+
"learning_rate": 0.0001,
|
162 |
+
"loss": 1.2314,
|
163 |
+
"step": 363
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 4.99,
|
167 |
+
"eval_bleu": 17.0718,
|
168 |
+
"eval_bp": 0.8235,
|
169 |
+
"eval_counts_1": 9240,
|
170 |
+
"eval_counts_2": 3869,
|
171 |
+
"eval_counts_3": 1994,
|
172 |
+
"eval_counts_4": 1076,
|
173 |
+
"eval_exact_match": 0.0363,
|
174 |
+
"eval_f1": 0.4256,
|
175 |
+
"eval_gen_len": 13.2137,
|
176 |
+
"eval_loss": 1.1193382740020752,
|
177 |
+
"eval_precisions_1": 51.9276,
|
178 |
+
"eval_precisions_2": 24.8172,
|
179 |
+
"eval_precisions_3": 14.8962,
|
180 |
+
"eval_precisions_4": 9.6226,
|
181 |
+
"eval_ref_len": 21250,
|
182 |
+
"eval_rouge1": 0.4336,
|
183 |
+
"eval_rouge2": 0.2413,
|
184 |
+
"eval_rougeL": 0.4183,
|
185 |
+
"eval_rougeLsum": 0.418,
|
186 |
+
"eval_runtime": 4116.6581,
|
187 |
+
"eval_samples_per_second": 0.535,
|
188 |
+
"eval_steps_per_second": 0.535,
|
189 |
+
"eval_sys_len": 17794,
|
190 |
+
"eval_totals_1": 17794,
|
191 |
+
"eval_totals_2": 15590,
|
192 |
+
"eval_totals_3": 13386,
|
193 |
+
"eval_totals_4": 11182,
|
194 |
+
"step": 363
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 5.99,
|
198 |
+
"learning_rate": 0.0001,
|
199 |
+
"loss": 1.1264,
|
200 |
+
"step": 436
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 5.99,
|
204 |
+
"eval_bleu": 17.4744,
|
205 |
+
"eval_bp": 0.8072,
|
206 |
+
"eval_counts_1": 9263,
|
207 |
+
"eval_counts_2": 3908,
|
208 |
+
"eval_counts_3": 2055,
|
209 |
+
"eval_counts_4": 1127,
|
210 |
+
"eval_exact_match": 0.0372,
|
211 |
+
"eval_f1": 0.4309,
|
212 |
+
"eval_gen_len": 13.034,
|
213 |
+
"eval_loss": 1.1085509061813354,
|
214 |
+
"eval_precisions_1": 52.9254,
|
215 |
+
"eval_precisions_2": 25.5458,
|
216 |
+
"eval_precisions_3": 15.6942,
|
217 |
+
"eval_precisions_4": 10.3489,
|
218 |
+
"eval_ref_len": 21250,
|
219 |
+
"eval_rouge1": 0.4383,
|
220 |
+
"eval_rouge2": 0.2452,
|
221 |
+
"eval_rougeL": 0.4239,
|
222 |
+
"eval_rougeLsum": 0.4237,
|
223 |
+
"eval_runtime": 3709.3886,
|
224 |
+
"eval_samples_per_second": 0.594,
|
225 |
+
"eval_steps_per_second": 0.594,
|
226 |
+
"eval_sys_len": 17502,
|
227 |
+
"eval_totals_1": 17502,
|
228 |
+
"eval_totals_2": 15298,
|
229 |
+
"eval_totals_3": 13094,
|
230 |
+
"eval_totals_4": 10890,
|
231 |
+
"step": 436
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 7.0,
|
235 |
+
"learning_rate": 0.0001,
|
236 |
+
"loss": 1.0469,
|
237 |
+
"step": 509
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 7.0,
|
241 |
+
"eval_bleu": 18.0906,
|
242 |
+
"eval_bp": 0.8363,
|
243 |
+
"eval_counts_1": 9434,
|
244 |
+
"eval_counts_2": 4034,
|
245 |
+
"eval_counts_3": 2146,
|
246 |
+
"eval_counts_4": 1189,
|
247 |
+
"eval_exact_match": 0.039,
|
248 |
+
"eval_f1": 0.4348,
|
249 |
+
"eval_gen_len": 13.422,
|
250 |
+
"eval_loss": 1.103752851486206,
|
251 |
+
"eval_precisions_1": 52.3297,
|
252 |
+
"eval_precisions_2": 25.4929,
|
253 |
+
"eval_precisions_3": 15.7562,
|
254 |
+
"eval_precisions_4": 10.4152,
|
255 |
+
"eval_ref_len": 21250,
|
256 |
+
"eval_rouge1": 0.4433,
|
257 |
+
"eval_rouge2": 0.2505,
|
258 |
+
"eval_rougeL": 0.4286,
|
259 |
+
"eval_rougeLsum": 0.4282,
|
260 |
+
"eval_runtime": 4081.2971,
|
261 |
+
"eval_samples_per_second": 0.54,
|
262 |
+
"eval_steps_per_second": 0.54,
|
263 |
+
"eval_sys_len": 18028,
|
264 |
+
"eval_totals_1": 18028,
|
265 |
+
"eval_totals_2": 15824,
|
266 |
+
"eval_totals_3": 13620,
|
267 |
+
"eval_totals_4": 11416,
|
268 |
+
"step": 509
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 8.0,
|
272 |
+
"learning_rate": 0.0001,
|
273 |
+
"loss": 0.9874,
|
274 |
+
"step": 582
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 8.0,
|
278 |
+
"eval_bleu": 19.1287,
|
279 |
+
"eval_bp": 0.8539,
|
280 |
+
"eval_counts_1": 9746,
|
281 |
+
"eval_counts_2": 4265,
|
282 |
+
"eval_counts_3": 2287,
|
283 |
+
"eval_counts_4": 1285,
|
284 |
+
"eval_exact_match": 0.0454,
|
285 |
+
"eval_f1": 0.4498,
|
286 |
+
"eval_gen_len": 13.6466,
|
287 |
+
"eval_loss": 1.0989724397659302,
|
288 |
+
"eval_precisions_1": 53.1088,
|
289 |
+
"eval_precisions_2": 26.4136,
|
290 |
+
"eval_precisions_3": 16.4025,
|
291 |
+
"eval_precisions_4": 10.9464,
|
292 |
+
"eval_ref_len": 21250,
|
293 |
+
"eval_rouge1": 0.457,
|
294 |
+
"eval_rouge2": 0.2627,
|
295 |
+
"eval_rougeL": 0.4417,
|
296 |
+
"eval_rougeLsum": 0.4416,
|
297 |
+
"eval_runtime": 2875.9709,
|
298 |
+
"eval_samples_per_second": 0.766,
|
299 |
+
"eval_steps_per_second": 0.766,
|
300 |
+
"eval_sys_len": 18351,
|
301 |
+
"eval_totals_1": 18351,
|
302 |
+
"eval_totals_2": 16147,
|
303 |
+
"eval_totals_3": 13943,
|
304 |
+
"eval_totals_4": 11739,
|
305 |
+
"step": 582
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 8.99,
|
309 |
+
"learning_rate": 0.0001,
|
310 |
+
"loss": 0.9488,
|
311 |
+
"step": 654
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 8.99,
|
315 |
+
"eval_bleu": 18.2172,
|
316 |
+
"eval_bp": 0.8255,
|
317 |
+
"eval_counts_1": 9484,
|
318 |
+
"eval_counts_2": 4062,
|
319 |
+
"eval_counts_3": 2158,
|
320 |
+
"eval_counts_4": 1197,
|
321 |
+
"eval_exact_match": 0.0431,
|
322 |
+
"eval_f1": 0.4399,
|
323 |
+
"eval_gen_len": 13.2763,
|
324 |
+
"eval_loss": 1.1175453662872314,
|
325 |
+
"eval_precisions_1": 53.1883,
|
326 |
+
"eval_precisions_2": 25.9935,
|
327 |
+
"eval_precisions_3": 16.0769,
|
328 |
+
"eval_precisions_4": 10.6694,
|
329 |
+
"eval_ref_len": 21250,
|
330 |
+
"eval_rouge1": 0.4482,
|
331 |
+
"eval_rouge2": 0.2548,
|
332 |
+
"eval_rougeL": 0.4338,
|
333 |
+
"eval_rougeLsum": 0.4333,
|
334 |
+
"eval_runtime": 4231.6184,
|
335 |
+
"eval_samples_per_second": 0.521,
|
336 |
+
"eval_steps_per_second": 0.521,
|
337 |
+
"eval_sys_len": 17831,
|
338 |
+
"eval_totals_1": 17831,
|
339 |
+
"eval_totals_2": 15627,
|
340 |
+
"eval_totals_3": 13423,
|
341 |
+
"eval_totals_4": 11219,
|
342 |
+
"step": 654
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 9.99,
|
346 |
+
"learning_rate": 0.0001,
|
347 |
+
"loss": 0.8893,
|
348 |
+
"step": 727
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 9.99,
|
352 |
+
"eval_bleu": 19.064,
|
353 |
+
"eval_bp": 0.8357,
|
354 |
+
"eval_counts_1": 9650,
|
355 |
+
"eval_counts_2": 4205,
|
356 |
+
"eval_counts_3": 2289,
|
357 |
+
"eval_counts_4": 1289,
|
358 |
+
"eval_exact_match": 0.0463,
|
359 |
+
"eval_f1": 0.4472,
|
360 |
+
"eval_gen_len": 13.4251,
|
361 |
+
"eval_loss": 1.1221915483474731,
|
362 |
+
"eval_precisions_1": 53.5605,
|
363 |
+
"eval_precisions_2": 26.592,
|
364 |
+
"eval_precisions_3": 16.8198,
|
365 |
+
"eval_precisions_4": 11.3021,
|
366 |
+
"eval_ref_len": 21250,
|
367 |
+
"eval_rouge1": 0.4543,
|
368 |
+
"eval_rouge2": 0.262,
|
369 |
+
"eval_rougeL": 0.4396,
|
370 |
+
"eval_rougeLsum": 0.4394,
|
371 |
+
"eval_runtime": 4369.7974,
|
372 |
+
"eval_samples_per_second": 0.504,
|
373 |
+
"eval_steps_per_second": 0.504,
|
374 |
+
"eval_sys_len": 18017,
|
375 |
+
"eval_totals_1": 18017,
|
376 |
+
"eval_totals_2": 15813,
|
377 |
+
"eval_totals_3": 13609,
|
378 |
+
"eval_totals_4": 11405,
|
379 |
+
"step": 727
|
380 |
+
},
|
381 |
+
{
|
382 |
+
"epoch": 10.99,
|
383 |
+
"learning_rate": 0.0001,
|
384 |
+
"loss": 0.8362,
|
385 |
+
"step": 800
|
386 |
+
},
|
387 |
+
{
|
388 |
+
"epoch": 10.99,
|
389 |
+
"eval_bleu": 19.052,
|
390 |
+
"eval_bp": 0.8474,
|
391 |
+
"eval_counts_1": 9706,
|
392 |
+
"eval_counts_2": 4232,
|
393 |
+
"eval_counts_3": 2279,
|
394 |
+
"eval_counts_4": 1281,
|
395 |
+
"eval_exact_match": 0.0472,
|
396 |
+
"eval_f1": 0.4473,
|
397 |
+
"eval_gen_len": 13.6021,
|
398 |
+
"eval_loss": 1.1342219114303589,
|
399 |
+
"eval_precisions_1": 53.2361,
|
400 |
+
"eval_precisions_2": 26.4038,
|
401 |
+
"eval_precisions_3": 16.4858,
|
402 |
+
"eval_precisions_4": 11.0241,
|
403 |
+
"eval_ref_len": 21250,
|
404 |
+
"eval_rouge1": 0.4551,
|
405 |
+
"eval_rouge2": 0.2632,
|
406 |
+
"eval_rougeL": 0.4395,
|
407 |
+
"eval_rougeLsum": 0.4393,
|
408 |
+
"eval_runtime": 4741.4712,
|
409 |
+
"eval_samples_per_second": 0.465,
|
410 |
+
"eval_steps_per_second": 0.465,
|
411 |
+
"eval_sys_len": 18232,
|
412 |
+
"eval_totals_1": 18232,
|
413 |
+
"eval_totals_2": 16028,
|
414 |
+
"eval_totals_3": 13824,
|
415 |
+
"eval_totals_4": 11620,
|
416 |
+
"step": 800
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 12.0,
|
420 |
+
"learning_rate": 0.0001,
|
421 |
+
"loss": 0.7835,
|
422 |
+
"step": 873
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 12.0,
|
426 |
+
"eval_bleu": 19.169,
|
427 |
+
"eval_bp": 0.8614,
|
428 |
+
"eval_counts_1": 9802,
|
429 |
+
"eval_counts_2": 4280,
|
430 |
+
"eval_counts_3": 2292,
|
431 |
+
"eval_counts_4": 1285,
|
432 |
+
"eval_exact_match": 0.0472,
|
433 |
+
"eval_f1": 0.4497,
|
434 |
+
"eval_gen_len": 14.0168,
|
435 |
+
"eval_loss": 1.1426819562911987,
|
436 |
+
"eval_precisions_1": 53.0096,
|
437 |
+
"eval_precisions_2": 26.2786,
|
438 |
+
"eval_precisions_3": 16.2749,
|
439 |
+
"eval_precisions_4": 10.8174,
|
440 |
+
"eval_ref_len": 21250,
|
441 |
+
"eval_rouge1": 0.458,
|
442 |
+
"eval_rouge2": 0.2634,
|
443 |
+
"eval_rougeL": 0.4414,
|
444 |
+
"eval_rougeLsum": 0.4412,
|
445 |
+
"eval_runtime": 2858.9204,
|
446 |
+
"eval_samples_per_second": 0.771,
|
447 |
+
"eval_steps_per_second": 0.771,
|
448 |
+
"eval_sys_len": 18491,
|
449 |
+
"eval_totals_1": 18491,
|
450 |
+
"eval_totals_2": 16287,
|
451 |
+
"eval_totals_3": 14083,
|
452 |
+
"eval_totals_4": 11879,
|
453 |
+
"step": 873
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 12.99,
|
457 |
+
"learning_rate": 0.0001,
|
458 |
+
"loss": 0.7441,
|
459 |
+
"step": 945
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 12.99,
|
463 |
+
"eval_bleu": 19.3443,
|
464 |
+
"eval_bp": 0.8618,
|
465 |
+
"eval_counts_1": 9816,
|
466 |
+
"eval_counts_2": 4323,
|
467 |
+
"eval_counts_3": 2334,
|
468 |
+
"eval_counts_4": 1294,
|
469 |
+
"eval_exact_match": 0.0463,
|
470 |
+
"eval_f1": 0.4493,
|
471 |
+
"eval_gen_len": 13.8348,
|
472 |
+
"eval_loss": 1.1669002771377563,
|
473 |
+
"eval_precisions_1": 53.0652,
|
474 |
+
"eval_precisions_2": 26.5312,
|
475 |
+
"eval_precisions_3": 16.5649,
|
476 |
+
"eval_precisions_4": 10.8868,
|
477 |
+
"eval_ref_len": 21250,
|
478 |
+
"eval_rouge1": 0.4577,
|
479 |
+
"eval_rouge2": 0.2659,
|
480 |
+
"eval_rougeL": 0.4418,
|
481 |
+
"eval_rougeLsum": 0.4417,
|
482 |
+
"eval_runtime": 2130.8,
|
483 |
+
"eval_samples_per_second": 1.034,
|
484 |
+
"eval_steps_per_second": 1.034,
|
485 |
+
"eval_sys_len": 18498,
|
486 |
+
"eval_totals_1": 18498,
|
487 |
+
"eval_totals_2": 16294,
|
488 |
+
"eval_totals_3": 14090,
|
489 |
+
"eval_totals_4": 11886,
|
490 |
+
"step": 945
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 13.99,
|
494 |
+
"learning_rate": 0.0001,
|
495 |
+
"loss": 0.7012,
|
496 |
+
"step": 1018
|
497 |
+
},
|
498 |
+
{
|
499 |
+
"epoch": 13.99,
|
500 |
+
"eval_bleu": 19.7341,
|
501 |
+
"eval_bp": 0.8639,
|
502 |
+
"eval_counts_1": 9856,
|
503 |
+
"eval_counts_2": 4364,
|
504 |
+
"eval_counts_3": 2375,
|
505 |
+
"eval_counts_4": 1360,
|
506 |
+
"eval_exact_match": 0.0476,
|
507 |
+
"eval_f1": 0.4514,
|
508 |
+
"eval_gen_len": 13.976,
|
509 |
+
"eval_loss": 1.1739834547042847,
|
510 |
+
"eval_precisions_1": 53.1693,
|
511 |
+
"eval_precisions_2": 26.7189,
|
512 |
+
"eval_precisions_3": 16.8094,
|
513 |
+
"eval_precisions_4": 11.4046,
|
514 |
+
"eval_ref_len": 21250,
|
515 |
+
"eval_rouge1": 0.4591,
|
516 |
+
"eval_rouge2": 0.2653,
|
517 |
+
"eval_rougeL": 0.443,
|
518 |
+
"eval_rougeLsum": 0.4428,
|
519 |
+
"eval_runtime": 2149.1056,
|
520 |
+
"eval_samples_per_second": 1.026,
|
521 |
+
"eval_steps_per_second": 1.026,
|
522 |
+
"eval_sys_len": 18537,
|
523 |
+
"eval_totals_1": 18537,
|
524 |
+
"eval_totals_2": 16333,
|
525 |
+
"eval_totals_3": 14129,
|
526 |
+
"eval_totals_4": 11925,
|
527 |
+
"step": 1018
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 14.99,
|
531 |
+
"learning_rate": 0.0001,
|
532 |
+
"loss": 0.6597,
|
533 |
+
"step": 1091
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 14.99,
|
537 |
+
"eval_bleu": 19.3289,
|
538 |
+
"eval_bp": 0.8602,
|
539 |
+
"eval_counts_1": 9780,
|
540 |
+
"eval_counts_2": 4292,
|
541 |
+
"eval_counts_3": 2336,
|
542 |
+
"eval_counts_4": 1302,
|
543 |
+
"eval_exact_match": 0.0485,
|
544 |
+
"eval_f1": 0.4492,
|
545 |
+
"eval_gen_len": 13.8802,
|
546 |
+
"eval_loss": 1.1987030506134033,
|
547 |
+
"eval_precisions_1": 52.9565,
|
548 |
+
"eval_precisions_2": 26.3896,
|
549 |
+
"eval_precisions_3": 16.6145,
|
550 |
+
"eval_precisions_4": 10.9818,
|
551 |
+
"eval_ref_len": 21250,
|
552 |
+
"eval_rouge1": 0.457,
|
553 |
+
"eval_rouge2": 0.2633,
|
554 |
+
"eval_rougeL": 0.4418,
|
555 |
+
"eval_rougeLsum": 0.4416,
|
556 |
+
"eval_runtime": 2149.2833,
|
557 |
+
"eval_samples_per_second": 1.025,
|
558 |
+
"eval_steps_per_second": 1.025,
|
559 |
+
"eval_sys_len": 18468,
|
560 |
+
"eval_totals_1": 18468,
|
561 |
+
"eval_totals_2": 16264,
|
562 |
+
"eval_totals_3": 14060,
|
563 |
+
"eval_totals_4": 11856,
|
564 |
+
"step": 1091
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 16.0,
|
568 |
+
"learning_rate": 0.0001,
|
569 |
+
"loss": 0.6236,
|
570 |
+
"step": 1164
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 16.0,
|
574 |
+
"eval_bleu": 19.8055,
|
575 |
+
"eval_bp": 0.8734,
|
576 |
+
"eval_counts_1": 9931,
|
577 |
+
"eval_counts_2": 4388,
|
578 |
+
"eval_counts_3": 2390,
|
579 |
+
"eval_counts_4": 1359,
|
580 |
+
"eval_exact_match": 0.0495,
|
581 |
+
"eval_f1": 0.4538,
|
582 |
+
"eval_gen_len": 14.044,
|
583 |
+
"eval_loss": 1.2135030031204224,
|
584 |
+
"eval_precisions_1": 53.0587,
|
585 |
+
"eval_precisions_2": 26.573,
|
586 |
+
"eval_precisions_3": 16.7028,
|
587 |
+
"eval_precisions_4": 11.2268,
|
588 |
+
"eval_ref_len": 21250,
|
589 |
+
"eval_rouge1": 0.4618,
|
590 |
+
"eval_rouge2": 0.2682,
|
591 |
+
"eval_rougeL": 0.4452,
|
592 |
+
"eval_rougeLsum": 0.445,
|
593 |
+
"eval_runtime": 2168.1341,
|
594 |
+
"eval_samples_per_second": 1.017,
|
595 |
+
"eval_steps_per_second": 1.017,
|
596 |
+
"eval_sys_len": 18717,
|
597 |
+
"eval_totals_1": 18717,
|
598 |
+
"eval_totals_2": 16513,
|
599 |
+
"eval_totals_3": 14309,
|
600 |
+
"eval_totals_4": 12105,
|
601 |
+
"step": 1164
|
602 |
+
},
|
603 |
+
{
|
604 |
+
"epoch": 17.0,
|
605 |
+
"learning_rate": 0.0001,
|
606 |
+
"loss": 0.5933,
|
607 |
+
"step": 1237
|
608 |
+
},
|
609 |
+
{
|
610 |
+
"epoch": 17.0,
|
611 |
+
"eval_bleu": 19.5893,
|
612 |
+
"eval_bp": 0.8654,
|
613 |
+
"eval_counts_1": 9806,
|
614 |
+
"eval_counts_2": 4316,
|
615 |
+
"eval_counts_3": 2366,
|
616 |
+
"eval_counts_4": 1348,
|
617 |
+
"eval_exact_match": 0.049,
|
618 |
+
"eval_f1": 0.4485,
|
619 |
+
"eval_gen_len": 14.0622,
|
620 |
+
"eval_loss": 1.2305341958999634,
|
621 |
+
"eval_precisions_1": 52.817,
|
622 |
+
"eval_precisions_2": 26.3782,
|
623 |
+
"eval_precisions_3": 16.7114,
|
624 |
+
"eval_precisions_4": 11.2766,
|
625 |
+
"eval_ref_len": 21250,
|
626 |
+
"eval_rouge1": 0.4571,
|
627 |
+
"eval_rouge2": 0.2628,
|
628 |
+
"eval_rougeL": 0.4407,
|
629 |
+
"eval_rougeLsum": 0.4409,
|
630 |
+
"eval_runtime": 2171.7325,
|
631 |
+
"eval_samples_per_second": 1.015,
|
632 |
+
"eval_steps_per_second": 1.015,
|
633 |
+
"eval_sys_len": 18566,
|
634 |
+
"eval_totals_1": 18566,
|
635 |
+
"eval_totals_2": 16362,
|
636 |
+
"eval_totals_3": 14158,
|
637 |
+
"eval_totals_4": 11954,
|
638 |
+
"step": 1237
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 17.99,
|
642 |
+
"learning_rate": 0.0001,
|
643 |
+
"loss": 0.5622,
|
644 |
+
"step": 1309
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 17.99,
|
648 |
+
"eval_bleu": 19.4914,
|
649 |
+
"eval_bp": 0.865,
|
650 |
+
"eval_counts_1": 9787,
|
651 |
+
"eval_counts_2": 4306,
|
652 |
+
"eval_counts_3": 2346,
|
653 |
+
"eval_counts_4": 1338,
|
654 |
+
"eval_exact_match": 0.0476,
|
655 |
+
"eval_f1": 0.447,
|
656 |
+
"eval_gen_len": 13.7763,
|
657 |
+
"eval_loss": 1.2796473503112793,
|
658 |
+
"eval_precisions_1": 52.7345,
|
659 |
+
"eval_precisions_2": 26.3283,
|
660 |
+
"eval_precisions_3": 16.5783,
|
661 |
+
"eval_precisions_4": 11.1995,
|
662 |
+
"eval_ref_len": 21250,
|
663 |
+
"eval_rouge1": 0.4549,
|
664 |
+
"eval_rouge2": 0.2609,
|
665 |
+
"eval_rougeL": 0.4383,
|
666 |
+
"eval_rougeLsum": 0.4382,
|
667 |
+
"eval_runtime": 2158.5699,
|
668 |
+
"eval_samples_per_second": 1.021,
|
669 |
+
"eval_steps_per_second": 1.021,
|
670 |
+
"eval_sys_len": 18559,
|
671 |
+
"eval_totals_1": 18559,
|
672 |
+
"eval_totals_2": 16355,
|
673 |
+
"eval_totals_3": 14151,
|
674 |
+
"eval_totals_4": 11947,
|
675 |
+
"step": 1309
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 18.99,
|
679 |
+
"learning_rate": 0.0001,
|
680 |
+
"loss": 0.5275,
|
681 |
+
"step": 1382
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 18.99,
|
685 |
+
"eval_bleu": 19.6947,
|
686 |
+
"eval_bp": 0.8857,
|
687 |
+
"eval_counts_1": 9918,
|
688 |
+
"eval_counts_2": 4363,
|
689 |
+
"eval_counts_3": 2374,
|
690 |
+
"eval_counts_4": 1355,
|
691 |
+
"eval_exact_match": 0.0508,
|
692 |
+
"eval_f1": 0.4499,
|
693 |
+
"eval_gen_len": 14.1647,
|
694 |
+
"eval_loss": 1.2833356857299805,
|
695 |
+
"eval_precisions_1": 52.3377,
|
696 |
+
"eval_precisions_2": 26.054,
|
697 |
+
"eval_precisions_3": 16.3251,
|
698 |
+
"eval_precisions_4": 10.9823,
|
699 |
+
"eval_ref_len": 21250,
|
700 |
+
"eval_rouge1": 0.4573,
|
701 |
+
"eval_rouge2": 0.2624,
|
702 |
+
"eval_rougeL": 0.441,
|
703 |
+
"eval_rougeLsum": 0.4408,
|
704 |
+
"eval_runtime": 2190.1704,
|
705 |
+
"eval_samples_per_second": 1.006,
|
706 |
+
"eval_steps_per_second": 1.006,
|
707 |
+
"eval_sys_len": 18950,
|
708 |
+
"eval_totals_1": 18950,
|
709 |
+
"eval_totals_2": 16746,
|
710 |
+
"eval_totals_3": 14542,
|
711 |
+
"eval_totals_4": 12338,
|
712 |
+
"step": 1382
|
713 |
+
},
|
714 |
+
{
|
715 |
+
"epoch": 19.79,
|
716 |
+
"learning_rate": 0.0001,
|
717 |
+
"loss": 0.4986,
|
718 |
+
"step": 1440
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 19.79,
|
722 |
+
"eval_bleu": 19.4544,
|
723 |
+
"eval_bp": 0.8847,
|
724 |
+
"eval_counts_1": 9879,
|
725 |
+
"eval_counts_2": 4315,
|
726 |
+
"eval_counts_3": 2347,
|
727 |
+
"eval_counts_4": 1324,
|
728 |
+
"eval_exact_match": 0.0495,
|
729 |
+
"eval_f1": 0.4478,
|
730 |
+
"eval_gen_len": 14.2827,
|
731 |
+
"eval_loss": 1.3059108257293701,
|
732 |
+
"eval_precisions_1": 52.1842,
|
733 |
+
"eval_precisions_2": 25.7966,
|
734 |
+
"eval_precisions_3": 16.1606,
|
735 |
+
"eval_precisions_4": 10.7476,
|
736 |
+
"eval_ref_len": 21250,
|
737 |
+
"eval_rouge1": 0.4564,
|
738 |
+
"eval_rouge2": 0.2622,
|
739 |
+
"eval_rougeL": 0.4407,
|
740 |
+
"eval_rougeLsum": 0.4403,
|
741 |
+
"eval_runtime": 3646.8693,
|
742 |
+
"eval_samples_per_second": 0.604,
|
743 |
+
"eval_steps_per_second": 0.604,
|
744 |
+
"eval_sys_len": 18931,
|
745 |
+
"eval_totals_1": 18931,
|
746 |
+
"eval_totals_2": 16727,
|
747 |
+
"eval_totals_3": 14523,
|
748 |
+
"eval_totals_4": 12319,
|
749 |
+
"step": 1440
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 19.79,
|
753 |
+
"step": 1440,
|
754 |
+
"total_flos": 1.102412878184448e+18,
|
755 |
+
"train_loss": 1.0667428798145717,
|
756 |
+
"train_runtime": 140813.6912,
|
757 |
+
"train_samples_per_second": 1.323,
|
758 |
+
"train_steps_per_second": 0.01
|
759 |
+
}
|
760 |
+
],
|
761 |
+
"logging_steps": 500,
|
762 |
+
"max_steps": 1440,
|
763 |
+
"num_train_epochs": 20,
|
764 |
+
"save_steps": 500,
|
765 |
+
"total_flos": 1.102412878184448e+18,
|
766 |
+
"trial_name": null,
|
767 |
+
"trial_params": null
|
768 |
+
}
|
training_args.bin
ADDED
Binary file (4.66 kB). View file
|
|