gsarti commited on
Commit
8ab6b9a
1 Parent(s): c3f1846

Initial commit

Browse files
README.md CHANGED
@@ -1,3 +1,83 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: mit
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - it5/datasets
7
+ metrics:
8
+ - rouge
9
+ model-index:
10
+ - name: it5-efficient-small-el32-qa-0.0003
11
+ results:
12
+ - task:
13
+ name: Summarization
14
+ type: summarization
15
+ dataset:
16
+ name: it5/datasets qa
17
+ type: it5/datasets
18
+ args: qa
19
+ metrics:
20
+ - name: Rouge1
21
+ type: rouge
22
+ value: 74.2234
23
  ---
24
+
25
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
26
+ should probably proofread and complete it, then remove this comment. -->
27
+
28
+ # it5-efficient-small-el32-qa-0.0003
29
+
30
+ This model is a fine-tuned version of [stefan-it/it5-efficient-small-el32](https://huggingface.co/stefan-it/it5-efficient-small-el32) on the it5/datasets qa dataset.
31
+ It achieves the following results on the evaluation set:
32
+ - Loss: 0.8225
33
+ - Rouge1: 74.2234
34
+ - Rouge2: 40.5909
35
+ - Rougel: 74.1327
36
+ - Rougelsum: 74.2081
37
+ - Gen Len: 4.7055
38
+
39
+ ## Model description
40
+
41
+ More information needed
42
+
43
+ ## Intended uses & limitations
44
+
45
+ More information needed
46
+
47
+ ## Training and evaluation data
48
+
49
+ More information needed
50
+
51
+ ## Training procedure
52
+
53
+ ### Training hyperparameters
54
+
55
+ The following hyperparameters were used during training:
56
+ - learning_rate: 0.0003
57
+ - train_batch_size: 8
58
+ - eval_batch_size: 8
59
+ - seed: 42
60
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
61
+ - lr_scheduler_type: linear
62
+ - num_epochs: 7.0
63
+
64
+ ### Training results
65
+
66
+ | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
67
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
68
+ | 1.1164 | 0.8 | 5000 | 0.8244 | 66.4678 | 35.3554 | 66.4543 | 66.4522 | 4.541 |
69
+ | 0.9097 | 1.59 | 10000 | 0.7299 | 70.0574 | 37.5535 | 69.9512 | 70.0084 | 4.5548 |
70
+ | 0.6637 | 2.39 | 15000 | 0.7314 | 72.0767 | 39.2263 | 72.0257 | 72.0473 | 4.703 |
71
+ | 0.5015 | 3.19 | 20000 | 0.7147 | 73.0185 | 39.9998 | 72.9347 | 72.9576 | 4.75 |
72
+ | 0.5101 | 3.99 | 25000 | 0.7055 | 73.7898 | 40.5481 | 73.7235 | 73.7901 | 4.8728 |
73
+ | 0.3903 | 4.78 | 30000 | 0.7442 | 74.0845 | 39.9841 | 74.0172 | 74.0635 | 4.5938 |
74
+ | 0.2993 | 5.58 | 35000 | 0.8184 | 73.8405 | 40.2569 | 73.7756 | 73.7972 | 4.7412 |
75
+ | 0.2227 | 6.38 | 40000 | 0.8278 | 74.0159 | 40.6403 | 73.9412 | 73.9722 | 4.742 |
76
+
77
+
78
+ ### Framework versions
79
+
80
+ - Transformers 4.15.0
81
+ - Pytorch 1.10.0+cu102
82
+ - Datasets 1.17.0
83
+ - Tokenizers 0.10.3
all_results.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "eval_gen_len": 4.7055,
4
+ "eval_loss": 0.8224949240684509,
5
+ "eval_rouge1": 74.2234,
6
+ "eval_rouge2": 40.5909,
7
+ "eval_rougeL": 74.1327,
8
+ "eval_rougeLsum": 74.2081,
9
+ "eval_runtime": 110.8699,
10
+ "eval_samples": 4000,
11
+ "eval_samples_per_second": 36.078,
12
+ "eval_steps_per_second": 4.51,
13
+ "train_loss": 0.639304427944139,
14
+ "train_runtime": 13849.1505,
15
+ "train_samples": 50159,
16
+ "train_samples_per_second": 25.353,
17
+ "train_steps_per_second": 3.169
18
+ }
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": ".",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 2048,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "relu",
13
+ "initializer_factor": 1.0,
14
+ "is_encoder_decoder": true,
15
+ "layer_norm_epsilon": 1e-06,
16
+ "model_type": "t5",
17
+ "n_positions": 512,
18
+ "num_decoder_layers": 6,
19
+ "num_heads": 8,
20
+ "num_layers": 32,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "relative_attention_max_distance": 128,
24
+ "relative_attention_num_buckets": 32,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.15.0",
27
+ "use_cache": true,
28
+ "vocab_size": 32100
29
+ }
eval_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "eval_gen_len": 4.7055,
4
+ "eval_loss": 0.8224949240684509,
5
+ "eval_rouge1": 74.2234,
6
+ "eval_rouge2": 40.5909,
7
+ "eval_rougeL": 74.1327,
8
+ "eval_rougeLsum": 74.2081,
9
+ "eval_runtime": 110.8699,
10
+ "eval_samples": 4000,
11
+ "eval_samples_per_second": 36.078,
12
+ "eval_steps_per_second": 4.51
13
+ }
flax_model.msgpack ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e915ac57bb558ddb9377c6dd1f2994d71525b430eb0b4f5a25926c3a157bb09
3
+ size 569246164
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55e95dbad8bab6a16f778003f3569f028d7dbd9647dea4866231543eeaebb408
3
+ size 569387035
runs/Apr26_12-34-52_pg-gpu27/1650969352.401718/events.out.tfevents.1650969352.pg-gpu27.29084.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38bf12be4326a32ac3ee73b11225826813a84f6458ad8d178ae3cf70ed6bbe74
3
+ size 5127
runs/Apr26_12-34-52_pg-gpu27/events.out.tfevents.1650969352.pg-gpu27.29084.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3084b5f9113e39251cc00d5a2385ed5c019be51844eb874cfbe707a457f6a9dd
3
+ size 22059
runs/Apr26_12-34-52_pg-gpu27/events.out.tfevents.1650983316.pg-gpu27.29084.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d481f04287119f49c15b2fd1a0021bcdddd851ed69235a8e65aeb6cf84d18cc
3
+ size 575
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dffd01fc009b7e92d98eddff8853983e271b41302ed0d363000e8581df12000
3
+ size 817200
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cbbcf69da0bb6912d11e103dcefa8680c366bb0e6cff3eaa0b5fadabd25ae89
3
+ size 569947488
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "special_tokens_map_file": null, "name_or_path": "stefan-it/it5-efficient-small-el32", "sp_model_kwargs": {}, "tokenizer_class": "T5Tokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 7.0,
3
+ "train_loss": 0.639304427944139,
4
+ "train_runtime": 13849.1505,
5
+ "train_samples": 50159,
6
+ "train_samples_per_second": 25.353,
7
+ "train_steps_per_second": 3.169
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,651 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 7.0,
5
+ "global_step": 43890,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.08,
12
+ "learning_rate": 0.0002965823650034176,
13
+ "loss": 3.7122,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.16,
18
+ "learning_rate": 0.00029316473000683526,
19
+ "loss": 1.7581,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.24,
24
+ "learning_rate": 0.0002897470950102529,
25
+ "loss": 1.5341,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.32,
30
+ "learning_rate": 0.00028632946001367054,
31
+ "loss": 1.4748,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.4,
36
+ "learning_rate": 0.0002829118250170881,
37
+ "loss": 1.3464,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.48,
42
+ "learning_rate": 0.0002794941900205058,
43
+ "loss": 1.2892,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.56,
48
+ "learning_rate": 0.0002760765550239234,
49
+ "loss": 1.2375,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.64,
54
+ "learning_rate": 0.00027265892002734105,
55
+ "loss": 1.2408,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.72,
60
+ "learning_rate": 0.0002692412850307587,
61
+ "loss": 1.1577,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.8,
66
+ "learning_rate": 0.0002658236500341763,
67
+ "loss": 1.1164,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.8,
72
+ "eval_gen_len": 4.541,
73
+ "eval_loss": 0.8244166970252991,
74
+ "eval_rouge1": 66.4678,
75
+ "eval_rouge2": 35.3554,
76
+ "eval_rougeL": 66.4543,
77
+ "eval_rougeLsum": 66.4522,
78
+ "eval_runtime": 109.1952,
79
+ "eval_samples_per_second": 36.632,
80
+ "eval_steps_per_second": 4.579,
81
+ "step": 5000
82
+ },
83
+ {
84
+ "epoch": 0.88,
85
+ "learning_rate": 0.00026240601503759397,
86
+ "loss": 1.1345,
87
+ "step": 5500
88
+ },
89
+ {
90
+ "epoch": 0.96,
91
+ "learning_rate": 0.0002589883800410116,
92
+ "loss": 1.1116,
93
+ "step": 6000
94
+ },
95
+ {
96
+ "epoch": 1.04,
97
+ "learning_rate": 0.00025557074504442925,
98
+ "loss": 1.0336,
99
+ "step": 6500
100
+ },
101
+ {
102
+ "epoch": 1.12,
103
+ "learning_rate": 0.0002521531100478469,
104
+ "loss": 0.9108,
105
+ "step": 7000
106
+ },
107
+ {
108
+ "epoch": 1.2,
109
+ "learning_rate": 0.0002487354750512645,
110
+ "loss": 0.92,
111
+ "step": 7500
112
+ },
113
+ {
114
+ "epoch": 1.28,
115
+ "learning_rate": 0.0002453178400546821,
116
+ "loss": 0.9071,
117
+ "step": 8000
118
+ },
119
+ {
120
+ "epoch": 1.36,
121
+ "learning_rate": 0.00024190020505809978,
122
+ "loss": 0.8936,
123
+ "step": 8500
124
+ },
125
+ {
126
+ "epoch": 1.44,
127
+ "learning_rate": 0.0002384825700615174,
128
+ "loss": 0.888,
129
+ "step": 9000
130
+ },
131
+ {
132
+ "epoch": 1.52,
133
+ "learning_rate": 0.00023506493506493504,
134
+ "loss": 0.8847,
135
+ "step": 9500
136
+ },
137
+ {
138
+ "epoch": 1.59,
139
+ "learning_rate": 0.00023164730006835268,
140
+ "loss": 0.9097,
141
+ "step": 10000
142
+ },
143
+ {
144
+ "epoch": 1.59,
145
+ "eval_gen_len": 4.5548,
146
+ "eval_loss": 0.7299422025680542,
147
+ "eval_rouge1": 70.0574,
148
+ "eval_rouge2": 37.5535,
149
+ "eval_rougeL": 69.9512,
150
+ "eval_rougeLsum": 70.0084,
151
+ "eval_runtime": 106.5141,
152
+ "eval_samples_per_second": 37.554,
153
+ "eval_steps_per_second": 4.694,
154
+ "step": 10000
155
+ },
156
+ {
157
+ "epoch": 1.67,
158
+ "learning_rate": 0.00022822966507177032,
159
+ "loss": 0.8895,
160
+ "step": 10500
161
+ },
162
+ {
163
+ "epoch": 1.75,
164
+ "learning_rate": 0.00022481203007518796,
165
+ "loss": 0.8544,
166
+ "step": 11000
167
+ },
168
+ {
169
+ "epoch": 1.83,
170
+ "learning_rate": 0.0002213943950786056,
171
+ "loss": 0.8807,
172
+ "step": 11500
173
+ },
174
+ {
175
+ "epoch": 1.91,
176
+ "learning_rate": 0.00021797676008202322,
177
+ "loss": 0.8451,
178
+ "step": 12000
179
+ },
180
+ {
181
+ "epoch": 1.99,
182
+ "learning_rate": 0.00021455912508544086,
183
+ "loss": 0.82,
184
+ "step": 12500
185
+ },
186
+ {
187
+ "epoch": 2.07,
188
+ "learning_rate": 0.00021114149008885847,
189
+ "loss": 0.6878,
190
+ "step": 13000
191
+ },
192
+ {
193
+ "epoch": 2.15,
194
+ "learning_rate": 0.00020772385509227614,
195
+ "loss": 0.6759,
196
+ "step": 13500
197
+ },
198
+ {
199
+ "epoch": 2.23,
200
+ "learning_rate": 0.00020430622009569378,
201
+ "loss": 0.6998,
202
+ "step": 14000
203
+ },
204
+ {
205
+ "epoch": 2.31,
206
+ "learning_rate": 0.0002008885850991114,
207
+ "loss": 0.6751,
208
+ "step": 14500
209
+ },
210
+ {
211
+ "epoch": 2.39,
212
+ "learning_rate": 0.00019747095010252903,
213
+ "loss": 0.6637,
214
+ "step": 15000
215
+ },
216
+ {
217
+ "epoch": 2.39,
218
+ "eval_gen_len": 4.703,
219
+ "eval_loss": 0.7314157485961914,
220
+ "eval_rouge1": 72.0767,
221
+ "eval_rouge2": 39.2263,
222
+ "eval_rougeL": 72.0257,
223
+ "eval_rougeLsum": 72.0473,
224
+ "eval_runtime": 110.4087,
225
+ "eval_samples_per_second": 36.229,
226
+ "eval_steps_per_second": 4.529,
227
+ "step": 15000
228
+ },
229
+ {
230
+ "epoch": 2.47,
231
+ "learning_rate": 0.00019405331510594667,
232
+ "loss": 0.6698,
233
+ "step": 15500
234
+ },
235
+ {
236
+ "epoch": 2.55,
237
+ "learning_rate": 0.0001906356801093643,
238
+ "loss": 0.672,
239
+ "step": 16000
240
+ },
241
+ {
242
+ "epoch": 2.63,
243
+ "learning_rate": 0.00018721804511278195,
244
+ "loss": 0.6431,
245
+ "step": 16500
246
+ },
247
+ {
248
+ "epoch": 2.71,
249
+ "learning_rate": 0.00018380041011619957,
250
+ "loss": 0.6653,
251
+ "step": 17000
252
+ },
253
+ {
254
+ "epoch": 2.79,
255
+ "learning_rate": 0.0001803827751196172,
256
+ "loss": 0.6824,
257
+ "step": 17500
258
+ },
259
+ {
260
+ "epoch": 2.87,
261
+ "learning_rate": 0.00017696514012303485,
262
+ "loss": 0.6668,
263
+ "step": 18000
264
+ },
265
+ {
266
+ "epoch": 2.95,
267
+ "learning_rate": 0.00017354750512645246,
268
+ "loss": 0.6318,
269
+ "step": 18500
270
+ },
271
+ {
272
+ "epoch": 3.03,
273
+ "learning_rate": 0.0001701298701298701,
274
+ "loss": 0.5934,
275
+ "step": 19000
276
+ },
277
+ {
278
+ "epoch": 3.11,
279
+ "learning_rate": 0.00016671223513328777,
280
+ "loss": 0.5086,
281
+ "step": 19500
282
+ },
283
+ {
284
+ "epoch": 3.19,
285
+ "learning_rate": 0.00016329460013670539,
286
+ "loss": 0.5015,
287
+ "step": 20000
288
+ },
289
+ {
290
+ "epoch": 3.19,
291
+ "eval_gen_len": 4.75,
292
+ "eval_loss": 0.7147404551506042,
293
+ "eval_rouge1": 73.0185,
294
+ "eval_rouge2": 39.9998,
295
+ "eval_rougeL": 72.9347,
296
+ "eval_rougeLsum": 72.9576,
297
+ "eval_runtime": 106.9579,
298
+ "eval_samples_per_second": 37.398,
299
+ "eval_steps_per_second": 4.675,
300
+ "step": 20000
301
+ },
302
+ {
303
+ "epoch": 3.27,
304
+ "learning_rate": 0.00015987696514012303,
305
+ "loss": 0.4909,
306
+ "step": 20500
307
+ },
308
+ {
309
+ "epoch": 3.35,
310
+ "learning_rate": 0.00015645933014354064,
311
+ "loss": 0.5114,
312
+ "step": 21000
313
+ },
314
+ {
315
+ "epoch": 3.43,
316
+ "learning_rate": 0.00015304169514695828,
317
+ "loss": 0.5314,
318
+ "step": 21500
319
+ },
320
+ {
321
+ "epoch": 3.51,
322
+ "learning_rate": 0.00014962406015037592,
323
+ "loss": 0.5089,
324
+ "step": 22000
325
+ },
326
+ {
327
+ "epoch": 3.59,
328
+ "learning_rate": 0.00014620642515379356,
329
+ "loss": 0.5133,
330
+ "step": 22500
331
+ },
332
+ {
333
+ "epoch": 3.67,
334
+ "learning_rate": 0.0001427887901572112,
335
+ "loss": 0.5057,
336
+ "step": 23000
337
+ },
338
+ {
339
+ "epoch": 3.75,
340
+ "learning_rate": 0.00013937115516062882,
341
+ "loss": 0.5181,
342
+ "step": 23500
343
+ },
344
+ {
345
+ "epoch": 3.83,
346
+ "learning_rate": 0.00013595352016404648,
347
+ "loss": 0.4826,
348
+ "step": 24000
349
+ },
350
+ {
351
+ "epoch": 3.91,
352
+ "learning_rate": 0.0001325358851674641,
353
+ "loss": 0.497,
354
+ "step": 24500
355
+ },
356
+ {
357
+ "epoch": 3.99,
358
+ "learning_rate": 0.00012911825017088174,
359
+ "loss": 0.5101,
360
+ "step": 25000
361
+ },
362
+ {
363
+ "epoch": 3.99,
364
+ "eval_gen_len": 4.8728,
365
+ "eval_loss": 0.7054756283760071,
366
+ "eval_rouge1": 73.7898,
367
+ "eval_rouge2": 40.5481,
368
+ "eval_rougeL": 73.7235,
369
+ "eval_rougeLsum": 73.7901,
370
+ "eval_runtime": 110.0456,
371
+ "eval_samples_per_second": 36.349,
372
+ "eval_steps_per_second": 4.544,
373
+ "step": 25000
374
+ },
375
+ {
376
+ "epoch": 4.07,
377
+ "learning_rate": 0.00012570061517429938,
378
+ "loss": 0.3961,
379
+ "step": 25500
380
+ },
381
+ {
382
+ "epoch": 4.15,
383
+ "learning_rate": 0.00012228298017771702,
384
+ "loss": 0.3725,
385
+ "step": 26000
386
+ },
387
+ {
388
+ "epoch": 4.23,
389
+ "learning_rate": 0.00011886534518113465,
390
+ "loss": 0.3698,
391
+ "step": 26500
392
+ },
393
+ {
394
+ "epoch": 4.31,
395
+ "learning_rate": 0.00011544771018455227,
396
+ "loss": 0.3946,
397
+ "step": 27000
398
+ },
399
+ {
400
+ "epoch": 4.39,
401
+ "learning_rate": 0.00011203007518796991,
402
+ "loss": 0.4009,
403
+ "step": 27500
404
+ },
405
+ {
406
+ "epoch": 4.47,
407
+ "learning_rate": 0.00010861244019138756,
408
+ "loss": 0.391,
409
+ "step": 28000
410
+ },
411
+ {
412
+ "epoch": 4.55,
413
+ "learning_rate": 0.00010519480519480518,
414
+ "loss": 0.3787,
415
+ "step": 28500
416
+ },
417
+ {
418
+ "epoch": 4.63,
419
+ "learning_rate": 0.00010177717019822282,
420
+ "loss": 0.3736,
421
+ "step": 29000
422
+ },
423
+ {
424
+ "epoch": 4.7,
425
+ "learning_rate": 9.835953520164045e-05,
426
+ "loss": 0.3842,
427
+ "step": 29500
428
+ },
429
+ {
430
+ "epoch": 4.78,
431
+ "learning_rate": 9.494190020505809e-05,
432
+ "loss": 0.3903,
433
+ "step": 30000
434
+ },
435
+ {
436
+ "epoch": 4.78,
437
+ "eval_gen_len": 4.5938,
438
+ "eval_loss": 0.7442232370376587,
439
+ "eval_rouge1": 74.0845,
440
+ "eval_rouge2": 39.9841,
441
+ "eval_rougeL": 74.0172,
442
+ "eval_rougeLsum": 74.0635,
443
+ "eval_runtime": 110.762,
444
+ "eval_samples_per_second": 36.113,
445
+ "eval_steps_per_second": 4.514,
446
+ "step": 30000
447
+ },
448
+ {
449
+ "epoch": 4.86,
450
+ "learning_rate": 9.152426520847573e-05,
451
+ "loss": 0.3945,
452
+ "step": 30500
453
+ },
454
+ {
455
+ "epoch": 4.94,
456
+ "learning_rate": 8.810663021189336e-05,
457
+ "loss": 0.4128,
458
+ "step": 31000
459
+ },
460
+ {
461
+ "epoch": 5.02,
462
+ "learning_rate": 8.468899521531099e-05,
463
+ "loss": 0.363,
464
+ "step": 31500
465
+ },
466
+ {
467
+ "epoch": 5.1,
468
+ "learning_rate": 8.127136021872864e-05,
469
+ "loss": 0.3144,
470
+ "step": 32000
471
+ },
472
+ {
473
+ "epoch": 5.18,
474
+ "learning_rate": 7.785372522214627e-05,
475
+ "loss": 0.3106,
476
+ "step": 32500
477
+ },
478
+ {
479
+ "epoch": 5.26,
480
+ "learning_rate": 7.44360902255639e-05,
481
+ "loss": 0.2982,
482
+ "step": 33000
483
+ },
484
+ {
485
+ "epoch": 5.34,
486
+ "learning_rate": 7.101845522898154e-05,
487
+ "loss": 0.3016,
488
+ "step": 33500
489
+ },
490
+ {
491
+ "epoch": 5.42,
492
+ "learning_rate": 6.760082023239918e-05,
493
+ "loss": 0.3095,
494
+ "step": 34000
495
+ },
496
+ {
497
+ "epoch": 5.5,
498
+ "learning_rate": 6.41831852358168e-05,
499
+ "loss": 0.2863,
500
+ "step": 34500
501
+ },
502
+ {
503
+ "epoch": 5.58,
504
+ "learning_rate": 6.076555023923445e-05,
505
+ "loss": 0.2993,
506
+ "step": 35000
507
+ },
508
+ {
509
+ "epoch": 5.58,
510
+ "eval_gen_len": 4.7412,
511
+ "eval_loss": 0.8183711171150208,
512
+ "eval_rouge1": 73.8405,
513
+ "eval_rouge2": 40.2569,
514
+ "eval_rougeL": 73.7756,
515
+ "eval_rougeLsum": 73.7972,
516
+ "eval_runtime": 109.7934,
517
+ "eval_samples_per_second": 36.432,
518
+ "eval_steps_per_second": 4.554,
519
+ "step": 35000
520
+ },
521
+ {
522
+ "epoch": 5.66,
523
+ "learning_rate": 5.734791524265208e-05,
524
+ "loss": 0.2975,
525
+ "step": 35500
526
+ },
527
+ {
528
+ "epoch": 5.74,
529
+ "learning_rate": 5.393028024606972e-05,
530
+ "loss": 0.3014,
531
+ "step": 36000
532
+ },
533
+ {
534
+ "epoch": 5.82,
535
+ "learning_rate": 5.051264524948735e-05,
536
+ "loss": 0.2996,
537
+ "step": 36500
538
+ },
539
+ {
540
+ "epoch": 5.9,
541
+ "learning_rate": 4.7095010252904986e-05,
542
+ "loss": 0.3068,
543
+ "step": 37000
544
+ },
545
+ {
546
+ "epoch": 5.98,
547
+ "learning_rate": 4.367737525632262e-05,
548
+ "loss": 0.2993,
549
+ "step": 37500
550
+ },
551
+ {
552
+ "epoch": 6.06,
553
+ "learning_rate": 4.025974025974026e-05,
554
+ "loss": 0.2447,
555
+ "step": 38000
556
+ },
557
+ {
558
+ "epoch": 6.14,
559
+ "learning_rate": 3.684210526315789e-05,
560
+ "loss": 0.2379,
561
+ "step": 38500
562
+ },
563
+ {
564
+ "epoch": 6.22,
565
+ "learning_rate": 3.342447026657553e-05,
566
+ "loss": 0.2447,
567
+ "step": 39000
568
+ },
569
+ {
570
+ "epoch": 6.3,
571
+ "learning_rate": 3.0006835269993163e-05,
572
+ "loss": 0.2452,
573
+ "step": 39500
574
+ },
575
+ {
576
+ "epoch": 6.38,
577
+ "learning_rate": 2.65892002734108e-05,
578
+ "loss": 0.2227,
579
+ "step": 40000
580
+ },
581
+ {
582
+ "epoch": 6.38,
583
+ "eval_gen_len": 4.742,
584
+ "eval_loss": 0.8277584910392761,
585
+ "eval_rouge1": 74.0159,
586
+ "eval_rouge2": 40.6403,
587
+ "eval_rougeL": 73.9412,
588
+ "eval_rougeLsum": 73.9722,
589
+ "eval_runtime": 108.0867,
590
+ "eval_samples_per_second": 37.007,
591
+ "eval_steps_per_second": 4.626,
592
+ "step": 40000
593
+ },
594
+ {
595
+ "epoch": 6.46,
596
+ "learning_rate": 2.3171565276828434e-05,
597
+ "loss": 0.2331,
598
+ "step": 40500
599
+ },
600
+ {
601
+ "epoch": 6.54,
602
+ "learning_rate": 1.9753930280246068e-05,
603
+ "loss": 0.2374,
604
+ "step": 41000
605
+ },
606
+ {
607
+ "epoch": 6.62,
608
+ "learning_rate": 1.6336295283663705e-05,
609
+ "loss": 0.2462,
610
+ "step": 41500
611
+ },
612
+ {
613
+ "epoch": 6.7,
614
+ "learning_rate": 1.2918660287081339e-05,
615
+ "loss": 0.24,
616
+ "step": 42000
617
+ },
618
+ {
619
+ "epoch": 6.78,
620
+ "learning_rate": 9.501025290498975e-06,
621
+ "loss": 0.2217,
622
+ "step": 42500
623
+ },
624
+ {
625
+ "epoch": 6.86,
626
+ "learning_rate": 6.083390293916609e-06,
627
+ "loss": 0.2283,
628
+ "step": 43000
629
+ },
630
+ {
631
+ "epoch": 6.94,
632
+ "learning_rate": 2.6657552973342446e-06,
633
+ "loss": 0.2345,
634
+ "step": 43500
635
+ },
636
+ {
637
+ "epoch": 7.0,
638
+ "step": 43890,
639
+ "total_flos": 8.113234147780608e+16,
640
+ "train_loss": 0.639304427944139,
641
+ "train_runtime": 13849.1505,
642
+ "train_samples_per_second": 25.353,
643
+ "train_steps_per_second": 3.169
644
+ }
645
+ ],
646
+ "max_steps": 43890,
647
+ "num_train_epochs": 7,
648
+ "total_flos": 8.113234147780608e+16,
649
+ "trial_name": null,
650
+ "trial_params": null
651
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76c2f097975531c8e6b1a18a860ae8bb85d0ad0373722e5b0d61e4b27dbedb65
3
+ size 3183