Marc Tomlinson
commited on
Commit
•
bb2d7e8
1
Parent(s):
70f8e66
pegasus XSUM model fine-tuned with 11500 iterations on gigaword to produce short coherant summaries
Browse files- config.json +57 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer_config.json +1 -0
- trainer_state.json +154 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/pegasus-xsum",
|
3 |
+
"activation_dropout": 0.1,
|
4 |
+
"activation_function": "relu",
|
5 |
+
"add_bias_logits": false,
|
6 |
+
"add_final_layer_norm": true,
|
7 |
+
"architectures": [
|
8 |
+
"PegasusForConditionalGeneration"
|
9 |
+
],
|
10 |
+
"attention_dropout": 0.1,
|
11 |
+
"bos_token_id": 0,
|
12 |
+
"classif_dropout": 0.0,
|
13 |
+
"classifier_dropout": 0.0,
|
14 |
+
"d_model": 1024,
|
15 |
+
"decoder_attention_heads": 16,
|
16 |
+
"decoder_ffn_dim": 4096,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 16,
|
19 |
+
"decoder_start_token_id": 0,
|
20 |
+
"do_blenderbot_90_layernorm": false,
|
21 |
+
"dropout": 0.1,
|
22 |
+
"encoder_attention_heads": 16,
|
23 |
+
"encoder_ffn_dim": 4096,
|
24 |
+
"encoder_layerdrop": 0.0,
|
25 |
+
"encoder_layers": 16,
|
26 |
+
"eos_token_id": 1,
|
27 |
+
"extra_pos_embeddings": 0,
|
28 |
+
"force_bos_token_to_be_generated": false,
|
29 |
+
"forced_eos_token_id": 1,
|
30 |
+
"gradient_checkpointing": false,
|
31 |
+
"id2label": {
|
32 |
+
"0": "LABEL_0",
|
33 |
+
"1": "LABEL_1",
|
34 |
+
"2": "LABEL_2"
|
35 |
+
},
|
36 |
+
"init_std": 0.02,
|
37 |
+
"is_encoder_decoder": true,
|
38 |
+
"label2id": {
|
39 |
+
"LABEL_0": 0,
|
40 |
+
"LABEL_1": 1,
|
41 |
+
"LABEL_2": 2
|
42 |
+
},
|
43 |
+
"length_penalty": 0.6,
|
44 |
+
"max_length": 64,
|
45 |
+
"max_position_embeddings": 512,
|
46 |
+
"model_type": "pegasus",
|
47 |
+
"normalize_before": true,
|
48 |
+
"normalize_embedding": false,
|
49 |
+
"num_beams": 8,
|
50 |
+
"num_hidden_layers": 16,
|
51 |
+
"pad_token_id": 0,
|
52 |
+
"scale_embedding": true,
|
53 |
+
"static_position_embeddings": true,
|
54 |
+
"transformers_version": "4.5.0.dev0",
|
55 |
+
"use_cache": true,
|
56 |
+
"vocab_size": 96103
|
57 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:88135d49b8ac110ef03fd97a28c56186e8f409db9a6916e62a1010cf18cc2976
|
3 |
+
size 4550012928
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f66dd28963cd6410fb63156defa5fe65cd4e02aa3f4f411d5b854a0bea8eff74
|
3 |
+
size 2279640522
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f72400869c8937c8330eb8067dbc7fd5c38bc80e224dc9d40bc460162381213e
|
3 |
+
size 623
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask_2>", "additional_special_tokens": ["<mask_1>", "<unk_2>", "<unk_3>", "<unk_4>", "<unk_5>", "<unk_6>", "<unk_7>", "<unk_8>", "<unk_9>", "<unk_10>", "<unk_11>", "<unk_12>", "<unk_13>", "<unk_14>", "<unk_15>", "<unk_16>", "<unk_17>", "<unk_18>", "<unk_19>", "<unk_20>", "<unk_21>", "<unk_22>", "<unk_23>", "<unk_24>", "<unk_25>", "<unk_26>", "<unk_27>", "<unk_28>", "<unk_29>", "<unk_30>", "<unk_31>", "<unk_32>", "<unk_33>", "<unk_34>", "<unk_35>", "<unk_36>", "<unk_37>", "<unk_38>", "<unk_39>", "<unk_40>", "<unk_41>", "<unk_42>", "<unk_43>", "<unk_44>", "<unk_45>", "<unk_46>", "<unk_47>", "<unk_48>", "<unk_49>", "<unk_50>", "<unk_51>", "<unk_52>", "<unk_53>", "<unk_54>", "<unk_55>", "<unk_56>", "<unk_57>", "<unk_58>", "<unk_59>", "<unk_60>", "<unk_61>", "<unk_62>", "<unk_63>", "<unk_64>", "<unk_65>", "<unk_66>", "<unk_67>", "<unk_68>", "<unk_69>", "<unk_70>", "<unk_71>", "<unk_72>", "<unk_73>", "<unk_74>", "<unk_75>", "<unk_76>", "<unk_77>", "<unk_78>", "<unk_79>", "<unk_80>", "<unk_81>", "<unk_82>", "<unk_83>", "<unk_84>", "<unk_85>", "<unk_86>", "<unk_87>", "<unk_88>", "<unk_89>", "<unk_90>", "<unk_91>", "<unk_92>", "<unk_93>", "<unk_94>", "<unk_95>", "<unk_96>", "<unk_97>", "<unk_98>", "<unk_99>", "<unk_100>", "<unk_101>", "<unk_102>"]}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0015189ef36359283fec8b93cf6d9ce51bca37eb1101defc68a53b394913b96c
|
3 |
+
size 1912529
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"pad_token": "<pad>", "eos_token": "</s>", "unk_token": "<unk>", "mask_token": "<mask_2>", "mask_token_sent": "<mask_1>", "additional_special_tokens": ["<mask_1>", "<unk_2>", "<unk_3>", "<unk_4>", "<unk_5>", "<unk_6>", "<unk_7>", "<unk_8>", "<unk_9>", "<unk_10>", "<unk_11>", "<unk_12>", "<unk_13>", "<unk_14>", "<unk_15>", "<unk_16>", "<unk_17>", "<unk_18>", "<unk_19>", "<unk_20>", "<unk_21>", "<unk_22>", "<unk_23>", "<unk_24>", "<unk_25>", "<unk_26>", "<unk_27>", "<unk_28>", "<unk_29>", "<unk_30>", "<unk_31>", "<unk_32>", "<unk_33>", "<unk_34>", "<unk_35>", "<unk_36>", "<unk_37>", "<unk_38>", "<unk_39>", "<unk_40>", "<unk_41>", "<unk_42>", "<unk_43>", "<unk_44>", "<unk_45>", "<unk_46>", "<unk_47>", "<unk_48>", "<unk_49>", "<unk_50>", "<unk_51>", "<unk_52>", "<unk_53>", "<unk_54>", "<unk_55>", "<unk_56>", "<unk_57>", "<unk_58>", "<unk_59>", "<unk_60>", "<unk_61>", "<unk_62>", "<unk_63>", "<unk_64>", "<unk_65>", "<unk_66>", "<unk_67>", "<unk_68>", "<unk_69>", "<unk_70>", "<unk_71>", "<unk_72>", "<unk_73>", "<unk_74>", "<unk_75>", "<unk_76>", "<unk_77>", "<unk_78>", "<unk_79>", "<unk_80>", "<unk_81>", "<unk_82>", "<unk_83>", "<unk_84>", "<unk_85>", "<unk_86>", "<unk_87>", "<unk_88>", "<unk_89>", "<unk_90>", "<unk_91>", "<unk_92>", "<unk_93>", "<unk_94>", "<unk_95>", "<unk_96>", "<unk_97>", "<unk_98>", "<unk_99>", "<unk_100>", "<unk_101>", "<unk_102>"], "model_max_length": 512, "special_tokens_map_file": null, "full_tokenizer_file": null, "name_or_path": "google/pegasus-xsum"}
|
trainer_state.json
ADDED
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.012092661331875204,
|
5 |
+
"global_step": 11500,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.0,
|
12 |
+
"learning_rate": 4.9991237201933425e-05,
|
13 |
+
"loss": 3.5502,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.0,
|
18 |
+
"learning_rate": 4.9982474403866854e-05,
|
19 |
+
"loss": 3.0742,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.0,
|
24 |
+
"learning_rate": 4.997371160580027e-05,
|
25 |
+
"loss": 2.9389,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.0,
|
30 |
+
"learning_rate": 4.99649488077337e-05,
|
31 |
+
"loss": 2.8079,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.0,
|
36 |
+
"learning_rate": 4.995618600966712e-05,
|
37 |
+
"loss": 2.7687,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.0,
|
42 |
+
"learning_rate": 4.9947423211600545e-05,
|
43 |
+
"loss": 2.7448,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.0,
|
48 |
+
"learning_rate": 4.993866041353397e-05,
|
49 |
+
"loss": 2.6388,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.0,
|
54 |
+
"learning_rate": 4.992989761546739e-05,
|
55 |
+
"loss": 2.6045,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.0,
|
60 |
+
"learning_rate": 4.992113481740082e-05,
|
61 |
+
"loss": 2.5276,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.01,
|
66 |
+
"learning_rate": 4.991237201933424e-05,
|
67 |
+
"loss": 2.5322,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.01,
|
72 |
+
"learning_rate": 4.9903609221267665e-05,
|
73 |
+
"loss": 2.4583,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.01,
|
78 |
+
"learning_rate": 4.989484642320109e-05,
|
79 |
+
"loss": 2.5075,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.01,
|
84 |
+
"learning_rate": 4.988608362513451e-05,
|
85 |
+
"loss": 2.4178,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.01,
|
90 |
+
"learning_rate": 4.987732082706794e-05,
|
91 |
+
"loss": 2.3881,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.01,
|
96 |
+
"learning_rate": 4.986855802900136e-05,
|
97 |
+
"loss": 2.3897,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.01,
|
102 |
+
"learning_rate": 4.985979523093478e-05,
|
103 |
+
"loss": 2.3356,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.01,
|
108 |
+
"learning_rate": 4.985103243286821e-05,
|
109 |
+
"loss": 2.3406,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.01,
|
114 |
+
"learning_rate": 4.984226963480163e-05,
|
115 |
+
"loss": 2.3413,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.01,
|
120 |
+
"learning_rate": 4.983350683673505e-05,
|
121 |
+
"loss": 2.3287,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.01,
|
126 |
+
"learning_rate": 4.9824744038668475e-05,
|
127 |
+
"loss": 2.3343,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.01,
|
132 |
+
"learning_rate": 4.98159812406019e-05,
|
133 |
+
"loss": 2.3305,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.01,
|
138 |
+
"learning_rate": 4.980721844253533e-05,
|
139 |
+
"loss": 2.3109,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.01,
|
144 |
+
"learning_rate": 4.979845564446875e-05,
|
145 |
+
"loss": 2.3059,
|
146 |
+
"step": 11500
|
147 |
+
}
|
148 |
+
],
|
149 |
+
"max_steps": 2852970,
|
150 |
+
"num_train_epochs": 3,
|
151 |
+
"total_flos": 1.082983380369408e+16,
|
152 |
+
"trial_name": null,
|
153 |
+
"trial_params": null
|
154 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d3b7caca996f4f96a3374083fc58a9d7cd924e57bdb8855c3795d439c63eaee1
|
3 |
+
size 2351
|