farleyknight
commited on
Commit
•
c17656f
1
Parent(s):
0434124
Training in progress, step 500
Browse files- .gitignore +1 -0
- config.json +43 -0
- pytorch_model.bin +3 -0
- runs/Sep16_02-33-04_nb91huwb97/1663295604.626464/events.out.tfevents.1663295604.nb91huwb97.3954.1 +3 -0
- runs/Sep16_02-33-04_nb91huwb97/events.out.tfevents.1663295604.nb91huwb97.3954.0 +3 -0
- runs/Sep16_02-34-32_nb91huwb97/1663295687.0001626/events.out.tfevents.1663295687.nb91huwb97.4156.1 +3 -0
- runs/Sep16_02-34-32_nb91huwb97/events.out.tfevents.1663295686.nb91huwb97.4156.0 +3 -0
- runs/Sep16_02-35-20_nb91huwb97/1663295736.5014257/events.out.tfevents.1663295736.nb91huwb97.4323.1 +3 -0
- runs/Sep16_02-35-20_nb91huwb97/events.out.tfevents.1663295736.nb91huwb97.4323.0 +3 -0
- runs/Sep16_02-36-13_nb91huwb97/1663295788.0597906/events.out.tfevents.1663295788.nb91huwb97.4504.1 +3 -0
- runs/Sep16_02-36-13_nb91huwb97/events.out.tfevents.1663295788.nb91huwb97.4504.0 +3 -0
- runs/Sep16_02-37-16_nb91huwb97/1663295850.0690334/events.out.tfevents.1663295850.nb91huwb97.4791.1 +3 -0
- runs/Sep16_02-37-16_nb91huwb97/events.out.tfevents.1663295850.nb91huwb97.4791.0 +3 -0
- runs/Sep16_02-38-56_nb91huwb97/1663295953.249989/events.out.tfevents.1663295953.nb91huwb97.5011.1 +3 -0
- runs/Sep16_02-38-56_nb91huwb97/events.out.tfevents.1663295953.nb91huwb97.5011.0 +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- training_args.bin +3 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/bigbird-pegasus-large-arxiv",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu_new",
|
5 |
+
"architectures": [
|
6 |
+
"BigBirdPegasusForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"attention_type": "block_sparse",
|
10 |
+
"block_size": 64,
|
11 |
+
"bos_token_id": 2,
|
12 |
+
"classifier_dropout": 0.0,
|
13 |
+
"d_model": 1024,
|
14 |
+
"decoder_attention_heads": 16,
|
15 |
+
"decoder_ffn_dim": 4096,
|
16 |
+
"decoder_layerdrop": 0.0,
|
17 |
+
"decoder_layers": 16,
|
18 |
+
"decoder_start_token_id": 2,
|
19 |
+
"dropout": 0.1,
|
20 |
+
"encoder_attention_heads": 16,
|
21 |
+
"encoder_ffn_dim": 4096,
|
22 |
+
"encoder_layerdrop": 0.0,
|
23 |
+
"encoder_layers": 16,
|
24 |
+
"eos_token_id": 1,
|
25 |
+
"gradient_checkpointing": false,
|
26 |
+
"init_std": 0.02,
|
27 |
+
"is_encoder_decoder": true,
|
28 |
+
"length_penalty": 0.8,
|
29 |
+
"max_length": 256,
|
30 |
+
"max_position_embeddings": 4096,
|
31 |
+
"model_type": "bigbird_pegasus",
|
32 |
+
"num_beams": 5,
|
33 |
+
"num_hidden_layers": 16,
|
34 |
+
"num_random_blocks": 3,
|
35 |
+
"pad_token_id": 0,
|
36 |
+
"scale_embedding": true,
|
37 |
+
"tokenizer_class": "PegasusTokenizer",
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.23.0.dev0",
|
40 |
+
"use_bias": false,
|
41 |
+
"use_cache": true,
|
42 |
+
"vocab_size": 96103
|
43 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:507f0872443f62d349c8ca67f6d09fbecd6209d11d981eafc6e5d30022d0a8ce
|
3 |
+
size 2308115505
|
runs/Sep16_02-33-04_nb91huwb97/1663295604.626464/events.out.tfevents.1663295604.nb91huwb97.3954.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ed3da4f899fc8c7b5d49caee2e4132b8d20173d65e184523176ef7f1b1e96de6
|
3 |
+
size 5892
|
runs/Sep16_02-33-04_nb91huwb97/events.out.tfevents.1663295604.nb91huwb97.3954.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7aa00b2399ba324e646a86f164973950a6a6bd252e1e11302d5300c197e79aba
|
3 |
+
size 4452
|
runs/Sep16_02-34-32_nb91huwb97/1663295687.0001626/events.out.tfevents.1663295687.nb91huwb97.4156.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ec1e4050b4f3540f347e7629ebf5825f30eaef597a396c019c27fd67267c9242
|
3 |
+
size 5892
|
runs/Sep16_02-34-32_nb91huwb97/events.out.tfevents.1663295686.nb91huwb97.4156.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c86348f21374da3f880e86d0bcde0d797ac48ef1ee7fa57ba0a3b9d54d4ca1e6
|
3 |
+
size 4452
|
runs/Sep16_02-35-20_nb91huwb97/1663295736.5014257/events.out.tfevents.1663295736.nb91huwb97.4323.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9a16e3843d54d99421570caaff5b6d3056a70199276c3cc05b20c8c74b996cb7
|
3 |
+
size 5892
|
runs/Sep16_02-35-20_nb91huwb97/events.out.tfevents.1663295736.nb91huwb97.4323.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6d40c35ea682944ab2215063de78a3bca1b05de84f52d57052758d0e784c745b
|
3 |
+
size 4452
|
runs/Sep16_02-36-13_nb91huwb97/1663295788.0597906/events.out.tfevents.1663295788.nb91huwb97.4504.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:42a495869cd5bac2bb895d53392ad7cc7b1e0d7e3a40a487e60f4636fa8efbb9
|
3 |
+
size 5892
|
runs/Sep16_02-36-13_nb91huwb97/events.out.tfevents.1663295788.nb91huwb97.4504.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aa7643217824e17d5b3aa3bb6c4d817943a2c38c10f8aa88edd0487954db8248
|
3 |
+
size 4452
|
runs/Sep16_02-37-16_nb91huwb97/1663295850.0690334/events.out.tfevents.1663295850.nb91huwb97.4791.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6d16245b20beaf2598efef8a439502a05cef38377fa4ddedc629243cd0b25bdd
|
3 |
+
size 5892
|
runs/Sep16_02-37-16_nb91huwb97/events.out.tfevents.1663295850.nb91huwb97.4791.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3e569e208d3269e894913d3713a69d1017c3a7134de06e4273a2d06e696c0590
|
3 |
+
size 4452
|
runs/Sep16_02-38-56_nb91huwb97/1663295953.249989/events.out.tfevents.1663295953.nb91huwb97.5011.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d16477e535fcb27655feae25889dbf2b36384c22731a6787ae70f8c9235c16ed
|
3 |
+
size 5892
|
runs/Sep16_02-38-56_nb91huwb97/events.out.tfevents.1663295953.nb91huwb97.5011.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5613658e7604c256e1e39e2b69f3a4272d52578c35750fce1fae98aad1a6df06
|
3 |
+
size 4609
|
special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": true,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "[CLS]",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": true,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": true,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "[MASK]",
|
25 |
+
"lstrip": true,
|
26 |
+
"normalized": true,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": true,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "[SEP]",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": true,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": true,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [],
|
3 |
+
"bos_token": {
|
4 |
+
"__type": "AddedToken",
|
5 |
+
"content": "<s>",
|
6 |
+
"lstrip": false,
|
7 |
+
"normalized": true,
|
8 |
+
"rstrip": false,
|
9 |
+
"single_word": false
|
10 |
+
},
|
11 |
+
"cls_token": {
|
12 |
+
"__type": "AddedToken",
|
13 |
+
"content": "[CLS]",
|
14 |
+
"lstrip": false,
|
15 |
+
"normalized": true,
|
16 |
+
"rstrip": false,
|
17 |
+
"single_word": false
|
18 |
+
},
|
19 |
+
"eos_token": {
|
20 |
+
"__type": "AddedToken",
|
21 |
+
"content": "</s>",
|
22 |
+
"lstrip": false,
|
23 |
+
"normalized": true,
|
24 |
+
"rstrip": false,
|
25 |
+
"single_word": false
|
26 |
+
},
|
27 |
+
"mask_token": "[MASK]",
|
28 |
+
"mask_token_sent": null,
|
29 |
+
"model_max_length": 4096,
|
30 |
+
"name_or_path": "google/bigbird-pegasus-large-arxiv",
|
31 |
+
"offset": 0,
|
32 |
+
"pad_token": {
|
33 |
+
"__type": "AddedToken",
|
34 |
+
"content": "<pad>",
|
35 |
+
"lstrip": false,
|
36 |
+
"normalized": true,
|
37 |
+
"rstrip": false,
|
38 |
+
"single_word": false
|
39 |
+
},
|
40 |
+
"sep_token": {
|
41 |
+
"__type": "AddedToken",
|
42 |
+
"content": "[SEP]",
|
43 |
+
"lstrip": false,
|
44 |
+
"normalized": true,
|
45 |
+
"rstrip": false,
|
46 |
+
"single_word": false
|
47 |
+
},
|
48 |
+
"special_tokens_map_file": "/home/patrick/.cache/huggingface/transformers/b548e984b09823ed5cea0e622c0ec194a7e07b3c5d2e6b48a7cde84bd179cc7c.4eda581f816a0a941629106f0338c957910ce4839ecf7e3e743bb79523bf7249",
|
49 |
+
"tokenizer_class": "PegasusTokenizer",
|
50 |
+
"unk_token": {
|
51 |
+
"__type": "AddedToken",
|
52 |
+
"content": "<unk>",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": true,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false
|
57 |
+
}
|
58 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0f79116f37b010a4d62d40d71a1b582ca19da5d9c4c1240378ab66f13509e802
|
3 |
+
size 3631
|