Hugging Face's logo

Nighter
/
Fine-tuned-longformer-base

Question Answering
Transformers
TensorBoard
Safetensors
longformer
Inference Endpoints
Model card Files Metrics Community
Nighter commited on
Commit
fec849f
•
1 Parent(s): cad3588

Upload 26 files

Browse files
Files changed (26) hide show
  1. args.json +1 -0
  2. models/checkpoint-1000/config.json +43 -0
  3. models/checkpoint-1000/model.safetensors +3 -0
  4. models/checkpoint-1000/optimizer.pt +3 -0
  5. models/checkpoint-1000/rng_state.pth +3 -0
  6. models/checkpoint-1000/scheduler.pt +3 -0
  7. models/checkpoint-1000/trainer_state.json +139 -0
  8. models/checkpoint-1000/training_args.bin +3 -0
  9. models/checkpoint-500/config.json +43 -0
  10. models/checkpoint-500/model.safetensors +3 -0
  11. models/checkpoint-500/optimizer.pt +3 -0
  12. models/checkpoint-500/rng_state.pth +3 -0
  13. models/checkpoint-500/scheduler.pt +3 -0
  14. models/checkpoint-500/trainer_state.json +79 -0
  15. models/checkpoint-500/training_args.bin +3 -0
  16. models/config.json +43 -0
  17. models/eval_results.txt +5 -0
  18. models/merges.txt +0 -0
  19. models/model.safetensors +3 -0
  20. models/runs/Dec02_17-07-54_47f49ef5dd0b/events.out.tfevents.1701536890.47f49ef5dd0b.102.0 +3 -0
  21. models/runs/Dec02_17-07-54_47f49ef5dd0b/events.out.tfevents.1701544466.47f49ef5dd0b.102.1 +3 -0
  22. models/special_tokens_map.json +15 -0
  23. models/tokenizer.json +0 -0
  24. models/tokenizer_config.json +57 -0
  25. models/training_args.bin +3 -0
  26. models/vocab.json +0 -0
args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"model_name_or_path": "allenai/longformer-base-4096", "max_len": 4096, "output_dir": "./models", "overwrite_output_dir": true, "per_gpu_train_batch_size": 1, "per_gpu_eval_batch_size": 1, "gradient_accumulation_steps": 2, "learning_rate": 0.0001, "num_train_epochs": 3, "do_train": true, "do_eval": true, "max_steps": 1000, "logging_steps": 50, "eval_steps": 50}
models/checkpoint-1000/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "architectures": [
4
+ "LongformerForQuestionAnswering"
5
+ ],
6
+ "attention_mode": "longformer",
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "attention_window": [
9
+ 512,
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512
21
+ ],
22
+ "bos_token_id": 0,
23
+ "eos_token_id": 2,
24
+ "gradient_checkpointing": false,
25
+ "hidden_act": "gelu",
26
+ "hidden_dropout_prob": 0.1,
27
+ "hidden_size": 768,
28
+ "ignore_attention_mask": false,
29
+ "initializer_range": 0.02,
30
+ "intermediate_size": 3072,
31
+ "layer_norm_eps": 1e-05,
32
+ "max_position_embeddings": 4098,
33
+ "model_type": "longformer",
34
+ "num_attention_heads": 12,
35
+ "num_hidden_layers": 12,
36
+ "onnx_export": false,
37
+ "pad_token_id": 1,
38
+ "sep_token_id": 2,
39
+ "torch_dtype": "float32",
40
+ "transformers_version": "4.35.0",
41
+ "type_vocab_size": 1,
42
+ "vocab_size": 50265
43
+ }
models/checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72b079e4512fc1ab489f928dd8507b9dffed6b66476cb6e0fee37fa23e031a33
3
+ size 592315616
models/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f54ec241c82997be03e2aaa9809317f72aafb5ad6482d58692c69452a27e433
3
+ size 1184789445
models/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f53b1d2a50f10a43ca81f48c5a81ecc9c58446114c9d8793e1fcacb686f82f64
3
+ size 14575
models/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72613c3db44bc51082dbd8daac1b7493525ab4d2e8ea3698a81b418a19bd4aba
3
+ size 627
models/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.024622051509331756,
5
+ "eval_steps": 50,
6
+ "global_step": 1000,
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": 9.5e-05,
14
+ "loss": 5.161,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 9e-05,
20
+ "loss": 5.2417,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 8.5e-05,
26
+ "loss": 5.2132,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.0,
31
+ "learning_rate": 8e-05,
32
+ "loss": 5.1672,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 7.500000000000001e-05,
38
+ "loss": 5.0757,
39
+ "step": 250
40
+ },
41
+ {
42
+ "epoch": 0.01,
43
+ "learning_rate": 7e-05,
44
+ "loss": 5.0972,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "learning_rate": 6.500000000000001e-05,
50
+ "loss": 5.2191,
51
+ "step": 350
52
+ },
53
+ {
54
+ "epoch": 0.01,
55
+ "learning_rate": 6e-05,
56
+ "loss": 5.1446,
57
+ "step": 400
58
+ },
59
+ {
60
+ "epoch": 0.01,
61
+ "learning_rate": 5.500000000000001e-05,
62
+ "loss": 5.1846,
63
+ "step": 450
64
+ },
65
+ {
66
+ "epoch": 0.01,
67
+ "learning_rate": 5e-05,
68
+ "loss": 5.1827,
69
+ "step": 500
70
+ },
71
+ {
72
+ "epoch": 0.01,
73
+ "learning_rate": 4.5e-05,
74
+ "loss": 4.9039,
75
+ "step": 550
76
+ },
77
+ {
78
+ "epoch": 0.01,
79
+ "learning_rate": 4e-05,
80
+ "loss": 5.0134,
81
+ "step": 600
82
+ },
83
+ {
84
+ "epoch": 0.02,
85
+ "learning_rate": 3.5e-05,
86
+ "loss": 5.1791,
87
+ "step": 650
88
+ },
89
+ {
90
+ "epoch": 0.02,
91
+ "learning_rate": 3e-05,
92
+ "loss": 5.0575,
93
+ "step": 700
94
+ },
95
+ {
96
+ "epoch": 0.02,
97
+ "learning_rate": 2.5e-05,
98
+ "loss": 5.0015,
99
+ "step": 750
100
+ },
101
+ {
102
+ "epoch": 0.02,
103
+ "learning_rate": 2e-05,
104
+ "loss": 5.0853,
105
+ "step": 800
106
+ },
107
+ {
108
+ "epoch": 0.02,
109
+ "learning_rate": 1.5e-05,
110
+ "loss": 5.0391,
111
+ "step": 850
112
+ },
113
+ {
114
+ "epoch": 0.02,
115
+ "learning_rate": 1e-05,
116
+ "loss": 5.1997,
117
+ "step": 900
118
+ },
119
+ {
120
+ "epoch": 0.02,
121
+ "learning_rate": 5e-06,
122
+ "loss": 5.0387,
123
+ "step": 950
124
+ },
125
+ {
126
+ "epoch": 0.02,
127
+ "learning_rate": 0.0,
128
+ "loss": 4.8919,
129
+ "step": 1000
130
+ }
131
+ ],
132
+ "logging_steps": 50,
133
+ "max_steps": 1000,
134
+ "num_train_epochs": 1,
135
+ "save_steps": 500,
136
+ "total_flos": 5225784115200000.0,
137
+ "trial_name": null,
138
+ "trial_params": null
139
+ }
models/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:858178eccbe5c575b20f211c1cdbf7566d777add66e5a8bb6a44678d1773e11c
3
+ size 4091
models/checkpoint-500/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "architectures": [
4
+ "LongformerForQuestionAnswering"
5
+ ],
6
+ "attention_mode": "longformer",
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "attention_window": [
9
+ 512,
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512
21
+ ],
22
+ "bos_token_id": 0,
23
+ "eos_token_id": 2,
24
+ "gradient_checkpointing": false,
25
+ "hidden_act": "gelu",
26
+ "hidden_dropout_prob": 0.1,
27
+ "hidden_size": 768,
28
+ "ignore_attention_mask": false,
29
+ "initializer_range": 0.02,
30
+ "intermediate_size": 3072,
31
+ "layer_norm_eps": 1e-05,
32
+ "max_position_embeddings": 4098,
33
+ "model_type": "longformer",
34
+ "num_attention_heads": 12,
35
+ "num_hidden_layers": 12,
36
+ "onnx_export": false,
37
+ "pad_token_id": 1,
38
+ "sep_token_id": 2,
39
+ "torch_dtype": "float32",
40
+ "transformers_version": "4.35.0",
41
+ "type_vocab_size": 1,
42
+ "vocab_size": 50265
43
+ }
models/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2460df31587bb48597110a1567445f0cacc27f2ee6e77a6d91f17c50177981bc
3
+ size 592315616
models/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f756aa84451538e2dce168e9a224c8155e48329cfb136a15b2049f907a7e3587
3
+ size 1184789445
models/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cfd58e02f4808773b2c3954bfc7fbba88d4d8075a08c8ec124654f3457993a9
3
+ size 14575
models/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1772beaff8e1a54e1110d9989a781f461d3e253f0c02cade590c2e900e7d022
3
+ size 627
models/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.012311025754665878,
5
+ "eval_steps": 50,
6
+ "global_step": 500,
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": 9.5e-05,
14
+ "loss": 5.161,
15
+ "step": 50
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 9e-05,
20
+ "loss": 5.2417,
21
+ "step": 100
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 8.5e-05,
26
+ "loss": 5.2132,
27
+ "step": 150
28
+ },
29
+ {
30
+ "epoch": 0.0,
31
+ "learning_rate": 8e-05,
32
+ "loss": 5.1672,
33
+ "step": 200
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 7.500000000000001e-05,
38
+ "loss": 5.0757,
39
+ "step": 250
40
+ },
41
+ {
42
+ "epoch": 0.01,
43
+ "learning_rate": 7e-05,
44
+ "loss": 5.0972,
45
+ "step": 300
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "learning_rate": 6.500000000000001e-05,
50
+ "loss": 5.2191,
51
+ "step": 350
52
+ },
53
+ {
54
+ "epoch": 0.01,
55
+ "learning_rate": 6e-05,
56
+ "loss": 5.1446,
57
+ "step": 400
58
+ },
59
+ {
60
+ "epoch": 0.01,
61
+ "learning_rate": 5.500000000000001e-05,
62
+ "loss": 5.1846,
63
+ "step": 450
64
+ },
65
+ {
66
+ "epoch": 0.01,
67
+ "learning_rate": 5e-05,
68
+ "loss": 5.1827,
69
+ "step": 500
70
+ }
71
+ ],
72
+ "logging_steps": 50,
73
+ "max_steps": 1000,
74
+ "num_train_epochs": 1,
75
+ "save_steps": 500,
76
+ "total_flos": 2612892057600000.0,
77
+ "trial_name": null,
78
+ "trial_params": null
79
+ }
models/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:858178eccbe5c575b20f211c1cdbf7566d777add66e5a8bb6a44678d1773e11c
3
+ size 4091
models/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/longformer-base-4096",
3
+ "architectures": [
4
+ "LongformerForQuestionAnswering"
5
+ ],
6
+ "attention_mode": "longformer",
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "attention_window": [
9
+ 512,
10
+ 512,
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512,
19
+ 512,
20
+ 512
21
+ ],
22
+ "bos_token_id": 0,
23
+ "eos_token_id": 2,
24
+ "gradient_checkpointing": false,
25
+ "hidden_act": "gelu",
26
+ "hidden_dropout_prob": 0.1,
27
+ "hidden_size": 768,
28
+ "ignore_attention_mask": false,
29
+ "initializer_range": 0.02,
30
+ "intermediate_size": 3072,
31
+ "layer_norm_eps": 1e-05,
32
+ "max_position_embeddings": 4098,
33
+ "model_type": "longformer",
34
+ "num_attention_heads": 12,
35
+ "num_hidden_layers": 12,
36
+ "onnx_export": false,
37
+ "pad_token_id": 1,
38
+ "sep_token_id": 2,
39
+ "torch_dtype": "float32",
40
+ "transformers_version": "4.35.0",
41
+ "type_vocab_size": 1,
42
+ "vocab_size": 50265
43
+ }
models/eval_results.txt ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ epoch = 0.02
2
+ eval_loss = 5.012203216552734
3
+ eval_runtime = 5098.5061
4
+ eval_samples_per_second = 3.983
5
+ eval_steps_per_second = 3.983
models/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
models/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72b079e4512fc1ab489f928dd8507b9dffed6b66476cb6e0fee37fa23e031a33
3
+ size 592315616
models/runs/Dec02_17-07-54_47f49ef5dd0b/events.out.tfevents.1701536890.47f49ef5dd0b.102.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac195f5620133debbc1489b20458444af76e3f1e4cfdef3e819d64559f1907c6
3
+ size 7836
models/runs/Dec02_17-07-54_47f49ef5dd0b/events.out.tfevents.1701544466.47f49ef5dd0b.102.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa9668cadd28a078978c436c550310718abb00a01d9fb3b1b4d335cfbce0bcfd
3
+ size 359
models/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
models/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
models/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 4096,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "LongformerTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
models/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:858178eccbe5c575b20f211c1cdbf7566d777add66e5a8bb6a44678d1773e11c
3
+ size 4091
models/vocab.json ADDED
The diff for this file is too large to render. See raw diff