Hugging Face's logo

Superar
/
Portuguese-Pun-Location

Token Classification
Safetensors
Portuguese
humor
puns
pun-location
Model card Files Community
cabralpinto commited on
Commit
1ccf6e0
·
1 Parent(s): bd5119e

Add albertina models

Browse files
Files changed (40) hide show
  1. albertina-900m-ptbr-all/checkpoint-500/config.json +46 -0
  2. albertina-900m-ptbr-all/checkpoint-500/model.safetensors +3 -0
  3. albertina-900m-ptbr-all/checkpoint-500/optimizer.pt +3 -0
  4. albertina-900m-ptbr-all/checkpoint-500/rng_state.pth +3 -0
  5. albertina-900m-ptbr-all/checkpoint-500/scheduler.pt +3 -0
  6. albertina-900m-ptbr-all/checkpoint-500/special_tokens_map.json +51 -0
  7. albertina-900m-ptbr-all/checkpoint-500/tokenizer.json +0 -0
  8. albertina-900m-ptbr-all/checkpoint-500/tokenizer_config.json +58 -0
  9. albertina-900m-ptbr-all/checkpoint-500/trainer_state.json +68 -0
  10. albertina-900m-ptbr-all/checkpoint-500/training_args.bin +3 -0
  11. albertina-900m-ptbr-positive/checkpoint-500/config.json +46 -0
  12. albertina-900m-ptbr-positive/checkpoint-500/model.safetensors +3 -0
  13. albertina-900m-ptbr-positive/checkpoint-500/optimizer.pt +3 -0
  14. albertina-900m-ptbr-positive/checkpoint-500/rng_state.pth +3 -0
  15. albertina-900m-ptbr-positive/checkpoint-500/scheduler.pt +3 -0
  16. albertina-900m-ptbr-positive/checkpoint-500/special_tokens_map.json +51 -0
  17. albertina-900m-ptbr-positive/checkpoint-500/tokenizer.json +0 -0
  18. albertina-900m-ptbr-positive/checkpoint-500/tokenizer_config.json +58 -0
  19. albertina-900m-ptbr-positive/checkpoint-500/trainer_state.json +68 -0
  20. albertina-900m-ptbr-positive/checkpoint-500/training_args.bin +3 -0
  21. albertina-900m-ptpt-all/checkpoint-500/config.json +46 -0
  22. albertina-900m-ptpt-all/checkpoint-500/model.safetensors +3 -0
  23. albertina-900m-ptpt-all/checkpoint-500/optimizer.pt +3 -0
  24. albertina-900m-ptpt-all/checkpoint-500/rng_state.pth +3 -0
  25. albertina-900m-ptpt-all/checkpoint-500/scheduler.pt +3 -0
  26. albertina-900m-ptpt-all/checkpoint-500/special_tokens_map.json +51 -0
  27. albertina-900m-ptpt-all/checkpoint-500/tokenizer.json +0 -0
  28. albertina-900m-ptpt-all/checkpoint-500/tokenizer_config.json +58 -0
  29. albertina-900m-ptpt-all/checkpoint-500/trainer_state.json +68 -0
  30. albertina-900m-ptpt-all/checkpoint-500/training_args.bin +3 -0
  31. albertina-900m-ptpt-positive/checkpoint-500/config.json +46 -0
  32. albertina-900m-ptpt-positive/checkpoint-500/model.safetensors +3 -0
  33. albertina-900m-ptpt-positive/checkpoint-500/optimizer.pt +3 -0
  34. albertina-900m-ptpt-positive/checkpoint-500/rng_state.pth +3 -0
  35. albertina-900m-ptpt-positive/checkpoint-500/scheduler.pt +3 -0
  36. albertina-900m-ptpt-positive/checkpoint-500/special_tokens_map.json +51 -0
  37. albertina-900m-ptpt-positive/checkpoint-500/tokenizer.json +0 -0
  38. albertina-900m-ptpt-positive/checkpoint-500/tokenizer_config.json +58 -0
  39. albertina-900m-ptpt-positive/checkpoint-500/trainer_state.json +68 -0
  40. albertina-900m-ptpt-positive/checkpoint-500/training_args.bin +3 -0
albertina-900m-ptbr-all/checkpoint-500/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PORTULAN/albertina-900m-portuguese-ptbr-encoder",
3
+ "architectures": [
4
+ "DebertaV2ForTokenClassification"
5
+ ],
6
+ "attention_head_size": 64,
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "conv_act": "gelu",
9
+ "conv_kernel_size": 3,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1536,
13
+ "id2label": {
14
+ "0": "Non-Pun",
15
+ "1": "Pun"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6144,
19
+ "label2id": {
20
+ "Non-Pun": 0,
21
+ "Pun": 1
22
+ },
23
+ "layer_norm_eps": 1e-07,
24
+ "max_position_embeddings": 512,
25
+ "max_relative_positions": -1,
26
+ "model_type": "deberta-v2",
27
+ "norm_rel_ebd": "layer_norm",
28
+ "num_attention_heads": 24,
29
+ "num_hidden_layers": 24,
30
+ "pad_token_id": 0,
31
+ "pooler_dropout": 0,
32
+ "pooler_hidden_act": "gelu",
33
+ "pooler_hidden_size": 1536,
34
+ "pos_att_type": [
35
+ "p2c",
36
+ "c2p"
37
+ ],
38
+ "position_biased_input": false,
39
+ "position_buckets": 256,
40
+ "relative_attention": true,
41
+ "share_att_key": true,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.39.3",
44
+ "type_vocab_size": 0,
45
+ "vocab_size": 128100
46
+ }
albertina-900m-ptbr-all/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad7ccbb21b27cf55e5d804ae7088887815dc7b9b471ea90af5915a05220a0b1d
3
+ size 3538434600
albertina-900m-ptbr-all/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b80d9a672ec1afa93494b36f8b09f1689117851a89754bcb91805a34c315837
3
+ size 7077110642
albertina-900m-ptbr-all/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1350614c01f5c56c92023d1aaacede6096fa3c76e469d71c9dbc0fa0cb424858
3
+ size 14244
albertina-900m-ptbr-all/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c01bd80bde58ba6c77db8a36ebac42b242ded79a9c8a353421be48f0baca0c41
3
+ size 1064
albertina-900m-ptbr-all/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
albertina-900m-ptbr-all/checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
albertina-900m-ptbr-all/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 512,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
albertina-900m-ptbr-all/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.13585524260997772,
3
+ "best_model_checkpoint": "results/pun_location/models/complete_corpus/albertina-900m-portuguese-ptbr-encoder/checkpoint-500",
4
+ "epoch": 2.004008016032064,
5
+ "eval_steps": 500,
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": 0.4,
13
+ "grad_norm": 2.011240005493164,
14
+ "learning_rate": 6e-06,
15
+ "loss": 0.2399,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.8,
20
+ "grad_norm": 1.6278150081634521,
21
+ "learning_rate": 1.2e-05,
22
+ "loss": 0.1176,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 1.2,
27
+ "grad_norm": 1.499725580215454,
28
+ "learning_rate": 1.8e-05,
29
+ "loss": 0.1079,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 1.6,
34
+ "grad_norm": 2.030484914779663,
35
+ "learning_rate": 2.4e-05,
36
+ "loss": 0.1016,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 2.0,
41
+ "grad_norm": 9.072067260742188,
42
+ "learning_rate": 3e-05,
43
+ "loss": 0.1064,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 2.0,
48
+ "eval_accuracy": 0.9520718921617574,
49
+ "eval_f1": 0.5126093669583119,
50
+ "eval_loss": 0.13585524260997772,
51
+ "eval_precision": 0.6043689320388349,
52
+ "eval_recall": 0.4450402144772118,
53
+ "eval_runtime": 3.556,
54
+ "eval_samples_per_second": 160.293,
55
+ "eval_steps_per_second": 20.248,
56
+ "step": 500
57
+ }
58
+ ],
59
+ "logging_steps": 100,
60
+ "max_steps": 996,
61
+ "num_input_tokens_seen": 0,
62
+ "num_train_epochs": 4,
63
+ "save_steps": 500,
64
+ "total_flos": 1166954105701776.0,
65
+ "train_batch_size": 8,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
albertina-900m-ptbr-all/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:819676478c8afd48f364a349a34be7761c1cc87b70e3693685552426119aad9d
3
+ size 4920
albertina-900m-ptbr-positive/checkpoint-500/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PORTULAN/albertina-900m-portuguese-ptbr-encoder",
3
+ "architectures": [
4
+ "DebertaV2ForTokenClassification"
5
+ ],
6
+ "attention_head_size": 64,
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "conv_act": "gelu",
9
+ "conv_kernel_size": 3,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1536,
13
+ "id2label": {
14
+ "0": "Non-Pun",
15
+ "1": "Pun"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6144,
19
+ "label2id": {
20
+ "Non-Pun": 0,
21
+ "Pun": 1
22
+ },
23
+ "layer_norm_eps": 1e-07,
24
+ "max_position_embeddings": 512,
25
+ "max_relative_positions": -1,
26
+ "model_type": "deberta-v2",
27
+ "norm_rel_ebd": "layer_norm",
28
+ "num_attention_heads": 24,
29
+ "num_hidden_layers": 24,
30
+ "pad_token_id": 0,
31
+ "pooler_dropout": 0,
32
+ "pooler_hidden_act": "gelu",
33
+ "pooler_hidden_size": 1536,
34
+ "pos_att_type": [
35
+ "p2c",
36
+ "c2p"
37
+ ],
38
+ "position_biased_input": false,
39
+ "position_buckets": 256,
40
+ "relative_attention": true,
41
+ "share_att_key": true,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.39.3",
44
+ "type_vocab_size": 0,
45
+ "vocab_size": 128100
46
+ }
albertina-900m-ptbr-positive/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f29a994f82cbd13b2ea1c671c4efeec1b81938594aaa297fb37d81b5b6caa6f
3
+ size 3538434600
albertina-900m-ptbr-positive/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33593fd898fd80fbaee7577cdfa76574e21a7b467c9048c9fd58405f7613cf84
3
+ size 7077110642
albertina-900m-ptbr-positive/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:559df40ef40bba20c8dacb7fd7439deac63b4c73d8ef3c10816b499c75f21521
3
+ size 14244
albertina-900m-ptbr-positive/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:804313fcf5eb7df1c2808319014a74abe62db8d2c2b80e85464884a6da5c013e
3
+ size 1064
albertina-900m-ptbr-positive/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
albertina-900m-ptbr-positive/checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
albertina-900m-ptbr-positive/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 512,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
albertina-900m-ptbr-positive/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.1587166041135788,
3
+ "best_model_checkpoint": "results/pun_location/models/only_positives/albertina-900m-portuguese-ptbr-encoder/checkpoint-500",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
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": 0.8,
13
+ "grad_norm": 1.9746909141540527,
14
+ "learning_rate": 6e-06,
15
+ "loss": 0.2685,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 1.6,
20
+ "grad_norm": 1.9112613201141357,
21
+ "learning_rate": 1.2e-05,
22
+ "loss": 0.1055,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 2.4,
27
+ "grad_norm": 3.209425210952759,
28
+ "learning_rate": 1.8e-05,
29
+ "loss": 0.0886,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 3.2,
34
+ "grad_norm": 2.3601949214935303,
35
+ "learning_rate": 2.4e-05,
36
+ "loss": 0.069,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 4.0,
41
+ "grad_norm": 3.6319262981414795,
42
+ "learning_rate": 0.0,
43
+ "loss": 0.0644,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 4.0,
48
+ "eval_accuracy": 0.9476284584980237,
49
+ "eval_f1": 0.646477132262052,
50
+ "eval_loss": 0.1587166041135788,
51
+ "eval_precision": 0.6670918367346939,
52
+ "eval_recall": 0.6270983213429256,
53
+ "eval_runtime": 1.801,
54
+ "eval_samples_per_second": 158.248,
55
+ "eval_steps_per_second": 19.989,
56
+ "step": 500
57
+ }
58
+ ],
59
+ "logging_steps": 100,
60
+ "max_steps": 500,
61
+ "num_input_tokens_seen": 0,
62
+ "num_train_epochs": 4,
63
+ "save_steps": 500,
64
+ "total_flos": 1176925923189492.0,
65
+ "train_batch_size": 8,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
albertina-900m-ptbr-positive/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f80de0f90916b9e88dc962c4ed64abee36884bd0b3dbe561df439cb8cbc1c4a8
3
+ size 4920
albertina-900m-ptpt-all/checkpoint-500/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PORTULAN/albertina-900m-portuguese-ptpt-encoder",
3
+ "architectures": [
4
+ "DebertaV2ForTokenClassification"
5
+ ],
6
+ "attention_head_size": 64,
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "conv_act": "gelu",
9
+ "conv_kernel_size": 3,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1536,
13
+ "id2label": {
14
+ "0": "Non-Pun",
15
+ "1": "Pun"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6144,
19
+ "label2id": {
20
+ "Non-Pun": 0,
21
+ "Pun": 1
22
+ },
23
+ "layer_norm_eps": 1e-07,
24
+ "max_position_embeddings": 512,
25
+ "max_relative_positions": -1,
26
+ "model_type": "deberta-v2",
27
+ "norm_rel_ebd": "layer_norm",
28
+ "num_attention_heads": 24,
29
+ "num_hidden_layers": 24,
30
+ "pad_token_id": 0,
31
+ "pooler_dropout": 0,
32
+ "pooler_hidden_act": "gelu",
33
+ "pooler_hidden_size": 1536,
34
+ "pos_att_type": [
35
+ "p2c",
36
+ "c2p"
37
+ ],
38
+ "position_biased_input": false,
39
+ "position_buckets": 256,
40
+ "relative_attention": true,
41
+ "share_att_key": true,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.39.3",
44
+ "type_vocab_size": 0,
45
+ "vocab_size": 128100
46
+ }
albertina-900m-ptpt-all/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bfe722001bd459916f6806b87fe973d668fe7ef2d7e93c469424c4a9d9a35f5
3
+ size 3538434600
albertina-900m-ptpt-all/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:302cb860dd2e31f25709aaaf4c4222410268787eedfc969d9168f443e55cdd6e
3
+ size 7077110642
albertina-900m-ptpt-all/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1350614c01f5c56c92023d1aaacede6096fa3c76e469d71c9dbc0fa0cb424858
3
+ size 14244
albertina-900m-ptpt-all/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c01bd80bde58ba6c77db8a36ebac42b242ded79a9c8a353421be48f0baca0c41
3
+ size 1064
albertina-900m-ptpt-all/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
albertina-900m-ptpt-all/checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
albertina-900m-ptpt-all/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 512,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
albertina-900m-ptpt-all/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.13921774923801422,
3
+ "best_model_checkpoint": "results/pun_location/models/complete_corpus/albertina-900m-portuguese-ptpt-encoder/checkpoint-500",
4
+ "epoch": 2.004008016032064,
5
+ "eval_steps": 500,
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": 0.4,
13
+ "grad_norm": 1.5726655721664429,
14
+ "learning_rate": 6e-06,
15
+ "loss": 0.3033,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 0.8,
20
+ "grad_norm": 1.3350284099578857,
21
+ "learning_rate": 1.2e-05,
22
+ "loss": 0.1165,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 1.2,
27
+ "grad_norm": 1.1118038892745972,
28
+ "learning_rate": 1.8e-05,
29
+ "loss": 0.1063,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 1.6,
34
+ "grad_norm": 2.201014518737793,
35
+ "learning_rate": 2.4e-05,
36
+ "loss": 0.101,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 2.0,
41
+ "grad_norm": 2.5950186252593994,
42
+ "learning_rate": 3e-05,
43
+ "loss": 0.1024,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 2.0,
48
+ "eval_accuracy": 0.9519470793809286,
49
+ "eval_f1": 0.5058286872782566,
50
+ "eval_loss": 0.13921774923801422,
51
+ "eval_precision": 0.5843091334894613,
52
+ "eval_recall": 0.4459338695263628,
53
+ "eval_runtime": 3.553,
54
+ "eval_samples_per_second": 160.429,
55
+ "eval_steps_per_second": 20.265,
56
+ "step": 500
57
+ }
58
+ ],
59
+ "logging_steps": 100,
60
+ "max_steps": 996,
61
+ "num_input_tokens_seen": 0,
62
+ "num_train_epochs": 4,
63
+ "save_steps": 500,
64
+ "total_flos": 1166954105701776.0,
65
+ "train_batch_size": 8,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
albertina-900m-ptpt-all/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:256471e658c9b818cd5e1cc02f1162d8d1793b6c9cfcdd72a97162ea2de8fc93
3
+ size 4920
albertina-900m-ptpt-positive/checkpoint-500/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PORTULAN/albertina-900m-portuguese-ptpt-encoder",
3
+ "architectures": [
4
+ "DebertaV2ForTokenClassification"
5
+ ],
6
+ "attention_head_size": 64,
7
+ "attention_probs_dropout_prob": 0.1,
8
+ "conv_act": "gelu",
9
+ "conv_kernel_size": 3,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 1536,
13
+ "id2label": {
14
+ "0": "Non-Pun",
15
+ "1": "Pun"
16
+ },
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6144,
19
+ "label2id": {
20
+ "Non-Pun": 0,
21
+ "Pun": 1
22
+ },
23
+ "layer_norm_eps": 1e-07,
24
+ "max_position_embeddings": 512,
25
+ "max_relative_positions": -1,
26
+ "model_type": "deberta-v2",
27
+ "norm_rel_ebd": "layer_norm",
28
+ "num_attention_heads": 24,
29
+ "num_hidden_layers": 24,
30
+ "pad_token_id": 0,
31
+ "pooler_dropout": 0,
32
+ "pooler_hidden_act": "gelu",
33
+ "pooler_hidden_size": 1536,
34
+ "pos_att_type": [
35
+ "p2c",
36
+ "c2p"
37
+ ],
38
+ "position_biased_input": false,
39
+ "position_buckets": 256,
40
+ "relative_attention": true,
41
+ "share_att_key": true,
42
+ "torch_dtype": "float32",
43
+ "transformers_version": "4.39.3",
44
+ "type_vocab_size": 0,
45
+ "vocab_size": 128100
46
+ }
albertina-900m-ptpt-positive/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:114ec0d05d3fbd2890b36b60dfa6c66d1ab640542e2032ec3873b9f585de43e0
3
+ size 3538434600
albertina-900m-ptpt-positive/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14fde5adb568a044cc70f8156f89ecb56b923da2230b1df23f72ab36af8c470c
3
+ size 7077110642
albertina-900m-ptpt-positive/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:559df40ef40bba20c8dacb7fd7439deac63b4c73d8ef3c10816b499c75f21521
3
+ size 14244
albertina-900m-ptpt-positive/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:804313fcf5eb7df1c2808319014a74abe62db8d2c2b80e85464884a6da5c013e
3
+ size 1064
albertina-900m-ptpt-positive/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "[CLS]",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "[CLS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "[SEP]",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "[MASK]",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "[PAD]",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "[SEP]",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "[UNK]",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
albertina-900m-ptpt-positive/checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
albertina-900m-ptpt-positive/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 512,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
albertina-900m-ptpt-positive/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.12566307187080383,
3
+ "best_model_checkpoint": "results/pun_location/models/only_positives/albertina-900m-portuguese-ptpt-encoder/checkpoint-500",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
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": 0.8,
13
+ "grad_norm": 2.195570945739746,
14
+ "learning_rate": 6e-06,
15
+ "loss": 0.2681,
16
+ "step": 100
17
+ },
18
+ {
19
+ "epoch": 1.6,
20
+ "grad_norm": 2.5788235664367676,
21
+ "learning_rate": 1.2e-05,
22
+ "loss": 0.1021,
23
+ "step": 200
24
+ },
25
+ {
26
+ "epoch": 2.4,
27
+ "grad_norm": 1.872805118560791,
28
+ "learning_rate": 1.8e-05,
29
+ "loss": 0.0994,
30
+ "step": 300
31
+ },
32
+ {
33
+ "epoch": 3.2,
34
+ "grad_norm": 1.4116291999816895,
35
+ "learning_rate": 2.4e-05,
36
+ "loss": 0.0662,
37
+ "step": 400
38
+ },
39
+ {
40
+ "epoch": 4.0,
41
+ "grad_norm": 1.0769490003585815,
42
+ "learning_rate": 0.0,
43
+ "loss": 0.0574,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 4.0,
48
+ "eval_accuracy": 0.9572628458498024,
49
+ "eval_f1": 0.7253396337861784,
50
+ "eval_loss": 0.12566307187080383,
51
+ "eval_precision": 0.7147846332945286,
52
+ "eval_recall": 0.7362110311750599,
53
+ "eval_runtime": 1.8014,
54
+ "eval_samples_per_second": 158.21,
55
+ "eval_steps_per_second": 19.984,
56
+ "step": 500
57
+ }
58
+ ],
59
+ "logging_steps": 100,
60
+ "max_steps": 500,
61
+ "num_input_tokens_seen": 0,
62
+ "num_train_epochs": 4,
63
+ "save_steps": 500,
64
+ "total_flos": 1176925923189492.0,
65
+ "train_batch_size": 8,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
albertina-900m-ptpt-positive/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa567bf4eb7ca3ad9a6e96e74ce4250dd6f1b01cc479a2ca6affd4d8793e5121
3
+ size 4920