Hugging Face's logo

Farmellion
/
antitoxic_model

Safetensors
PyTorch
russian
toxicity
classification
moderation
bert
Model card Files
xet
 Community
Farmellion commited on
Commit
2e539fd
·
verified ·
1 Parent(s): 06b68e4

Upload 4 files

Browse files
Files changed (4) hide show
  1. toxic_model/config.json +115 -0
  2. toxic_model/model.safetensors +3 -0
  3. toxic_model/tokenizer.json +0 -0
  4. toxic_model/tokenizer_config.json +17 -0
toxic_model/config.json ADDED
@@ -0,0 +1,115 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu",
3
+ "allow_embedding_resizing": true,
4
+ "architectures": [
5
+ "ModernBertForSequenceClassification"
6
+ ],
7
+ "attention_bias": false,
8
+ "attention_dropout": 0.0,
9
+ "attention_layer": "rope",
10
+ "attention_probs_dropout_prob": 0.0,
11
+ "attn_out_bias": false,
12
+ "attn_out_dropout_prob": 0.1,
13
+ "attn_qkv_bias": false,
14
+ "bert_layer": "prenorm",
15
+ "bos_token_id": null,
16
+ "classifier_activation": "gelu",
17
+ "classifier_bias": false,
18
+ "classifier_dropout": 0.0,
19
+ "classifier_pooling": "cls",
20
+ "cls_token_id": 50281,
21
+ "compile_model": true,
22
+ "decoder_bias": true,
23
+ "deterministic_flash_attn": false,
24
+ "dtype": "float32",
25
+ "embed_dropout_prob": 0.0,
26
+ "embed_norm": true,
27
+ "embedding_dropout": 0.0,
28
+ "embedding_layer": "sans_pos",
29
+ "eos_token_id": null,
30
+ "final_norm": true,
31
+ "global_attn_every_n_layers": 3,
32
+ "head_pred_act": "gelu",
33
+ "hidden_act": "gelu",
34
+ "hidden_activation": "gelu",
35
+ "hidden_size": 768,
36
+ "init_method": "full_megatron",
37
+ "initializer_cutoff_factor": 2.0,
38
+ "initializer_range": 0.02,
39
+ "intermediate_size": 1152,
40
+ "layer_types": [
41
+ "full_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "full_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "full_attention",
48
+ "sliding_attention",
49
+ "sliding_attention",
50
+ "full_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "full_attention",
54
+ "sliding_attention",
55
+ "sliding_attention",
56
+ "full_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "full_attention",
60
+ "sliding_attention",
61
+ "sliding_attention",
62
+ "full_attention"
63
+ ],
64
+ "local_attention": 256,
65
+ "local_attn_rotary_emb_base": 10000.0,
66
+ "loss_function": "fa_cross_entropy",
67
+ "loss_kwargs": {
68
+ "reduction": "mean"
69
+ },
70
+ "masked_prediction": true,
71
+ "max_position_embeddings": 8192,
72
+ "mlp_bias": false,
73
+ "mlp_dropout": 0.0,
74
+ "mlp_dropout_prob": 0.0,
75
+ "mlp_in_bias": false,
76
+ "mlp_layer": "glu",
77
+ "mlp_out_bias": false,
78
+ "model_type": "modernbert",
79
+ "norm_bias": false,
80
+ "norm_eps": 1e-05,
81
+ "norm_kwargs": {
82
+ "bias": false,
83
+ "eps": 1e-05
84
+ },
85
+ "normalization": "layernorm",
86
+ "num_attention_heads": 12,
87
+ "num_hidden_layers": 22,
88
+ "pad_token_id": 50283,
89
+ "padding": "unpadded",
90
+ "problem_type": "single_label_classification",
91
+ "repad_logits_with_grad": false,
92
+ "rope_parameters": {
93
+ "full_attention": {
94
+ "rope_theta": 160000.0,
95
+ "rope_type": "default"
96
+ },
97
+ "sliding_attention": {
98
+ "rope_theta": 10000.0,
99
+ "rope_type": "default"
100
+ }
101
+ },
102
+ "rotary_emb_base": 160000.0,
103
+ "rotary_emb_dim": null,
104
+ "rotary_emb_interleaved": false,
105
+ "rotary_emb_scale_base": null,
106
+ "sep_token_id": 50282,
107
+ "skip_first_prenorm": true,
108
+ "sparse_pred_ignore_index": -100,
109
+ "sparse_prediction": false,
110
+ "tie_word_embeddings": true,
111
+ "transformers_version": "5.12.1",
112
+ "unpad_embeddings": true,
113
+ "use_cache": false,
114
+ "vocab_size": 50368
115
+ }
toxic_model/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc4f882f2870719a4344e757b3d8d1175e7aa1634a96730dfcab6d0c5783bac2
3
+ size 598439784
toxic_model/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
toxic_model/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "[CLS]",
5
+ "is_local": false,
6
+ "local_files_only": false,
7
+ "mask_token": "[MASK]",
8
+ "model_input_names": [
9
+ "input_ids",
10
+ "attention_mask"
11
+ ],
12
+ "model_max_length": 1000000000000000019884624838656,
13
+ "pad_token": "[PAD]",
14
+ "sep_token": "[SEP]",
15
+ "tokenizer_class": "TokenizersBackend",
16
+ "unk_token": "[UNK]"
17
+ }