SKNahin commited on
Commit
dc4343b
1 Parent(s): ef44adc

Training in progress, epoch 1

Browse files
config.json ADDED
@@ -0,0 +1,260 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "uw-madison/yoso-4096",
3
+ "architectures": [
4
+ "YosoForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "conv_window": null,
9
+ "eos_token_id": 2,
10
+ "hash_code_len": 9,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "LABEL_0",
16
+ "1": "LABEL_1",
17
+ "2": "LABEL_2",
18
+ "3": "LABEL_3",
19
+ "4": "LABEL_4",
20
+ "5": "LABEL_5",
21
+ "6": "LABEL_6",
22
+ "7": "LABEL_7",
23
+ "8": "LABEL_8",
24
+ "9": "LABEL_9",
25
+ "10": "LABEL_10",
26
+ "11": "LABEL_11",
27
+ "12": "LABEL_12",
28
+ "13": "LABEL_13",
29
+ "14": "LABEL_14",
30
+ "15": "LABEL_15",
31
+ "16": "LABEL_16",
32
+ "17": "LABEL_17",
33
+ "18": "LABEL_18",
34
+ "19": "LABEL_19",
35
+ "20": "LABEL_20",
36
+ "21": "LABEL_21",
37
+ "22": "LABEL_22",
38
+ "23": "LABEL_23",
39
+ "24": "LABEL_24",
40
+ "25": "LABEL_25",
41
+ "26": "LABEL_26",
42
+ "27": "LABEL_27",
43
+ "28": "LABEL_28",
44
+ "29": "LABEL_29",
45
+ "30": "LABEL_30",
46
+ "31": "LABEL_31",
47
+ "32": "LABEL_32",
48
+ "33": "LABEL_33",
49
+ "34": "LABEL_34",
50
+ "35": "LABEL_35",
51
+ "36": "LABEL_36",
52
+ "37": "LABEL_37",
53
+ "38": "LABEL_38",
54
+ "39": "LABEL_39",
55
+ "40": "LABEL_40",
56
+ "41": "LABEL_41",
57
+ "42": "LABEL_42",
58
+ "43": "LABEL_43",
59
+ "44": "LABEL_44",
60
+ "45": "LABEL_45",
61
+ "46": "LABEL_46",
62
+ "47": "LABEL_47",
63
+ "48": "LABEL_48",
64
+ "49": "LABEL_49",
65
+ "50": "LABEL_50",
66
+ "51": "LABEL_51",
67
+ "52": "LABEL_52",
68
+ "53": "LABEL_53",
69
+ "54": "LABEL_54",
70
+ "55": "LABEL_55",
71
+ "56": "LABEL_56",
72
+ "57": "LABEL_57",
73
+ "58": "LABEL_58",
74
+ "59": "LABEL_59",
75
+ "60": "LABEL_60",
76
+ "61": "LABEL_61",
77
+ "62": "LABEL_62",
78
+ "63": "LABEL_63",
79
+ "64": "LABEL_64",
80
+ "65": "LABEL_65",
81
+ "66": "LABEL_66",
82
+ "67": "LABEL_67",
83
+ "68": "LABEL_68",
84
+ "69": "LABEL_69",
85
+ "70": "LABEL_70",
86
+ "71": "LABEL_71",
87
+ "72": "LABEL_72",
88
+ "73": "LABEL_73",
89
+ "74": "LABEL_74",
90
+ "75": "LABEL_75",
91
+ "76": "LABEL_76",
92
+ "77": "LABEL_77",
93
+ "78": "LABEL_78",
94
+ "79": "LABEL_79",
95
+ "80": "LABEL_80",
96
+ "81": "LABEL_81",
97
+ "82": "LABEL_82",
98
+ "83": "LABEL_83",
99
+ "84": "LABEL_84",
100
+ "85": "LABEL_85",
101
+ "86": "LABEL_86",
102
+ "87": "LABEL_87",
103
+ "88": "LABEL_88",
104
+ "89": "LABEL_89",
105
+ "90": "LABEL_90",
106
+ "91": "LABEL_91",
107
+ "92": "LABEL_92",
108
+ "93": "LABEL_93",
109
+ "94": "LABEL_94",
110
+ "95": "LABEL_95",
111
+ "96": "LABEL_96",
112
+ "97": "LABEL_97",
113
+ "98": "LABEL_98",
114
+ "99": "LABEL_99",
115
+ "100": "LABEL_100",
116
+ "101": "LABEL_101",
117
+ "102": "LABEL_102",
118
+ "103": "LABEL_103",
119
+ "104": "LABEL_104",
120
+ "105": "LABEL_105",
121
+ "106": "LABEL_106",
122
+ "107": "LABEL_107",
123
+ "108": "LABEL_108",
124
+ "109": "LABEL_109",
125
+ "110": "LABEL_110",
126
+ "111": "LABEL_111"
127
+ },
128
+ "initializer_range": 0.02,
129
+ "intermediate_size": 3072,
130
+ "label2id": {
131
+ "LABEL_0": 0,
132
+ "LABEL_1": 1,
133
+ "LABEL_10": 10,
134
+ "LABEL_100": 100,
135
+ "LABEL_101": 101,
136
+ "LABEL_102": 102,
137
+ "LABEL_103": 103,
138
+ "LABEL_104": 104,
139
+ "LABEL_105": 105,
140
+ "LABEL_106": 106,
141
+ "LABEL_107": 107,
142
+ "LABEL_108": 108,
143
+ "LABEL_109": 109,
144
+ "LABEL_11": 11,
145
+ "LABEL_110": 110,
146
+ "LABEL_111": 111,
147
+ "LABEL_12": 12,
148
+ "LABEL_13": 13,
149
+ "LABEL_14": 14,
150
+ "LABEL_15": 15,
151
+ "LABEL_16": 16,
152
+ "LABEL_17": 17,
153
+ "LABEL_18": 18,
154
+ "LABEL_19": 19,
155
+ "LABEL_2": 2,
156
+ "LABEL_20": 20,
157
+ "LABEL_21": 21,
158
+ "LABEL_22": 22,
159
+ "LABEL_23": 23,
160
+ "LABEL_24": 24,
161
+ "LABEL_25": 25,
162
+ "LABEL_26": 26,
163
+ "LABEL_27": 27,
164
+ "LABEL_28": 28,
165
+ "LABEL_29": 29,
166
+ "LABEL_3": 3,
167
+ "LABEL_30": 30,
168
+ "LABEL_31": 31,
169
+ "LABEL_32": 32,
170
+ "LABEL_33": 33,
171
+ "LABEL_34": 34,
172
+ "LABEL_35": 35,
173
+ "LABEL_36": 36,
174
+ "LABEL_37": 37,
175
+ "LABEL_38": 38,
176
+ "LABEL_39": 39,
177
+ "LABEL_4": 4,
178
+ "LABEL_40": 40,
179
+ "LABEL_41": 41,
180
+ "LABEL_42": 42,
181
+ "LABEL_43": 43,
182
+ "LABEL_44": 44,
183
+ "LABEL_45": 45,
184
+ "LABEL_46": 46,
185
+ "LABEL_47": 47,
186
+ "LABEL_48": 48,
187
+ "LABEL_49": 49,
188
+ "LABEL_5": 5,
189
+ "LABEL_50": 50,
190
+ "LABEL_51": 51,
191
+ "LABEL_52": 52,
192
+ "LABEL_53": 53,
193
+ "LABEL_54": 54,
194
+ "LABEL_55": 55,
195
+ "LABEL_56": 56,
196
+ "LABEL_57": 57,
197
+ "LABEL_58": 58,
198
+ "LABEL_59": 59,
199
+ "LABEL_6": 6,
200
+ "LABEL_60": 60,
201
+ "LABEL_61": 61,
202
+ "LABEL_62": 62,
203
+ "LABEL_63": 63,
204
+ "LABEL_64": 64,
205
+ "LABEL_65": 65,
206
+ "LABEL_66": 66,
207
+ "LABEL_67": 67,
208
+ "LABEL_68": 68,
209
+ "LABEL_69": 69,
210
+ "LABEL_7": 7,
211
+ "LABEL_70": 70,
212
+ "LABEL_71": 71,
213
+ "LABEL_72": 72,
214
+ "LABEL_73": 73,
215
+ "LABEL_74": 74,
216
+ "LABEL_75": 75,
217
+ "LABEL_76": 76,
218
+ "LABEL_77": 77,
219
+ "LABEL_78": 78,
220
+ "LABEL_79": 79,
221
+ "LABEL_8": 8,
222
+ "LABEL_80": 80,
223
+ "LABEL_81": 81,
224
+ "LABEL_82": 82,
225
+ "LABEL_83": 83,
226
+ "LABEL_84": 84,
227
+ "LABEL_85": 85,
228
+ "LABEL_86": 86,
229
+ "LABEL_87": 87,
230
+ "LABEL_88": 88,
231
+ "LABEL_89": 89,
232
+ "LABEL_9": 9,
233
+ "LABEL_90": 90,
234
+ "LABEL_91": 91,
235
+ "LABEL_92": 92,
236
+ "LABEL_93": 93,
237
+ "LABEL_94": 94,
238
+ "LABEL_95": 95,
239
+ "LABEL_96": 96,
240
+ "LABEL_97": 97,
241
+ "LABEL_98": 98,
242
+ "LABEL_99": 99
243
+ },
244
+ "layer_norm_eps": 1e-12,
245
+ "lsh_backward": true,
246
+ "max_position_embeddings": 4096,
247
+ "model_type": "yoso",
248
+ "num_attention_heads": 12,
249
+ "num_hash": 64,
250
+ "num_hidden_layers": 12,
251
+ "pad_token_id": 1,
252
+ "position_embedding_type": "absolute",
253
+ "tokenizer_class": "AlbertTokenizer",
254
+ "torch_dtype": "float32",
255
+ "transformers_version": "4.37.0",
256
+ "type_vocab_size": 1,
257
+ "use_expectation": true,
258
+ "use_fast_hash": true,
259
+ "vocab_size": 50265
260
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1bbc34e7124c87cf7da3854596d6aab976907cd1061b7462efac4f7b15cde67
3
+ size 507598072
runs/Feb25_05-10-50_0601eb897907/events.out.tfevents.1708837851.0601eb897907.34.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6319ac43182eb96f1bee018b1c7e17e02380357f04288c77ec22acba671ae881
3
+ size 42967
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
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": "</s>",
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
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
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": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": false,
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": 512,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "RobertaTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b97e0059fd7206e6dc8708dd53c3617f1cb2a0bad6d434105fb1bff0f421dcc5
3
+ size 4728
vocab.json ADDED
The diff for this file is too large to render. See raw diff