benjamin commited on
Commit
a766c6f
1 Parent(s): 457b154

initial commit

Browse files
Files changed (7) hide show
  1. config.json +295 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scheduler.pt +3 -0
  6. trainer_state.json +0 -0
  7. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,295 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/canine-s",
3
+ "architectures": [
4
+ "CanineForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "mixture_name": "wtp-canine-s-6-layers-no-adapters-clfs",
8
+ "bos_token_id": 57344,
9
+ "bottleneck_factor": 2,
10
+ "downsampling_rate": 4,
11
+ "eos_token_id": 57345,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "LABEL_0",
17
+ "1": "LABEL_1",
18
+ "2": "LABEL_2",
19
+ "3": "LABEL_3",
20
+ "4": "LABEL_4",
21
+ "5": "LABEL_5",
22
+ "6": "LABEL_6",
23
+ "7": "LABEL_7",
24
+ "8": "LABEL_8",
25
+ "9": "LABEL_9",
26
+ "10": "LABEL_10",
27
+ "11": "LABEL_11",
28
+ "12": "LABEL_12",
29
+ "13": "LABEL_13",
30
+ "14": "LABEL_14",
31
+ "15": "LABEL_15",
32
+ "16": "LABEL_16",
33
+ "17": "LABEL_17",
34
+ "18": "LABEL_18",
35
+ "19": "LABEL_19",
36
+ "20": "LABEL_20",
37
+ "21": "LABEL_21",
38
+ "22": "LABEL_22",
39
+ "23": "LABEL_23",
40
+ "24": "LABEL_24",
41
+ "25": "LABEL_25",
42
+ "26": "LABEL_26",
43
+ "27": "LABEL_27",
44
+ "28": "LABEL_28",
45
+ "29": "LABEL_29",
46
+ "30": "LABEL_30",
47
+ "31": "LABEL_31",
48
+ "32": "LABEL_32",
49
+ "33": "LABEL_33",
50
+ "34": "LABEL_34",
51
+ "35": "LABEL_35",
52
+ "36": "LABEL_36",
53
+ "37": "LABEL_37",
54
+ "38": "LABEL_38",
55
+ "39": "LABEL_39",
56
+ "40": "LABEL_40",
57
+ "41": "LABEL_41",
58
+ "42": "LABEL_42",
59
+ "43": "LABEL_43",
60
+ "44": "LABEL_44",
61
+ "45": "LABEL_45",
62
+ "46": "LABEL_46",
63
+ "47": "LABEL_47",
64
+ "48": "LABEL_48",
65
+ "49": "LABEL_49",
66
+ "50": "LABEL_50",
67
+ "51": "LABEL_51",
68
+ "52": "LABEL_52",
69
+ "53": "LABEL_53",
70
+ "54": "LABEL_54",
71
+ "55": "LABEL_55",
72
+ "56": "LABEL_56",
73
+ "57": "LABEL_57",
74
+ "58": "LABEL_58",
75
+ "59": "LABEL_59",
76
+ "60": "LABEL_60",
77
+ "61": "LABEL_61",
78
+ "62": "LABEL_62",
79
+ "63": "LABEL_63",
80
+ "64": "LABEL_64",
81
+ "65": "LABEL_65",
82
+ "66": "LABEL_66",
83
+ "67": "LABEL_67",
84
+ "68": "LABEL_68",
85
+ "69": "LABEL_69",
86
+ "70": "LABEL_70",
87
+ "71": "LABEL_71",
88
+ "72": "LABEL_72",
89
+ "73": "LABEL_73",
90
+ "74": "LABEL_74",
91
+ "75": "LABEL_75",
92
+ "76": "LABEL_76",
93
+ "77": "LABEL_77",
94
+ "78": "LABEL_78",
95
+ "79": "LABEL_79",
96
+ "80": "LABEL_80",
97
+ "81": "LABEL_81",
98
+ "82": "LABEL_82",
99
+ "83": "LABEL_83",
100
+ "84": "LABEL_84",
101
+ "85": "LABEL_85",
102
+ "86": "LABEL_86",
103
+ "87": "LABEL_87",
104
+ "88": "LABEL_88",
105
+ "89": "LABEL_89",
106
+ "90": "LABEL_90",
107
+ "91": "LABEL_91",
108
+ "92": "LABEL_92",
109
+ "93": "LABEL_93",
110
+ "94": "LABEL_94",
111
+ "95": "LABEL_95",
112
+ "96": "LABEL_96",
113
+ "97": "LABEL_97",
114
+ "98": "LABEL_98",
115
+ "99": "LABEL_99",
116
+ "100": "LABEL_100",
117
+ "101": "LABEL_101",
118
+ "102": "LABEL_102",
119
+ "103": "LABEL_103",
120
+ "104": "LABEL_104",
121
+ "105": "LABEL_105",
122
+ "106": "LABEL_106",
123
+ "107": "LABEL_107",
124
+ "108": "LABEL_108",
125
+ "109": "LABEL_109",
126
+ "110": "LABEL_110",
127
+ "111": "LABEL_111",
128
+ "112": "LABEL_112",
129
+ "113": "LABEL_113",
130
+ "114": "LABEL_114",
131
+ "115": "LABEL_115",
132
+ "116": "LABEL_116",
133
+ "117": "LABEL_117",
134
+ "118": "LABEL_118",
135
+ "119": "LABEL_119",
136
+ "120": "LABEL_120",
137
+ "121": "LABEL_121",
138
+ "122": "LABEL_122",
139
+ "123": "LABEL_123",
140
+ "124": "LABEL_124",
141
+ "125": "LABEL_125",
142
+ "126": "LABEL_126"
143
+ },
144
+ "initializer_range": 0.02,
145
+ "intermediate_size": 3072,
146
+ "label2id": {
147
+ "LABEL_0": 0,
148
+ "LABEL_1": 1,
149
+ "LABEL_10": 10,
150
+ "LABEL_100": 100,
151
+ "LABEL_101": 101,
152
+ "LABEL_102": 102,
153
+ "LABEL_103": 103,
154
+ "LABEL_104": 104,
155
+ "LABEL_105": 105,
156
+ "LABEL_106": 106,
157
+ "LABEL_107": 107,
158
+ "LABEL_108": 108,
159
+ "LABEL_109": 109,
160
+ "LABEL_11": 11,
161
+ "LABEL_110": 110,
162
+ "LABEL_111": 111,
163
+ "LABEL_112": 112,
164
+ "LABEL_113": 113,
165
+ "LABEL_114": 114,
166
+ "LABEL_115": 115,
167
+ "LABEL_116": 116,
168
+ "LABEL_117": 117,
169
+ "LABEL_118": 118,
170
+ "LABEL_119": 119,
171
+ "LABEL_12": 12,
172
+ "LABEL_120": 120,
173
+ "LABEL_121": 121,
174
+ "LABEL_122": 122,
175
+ "LABEL_123": 123,
176
+ "LABEL_124": 124,
177
+ "LABEL_125": 125,
178
+ "LABEL_126": 126,
179
+ "LABEL_13": 13,
180
+ "LABEL_14": 14,
181
+ "LABEL_15": 15,
182
+ "LABEL_16": 16,
183
+ "LABEL_17": 17,
184
+ "LABEL_18": 18,
185
+ "LABEL_19": 19,
186
+ "LABEL_2": 2,
187
+ "LABEL_20": 20,
188
+ "LABEL_21": 21,
189
+ "LABEL_22": 22,
190
+ "LABEL_23": 23,
191
+ "LABEL_24": 24,
192
+ "LABEL_25": 25,
193
+ "LABEL_26": 26,
194
+ "LABEL_27": 27,
195
+ "LABEL_28": 28,
196
+ "LABEL_29": 29,
197
+ "LABEL_3": 3,
198
+ "LABEL_30": 30,
199
+ "LABEL_31": 31,
200
+ "LABEL_32": 32,
201
+ "LABEL_33": 33,
202
+ "LABEL_34": 34,
203
+ "LABEL_35": 35,
204
+ "LABEL_36": 36,
205
+ "LABEL_37": 37,
206
+ "LABEL_38": 38,
207
+ "LABEL_39": 39,
208
+ "LABEL_4": 4,
209
+ "LABEL_40": 40,
210
+ "LABEL_41": 41,
211
+ "LABEL_42": 42,
212
+ "LABEL_43": 43,
213
+ "LABEL_44": 44,
214
+ "LABEL_45": 45,
215
+ "LABEL_46": 46,
216
+ "LABEL_47": 47,
217
+ "LABEL_48": 48,
218
+ "LABEL_49": 49,
219
+ "LABEL_5": 5,
220
+ "LABEL_50": 50,
221
+ "LABEL_51": 51,
222
+ "LABEL_52": 52,
223
+ "LABEL_53": 53,
224
+ "LABEL_54": 54,
225
+ "LABEL_55": 55,
226
+ "LABEL_56": 56,
227
+ "LABEL_57": 57,
228
+ "LABEL_58": 58,
229
+ "LABEL_59": 59,
230
+ "LABEL_6": 6,
231
+ "LABEL_60": 60,
232
+ "LABEL_61": 61,
233
+ "LABEL_62": 62,
234
+ "LABEL_63": 63,
235
+ "LABEL_64": 64,
236
+ "LABEL_65": 65,
237
+ "LABEL_66": 66,
238
+ "LABEL_67": 67,
239
+ "LABEL_68": 68,
240
+ "LABEL_69": 69,
241
+ "LABEL_7": 7,
242
+ "LABEL_70": 70,
243
+ "LABEL_71": 71,
244
+ "LABEL_72": 72,
245
+ "LABEL_73": 73,
246
+ "LABEL_74": 74,
247
+ "LABEL_75": 75,
248
+ "LABEL_76": 76,
249
+ "LABEL_77": 77,
250
+ "LABEL_78": 78,
251
+ "LABEL_79": 79,
252
+ "LABEL_8": 8,
253
+ "LABEL_80": 80,
254
+ "LABEL_81": 81,
255
+ "LABEL_82": 82,
256
+ "LABEL_83": 83,
257
+ "LABEL_84": 84,
258
+ "LABEL_85": 85,
259
+ "LABEL_86": 86,
260
+ "LABEL_87": 87,
261
+ "LABEL_88": 88,
262
+ "LABEL_89": 89,
263
+ "LABEL_9": 9,
264
+ "LABEL_90": 90,
265
+ "LABEL_91": 91,
266
+ "LABEL_92": 92,
267
+ "LABEL_93": 93,
268
+ "LABEL_94": 94,
269
+ "LABEL_95": 95,
270
+ "LABEL_96": 96,
271
+ "LABEL_97": 97,
272
+ "LABEL_98": 98,
273
+ "LABEL_99": 99
274
+ },
275
+ "language_adapter": "off",
276
+ "layer_norm_eps": 1e-12,
277
+ "local_transformer_stride": 128,
278
+ "lookahead": null,
279
+ "lookahead_block_size": 1,
280
+ "max_position_embeddings": 16384,
281
+ "model_type": "canine",
282
+ "n_languages": 85,
283
+ "ngram_order": 1,
284
+ "num_attention_heads": 12,
285
+ "num_hash_buckets": 16384,
286
+ "num_hash_functions": 8,
287
+ "num_hidden_layers": 6,
288
+ "pad_token_id": 0,
289
+ "torch_dtype": "float32",
290
+ "transformers_version": "4.27.4",
291
+ "type_vocab_size": 16,
292
+ "upsampling_kernel_size": 1,
293
+ "use_cache": true,
294
+ "wandb_run_id": "1hmnjhg4"
295
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f94a402b928891303919a701b709bbb307d4994b5d1faea60f728d1d8e98bc4
3
+ size 684274885
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a79aae184f631eb3b83dfc1fccd71960ef16a6f056e3adb729ef5e7a9e15373
3
+ size 344639281
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc3dbc09dc99f7d21960bdf26c6bca23196b93a992a666b6eb98d8e7f70b5702
3
+ size 13617
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9aaa4d51f45e08c37d93924cb66454728f2367839959edc52af953284fd50c9
3
+ size 627
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5111281b5873c58d56c2e446b2d48b94a7d341faf95b7a3e1d5bd2940953b194
3
+ size 3643