Upload folder using huggingface_hub
Browse files- expdecayforward_123/epoch0/metadata.json +6 -0
- expdecayforward_123/epoch0/metrics.json +562 -0
- expdecayforward_123/epoch0/model.pt +3 -0
- expdecayforward_123/epoch0/optimizer.pt +3 -0
- expdecayforward_123/epoch0/scheduler.pt +3 -0
- expdecayforward_123/epoch0/tokenizer/merges.txt +0 -0
- expdecayforward_123/epoch0/tokenizer/special_tokens_map.json +6 -0
- expdecayforward_123/epoch0/tokenizer/tokenizer.json +0 -0
- expdecayforward_123/epoch0/tokenizer/tokenizer_config.json +21 -0
- expdecayforward_123/epoch0/tokenizer/vocab.json +0 -0
expdecayforward_123/epoch0/metadata.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch_num": 0,
|
| 3 |
+
"global_batch_num": 134,
|
| 4 |
+
"device": "cuda",
|
| 5 |
+
"dtype": "bfloat16"
|
| 6 |
+
}
|
expdecayforward_123/epoch0/metrics.json
ADDED
|
@@ -0,0 +1,562 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"training": {
|
| 3 |
+
"effective_batch_nums": [
|
| 4 |
+
1,
|
| 5 |
+
2,
|
| 6 |
+
3,
|
| 7 |
+
4,
|
| 8 |
+
5,
|
| 9 |
+
6,
|
| 10 |
+
7,
|
| 11 |
+
8,
|
| 12 |
+
9,
|
| 13 |
+
10,
|
| 14 |
+
11,
|
| 15 |
+
12,
|
| 16 |
+
13,
|
| 17 |
+
14,
|
| 18 |
+
15,
|
| 19 |
+
16,
|
| 20 |
+
17,
|
| 21 |
+
18,
|
| 22 |
+
19,
|
| 23 |
+
20,
|
| 24 |
+
21,
|
| 25 |
+
22,
|
| 26 |
+
23,
|
| 27 |
+
24,
|
| 28 |
+
25,
|
| 29 |
+
26,
|
| 30 |
+
27,
|
| 31 |
+
28,
|
| 32 |
+
29,
|
| 33 |
+
30,
|
| 34 |
+
31,
|
| 35 |
+
32,
|
| 36 |
+
33,
|
| 37 |
+
34,
|
| 38 |
+
35,
|
| 39 |
+
36,
|
| 40 |
+
37,
|
| 41 |
+
38,
|
| 42 |
+
39,
|
| 43 |
+
40,
|
| 44 |
+
41,
|
| 45 |
+
42,
|
| 46 |
+
43,
|
| 47 |
+
44,
|
| 48 |
+
45,
|
| 49 |
+
46,
|
| 50 |
+
47,
|
| 51 |
+
48,
|
| 52 |
+
49,
|
| 53 |
+
50,
|
| 54 |
+
51,
|
| 55 |
+
52,
|
| 56 |
+
53,
|
| 57 |
+
54,
|
| 58 |
+
55,
|
| 59 |
+
56,
|
| 60 |
+
57,
|
| 61 |
+
58,
|
| 62 |
+
59,
|
| 63 |
+
60,
|
| 64 |
+
61,
|
| 65 |
+
62,
|
| 66 |
+
63,
|
| 67 |
+
64,
|
| 68 |
+
65,
|
| 69 |
+
66,
|
| 70 |
+
67,
|
| 71 |
+
68,
|
| 72 |
+
69,
|
| 73 |
+
70,
|
| 74 |
+
71,
|
| 75 |
+
72,
|
| 76 |
+
73,
|
| 77 |
+
74,
|
| 78 |
+
75,
|
| 79 |
+
76,
|
| 80 |
+
77,
|
| 81 |
+
78,
|
| 82 |
+
79,
|
| 83 |
+
80,
|
| 84 |
+
81,
|
| 85 |
+
82,
|
| 86 |
+
83,
|
| 87 |
+
84,
|
| 88 |
+
85,
|
| 89 |
+
86,
|
| 90 |
+
87,
|
| 91 |
+
88,
|
| 92 |
+
89,
|
| 93 |
+
90,
|
| 94 |
+
91,
|
| 95 |
+
92,
|
| 96 |
+
93,
|
| 97 |
+
94,
|
| 98 |
+
95,
|
| 99 |
+
96,
|
| 100 |
+
97,
|
| 101 |
+
98,
|
| 102 |
+
99,
|
| 103 |
+
100,
|
| 104 |
+
101,
|
| 105 |
+
102,
|
| 106 |
+
103,
|
| 107 |
+
104,
|
| 108 |
+
105,
|
| 109 |
+
106,
|
| 110 |
+
107,
|
| 111 |
+
108,
|
| 112 |
+
109,
|
| 113 |
+
110,
|
| 114 |
+
111,
|
| 115 |
+
112,
|
| 116 |
+
113,
|
| 117 |
+
114,
|
| 118 |
+
115,
|
| 119 |
+
116,
|
| 120 |
+
117,
|
| 121 |
+
118,
|
| 122 |
+
119,
|
| 123 |
+
120,
|
| 124 |
+
121,
|
| 125 |
+
122,
|
| 126 |
+
123,
|
| 127 |
+
124,
|
| 128 |
+
125,
|
| 129 |
+
126,
|
| 130 |
+
127,
|
| 131 |
+
128,
|
| 132 |
+
129,
|
| 133 |
+
130,
|
| 134 |
+
131,
|
| 135 |
+
132,
|
| 136 |
+
133,
|
| 137 |
+
134
|
| 138 |
+
],
|
| 139 |
+
"losses": [
|
| 140 |
+
11.0,
|
| 141 |
+
10.625,
|
| 142 |
+
9.875,
|
| 143 |
+
9.4375,
|
| 144 |
+
9.125,
|
| 145 |
+
8.875,
|
| 146 |
+
8.6875,
|
| 147 |
+
8.5625,
|
| 148 |
+
8.4375,
|
| 149 |
+
8.5,
|
| 150 |
+
8.5,
|
| 151 |
+
8.4375,
|
| 152 |
+
8.375,
|
| 153 |
+
8.375,
|
| 154 |
+
8.3125,
|
| 155 |
+
8.3125,
|
| 156 |
+
8.25,
|
| 157 |
+
8.25,
|
| 158 |
+
8.125,
|
| 159 |
+
8.125,
|
| 160 |
+
8.1875,
|
| 161 |
+
8.1875,
|
| 162 |
+
8.1875,
|
| 163 |
+
8.0625,
|
| 164 |
+
8.125,
|
| 165 |
+
8.0625,
|
| 166 |
+
8.125,
|
| 167 |
+
8.125,
|
| 168 |
+
8.0625,
|
| 169 |
+
8.0,
|
| 170 |
+
8.0625,
|
| 171 |
+
7.96875,
|
| 172 |
+
8.0625,
|
| 173 |
+
7.96875,
|
| 174 |
+
7.96875,
|
| 175 |
+
7.90625,
|
| 176 |
+
7.96875,
|
| 177 |
+
8.0,
|
| 178 |
+
7.9375,
|
| 179 |
+
7.96875,
|
| 180 |
+
7.96875,
|
| 181 |
+
7.9375,
|
| 182 |
+
7.90625,
|
| 183 |
+
7.9375,
|
| 184 |
+
7.84375,
|
| 185 |
+
7.875,
|
| 186 |
+
7.90625,
|
| 187 |
+
7.84375,
|
| 188 |
+
7.84375,
|
| 189 |
+
7.875,
|
| 190 |
+
7.90625,
|
| 191 |
+
7.84375,
|
| 192 |
+
7.8125,
|
| 193 |
+
7.8125,
|
| 194 |
+
7.8125,
|
| 195 |
+
7.78125,
|
| 196 |
+
7.75,
|
| 197 |
+
7.78125,
|
| 198 |
+
7.75,
|
| 199 |
+
7.75,
|
| 200 |
+
7.75,
|
| 201 |
+
7.71875,
|
| 202 |
+
7.6875,
|
| 203 |
+
7.71875,
|
| 204 |
+
7.71875,
|
| 205 |
+
7.71875,
|
| 206 |
+
7.65625,
|
| 207 |
+
7.65625,
|
| 208 |
+
7.71875,
|
| 209 |
+
7.71875,
|
| 210 |
+
7.6875,
|
| 211 |
+
7.71875,
|
| 212 |
+
7.625,
|
| 213 |
+
7.625,
|
| 214 |
+
7.59375,
|
| 215 |
+
7.59375,
|
| 216 |
+
7.625,
|
| 217 |
+
7.59375,
|
| 218 |
+
7.625,
|
| 219 |
+
7.59375,
|
| 220 |
+
7.5625,
|
| 221 |
+
7.65625,
|
| 222 |
+
7.5625,
|
| 223 |
+
7.5,
|
| 224 |
+
7.53125,
|
| 225 |
+
7.5,
|
| 226 |
+
7.46875,
|
| 227 |
+
7.46875,
|
| 228 |
+
7.46875,
|
| 229 |
+
7.53125,
|
| 230 |
+
7.53125,
|
| 231 |
+
7.5,
|
| 232 |
+
7.46875,
|
| 233 |
+
7.46875,
|
| 234 |
+
7.46875,
|
| 235 |
+
7.46875,
|
| 236 |
+
7.40625,
|
| 237 |
+
7.4375,
|
| 238 |
+
7.40625,
|
| 239 |
+
7.46875,
|
| 240 |
+
7.375,
|
| 241 |
+
7.40625,
|
| 242 |
+
7.375,
|
| 243 |
+
7.3125,
|
| 244 |
+
7.40625,
|
| 245 |
+
7.3125,
|
| 246 |
+
7.34375,
|
| 247 |
+
7.4375,
|
| 248 |
+
7.28125,
|
| 249 |
+
7.3125,
|
| 250 |
+
7.375,
|
| 251 |
+
7.375,
|
| 252 |
+
7.40625,
|
| 253 |
+
7.34375,
|
| 254 |
+
7.34375,
|
| 255 |
+
7.28125,
|
| 256 |
+
7.3125,
|
| 257 |
+
7.28125,
|
| 258 |
+
7.28125,
|
| 259 |
+
7.3125,
|
| 260 |
+
7.28125,
|
| 261 |
+
7.34375,
|
| 262 |
+
7.28125,
|
| 263 |
+
7.3125,
|
| 264 |
+
7.28125,
|
| 265 |
+
7.1875,
|
| 266 |
+
7.28125,
|
| 267 |
+
7.28125,
|
| 268 |
+
7.25,
|
| 269 |
+
7.21875,
|
| 270 |
+
7.15625,
|
| 271 |
+
7.3125,
|
| 272 |
+
7.21875,
|
| 273 |
+
7.21875
|
| 274 |
+
],
|
| 275 |
+
"accuracies": [
|
| 276 |
+
0.0,
|
| 277 |
+
0.02812463531333878,
|
| 278 |
+
0.05580243778567801,
|
| 279 |
+
0.03142387455741022,
|
| 280 |
+
0.035053805013209614,
|
| 281 |
+
0.055887110268280436,
|
| 282 |
+
0.06799605819952466,
|
| 283 |
+
0.0699074074074074,
|
| 284 |
+
0.07906027994654287,
|
| 285 |
+
0.08700605624512803,
|
| 286 |
+
0.08394011766761811,
|
| 287 |
+
0.08920158690800893,
|
| 288 |
+
0.10513377544020124,
|
| 289 |
+
0.10986180904522613,
|
| 290 |
+
0.11102376489085082,
|
| 291 |
+
0.10971212389903648,
|
| 292 |
+
0.10623966689118854,
|
| 293 |
+
0.10743702278602778,
|
| 294 |
+
0.11369019422074846,
|
| 295 |
+
0.11696793002915452,
|
| 296 |
+
0.11572271756875549,
|
| 297 |
+
0.11281489594742607,
|
| 298 |
+
0.11351891003053793,
|
| 299 |
+
0.11614577048389044,
|
| 300 |
+
0.11504476223694919,
|
| 301 |
+
0.12112332112332112,
|
| 302 |
+
0.11592799361968782,
|
| 303 |
+
0.11655355562112718,
|
| 304 |
+
0.11753831474745065,
|
| 305 |
+
0.12237105279107825,
|
| 306 |
+
0.11582833410244578,
|
| 307 |
+
0.11945372486958561,
|
| 308 |
+
0.11941396432293264,
|
| 309 |
+
0.12413708839880261,
|
| 310 |
+
0.1270426690876078,
|
| 311 |
+
0.12311862061011476,
|
| 312 |
+
0.12077236747421444,
|
| 313 |
+
0.1218741017533774,
|
| 314 |
+
0.12517136555999284,
|
| 315 |
+
0.1198767513516656,
|
| 316 |
+
0.12622848378117366,
|
| 317 |
+
0.12521938706628866,
|
| 318 |
+
0.13259741986966353,
|
| 319 |
+
0.13140794223826716,
|
| 320 |
+
0.13357336796127967,
|
| 321 |
+
0.12674021539269767,
|
| 322 |
+
0.12401055408970976,
|
| 323 |
+
0.13201035375323555,
|
| 324 |
+
0.13706772334293948,
|
| 325 |
+
0.12946058091286308,
|
| 326 |
+
0.12796864191837676,
|
| 327 |
+
0.13551148351317138,
|
| 328 |
+
0.13779205239624662,
|
| 329 |
+
0.13606641269119235,
|
| 330 |
+
0.1377235500584991,
|
| 331 |
+
0.13321266968325793,
|
| 332 |
+
0.14041115834837808,
|
| 333 |
+
0.1370464936761475,
|
| 334 |
+
0.14307588656826117,
|
| 335 |
+
0.14548791679412665,
|
| 336 |
+
0.1431086076395001,
|
| 337 |
+
0.14743961352657003,
|
| 338 |
+
0.14567544113393116,
|
| 339 |
+
0.14503392967304132,
|
| 340 |
+
0.14961971402494675,
|
| 341 |
+
0.14646360705220554,
|
| 342 |
+
0.15096714361420244,
|
| 343 |
+
0.15363263335747043,
|
| 344 |
+
0.14961543157123672,
|
| 345 |
+
0.1508825204023534,
|
| 346 |
+
0.15134191840173306,
|
| 347 |
+
0.15064431520127722,
|
| 348 |
+
0.15058211689616452,
|
| 349 |
+
0.1512258908139915,
|
| 350 |
+
0.1603488982161595,
|
| 351 |
+
0.15594261011214836,
|
| 352 |
+
0.15871363294580368,
|
| 353 |
+
0.16005055437467686,
|
| 354 |
+
0.15601818616893995,
|
| 355 |
+
0.15460653831244348,
|
| 356 |
+
0.15866980505763775,
|
| 357 |
+
0.15525712414018997,
|
| 358 |
+
0.16115973090432814,
|
| 359 |
+
0.1717459027495598,
|
| 360 |
+
0.16668777707409754,
|
| 361 |
+
0.16284023668639053,
|
| 362 |
+
0.1686798700104237,
|
| 363 |
+
0.16875613212689414,
|
| 364 |
+
0.16690728767811186,
|
| 365 |
+
0.16874632136550913,
|
| 366 |
+
0.16761029205672964,
|
| 367 |
+
0.16794294733150172,
|
| 368 |
+
0.16723809523809524,
|
| 369 |
+
0.16859978476623222,
|
| 370 |
+
0.17757758993027703,
|
| 371 |
+
0.17602841077241788,
|
| 372 |
+
0.17756889003347928,
|
| 373 |
+
0.1747295915621805,
|
| 374 |
+
0.17352185089974292,
|
| 375 |
+
0.17880794701986755,
|
| 376 |
+
0.17536126764492552,
|
| 377 |
+
0.17838720208945477,
|
| 378 |
+
0.18021246640215027,
|
| 379 |
+
0.19009545060121483,
|
| 380 |
+
0.17655180348616464,
|
| 381 |
+
0.19052537354541071,
|
| 382 |
+
0.18011749272799862,
|
| 383 |
+
0.1797315114467218,
|
| 384 |
+
0.186284046692607,
|
| 385 |
+
0.18001733316825555,
|
| 386 |
+
0.18213757396449703,
|
| 387 |
+
0.18563011269378182,
|
| 388 |
+
0.17573272222892292,
|
| 389 |
+
0.1841655266757866,
|
| 390 |
+
0.18270128899682422,
|
| 391 |
+
0.19241367180629654,
|
| 392 |
+
0.1895561670056227,
|
| 393 |
+
0.18808076172995297,
|
| 394 |
+
0.18585790884718498,
|
| 395 |
+
0.18848987642861287,
|
| 396 |
+
0.19080743726433494,
|
| 397 |
+
0.18423047763457165,
|
| 398 |
+
0.18567655180693962,
|
| 399 |
+
0.19176319176319176,
|
| 400 |
+
0.19318568994889268,
|
| 401 |
+
0.1997583081570997,
|
| 402 |
+
0.19179055855002072,
|
| 403 |
+
0.1922313229089104,
|
| 404 |
+
0.19317847691855336,
|
| 405 |
+
0.20124804992199688,
|
| 406 |
+
0.19785691774346045,
|
| 407 |
+
0.19172475538786213,
|
| 408 |
+
0.19011516870189246,
|
| 409 |
+
0.19933653219595995
|
| 410 |
+
],
|
| 411 |
+
"grad_norms": [
|
| 412 |
+
0.2255859375,
|
| 413 |
+
0.447265625,
|
| 414 |
+
0.8203125,
|
| 415 |
+
0.81640625,
|
| 416 |
+
0.71875,
|
| 417 |
+
0.54296875,
|
| 418 |
+
0.357421875,
|
| 419 |
+
0.384765625,
|
| 420 |
+
0.2470703125,
|
| 421 |
+
0.3515625,
|
| 422 |
+
0.361328125,
|
| 423 |
+
0.2578125,
|
| 424 |
+
0.34765625,
|
| 425 |
+
0.251953125,
|
| 426 |
+
0.1943359375,
|
| 427 |
+
0.1943359375,
|
| 428 |
+
0.169921875,
|
| 429 |
+
0.25390625,
|
| 430 |
+
0.171875,
|
| 431 |
+
0.177734375,
|
| 432 |
+
0.16796875,
|
| 433 |
+
0.1484375,
|
| 434 |
+
0.197265625,
|
| 435 |
+
0.15234375,
|
| 436 |
+
0.15234375,
|
| 437 |
+
0.15625,
|
| 438 |
+
0.11962890625,
|
| 439 |
+
0.12353515625,
|
| 440 |
+
0.1298828125,
|
| 441 |
+
0.1455078125,
|
| 442 |
+
0.154296875,
|
| 443 |
+
0.12255859375,
|
| 444 |
+
0.12060546875,
|
| 445 |
+
0.140625,
|
| 446 |
+
0.11181640625,
|
| 447 |
+
0.1318359375,
|
| 448 |
+
0.1416015625,
|
| 449 |
+
0.1376953125,
|
| 450 |
+
0.1318359375,
|
| 451 |
+
0.138671875,
|
| 452 |
+
0.1064453125,
|
| 453 |
+
0.12158203125,
|
| 454 |
+
0.11669921875,
|
| 455 |
+
0.1064453125,
|
| 456 |
+
0.1376953125,
|
| 457 |
+
0.142578125,
|
| 458 |
+
0.16015625,
|
| 459 |
+
0.1708984375,
|
| 460 |
+
0.1865234375,
|
| 461 |
+
0.1513671875,
|
| 462 |
+
0.13671875,
|
| 463 |
+
0.2470703125,
|
| 464 |
+
0.2431640625,
|
| 465 |
+
0.119140625,
|
| 466 |
+
0.171875,
|
| 467 |
+
0.140625,
|
| 468 |
+
0.142578125,
|
| 469 |
+
0.142578125,
|
| 470 |
+
0.12451171875,
|
| 471 |
+
0.1689453125,
|
| 472 |
+
0.11328125,
|
| 473 |
+
0.162109375,
|
| 474 |
+
0.1328125,
|
| 475 |
+
0.11572265625,
|
| 476 |
+
0.1865234375,
|
| 477 |
+
0.208984375,
|
| 478 |
+
0.1708984375,
|
| 479 |
+
0.13671875,
|
| 480 |
+
0.1279296875,
|
| 481 |
+
0.12255859375,
|
| 482 |
+
0.140625,
|
| 483 |
+
0.119140625,
|
| 484 |
+
0.12451171875,
|
| 485 |
+
0.12353515625,
|
| 486 |
+
0.1328125,
|
| 487 |
+
0.12353515625,
|
| 488 |
+
0.138671875,
|
| 489 |
+
0.119140625,
|
| 490 |
+
0.150390625,
|
| 491 |
+
0.1474609375,
|
| 492 |
+
0.2294921875,
|
| 493 |
+
0.263671875,
|
| 494 |
+
0.169921875,
|
| 495 |
+
0.1630859375,
|
| 496 |
+
0.173828125,
|
| 497 |
+
0.169921875,
|
| 498 |
+
0.1650390625,
|
| 499 |
+
0.1416015625,
|
| 500 |
+
0.1533203125,
|
| 501 |
+
0.125,
|
| 502 |
+
0.15234375,
|
| 503 |
+
0.1328125,
|
| 504 |
+
0.1201171875,
|
| 505 |
+
0.171875,
|
| 506 |
+
0.1416015625,
|
| 507 |
+
0.11962890625,
|
| 508 |
+
0.1318359375,
|
| 509 |
+
0.11767578125,
|
| 510 |
+
0.1611328125,
|
| 511 |
+
0.1328125,
|
| 512 |
+
0.11376953125,
|
| 513 |
+
0.1435546875,
|
| 514 |
+
0.17578125,
|
| 515 |
+
0.123046875,
|
| 516 |
+
0.123046875,
|
| 517 |
+
0.1318359375,
|
| 518 |
+
0.11083984375,
|
| 519 |
+
0.1298828125,
|
| 520 |
+
0.1171875,
|
| 521 |
+
0.1201171875,
|
| 522 |
+
0.1513671875,
|
| 523 |
+
0.1494140625,
|
| 524 |
+
0.1064453125,
|
| 525 |
+
0.1611328125,
|
| 526 |
+
0.18359375,
|
| 527 |
+
0.103515625,
|
| 528 |
+
0.1708984375,
|
| 529 |
+
0.1376953125,
|
| 530 |
+
0.1376953125,
|
| 531 |
+
0.1259765625,
|
| 532 |
+
0.1171875,
|
| 533 |
+
0.16015625,
|
| 534 |
+
0.11669921875,
|
| 535 |
+
0.171875,
|
| 536 |
+
0.1552734375,
|
| 537 |
+
0.12890625,
|
| 538 |
+
0.16796875,
|
| 539 |
+
0.11474609375,
|
| 540 |
+
0.126953125,
|
| 541 |
+
0.1181640625,
|
| 542 |
+
0.1083984375,
|
| 543 |
+
0.1298828125,
|
| 544 |
+
0.119140625,
|
| 545 |
+
0.107421875
|
| 546 |
+
]
|
| 547 |
+
},
|
| 548 |
+
"eval": {
|
| 549 |
+
"effective_batch_nums": [
|
| 550 |
+
133
|
| 551 |
+
],
|
| 552 |
+
"losses": [
|
| 553 |
+
7.40625
|
| 554 |
+
],
|
| 555 |
+
"perplexities": [
|
| 556 |
+
1646.2413521768196
|
| 557 |
+
],
|
| 558 |
+
"accuracies": [
|
| 559 |
+
0.1884248591132083
|
| 560 |
+
]
|
| 561 |
+
}
|
| 562 |
+
}
|
expdecayforward_123/epoch0/model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4048b889d7664e7119f4ba04ae0bd8558e8ddf93515db97da6db435ba8444e4
|
| 3 |
+
size 32139875
|
expdecayforward_123/epoch0/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:906f0175cd2d2aed92dafb42f7113b2fffc4b1362f84ed71ea15644068f61950
|
| 3 |
+
size 64342347
|
expdecayforward_123/epoch0/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ca4571aa547116e8ff16140914a30e8655115115b67de6445f70560baa715c9
|
| 3 |
+
size 1465
|
expdecayforward_123/epoch0/tokenizer/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
expdecayforward_123/epoch0/tokenizer/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
expdecayforward_123/epoch0/tokenizer/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
expdecayforward_123/epoch0/tokenizer/tokenizer_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": false,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"extra_special_tokens": {},
|
| 17 |
+
"model_max_length": 1024,
|
| 18 |
+
"pad_token": "<|endoftext|>",
|
| 19 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 20 |
+
"unk_token": "<|endoftext|>"
|
| 21 |
+
}
|
expdecayforward_123/epoch0/tokenizer/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|