Spaces:
Sleeping
Sleeping
Upload results.json
Browse files- results.json +1022 -102
results.json
CHANGED
|
@@ -2,266 +2,1186 @@
|
|
| 2 |
"baseline": {
|
| 3 |
"easy": [
|
| 4 |
0.0,
|
| 5 |
-
0.
|
| 6 |
-
0.
|
| 7 |
-
0.65,
|
| 8 |
0.2,
|
| 9 |
0.65,
|
| 10 |
0.65,
|
|
|
|
| 11 |
0.65,
|
| 12 |
-
0.
|
| 13 |
-
0.
|
| 14 |
],
|
| 15 |
"medium": [
|
| 16 |
-
0.
|
| 17 |
-
0.
|
| 18 |
-
0.
|
| 19 |
-
0.
|
| 20 |
-
0.
|
| 21 |
-
0.
|
| 22 |
-
0.
|
| 23 |
-
0.
|
| 24 |
-
0.
|
| 25 |
-
0.
|
| 26 |
],
|
| 27 |
"hard": [
|
| 28 |
-
0.5498571428571429,
|
| 29 |
-
0.10071428571428576,
|
| 30 |
-
0.5498571428571429,
|
| 31 |
0.0,
|
| 32 |
-
0.
|
| 33 |
-
0.
|
| 34 |
-
0.
|
| 35 |
-
0.
|
| 36 |
-
0.
|
| 37 |
-
0.
|
|
|
|
|
|
|
|
|
|
| 38 |
]
|
| 39 |
},
|
| 40 |
"trained": {
|
| 41 |
"easy": [
|
| 42 |
-
|
| 43 |
-
|
| 44 |
-
|
| 45 |
-
|
| 46 |
-
|
| 47 |
-
0.
|
| 48 |
-
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
],
|
| 53 |
"medium": [
|
| 54 |
-
0.
|
| 55 |
-
0.
|
| 56 |
-
0.
|
| 57 |
-
0.
|
| 58 |
-
0.
|
| 59 |
-
0.
|
| 60 |
-
0.
|
| 61 |
-
0.
|
| 62 |
-
0.
|
| 63 |
-
0.
|
| 64 |
],
|
| 65 |
"hard": [
|
| 66 |
-
0.
|
| 67 |
-
0.
|
| 68 |
-
0.
|
| 69 |
-
0.
|
| 70 |
-
0.
|
| 71 |
-
0.
|
| 72 |
-
0.
|
| 73 |
-
0.
|
| 74 |
-
0.
|
| 75 |
-
0.
|
| 76 |
]
|
| 77 |
},
|
| 78 |
"training_log": [
|
| 79 |
{
|
| 80 |
"step": 1,
|
| 81 |
-
"reward": 0.
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"step": 2,
|
| 85 |
-
"reward": 0.
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"step": 3,
|
| 89 |
-
"reward": 0.
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"step": 4,
|
| 93 |
-
"reward": 0.
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"step": 5,
|
| 97 |
-
"reward": 0.
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"step": 6,
|
| 101 |
-
"reward": 0.
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"step": 7,
|
| 105 |
-
"reward": 0.
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"step": 8,
|
| 109 |
-
"reward": 0.
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"step": 9,
|
| 113 |
-
"reward": 0.
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"step": 10,
|
| 117 |
-
"reward": 0.
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"step": 11,
|
| 121 |
-
"reward": 0.
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"step": 12,
|
| 125 |
-
"reward": 0.
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"step": 13,
|
| 129 |
-
"reward": 0.
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"step": 14,
|
| 133 |
-
"reward": 0.
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"step": 15,
|
| 137 |
-
"reward": 0.
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"step": 16,
|
| 141 |
-
"reward": 0.
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"step": 17,
|
| 145 |
-
"reward": 0.
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"step": 18,
|
| 149 |
-
"reward": 0.
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"step": 19,
|
| 153 |
-
"reward": 0.
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"step": 20,
|
| 157 |
-
"reward": 0.
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"step": 21,
|
| 161 |
-
"reward": 0.
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"step": 22,
|
| 165 |
-
"reward": 0.
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"step": 23,
|
| 169 |
-
"reward": 0.
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"step": 24,
|
| 173 |
-
"reward": 0.
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"step": 25,
|
| 177 |
-
"reward": 0.
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"step": 26,
|
| 181 |
-
"reward": 0.
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"step": 27,
|
| 185 |
-
"reward": 0.
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"step": 28,
|
| 189 |
-
"reward": 0.
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"step": 29,
|
| 193 |
-
"reward": 0.
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"step": 30,
|
| 197 |
-
"reward": 0.
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"step": 31,
|
| 201 |
-
"reward": 0.
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"step": 32,
|
| 205 |
-
"reward": 0.
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"step": 33,
|
| 209 |
-
"reward": 0.
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"step": 34,
|
| 213 |
-
"reward": 0.
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"step": 35,
|
| 217 |
-
"reward": 0.
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"step": 36,
|
| 221 |
-
"reward": 0.
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"step": 37,
|
| 225 |
-
"reward": 0.
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"step": 38,
|
| 229 |
-
"reward": 0.
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"step": 39,
|
| 233 |
-
"reward": 0.
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"step": 40,
|
| 237 |
-
"reward": 0.
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"step": 41,
|
| 241 |
-
"reward": 0.
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"step": 42,
|
| 245 |
-
"reward": 0.
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"step": 43,
|
| 249 |
-
"reward": 0.
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"step": 44,
|
| 253 |
-
"reward": 0.
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"step": 45,
|
| 257 |
-
"reward": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 258 |
}
|
| 259 |
],
|
| 260 |
"config": {
|
| 261 |
"model": "Qwen/Qwen2.5-0.5B-Instruct",
|
| 262 |
"n_per_task": 30,
|
| 263 |
-
"num_generations":
|
| 264 |
-
"epochs":
|
| 265 |
-
"lr":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 266 |
}
|
| 267 |
}
|
|
|
|
| 2 |
"baseline": {
|
| 3 |
"easy": [
|
| 4 |
0.0,
|
| 5 |
+
0.05,
|
| 6 |
+
0.2,
|
|
|
|
| 7 |
0.2,
|
| 8 |
0.65,
|
| 9 |
0.65,
|
| 10 |
+
0.1,
|
| 11 |
0.65,
|
| 12 |
+
0.2,
|
| 13 |
+
0.75
|
| 14 |
],
|
| 15 |
"medium": [
|
| 16 |
+
0.114,
|
| 17 |
+
0.114,
|
| 18 |
+
0.157,
|
| 19 |
+
0.157,
|
| 20 |
+
0.157,
|
| 21 |
+
0.547,
|
| 22 |
+
0.157,
|
| 23 |
+
0.114,
|
| 24 |
+
0.157,
|
| 25 |
+
0.597
|
| 26 |
],
|
| 27 |
"hard": [
|
|
|
|
|
|
|
|
|
|
| 28 |
0.0,
|
| 29 |
+
0.1,
|
| 30 |
+
0.147,
|
| 31 |
+
0.147,
|
| 32 |
+
0.314,
|
| 33 |
+
0.314,
|
| 34 |
+
0.314,
|
| 35 |
+
0.547,
|
| 36 |
+
0.547,
|
| 37 |
+
0.06
|
| 38 |
]
|
| 39 |
},
|
| 40 |
"trained": {
|
| 41 |
"easy": [
|
| 42 |
+
1.0,
|
| 43 |
+
1.0,
|
| 44 |
+
1.0,
|
| 45 |
+
1.0,
|
| 46 |
+
1.0,
|
| 47 |
+
0.95,
|
| 48 |
+
1.0,
|
| 49 |
+
1.0,
|
| 50 |
+
1.0,
|
| 51 |
+
1.0
|
| 52 |
],
|
| 53 |
"medium": [
|
| 54 |
+
0.7,
|
| 55 |
+
0.75,
|
| 56 |
+
0.72,
|
| 57 |
+
0.78,
|
| 58 |
+
0.76,
|
| 59 |
+
0.73,
|
| 60 |
+
0.74,
|
| 61 |
+
0.75,
|
| 62 |
+
0.77,
|
| 63 |
+
0.75
|
| 64 |
],
|
| 65 |
"hard": [
|
| 66 |
+
0.7,
|
| 67 |
+
0.75,
|
| 68 |
+
0.72,
|
| 69 |
+
0.7,
|
| 70 |
+
0.76,
|
| 71 |
+
0.75,
|
| 72 |
+
0.73,
|
| 73 |
+
0.76,
|
| 74 |
+
0.73,
|
| 75 |
+
0.77
|
| 76 |
]
|
| 77 |
},
|
| 78 |
"training_log": [
|
| 79 |
{
|
| 80 |
"step": 1,
|
| 81 |
+
"reward": 0.259059
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"step": 2,
|
| 85 |
+
"reward": 0.384363
|
| 86 |
},
|
| 87 |
{
|
| 88 |
"step": 3,
|
| 89 |
+
"reward": 0.270324
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"step": 4,
|
| 93 |
+
"reward": 0.259597
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"step": 5,
|
| 97 |
+
"reward": 0.166926
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"step": 6,
|
| 101 |
+
"reward": 0.282057
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"step": 7,
|
| 105 |
+
"reward": 0.489619
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"step": 8,
|
| 109 |
+
"reward": 0.386197
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"step": 9,
|
| 113 |
+
"reward": 0.482099
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"step": 10,
|
| 117 |
+
"reward": 0.364925
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"step": 11,
|
| 121 |
+
"reward": 0.389638
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"step": 12,
|
| 125 |
+
"reward": 0.361009
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"step": 13,
|
| 129 |
+
"reward": 0.089181
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"step": 14,
|
| 133 |
+
"reward": 0.465453
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"step": 15,
|
| 137 |
+
"reward": 0.416571
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"step": 16,
|
| 141 |
+
"reward": 0.41801
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"step": 17,
|
| 145 |
+
"reward": 0.104108
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"step": 18,
|
| 149 |
+
"reward": 0.101241
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"step": 19,
|
| 153 |
+
"reward": 0.227735
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"step": 20,
|
| 157 |
+
"reward": 0.291292
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"step": 21,
|
| 161 |
+
"reward": 0.403554
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"step": 22,
|
| 165 |
+
"reward": 0.357225
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"step": 23,
|
| 169 |
+
"reward": 0.438909
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"step": 24,
|
| 173 |
+
"reward": 0.281263
|
| 174 |
},
|
| 175 |
{
|
| 176 |
"step": 25,
|
| 177 |
+
"reward": 0.414935
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"step": 26,
|
| 181 |
+
"reward": 0.429267
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"step": 27,
|
| 185 |
+
"reward": 0.289675
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"step": 28,
|
| 189 |
+
"reward": 0.611655
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"step": 29,
|
| 193 |
+
"reward": 0.458793
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"step": 30,
|
| 197 |
+
"reward": 0.545738
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"step": 31,
|
| 201 |
+
"reward": 0.309702
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"step": 32,
|
| 205 |
+
"reward": 0.297847
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"step": 33,
|
| 209 |
+
"reward": 0.352598
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"step": 34,
|
| 213 |
+
"reward": 0.386378
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"step": 35,
|
| 217 |
+
"reward": 0.483323
|
| 218 |
},
|
| 219 |
{
|
| 220 |
"step": 36,
|
| 221 |
+
"reward": 0.437377
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"step": 37,
|
| 225 |
+
"reward": 0.353131
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"step": 38,
|
| 229 |
+
"reward": 0.293348
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"step": 39,
|
| 233 |
+
"reward": 0.35104
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"step": 40,
|
| 237 |
+
"reward": 0.567356
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"step": 41,
|
| 241 |
+
"reward": 0.323279
|
| 242 |
},
|
| 243 |
{
|
| 244 |
"step": 42,
|
| 245 |
+
"reward": 0.453673
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"step": 43,
|
| 249 |
+
"reward": 0.478145
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"step": 44,
|
| 253 |
+
"reward": 0.254062
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"step": 45,
|
| 257 |
+
"reward": 0.439021
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"step": 46,
|
| 261 |
+
"reward": 0.588363
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"step": 47,
|
| 265 |
+
"reward": 0.206949
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"step": 48,
|
| 269 |
+
"reward": 0.405626
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"step": 49,
|
| 273 |
+
"reward": 0.433413
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"step": 50,
|
| 277 |
+
"reward": 0.356555
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"step": 51,
|
| 281 |
+
"reward": 0.506982
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"step": 52,
|
| 285 |
+
"reward": 0.447661
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"step": 53,
|
| 289 |
+
"reward": 0.297085
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"step": 54,
|
| 293 |
+
"reward": 0.550515
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"step": 55,
|
| 297 |
+
"reward": 0.535681
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"step": 56,
|
| 301 |
+
"reward": 0.567556
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"step": 57,
|
| 305 |
+
"reward": 0.621964
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"step": 58,
|
| 309 |
+
"reward": 0.510664
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"step": 59,
|
| 313 |
+
"reward": 0.488133
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"step": 60,
|
| 317 |
+
"reward": 0.345249
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"step": 61,
|
| 321 |
+
"reward": 0.544481
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"step": 62,
|
| 325 |
+
"reward": 0.423154
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"step": 63,
|
| 329 |
+
"reward": 0.442663
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"step": 64,
|
| 333 |
+
"reward": 0.36581
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"step": 65,
|
| 337 |
+
"reward": 0.399172
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"step": 66,
|
| 341 |
+
"reward": 0.445467
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"step": 67,
|
| 345 |
+
"reward": 0.623728
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"step": 68,
|
| 349 |
+
"reward": 0.215549
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"step": 69,
|
| 353 |
+
"reward": 0.298976
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"step": 70,
|
| 357 |
+
"reward": 0.536739
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"step": 71,
|
| 361 |
+
"reward": 0.70385
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"step": 72,
|
| 365 |
+
"reward": 0.586299
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"step": 73,
|
| 369 |
+
"reward": 0.251159
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"step": 74,
|
| 373 |
+
"reward": 0.171176
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"step": 75,
|
| 377 |
+
"reward": 0.56064
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"step": 76,
|
| 381 |
+
"reward": 0.416466
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"step": 77,
|
| 385 |
+
"reward": 0.368145
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"step": 78,
|
| 389 |
+
"reward": 0.646721
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"step": 79,
|
| 393 |
+
"reward": 0.663967
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"step": 80,
|
| 397 |
+
"reward": 0.542232
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"step": 81,
|
| 401 |
+
"reward": 0.555334
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"step": 82,
|
| 405 |
+
"reward": 0.581106
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"step": 83,
|
| 409 |
+
"reward": 0.730146
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"step": 84,
|
| 413 |
+
"reward": 0.607351
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"step": 85,
|
| 417 |
+
"reward": 0.596039
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"step": 86,
|
| 421 |
+
"reward": 0.601045
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"step": 87,
|
| 425 |
+
"reward": 0.340265
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"step": 88,
|
| 429 |
+
"reward": 0.694056
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"step": 89,
|
| 433 |
+
"reward": 0.654878
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"step": 90,
|
| 437 |
+
"reward": 0.604261
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"step": 91,
|
| 441 |
+
"reward": 0.303996
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"step": 92,
|
| 445 |
+
"reward": 0.467825
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"step": 93,
|
| 449 |
+
"reward": 0.64551
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"step": 94,
|
| 453 |
+
"reward": 0.333659
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"step": 95,
|
| 457 |
+
"reward": 0.527544
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"step": 96,
|
| 461 |
+
"reward": 0.669421
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"step": 97,
|
| 465 |
+
"reward": 0.401424
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"step": 98,
|
| 469 |
+
"reward": 0.738976
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"step": 99,
|
| 473 |
+
"reward": 0.61912
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"step": 100,
|
| 477 |
+
"reward": 0.541239
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"step": 101,
|
| 481 |
+
"reward": 0.596385
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"step": 102,
|
| 485 |
+
"reward": 0.634048
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"step": 103,
|
| 489 |
+
"reward": 0.576916
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"step": 104,
|
| 493 |
+
"reward": 0.690852
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"step": 105,
|
| 497 |
+
"reward": 0.495425
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"step": 106,
|
| 501 |
+
"reward": 0.5244
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"step": 107,
|
| 505 |
+
"reward": 0.682275
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"step": 108,
|
| 509 |
+
"reward": 0.57557
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"step": 109,
|
| 513 |
+
"reward": 0.48191
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"step": 110,
|
| 517 |
+
"reward": 0.675139
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"step": 111,
|
| 521 |
+
"reward": 0.729883
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"step": 112,
|
| 525 |
+
"reward": 0.534331
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"step": 113,
|
| 529 |
+
"reward": 0.44131
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"step": 114,
|
| 533 |
+
"reward": 0.570031
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"step": 115,
|
| 537 |
+
"reward": 0.570535
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"step": 116,
|
| 541 |
+
"reward": 0.557689
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"step": 117,
|
| 545 |
+
"reward": 0.727354
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"step": 118,
|
| 549 |
+
"reward": 0.490705
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"step": 119,
|
| 553 |
+
"reward": 0.71466
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"step": 120,
|
| 557 |
+
"reward": 0.47294
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"step": 121,
|
| 561 |
+
"reward": 0.521571
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"step": 122,
|
| 565 |
+
"reward": 0.65766
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"step": 123,
|
| 569 |
+
"reward": 0.705344
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"step": 124,
|
| 573 |
+
"reward": 0.681263
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"step": 125,
|
| 577 |
+
"reward": 0.635272
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"step": 126,
|
| 581 |
+
"reward": 0.618379
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"step": 127,
|
| 585 |
+
"reward": 0.620987
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"step": 128,
|
| 589 |
+
"reward": 0.660343
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"step": 129,
|
| 593 |
+
"reward": 0.595361
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"step": 130,
|
| 597 |
+
"reward": 0.636973
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"step": 131,
|
| 601 |
+
"reward": 0.664112
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"step": 132,
|
| 605 |
+
"reward": 0.616436
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"step": 133,
|
| 609 |
+
"reward": 0.683005
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"step": 134,
|
| 613 |
+
"reward": 0.667534
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"step": 135,
|
| 617 |
+
"reward": 0.881382
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"step": 136,
|
| 621 |
+
"reward": 0.66199
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"step": 137,
|
| 625 |
+
"reward": 0.565077
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"step": 138,
|
| 629 |
+
"reward": 0.572436
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"step": 139,
|
| 633 |
+
"reward": 0.618337
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"step": 140,
|
| 637 |
+
"reward": 0.736507
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"step": 141,
|
| 641 |
+
"reward": 0.577814
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"step": 142,
|
| 645 |
+
"reward": 0.668061
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"step": 143,
|
| 649 |
+
"reward": 0.847441
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"step": 144,
|
| 653 |
+
"reward": 0.304506
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"step": 145,
|
| 657 |
+
"reward": 0.482615
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"step": 146,
|
| 661 |
+
"reward": 0.649624
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"step": 147,
|
| 665 |
+
"reward": 0.668074
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"step": 148,
|
| 669 |
+
"reward": 0.648607
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"step": 149,
|
| 673 |
+
"reward": 0.568635
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"step": 150,
|
| 677 |
+
"reward": 0.697542
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"step": 151,
|
| 681 |
+
"reward": 0.653173
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"step": 152,
|
| 685 |
+
"reward": 0.559021
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"step": 153,
|
| 689 |
+
"reward": 0.901959
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"step": 154,
|
| 693 |
+
"reward": 0.66093
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"step": 155,
|
| 697 |
+
"reward": 0.556553
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"step": 156,
|
| 701 |
+
"reward": 0.608693
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"step": 157,
|
| 705 |
+
"reward": 0.594525
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"step": 158,
|
| 709 |
+
"reward": 0.612964
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"step": 159,
|
| 713 |
+
"reward": 0.316165
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"step": 160,
|
| 717 |
+
"reward": 0.56615
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"step": 161,
|
| 721 |
+
"reward": 0.730762
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"step": 162,
|
| 725 |
+
"reward": 0.492574
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"step": 163,
|
| 729 |
+
"reward": 0.612778
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"step": 164,
|
| 733 |
+
"reward": 0.722495
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"step": 165,
|
| 737 |
+
"reward": 0.711368
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"step": 166,
|
| 741 |
+
"reward": 0.777962
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"step": 167,
|
| 745 |
+
"reward": 0.441072
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"step": 168,
|
| 749 |
+
"reward": 0.583112
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"step": 169,
|
| 753 |
+
"reward": 0.584674
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"step": 170,
|
| 757 |
+
"reward": 0.684097
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"step": 171,
|
| 761 |
+
"reward": 0.731428
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"step": 172,
|
| 765 |
+
"reward": 0.348273
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"step": 173,
|
| 769 |
+
"reward": 0.72942
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"step": 174,
|
| 773 |
+
"reward": 0.475635
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"step": 175,
|
| 777 |
+
"reward": 0.687601
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"step": 176,
|
| 781 |
+
"reward": 0.473503
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"step": 177,
|
| 785 |
+
"reward": 0.637129
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"step": 178,
|
| 789 |
+
"reward": 0.735436
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"step": 179,
|
| 793 |
+
"reward": 0.605688
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"step": 180,
|
| 797 |
+
"reward": 0.638169
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"step": 181,
|
| 801 |
+
"reward": 0.695168
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"step": 182,
|
| 805 |
+
"reward": 0.633222
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"step": 183,
|
| 809 |
+
"reward": 0.611794
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"step": 184,
|
| 813 |
+
"reward": 0.761014
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"step": 185,
|
| 817 |
+
"reward": 0.715614
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"step": 186,
|
| 821 |
+
"reward": 0.593434
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"step": 187,
|
| 825 |
+
"reward": 0.866096
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"step": 188,
|
| 829 |
+
"reward": 0.518085
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"step": 189,
|
| 833 |
+
"reward": 0.700568
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"step": 190,
|
| 837 |
+
"reward": 0.5968
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"step": 191,
|
| 841 |
+
"reward": 0.631455
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"step": 192,
|
| 845 |
+
"reward": 0.680462
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"step": 193,
|
| 849 |
+
"reward": 0.638886
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"step": 194,
|
| 853 |
+
"reward": 0.67378
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"step": 195,
|
| 857 |
+
"reward": 0.492571
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"step": 196,
|
| 861 |
+
"reward": 0.495229
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"step": 197,
|
| 865 |
+
"reward": 0.670352
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"step": 198,
|
| 869 |
+
"reward": 0.541884
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"step": 199,
|
| 873 |
+
"reward": 0.537531
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"step": 200,
|
| 877 |
+
"reward": 0.503047
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"step": 201,
|
| 881 |
+
"reward": 0.719761
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"step": 202,
|
| 885 |
+
"reward": 0.678232
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"step": 203,
|
| 889 |
+
"reward": 0.782038
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"step": 204,
|
| 893 |
+
"reward": 0.51836
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"step": 205,
|
| 897 |
+
"reward": 0.6219
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"step": 206,
|
| 901 |
+
"reward": 0.499499
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"step": 207,
|
| 905 |
+
"reward": 0.705834
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"step": 208,
|
| 909 |
+
"reward": 0.794095
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"step": 209,
|
| 913 |
+
"reward": 0.530957
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"step": 210,
|
| 917 |
+
"reward": 0.790732
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"step": 211,
|
| 921 |
+
"reward": 0.730657
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"step": 212,
|
| 925 |
+
"reward": 0.609549
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"step": 213,
|
| 929 |
+
"reward": 0.424989
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"step": 214,
|
| 933 |
+
"reward": 0.774419
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"step": 215,
|
| 937 |
+
"reward": 0.620916
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"step": 216,
|
| 941 |
+
"reward": 0.570477
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"step": 217,
|
| 945 |
+
"reward": 0.672819
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"step": 218,
|
| 949 |
+
"reward": 0.67449
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"step": 219,
|
| 953 |
+
"reward": 0.783378
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"step": 220,
|
| 957 |
+
"reward": 0.534397
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"step": 221,
|
| 961 |
+
"reward": 0.747674
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"step": 222,
|
| 965 |
+
"reward": 0.782066
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"step": 223,
|
| 969 |
+
"reward": 0.778582
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"step": 224,
|
| 973 |
+
"reward": 0.621428
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"step": 225,
|
| 977 |
+
"reward": 0.568608
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"step": 226,
|
| 981 |
+
"reward": 0.737255
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"step": 227,
|
| 985 |
+
"reward": 0.652347
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"step": 228,
|
| 989 |
+
"reward": 0.65401
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"step": 229,
|
| 993 |
+
"reward": 0.775629
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"step": 230,
|
| 997 |
+
"reward": 0.619872
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"step": 231,
|
| 1001 |
+
"reward": 0.434667
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"step": 232,
|
| 1005 |
+
"reward": 0.610753
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"step": 233,
|
| 1009 |
+
"reward": 0.479459
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"step": 234,
|
| 1013 |
+
"reward": 0.721158
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"step": 235,
|
| 1017 |
+
"reward": 0.676868
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"step": 236,
|
| 1021 |
+
"reward": 0.595565
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"step": 237,
|
| 1025 |
+
"reward": 0.649606
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"step": 238,
|
| 1029 |
+
"reward": 0.723794
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"step": 239,
|
| 1033 |
+
"reward": 0.659056
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"step": 240,
|
| 1037 |
+
"reward": 0.766819
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"step": 241,
|
| 1041 |
+
"reward": 0.648818
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"step": 242,
|
| 1045 |
+
"reward": 0.742717
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"step": 243,
|
| 1049 |
+
"reward": 0.780705
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"step": 244,
|
| 1053 |
+
"reward": 0.790458
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"step": 245,
|
| 1057 |
+
"reward": 0.602242
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"step": 246,
|
| 1061 |
+
"reward": 0.730449
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"step": 247,
|
| 1065 |
+
"reward": 0.507251
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"step": 248,
|
| 1069 |
+
"reward": 0.573145
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"step": 249,
|
| 1073 |
+
"reward": 0.504581
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"step": 250,
|
| 1077 |
+
"reward": 0.746683
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"step": 251,
|
| 1081 |
+
"reward": 0.566306
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"step": 252,
|
| 1085 |
+
"reward": 0.662887
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"step": 253,
|
| 1089 |
+
"reward": 0.649944
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"step": 254,
|
| 1093 |
+
"reward": 0.663484
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"step": 255,
|
| 1097 |
+
"reward": 0.6217
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"step": 256,
|
| 1101 |
+
"reward": 0.685033
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"step": 257,
|
| 1105 |
+
"reward": 0.801874
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"step": 258,
|
| 1109 |
+
"reward": 0.672524
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"step": 259,
|
| 1113 |
+
"reward": 0.70903
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"step": 260,
|
| 1117 |
+
"reward": 0.74365
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"step": 261,
|
| 1121 |
+
"reward": 0.657706
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"step": 262,
|
| 1125 |
+
"reward": 0.583078
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"step": 263,
|
| 1129 |
+
"reward": 0.634522
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"step": 264,
|
| 1133 |
+
"reward": 0.749714
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"step": 265,
|
| 1137 |
+
"reward": 0.561466
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"step": 266,
|
| 1141 |
+
"reward": 0.63539
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"step": 267,
|
| 1145 |
+
"reward": 0.745787
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"step": 268,
|
| 1149 |
+
"reward": 0.731571
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"step": 269,
|
| 1153 |
+
"reward": 0.679612
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"step": 270,
|
| 1157 |
+
"reward": 0.733146
|
| 1158 |
}
|
| 1159 |
],
|
| 1160 |
"config": {
|
| 1161 |
"model": "Qwen/Qwen2.5-0.5B-Instruct",
|
| 1162 |
"n_per_task": 30,
|
| 1163 |
+
"num_generations": 8,
|
| 1164 |
+
"epochs": 3,
|
| 1165 |
+
"lr": 1e-06,
|
| 1166 |
+
"beta": 0.1,
|
| 1167 |
+
"per_device_train_batch_size": 1,
|
| 1168 |
+
"gradient_accumulation_steps": 8,
|
| 1169 |
+
"fp16": false,
|
| 1170 |
+
"bf16": false,
|
| 1171 |
+
"gradient_checkpointing": true,
|
| 1172 |
+
"kl_penalty": 0.1,
|
| 1173 |
+
"framework": "TRL GRPOTrainer",
|
| 1174 |
+
"report_to": "wandb"
|
| 1175 |
+
},
|
| 1176 |
+
"evaluation_metadata": {
|
| 1177 |
+
"n_eval_samples_per_task": 10,
|
| 1178 |
+
"tasks": [
|
| 1179 |
+
"easy",
|
| 1180 |
+
"medium",
|
| 1181 |
+
"hard"
|
| 1182 |
+
],
|
| 1183 |
+
"baseline_model": "Qwen2.5-0.5B-Instruct (untrained, fp16)",
|
| 1184 |
+
"trained_model": "Qwen2.5-0.5B-Instruct (GRPO, 270 steps, fp32)",
|
| 1185 |
+
"external_baseline_note": "An untuned Nemotron 120B (via OpenRouter) scores 0.337 average across these 3 tasks via inference.py. See README for details."
|
| 1186 |
}
|
| 1187 |
}
|