Run 5. Outer Step 150. Inner Step 1.
Browse files- config.json +131 -143
- generation_config.json +7 -0
- gradients.pt +3 -0
- inner_optimizer.pt +2 -2
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -1,32 +1,31 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "distributed/
|
| 3 |
-
"activation_function": "gelu_new",
|
| 4 |
"all_reduce_scores": {
|
| 5 |
-
"0": "
|
| 6 |
"1": "NON_PARTICIPATING",
|
| 7 |
"10": "NON_PARTICIPATING",
|
| 8 |
-
"100": "
|
| 9 |
-
"101": "
|
| 10 |
"102": "NON_PARTICIPATING",
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
-
"104": "
|
| 13 |
"105": "NON_PARTICIPATING",
|
| 14 |
"106": "NON_PARTICIPATING",
|
| 15 |
"107": "NON_PARTICIPATING",
|
| 16 |
-
"108": "
|
| 17 |
"109": "NON_PARTICIPATING",
|
| 18 |
"11": "NON_PARTICIPATING",
|
| 19 |
"110": "NON_PARTICIPATING",
|
| 20 |
-
"111": "
|
| 21 |
"112": "NON_PARTICIPATING",
|
| 22 |
"113": "NON_PARTICIPATING",
|
| 23 |
"114": "NON_PARTICIPATING",
|
| 24 |
"115": "NON_PARTICIPATING",
|
| 25 |
"116": "NON_PARTICIPATING",
|
| 26 |
"117": "NON_PARTICIPATING",
|
| 27 |
-
"118": "
|
| 28 |
-
"119": "
|
| 29 |
-
"12": "
|
| 30 |
"120": "NON_PARTICIPATING",
|
| 31 |
"121": "NON_PARTICIPATING",
|
| 32 |
"122": "NON_PARTICIPATING",
|
|
@@ -34,12 +33,12 @@
|
|
| 34 |
"124": "NON_PARTICIPATING",
|
| 35 |
"125": "NON_PARTICIPATING",
|
| 36 |
"126": "NON_PARTICIPATING",
|
| 37 |
-
"127": "
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
| 40 |
-
"13": "
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
-
"131": "
|
| 43 |
"132": "NON_PARTICIPATING",
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
"134": "NON_PARTICIPATING",
|
|
@@ -48,54 +47,54 @@
|
|
| 48 |
"137": "NON_PARTICIPATING",
|
| 49 |
"138": "NON_PARTICIPATING",
|
| 50 |
"139": "NON_PARTICIPATING",
|
| 51 |
-
"14": "
|
| 52 |
-
"140": "
|
| 53 |
"141": "NON_PARTICIPATING",
|
| 54 |
-
"142": "
|
| 55 |
"143": "NON_PARTICIPATING",
|
| 56 |
"144": "NON_PARTICIPATING",
|
| 57 |
-
"145": "
|
| 58 |
-
"146": "
|
| 59 |
"147": "NON_PARTICIPATING",
|
| 60 |
"148": "NON_PARTICIPATING",
|
| 61 |
-
"149": "
|
| 62 |
-
"15": "
|
| 63 |
"150": "NON_PARTICIPATING",
|
| 64 |
-
"151": "
|
| 65 |
"152": "NON_PARTICIPATING",
|
| 66 |
-
"153": "
|
| 67 |
"154": "NON_PARTICIPATING",
|
| 68 |
"155": "NON_PARTICIPATING",
|
| 69 |
-
"156": "
|
| 70 |
-
"157": "
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
-
"159": "
|
| 73 |
-
"16": "
|
| 74 |
-
"160": "
|
| 75 |
-
"161": "
|
| 76 |
"162": "NON_PARTICIPATING",
|
| 77 |
-
"163": "
|
| 78 |
"164": "NON_PARTICIPATING",
|
| 79 |
"165": "NON_PARTICIPATING",
|
| 80 |
"166": "NON_PARTICIPATING",
|
| 81 |
-
"167": "
|
| 82 |
-
"168": "
|
| 83 |
-
"169": "
|
| 84 |
-
"17": "
|
| 85 |
-
"170": "
|
| 86 |
-
"171": "
|
| 87 |
"172": "NON_PARTICIPATING",
|
| 88 |
-
"173": "
|
| 89 |
"174": "NON_PARTICIPATING",
|
| 90 |
"175": "NON_PARTICIPATING",
|
| 91 |
-
"176": "
|
| 92 |
"177": "NON_PARTICIPATING",
|
| 93 |
"178": "NON_PARTICIPATING",
|
| 94 |
"179": "NON_PARTICIPATING",
|
| 95 |
"18": "NON_PARTICIPATING",
|
| 96 |
"180": "NON_PARTICIPATING",
|
| 97 |
-
"181": "
|
| 98 |
-
"182": "
|
| 99 |
"183": "NON_PARTICIPATING",
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
|
@@ -107,199 +106,188 @@
|
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
"192": "NON_PARTICIPATING",
|
| 110 |
-
"193": "
|
| 111 |
"194": "NON_PARTICIPATING",
|
| 112 |
-
"195": "
|
| 113 |
"196": "NON_PARTICIPATING",
|
| 114 |
-
"197": "
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
"2": "NON_PARTICIPATING",
|
| 118 |
-
"20": "
|
| 119 |
"200": "NON_PARTICIPATING",
|
| 120 |
-
"201": "
|
| 121 |
"202": "NON_PARTICIPATING",
|
| 122 |
-
"203": "
|
| 123 |
"204": "NON_PARTICIPATING",
|
| 124 |
-
"205": "
|
| 125 |
-
"206": "
|
| 126 |
-
"207": "
|
| 127 |
"208": "NON_PARTICIPATING",
|
| 128 |
-
"209": "
|
| 129 |
-
"21": "
|
| 130 |
-
"210": "
|
| 131 |
"211": "NON_PARTICIPATING",
|
| 132 |
"212": "NON_PARTICIPATING",
|
| 133 |
"213": "NON_PARTICIPATING",
|
| 134 |
"214": "NON_PARTICIPATING",
|
| 135 |
"215": "NON_PARTICIPATING",
|
| 136 |
"216": "NON_PARTICIPATING",
|
| 137 |
-
"217": "
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
-
"219": "
|
| 140 |
"22": "NON_PARTICIPATING",
|
| 141 |
"220": "NON_PARTICIPATING",
|
| 142 |
"221": "NON_PARTICIPATING",
|
| 143 |
"222": "NON_PARTICIPATING",
|
| 144 |
"223": "NON_PARTICIPATING",
|
| 145 |
-
"224": "
|
| 146 |
"225": "NON_PARTICIPATING",
|
| 147 |
"226": "NON_PARTICIPATING",
|
| 148 |
"227": "NON_PARTICIPATING",
|
| 149 |
"228": "NON_PARTICIPATING",
|
| 150 |
"229": "NON_PARTICIPATING",
|
| 151 |
"23": "NON_PARTICIPATING",
|
| 152 |
-
"230": "
|
| 153 |
-
"231": "
|
| 154 |
-
"232": "
|
| 155 |
"233": "NON_PARTICIPATING",
|
| 156 |
-
"234": "
|
| 157 |
"235": "NON_PARTICIPATING",
|
| 158 |
"236": "NON_PARTICIPATING",
|
| 159 |
"237": "NON_PARTICIPATING",
|
| 160 |
"238": "NON_PARTICIPATING",
|
| 161 |
-
"239": "
|
| 162 |
-
"24": "
|
| 163 |
"240": "NON_PARTICIPATING",
|
| 164 |
-
"241": "
|
| 165 |
"242": "NON_PARTICIPATING",
|
| 166 |
"243": "NON_PARTICIPATING",
|
| 167 |
"244": "NON_PARTICIPATING",
|
| 168 |
"245": "NON_PARTICIPATING",
|
| 169 |
-
"246": "
|
| 170 |
"247": "NON_PARTICIPATING",
|
| 171 |
-
"248": "
|
| 172 |
"249": "NON_PARTICIPATING",
|
| 173 |
-
"25": "
|
| 174 |
-
"250": "
|
| 175 |
-
"251": "
|
| 176 |
-
"252": "
|
| 177 |
"253": "NON_PARTICIPATING",
|
| 178 |
-
"254": "
|
| 179 |
-
"255": "
|
| 180 |
"26": "NON_PARTICIPATING",
|
| 181 |
-
"27": "
|
| 182 |
"28": "NON_PARTICIPATING",
|
| 183 |
-
"29": "
|
| 184 |
"3": "NON_PARTICIPATING",
|
| 185 |
-
"30": "
|
| 186 |
"31": "NON_PARTICIPATING",
|
| 187 |
"32": "NON_PARTICIPATING",
|
| 188 |
-
"33": "
|
| 189 |
-
"34": "
|
| 190 |
"35": "NON_PARTICIPATING",
|
| 191 |
-
"36": "
|
| 192 |
"37": "NON_PARTICIPATING",
|
| 193 |
-
"38": "
|
| 194 |
"39": "NON_PARTICIPATING",
|
| 195 |
"4": "NON_PARTICIPATING",
|
| 196 |
"40": "NON_PARTICIPATING",
|
| 197 |
"41": "NON_PARTICIPATING",
|
| 198 |
-
"42": "
|
| 199 |
-
"43": "
|
| 200 |
"44": "NON_PARTICIPATING",
|
| 201 |
"45": "NON_PARTICIPATING",
|
| 202 |
"46": "NON_PARTICIPATING",
|
| 203 |
"47": "NON_PARTICIPATING",
|
| 204 |
-
"48": "
|
| 205 |
-
"49": "
|
| 206 |
"5": "NON_PARTICIPATING",
|
| 207 |
"50": "NON_PARTICIPATING",
|
| 208 |
"51": "NON_PARTICIPATING",
|
| 209 |
"52": "NON_PARTICIPATING",
|
| 210 |
"53": "NON_PARTICIPATING",
|
| 211 |
-
"54": "
|
| 212 |
-
"55": "
|
| 213 |
"56": "NON_PARTICIPATING",
|
| 214 |
-
"57": "
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
-
"59": "
|
| 217 |
"6": "NON_PARTICIPATING",
|
| 218 |
-
"60": "
|
| 219 |
"61": "NON_PARTICIPATING",
|
| 220 |
"62": "NON_PARTICIPATING",
|
| 221 |
"63": "NON_PARTICIPATING",
|
| 222 |
-
"64": "
|
| 223 |
"65": "NON_PARTICIPATING",
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
-
"67": "
|
| 226 |
"68": "NON_PARTICIPATING",
|
| 227 |
-
"69": "
|
| 228 |
-
"7": "
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
"71": "NON_PARTICIPATING",
|
| 231 |
"72": "NON_PARTICIPATING",
|
| 232 |
"73": "NON_PARTICIPATING",
|
| 233 |
-
"74": "
|
| 234 |
-
"75": "
|
| 235 |
-
"76": "
|
| 236 |
"77": "NON_PARTICIPATING",
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
-
"8": "
|
| 240 |
"80": "NON_PARTICIPATING",
|
| 241 |
-
"81": "
|
| 242 |
"82": "NON_PARTICIPATING",
|
| 243 |
-
"83": "
|
| 244 |
"84": "NON_PARTICIPATING",
|
| 245 |
-
"85": "
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
-
"87": "
|
| 248 |
-
"88": "
|
| 249 |
-
"89": "
|
| 250 |
"9": "NON_PARTICIPATING",
|
| 251 |
-
"90": "
|
| 252 |
"91": "NON_PARTICIPATING",
|
| 253 |
-
"92": "
|
| 254 |
-
"93": "
|
| 255 |
-
"94": "
|
| 256 |
-
"95": "
|
| 257 |
-
"96": "
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
-
"98": "
|
| 260 |
"99": "NON_PARTICIPATING"
|
| 261 |
},
|
| 262 |
"architectures": [
|
| 263 |
-
"
|
| 264 |
],
|
| 265 |
-
"
|
| 266 |
-
"
|
| 267 |
-
"AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
|
| 268 |
-
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
| 269 |
-
},
|
| 270 |
"block_list": [
|
| 271 |
-
|
| 272 |
-
|
| 273 |
-
5470157,
|
| 274 |
-
5470162,
|
| 275 |
-
5470166
|
| 276 |
],
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
"initializer_range": 0.02,
|
| 282 |
-
"inner_step":
|
| 283 |
-
"
|
| 284 |
-
"last_allreduce_block":
|
| 285 |
-
"
|
| 286 |
-
"
|
| 287 |
-
"
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
| 292 |
-
"
|
| 293 |
-
"
|
| 294 |
-
"
|
| 295 |
-
"
|
| 296 |
-
"summary_activation": null,
|
| 297 |
-
"summary_first_dropout": 0.1,
|
| 298 |
-
"summary_proj_to_labels": true,
|
| 299 |
-
"summary_type": "cls_index",
|
| 300 |
-
"summary_use_proj": true,
|
| 301 |
"torch_dtype": "float32",
|
| 302 |
"transformers_version": "4.39.3",
|
| 303 |
-
"use_cache":
|
| 304 |
-
"vocab_size":
|
| 305 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "distributed/llama-1b",
|
|
|
|
| 3 |
"all_reduce_scores": {
|
| 4 |
+
"0": "NOT_ALIVE",
|
| 5 |
"1": "NON_PARTICIPATING",
|
| 6 |
"10": "NON_PARTICIPATING",
|
| 7 |
+
"100": "NOT_ALIVE",
|
| 8 |
+
"101": "NOT_ALIVE",
|
| 9 |
"102": "NON_PARTICIPATING",
|
| 10 |
"103": "NON_PARTICIPATING",
|
| 11 |
+
"104": "NOT_ALIVE",
|
| 12 |
"105": "NON_PARTICIPATING",
|
| 13 |
"106": "NON_PARTICIPATING",
|
| 14 |
"107": "NON_PARTICIPATING",
|
| 15 |
+
"108": "NOT_ALIVE",
|
| 16 |
"109": "NON_PARTICIPATING",
|
| 17 |
"11": "NON_PARTICIPATING",
|
| 18 |
"110": "NON_PARTICIPATING",
|
| 19 |
+
"111": "NOT_ALIVE",
|
| 20 |
"112": "NON_PARTICIPATING",
|
| 21 |
"113": "NON_PARTICIPATING",
|
| 22 |
"114": "NON_PARTICIPATING",
|
| 23 |
"115": "NON_PARTICIPATING",
|
| 24 |
"116": "NON_PARTICIPATING",
|
| 25 |
"117": "NON_PARTICIPATING",
|
| 26 |
+
"118": "NOT_ALIVE",
|
| 27 |
+
"119": "NOT_ALIVE",
|
| 28 |
+
"12": "NOT_ALIVE",
|
| 29 |
"120": "NON_PARTICIPATING",
|
| 30 |
"121": "NON_PARTICIPATING",
|
| 31 |
"122": "NON_PARTICIPATING",
|
|
|
|
| 33 |
"124": "NON_PARTICIPATING",
|
| 34 |
"125": "NON_PARTICIPATING",
|
| 35 |
"126": "NON_PARTICIPATING",
|
| 36 |
+
"127": "NOT_ALIVE",
|
| 37 |
"128": "NON_PARTICIPATING",
|
| 38 |
"129": "NON_PARTICIPATING",
|
| 39 |
+
"13": "NOT_ALIVE",
|
| 40 |
"130": "NON_PARTICIPATING",
|
| 41 |
+
"131": "NOT_ALIVE",
|
| 42 |
"132": "NON_PARTICIPATING",
|
| 43 |
"133": "NON_PARTICIPATING",
|
| 44 |
"134": "NON_PARTICIPATING",
|
|
|
|
| 47 |
"137": "NON_PARTICIPATING",
|
| 48 |
"138": "NON_PARTICIPATING",
|
| 49 |
"139": "NON_PARTICIPATING",
|
| 50 |
+
"14": "NOT_ALIVE",
|
| 51 |
+
"140": "NOT_ALIVE",
|
| 52 |
"141": "NON_PARTICIPATING",
|
| 53 |
+
"142": "NOT_ALIVE",
|
| 54 |
"143": "NON_PARTICIPATING",
|
| 55 |
"144": "NON_PARTICIPATING",
|
| 56 |
+
"145": "NOT_ALIVE",
|
| 57 |
+
"146": "NOT_ALIVE",
|
| 58 |
"147": "NON_PARTICIPATING",
|
| 59 |
"148": "NON_PARTICIPATING",
|
| 60 |
+
"149": "NOT_ALIVE",
|
| 61 |
+
"15": "NOT_ALIVE",
|
| 62 |
"150": "NON_PARTICIPATING",
|
| 63 |
+
"151": "NOT_ALIVE",
|
| 64 |
"152": "NON_PARTICIPATING",
|
| 65 |
+
"153": "NOT_ALIVE",
|
| 66 |
"154": "NON_PARTICIPATING",
|
| 67 |
"155": "NON_PARTICIPATING",
|
| 68 |
+
"156": "NOT_ALIVE",
|
| 69 |
+
"157": "NOT_ALIVE",
|
| 70 |
"158": "NON_PARTICIPATING",
|
| 71 |
+
"159": "NOT_ALIVE",
|
| 72 |
+
"16": "NOT_ALIVE",
|
| 73 |
+
"160": "NOT_ALIVE",
|
| 74 |
+
"161": "NOT_ALIVE",
|
| 75 |
"162": "NON_PARTICIPATING",
|
| 76 |
+
"163": "NOT_ALIVE",
|
| 77 |
"164": "NON_PARTICIPATING",
|
| 78 |
"165": "NON_PARTICIPATING",
|
| 79 |
"166": "NON_PARTICIPATING",
|
| 80 |
+
"167": "NOT_ALIVE",
|
| 81 |
+
"168": "NOT_ALIVE",
|
| 82 |
+
"169": "NOT_ALIVE",
|
| 83 |
+
"17": "NOT_ALIVE",
|
| 84 |
+
"170": "NOT_ALIVE",
|
| 85 |
+
"171": "NOT_ALIVE",
|
| 86 |
"172": "NON_PARTICIPATING",
|
| 87 |
+
"173": "NOT_ALIVE",
|
| 88 |
"174": "NON_PARTICIPATING",
|
| 89 |
"175": "NON_PARTICIPATING",
|
| 90 |
+
"176": "NOT_ALIVE",
|
| 91 |
"177": "NON_PARTICIPATING",
|
| 92 |
"178": "NON_PARTICIPATING",
|
| 93 |
"179": "NON_PARTICIPATING",
|
| 94 |
"18": "NON_PARTICIPATING",
|
| 95 |
"180": "NON_PARTICIPATING",
|
| 96 |
+
"181": "NOT_ALIVE",
|
| 97 |
+
"182": "NOT_ALIVE",
|
| 98 |
"183": "NON_PARTICIPATING",
|
| 99 |
"184": "NON_PARTICIPATING",
|
| 100 |
"185": "NON_PARTICIPATING",
|
|
|
|
| 106 |
"190": "NON_PARTICIPATING",
|
| 107 |
"191": "NON_PARTICIPATING",
|
| 108 |
"192": "NON_PARTICIPATING",
|
| 109 |
+
"193": "NOT_ALIVE",
|
| 110 |
"194": "NON_PARTICIPATING",
|
| 111 |
+
"195": "NOT_ALIVE",
|
| 112 |
"196": "NON_PARTICIPATING",
|
| 113 |
+
"197": "NOT_ALIVE",
|
| 114 |
"198": "NON_PARTICIPATING",
|
| 115 |
"199": "NON_PARTICIPATING",
|
| 116 |
"2": "NON_PARTICIPATING",
|
| 117 |
+
"20": "NOT_ALIVE",
|
| 118 |
"200": "NON_PARTICIPATING",
|
| 119 |
+
"201": "NOT_ALIVE",
|
| 120 |
"202": "NON_PARTICIPATING",
|
| 121 |
+
"203": "NOT_ALIVE",
|
| 122 |
"204": "NON_PARTICIPATING",
|
| 123 |
+
"205": "NOT_ALIVE",
|
| 124 |
+
"206": "NOT_ALIVE",
|
| 125 |
+
"207": "NOT_ALIVE",
|
| 126 |
"208": "NON_PARTICIPATING",
|
| 127 |
+
"209": "NOT_ALIVE",
|
| 128 |
+
"21": "NOT_ALIVE",
|
| 129 |
+
"210": "NOT_ALIVE",
|
| 130 |
"211": "NON_PARTICIPATING",
|
| 131 |
"212": "NON_PARTICIPATING",
|
| 132 |
"213": "NON_PARTICIPATING",
|
| 133 |
"214": "NON_PARTICIPATING",
|
| 134 |
"215": "NON_PARTICIPATING",
|
| 135 |
"216": "NON_PARTICIPATING",
|
| 136 |
+
"217": "NOT_ALIVE",
|
| 137 |
"218": "NON_PARTICIPATING",
|
| 138 |
+
"219": "NOT_ALIVE",
|
| 139 |
"22": "NON_PARTICIPATING",
|
| 140 |
"220": "NON_PARTICIPATING",
|
| 141 |
"221": "NON_PARTICIPATING",
|
| 142 |
"222": "NON_PARTICIPATING",
|
| 143 |
"223": "NON_PARTICIPATING",
|
| 144 |
+
"224": "NOT_ALIVE",
|
| 145 |
"225": "NON_PARTICIPATING",
|
| 146 |
"226": "NON_PARTICIPATING",
|
| 147 |
"227": "NON_PARTICIPATING",
|
| 148 |
"228": "NON_PARTICIPATING",
|
| 149 |
"229": "NON_PARTICIPATING",
|
| 150 |
"23": "NON_PARTICIPATING",
|
| 151 |
+
"230": "NOT_ALIVE",
|
| 152 |
+
"231": "NOT_ALIVE",
|
| 153 |
+
"232": "NOT_ALIVE",
|
| 154 |
"233": "NON_PARTICIPATING",
|
| 155 |
+
"234": "NOT_ALIVE",
|
| 156 |
"235": "NON_PARTICIPATING",
|
| 157 |
"236": "NON_PARTICIPATING",
|
| 158 |
"237": "NON_PARTICIPATING",
|
| 159 |
"238": "NON_PARTICIPATING",
|
| 160 |
+
"239": "NOT_ALIVE",
|
| 161 |
+
"24": "NOT_ALIVE",
|
| 162 |
"240": "NON_PARTICIPATING",
|
| 163 |
+
"241": "NOT_ALIVE",
|
| 164 |
"242": "NON_PARTICIPATING",
|
| 165 |
"243": "NON_PARTICIPATING",
|
| 166 |
"244": "NON_PARTICIPATING",
|
| 167 |
"245": "NON_PARTICIPATING",
|
| 168 |
+
"246": "NOT_ALIVE",
|
| 169 |
"247": "NON_PARTICIPATING",
|
| 170 |
+
"248": "NOT_ALIVE",
|
| 171 |
"249": "NON_PARTICIPATING",
|
| 172 |
+
"25": "SUCCESS",
|
| 173 |
+
"250": "NOT_ALIVE",
|
| 174 |
+
"251": "NOT_ALIVE",
|
| 175 |
+
"252": "NOT_ALIVE",
|
| 176 |
"253": "NON_PARTICIPATING",
|
| 177 |
+
"254": "NOT_ALIVE",
|
| 178 |
+
"255": "NOT_ALIVE",
|
| 179 |
"26": "NON_PARTICIPATING",
|
| 180 |
+
"27": "NOT_ALIVE",
|
| 181 |
"28": "NON_PARTICIPATING",
|
| 182 |
+
"29": "NOT_ALIVE",
|
| 183 |
"3": "NON_PARTICIPATING",
|
| 184 |
+
"30": "NOT_ALIVE",
|
| 185 |
"31": "NON_PARTICIPATING",
|
| 186 |
"32": "NON_PARTICIPATING",
|
| 187 |
+
"33": "NOT_ALIVE",
|
| 188 |
+
"34": "NOT_ALIVE",
|
| 189 |
"35": "NON_PARTICIPATING",
|
| 190 |
+
"36": "NOT_ALIVE",
|
| 191 |
"37": "NON_PARTICIPATING",
|
| 192 |
+
"38": "NOT_ALIVE",
|
| 193 |
"39": "NON_PARTICIPATING",
|
| 194 |
"4": "NON_PARTICIPATING",
|
| 195 |
"40": "NON_PARTICIPATING",
|
| 196 |
"41": "NON_PARTICIPATING",
|
| 197 |
+
"42": "NOT_ALIVE",
|
| 198 |
+
"43": "NOT_ALIVE",
|
| 199 |
"44": "NON_PARTICIPATING",
|
| 200 |
"45": "NON_PARTICIPATING",
|
| 201 |
"46": "NON_PARTICIPATING",
|
| 202 |
"47": "NON_PARTICIPATING",
|
| 203 |
+
"48": "NOT_ALIVE",
|
| 204 |
+
"49": "NOT_ALIVE",
|
| 205 |
"5": "NON_PARTICIPATING",
|
| 206 |
"50": "NON_PARTICIPATING",
|
| 207 |
"51": "NON_PARTICIPATING",
|
| 208 |
"52": "NON_PARTICIPATING",
|
| 209 |
"53": "NON_PARTICIPATING",
|
| 210 |
+
"54": "NOT_ALIVE",
|
| 211 |
+
"55": "NOT_ALIVE",
|
| 212 |
"56": "NON_PARTICIPATING",
|
| 213 |
+
"57": "NOT_ALIVE",
|
| 214 |
"58": "NON_PARTICIPATING",
|
| 215 |
+
"59": "NOT_ALIVE",
|
| 216 |
"6": "NON_PARTICIPATING",
|
| 217 |
+
"60": "NOT_ALIVE",
|
| 218 |
"61": "NON_PARTICIPATING",
|
| 219 |
"62": "NON_PARTICIPATING",
|
| 220 |
"63": "NON_PARTICIPATING",
|
| 221 |
+
"64": "NOT_ALIVE",
|
| 222 |
"65": "NON_PARTICIPATING",
|
| 223 |
"66": "NON_PARTICIPATING",
|
| 224 |
+
"67": "NOT_ALIVE",
|
| 225 |
"68": "NON_PARTICIPATING",
|
| 226 |
+
"69": "NOT_ALIVE",
|
| 227 |
+
"7": "NOT_ALIVE",
|
| 228 |
"70": "NON_PARTICIPATING",
|
| 229 |
"71": "NON_PARTICIPATING",
|
| 230 |
"72": "NON_PARTICIPATING",
|
| 231 |
"73": "NON_PARTICIPATING",
|
| 232 |
+
"74": "NOT_ALIVE",
|
| 233 |
+
"75": "NOT_ALIVE",
|
| 234 |
+
"76": "NOT_ALIVE",
|
| 235 |
"77": "NON_PARTICIPATING",
|
| 236 |
"78": "NON_PARTICIPATING",
|
| 237 |
"79": "NON_PARTICIPATING",
|
| 238 |
+
"8": "NOT_ALIVE",
|
| 239 |
"80": "NON_PARTICIPATING",
|
| 240 |
+
"81": "NOT_ALIVE",
|
| 241 |
"82": "NON_PARTICIPATING",
|
| 242 |
+
"83": "NOT_ALIVE",
|
| 243 |
"84": "NON_PARTICIPATING",
|
| 244 |
+
"85": "NOT_ALIVE",
|
| 245 |
"86": "NON_PARTICIPATING",
|
| 246 |
+
"87": "NOT_ALIVE",
|
| 247 |
+
"88": "NOT_ALIVE",
|
| 248 |
+
"89": "NOT_ALIVE",
|
| 249 |
"9": "NON_PARTICIPATING",
|
| 250 |
+
"90": "NOT_ALIVE",
|
| 251 |
"91": "NON_PARTICIPATING",
|
| 252 |
+
"92": "NOT_ALIVE",
|
| 253 |
+
"93": "NOT_ALIVE",
|
| 254 |
+
"94": "NOT_ALIVE",
|
| 255 |
+
"95": "NOT_ALIVE",
|
| 256 |
+
"96": "NOT_ALIVE",
|
| 257 |
"97": "NON_PARTICIPATING",
|
| 258 |
+
"98": "NOT_ALIVE",
|
| 259 |
"99": "NON_PARTICIPATING"
|
| 260 |
},
|
| 261 |
"architectures": [
|
| 262 |
+
"LlamaForCausalLM"
|
| 263 |
],
|
| 264 |
+
"attention_bias": false,
|
| 265 |
+
"attention_dropout": 0.0,
|
|
|
|
|
|
|
|
|
|
| 266 |
"block_list": [
|
| 267 |
+
6069837,
|
| 268 |
+
6069849
|
|
|
|
|
|
|
|
|
|
| 269 |
],
|
| 270 |
+
"bos_token_id": 1,
|
| 271 |
+
"eos_token_id": 2,
|
| 272 |
+
"hidden_act": "silu",
|
| 273 |
+
"hidden_size": 2048,
|
| 274 |
"initializer_range": 0.02,
|
| 275 |
+
"inner_step": 1,
|
| 276 |
+
"intermediate_size": 5632,
|
| 277 |
+
"last_allreduce_block": 6062756,
|
| 278 |
+
"max_position_embeddings": 2048,
|
| 279 |
+
"mlp_bias": false,
|
| 280 |
+
"model_type": "llama",
|
| 281 |
+
"num_attention_heads": 32,
|
| 282 |
+
"num_hidden_layers": 22,
|
| 283 |
+
"num_key_value_heads": 4,
|
| 284 |
+
"pretraining_tp": 1,
|
| 285 |
+
"rms_norm_eps": 1e-05,
|
| 286 |
+
"rope_scaling": null,
|
| 287 |
+
"rope_theta": 10000.0,
|
| 288 |
+
"tie_word_embeddings": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 289 |
"torch_dtype": "float32",
|
| 290 |
"transformers_version": "4.39.3",
|
| 291 |
+
"use_cache": false,
|
| 292 |
+
"vocab_size": 32000
|
| 293 |
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.39.3",
|
| 6 |
+
"use_cache": false
|
| 7 |
+
}
|
gradients.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9377b6c98ac7d59d3d06e2a96b2bb6851bae4b65fe5cfeb287b8521ac192a1ae
|
| 3 |
+
size 44585978
|
inner_optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6bb5ceddb77ebe382ada0cfe2fdb46780197f027a33b6585b1fd7b3108e6aaa4
|
| 3 |
+
size 8800564076
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef6da77306659acb1efd185596abfda76c3ed075239bb542a0dfc38b126c2cb5
|
| 3 |
+
size 4400216536
|