Outer Step 1. Inner Step 0. Batch Size 0
Browse files- config.json +48 -59
- inner_optimizer.pt +2 -2
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -13,36 +13,36 @@
|
|
| 13 |
"105": "NON_PARTICIPATING",
|
| 14 |
"106": "NON_PARTICIPATING",
|
| 15 |
"107": "NON_PARTICIPATING",
|
| 16 |
-
"108": "
|
| 17 |
"109": "NON_PARTICIPATING",
|
| 18 |
"11": "NON_PARTICIPATING",
|
| 19 |
-
"110": "
|
| 20 |
"111": "NON_PARTICIPATING",
|
| 21 |
"112": "NON_PARTICIPATING",
|
| 22 |
"113": "NON_PARTICIPATING",
|
| 23 |
"114": "NON_PARTICIPATING",
|
| 24 |
"115": "NON_PARTICIPATING",
|
| 25 |
"116": "NON_PARTICIPATING",
|
| 26 |
-
"117": "
|
| 27 |
"118": "NON_PARTICIPATING",
|
| 28 |
"119": "NON_PARTICIPATING",
|
| 29 |
-
"12": "
|
| 30 |
"120": "NON_PARTICIPATING",
|
| 31 |
-
"121": "
|
| 32 |
"122": "NON_PARTICIPATING",
|
| 33 |
"123": "NON_PARTICIPATING",
|
| 34 |
"124": "NON_PARTICIPATING",
|
| 35 |
"125": "NON_PARTICIPATING",
|
| 36 |
-
"126": "
|
| 37 |
-
"127": "
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
| 40 |
"13": "NON_PARTICIPATING",
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
"131": "NON_PARTICIPATING",
|
| 43 |
-
"132": "
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
-
"134": "
|
| 46 |
"135": "NON_PARTICIPATING",
|
| 47 |
"136": "NON_PARTICIPATING",
|
| 48 |
"137": "NON_PARTICIPATING",
|
|
@@ -56,7 +56,7 @@
|
|
| 56 |
"144": "NON_PARTICIPATING",
|
| 57 |
"145": "NON_PARTICIPATING",
|
| 58 |
"146": "NON_PARTICIPATING",
|
| 59 |
-
"147": "
|
| 60 |
"148": "NON_PARTICIPATING",
|
| 61 |
"149": "NON_PARTICIPATING",
|
| 62 |
"15": "NON_PARTICIPATING",
|
|
@@ -71,7 +71,7 @@
|
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
"16": "NON_PARTICIPATING",
|
| 74 |
-
"160": "
|
| 75 |
"161": "NON_PARTICIPATING",
|
| 76 |
"162": "NON_PARTICIPATING",
|
| 77 |
"163": "NON_PARTICIPATING",
|
|
@@ -82,12 +82,12 @@
|
|
| 82 |
"168": "NON_PARTICIPATING",
|
| 83 |
"169": "NON_PARTICIPATING",
|
| 84 |
"17": "NON_PARTICIPATING",
|
| 85 |
-
"170": "
|
| 86 |
"171": "NON_PARTICIPATING",
|
| 87 |
-
"172": "
|
| 88 |
"173": "NON_PARTICIPATING",
|
| 89 |
"174": "NON_PARTICIPATING",
|
| 90 |
-
"175": "
|
| 91 |
"176": "NON_PARTICIPATING",
|
| 92 |
"177": "NON_PARTICIPATING",
|
| 93 |
"178": "NON_PARTICIPATING",
|
|
@@ -99,24 +99,24 @@
|
|
| 99 |
"183": "NON_PARTICIPATING",
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
| 102 |
-
"186": "
|
| 103 |
-
"187": "
|
| 104 |
"188": "NON_PARTICIPATING",
|
| 105 |
"189": "NON_PARTICIPATING",
|
| 106 |
"19": "NON_PARTICIPATING",
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
-
"192": "
|
| 110 |
"193": "NON_PARTICIPATING",
|
| 111 |
"194": "NON_PARTICIPATING",
|
| 112 |
"195": "NON_PARTICIPATING",
|
| 113 |
-
"196": "
|
| 114 |
"197": "NON_PARTICIPATING",
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
"2": "NON_PARTICIPATING",
|
| 118 |
"20": "NON_PARTICIPATING",
|
| 119 |
-
"200": "
|
| 120 |
"201": "NON_PARTICIPATING",
|
| 121 |
"202": "NON_PARTICIPATING",
|
| 122 |
"203": "NON_PARTICIPATING",
|
|
@@ -130,38 +130,38 @@
|
|
| 130 |
"210": "NON_PARTICIPATING",
|
| 131 |
"211": "NON_PARTICIPATING",
|
| 132 |
"212": "NON_PARTICIPATING",
|
| 133 |
-
"213": "
|
| 134 |
"214": "NON_PARTICIPATING",
|
| 135 |
"215": "NON_PARTICIPATING",
|
| 136 |
"216": "NON_PARTICIPATING",
|
| 137 |
"217": "NON_PARTICIPATING",
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
"219": "NON_PARTICIPATING",
|
| 140 |
-
"22": "
|
| 141 |
"220": "NON_PARTICIPATING",
|
| 142 |
"221": "NON_PARTICIPATING",
|
| 143 |
"222": "NON_PARTICIPATING",
|
| 144 |
"223": "NON_PARTICIPATING",
|
| 145 |
"224": "NON_PARTICIPATING",
|
| 146 |
"225": "NON_PARTICIPATING",
|
| 147 |
-
"226": "
|
| 148 |
"227": "NON_PARTICIPATING",
|
| 149 |
"228": "NON_PARTICIPATING",
|
| 150 |
"229": "NON_PARTICIPATING",
|
| 151 |
"23": "NON_PARTICIPATING",
|
| 152 |
"230": "NON_PARTICIPATING",
|
| 153 |
"231": "NON_PARTICIPATING",
|
| 154 |
-
"232": "
|
| 155 |
"233": "NON_PARTICIPATING",
|
| 156 |
-
"234": "
|
| 157 |
"235": "SUCCESS",
|
| 158 |
"236": "NON_PARTICIPATING",
|
| 159 |
-
"237": "
|
| 160 |
"238": "NON_PARTICIPATING",
|
| 161 |
"239": "NON_PARTICIPATING",
|
| 162 |
"24": "NON_PARTICIPATING",
|
| 163 |
-
"240": "
|
| 164 |
-
"241": "
|
| 165 |
"242": "NON_PARTICIPATING",
|
| 166 |
"243": "NON_PARTICIPATING",
|
| 167 |
"244": "NON_PARTICIPATING",
|
|
@@ -169,12 +169,12 @@
|
|
| 169 |
"246": "NON_PARTICIPATING",
|
| 170 |
"247": "NON_PARTICIPATING",
|
| 171 |
"248": "NON_PARTICIPATING",
|
| 172 |
-
"249": "
|
| 173 |
"25": "SUCCESS",
|
| 174 |
-
"250": "
|
| 175 |
"251": "NON_PARTICIPATING",
|
| 176 |
"252": "NON_PARTICIPATING",
|
| 177 |
-
"253": "
|
| 178 |
"254": "NON_PARTICIPATING",
|
| 179 |
"255": "NON_PARTICIPATING",
|
| 180 |
"26": "NON_PARTICIPATING",
|
|
@@ -193,7 +193,7 @@
|
|
| 193 |
"38": "NON_PARTICIPATING",
|
| 194 |
"39": "NON_PARTICIPATING",
|
| 195 |
"4": "NON_PARTICIPATING",
|
| 196 |
-
"40": "
|
| 197 |
"41": "NON_PARTICIPATING",
|
| 198 |
"42": "NON_PARTICIPATING",
|
| 199 |
"43": "NON_PARTICIPATING",
|
|
@@ -205,59 +205,59 @@
|
|
| 205 |
"49": "NON_PARTICIPATING",
|
| 206 |
"5": "NON_PARTICIPATING",
|
| 207 |
"50": "NON_PARTICIPATING",
|
| 208 |
-
"51": "
|
| 209 |
"52": "NON_PARTICIPATING",
|
| 210 |
"53": "NON_PARTICIPATING",
|
| 211 |
"54": "NON_PARTICIPATING",
|
| 212 |
"55": "NON_PARTICIPATING",
|
| 213 |
-
"56": "
|
| 214 |
"57": "NON_PARTICIPATING",
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
"59": "NON_PARTICIPATING",
|
| 217 |
"6": "NON_PARTICIPATING",
|
| 218 |
"60": "NON_PARTICIPATING",
|
| 219 |
"61": "NON_PARTICIPATING",
|
| 220 |
-
"62": "
|
| 221 |
"63": "NON_PARTICIPATING",
|
| 222 |
"64": "NON_PARTICIPATING",
|
| 223 |
-
"65": "
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "SUCCESS",
|
| 226 |
-
"68": "
|
| 227 |
-
"69": "
|
| 228 |
"7": "NON_PARTICIPATING",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
"71": "NON_PARTICIPATING",
|
| 231 |
"72": "NON_PARTICIPATING",
|
| 232 |
"73": "NON_PARTICIPATING",
|
| 233 |
"74": "NON_PARTICIPATING",
|
| 234 |
-
"75": "
|
| 235 |
"76": "NON_PARTICIPATING",
|
| 236 |
"77": "NON_PARTICIPATING",
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
"8": "NON_PARTICIPATING",
|
| 240 |
"80": "NON_PARTICIPATING",
|
| 241 |
-
"81": "
|
| 242 |
-
"82": "
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
-
"84": "
|
| 245 |
-
"85": "
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
"87": "NON_PARTICIPATING",
|
| 248 |
"88": "NON_PARTICIPATING",
|
| 249 |
"89": "NON_PARTICIPATING",
|
| 250 |
"9": "NON_PARTICIPATING",
|
| 251 |
-
"90": "
|
| 252 |
"91": "NON_PARTICIPATING",
|
| 253 |
"92": "NON_PARTICIPATING",
|
| 254 |
"93": "NON_PARTICIPATING",
|
| 255 |
"94": "NON_PARTICIPATING",
|
| 256 |
-
"95": "
|
| 257 |
"96": "NON_PARTICIPATING",
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
"98": "NON_PARTICIPATING",
|
| 260 |
-
"99": "
|
| 261 |
},
|
| 262 |
"architectures": [
|
| 263 |
"GPTOptim"
|
|
@@ -267,26 +267,15 @@
|
|
| 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 |
-
5141919,
|
| 272 |
-
5141931,
|
| 273 |
-
5141938,
|
| 274 |
-
5141947,
|
| 275 |
-
5141962,
|
| 276 |
-
5141966,
|
| 277 |
-
5141973,
|
| 278 |
-
5141984,
|
| 279 |
-
5141993,
|
| 280 |
-
5142001
|
| 281 |
-
],
|
| 282 |
"block_size": 1024,
|
| 283 |
"bos_token_id": 50256,
|
| 284 |
"embd_pdrop": 0.1,
|
| 285 |
"eos_token_id": 50256,
|
| 286 |
"initializer_range": 0.02,
|
| 287 |
-
"inner_step":
|
| 288 |
"inner_steps": 0,
|
| 289 |
-
"last_allreduce_block":
|
| 290 |
"layer_norm_epsilon": 1e-05,
|
| 291 |
"model_type": "gpt_optimized",
|
| 292 |
"n_embd": 1280,
|
|
|
|
| 13 |
"105": "NON_PARTICIPATING",
|
| 14 |
"106": "NON_PARTICIPATING",
|
| 15 |
"107": "NON_PARTICIPATING",
|
| 16 |
+
"108": "NON_PARTICIPATING",
|
| 17 |
"109": "NON_PARTICIPATING",
|
| 18 |
"11": "NON_PARTICIPATING",
|
| 19 |
+
"110": "SUCCESS",
|
| 20 |
"111": "NON_PARTICIPATING",
|
| 21 |
"112": "NON_PARTICIPATING",
|
| 22 |
"113": "NON_PARTICIPATING",
|
| 23 |
"114": "NON_PARTICIPATING",
|
| 24 |
"115": "NON_PARTICIPATING",
|
| 25 |
"116": "NON_PARTICIPATING",
|
| 26 |
+
"117": "SUCCESS",
|
| 27 |
"118": "NON_PARTICIPATING",
|
| 28 |
"119": "NON_PARTICIPATING",
|
| 29 |
+
"12": "NON_PARTICIPATING",
|
| 30 |
"120": "NON_PARTICIPATING",
|
| 31 |
+
"121": "NON_PARTICIPATING",
|
| 32 |
"122": "NON_PARTICIPATING",
|
| 33 |
"123": "NON_PARTICIPATING",
|
| 34 |
"124": "NON_PARTICIPATING",
|
| 35 |
"125": "NON_PARTICIPATING",
|
| 36 |
+
"126": "SUCCESS",
|
| 37 |
+
"127": "SUCCESS",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
| 40 |
"13": "NON_PARTICIPATING",
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
"131": "NON_PARTICIPATING",
|
| 43 |
+
"132": "SUCCESS",
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
+
"134": "NON_PARTICIPATING",
|
| 46 |
"135": "NON_PARTICIPATING",
|
| 47 |
"136": "NON_PARTICIPATING",
|
| 48 |
"137": "NON_PARTICIPATING",
|
|
|
|
| 56 |
"144": "NON_PARTICIPATING",
|
| 57 |
"145": "NON_PARTICIPATING",
|
| 58 |
"146": "NON_PARTICIPATING",
|
| 59 |
+
"147": "NON_PARTICIPATING",
|
| 60 |
"148": "NON_PARTICIPATING",
|
| 61 |
"149": "NON_PARTICIPATING",
|
| 62 |
"15": "NON_PARTICIPATING",
|
|
|
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
"16": "NON_PARTICIPATING",
|
| 74 |
+
"160": "SUCCESS",
|
| 75 |
"161": "NON_PARTICIPATING",
|
| 76 |
"162": "NON_PARTICIPATING",
|
| 77 |
"163": "NON_PARTICIPATING",
|
|
|
|
| 82 |
"168": "NON_PARTICIPATING",
|
| 83 |
"169": "NON_PARTICIPATING",
|
| 84 |
"17": "NON_PARTICIPATING",
|
| 85 |
+
"170": "SUCCESS",
|
| 86 |
"171": "NON_PARTICIPATING",
|
| 87 |
+
"172": "SUCCESS",
|
| 88 |
"173": "NON_PARTICIPATING",
|
| 89 |
"174": "NON_PARTICIPATING",
|
| 90 |
+
"175": "SUCCESS",
|
| 91 |
"176": "NON_PARTICIPATING",
|
| 92 |
"177": "NON_PARTICIPATING",
|
| 93 |
"178": "NON_PARTICIPATING",
|
|
|
|
| 99 |
"183": "NON_PARTICIPATING",
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
| 102 |
+
"186": "SUCCESS",
|
| 103 |
+
"187": "NON_PARTICIPATING",
|
| 104 |
"188": "NON_PARTICIPATING",
|
| 105 |
"189": "NON_PARTICIPATING",
|
| 106 |
"19": "NON_PARTICIPATING",
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
+
"192": "SUCCESS",
|
| 110 |
"193": "NON_PARTICIPATING",
|
| 111 |
"194": "NON_PARTICIPATING",
|
| 112 |
"195": "NON_PARTICIPATING",
|
| 113 |
+
"196": "NON_PARTICIPATING",
|
| 114 |
"197": "NON_PARTICIPATING",
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
"2": "NON_PARTICIPATING",
|
| 118 |
"20": "NON_PARTICIPATING",
|
| 119 |
+
"200": "SUCCESS",
|
| 120 |
"201": "NON_PARTICIPATING",
|
| 121 |
"202": "NON_PARTICIPATING",
|
| 122 |
"203": "NON_PARTICIPATING",
|
|
|
|
| 130 |
"210": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
"219": "NON_PARTICIPATING",
|
| 140 |
+
"22": "SUCCESS",
|
| 141 |
"220": "NON_PARTICIPATING",
|
| 142 |
"221": "NON_PARTICIPATING",
|
| 143 |
"222": "NON_PARTICIPATING",
|
| 144 |
"223": "NON_PARTICIPATING",
|
| 145 |
"224": "NON_PARTICIPATING",
|
| 146 |
"225": "NON_PARTICIPATING",
|
| 147 |
+
"226": "SUCCESS",
|
| 148 |
"227": "NON_PARTICIPATING",
|
| 149 |
"228": "NON_PARTICIPATING",
|
| 150 |
"229": "NON_PARTICIPATING",
|
| 151 |
"23": "NON_PARTICIPATING",
|
| 152 |
"230": "NON_PARTICIPATING",
|
| 153 |
"231": "NON_PARTICIPATING",
|
| 154 |
+
"232": "NON_PARTICIPATING",
|
| 155 |
"233": "NON_PARTICIPATING",
|
| 156 |
+
"234": "NON_PARTICIPATING",
|
| 157 |
"235": "SUCCESS",
|
| 158 |
"236": "NON_PARTICIPATING",
|
| 159 |
+
"237": "NON_PARTICIPATING",
|
| 160 |
"238": "NON_PARTICIPATING",
|
| 161 |
"239": "NON_PARTICIPATING",
|
| 162 |
"24": "NON_PARTICIPATING",
|
| 163 |
+
"240": "SUCCESS",
|
| 164 |
+
"241": "NON_PARTICIPATING",
|
| 165 |
"242": "NON_PARTICIPATING",
|
| 166 |
"243": "NON_PARTICIPATING",
|
| 167 |
"244": "NON_PARTICIPATING",
|
|
|
|
| 169 |
"246": "NON_PARTICIPATING",
|
| 170 |
"247": "NON_PARTICIPATING",
|
| 171 |
"248": "NON_PARTICIPATING",
|
| 172 |
+
"249": "SUCCESS",
|
| 173 |
"25": "SUCCESS",
|
| 174 |
+
"250": "SUCCESS",
|
| 175 |
"251": "NON_PARTICIPATING",
|
| 176 |
"252": "NON_PARTICIPATING",
|
| 177 |
+
"253": "NON_PARTICIPATING",
|
| 178 |
"254": "NON_PARTICIPATING",
|
| 179 |
"255": "NON_PARTICIPATING",
|
| 180 |
"26": "NON_PARTICIPATING",
|
|
|
|
| 193 |
"38": "NON_PARTICIPATING",
|
| 194 |
"39": "NON_PARTICIPATING",
|
| 195 |
"4": "NON_PARTICIPATING",
|
| 196 |
+
"40": "SUCCESS",
|
| 197 |
"41": "NON_PARTICIPATING",
|
| 198 |
"42": "NON_PARTICIPATING",
|
| 199 |
"43": "NON_PARTICIPATING",
|
|
|
|
| 205 |
"49": "NON_PARTICIPATING",
|
| 206 |
"5": "NON_PARTICIPATING",
|
| 207 |
"50": "NON_PARTICIPATING",
|
| 208 |
+
"51": "NON_PARTICIPATING",
|
| 209 |
"52": "NON_PARTICIPATING",
|
| 210 |
"53": "NON_PARTICIPATING",
|
| 211 |
"54": "NON_PARTICIPATING",
|
| 212 |
"55": "NON_PARTICIPATING",
|
| 213 |
+
"56": "SUCCESS",
|
| 214 |
"57": "NON_PARTICIPATING",
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
"59": "NON_PARTICIPATING",
|
| 217 |
"6": "NON_PARTICIPATING",
|
| 218 |
"60": "NON_PARTICIPATING",
|
| 219 |
"61": "NON_PARTICIPATING",
|
| 220 |
+
"62": "SUCCESS",
|
| 221 |
"63": "NON_PARTICIPATING",
|
| 222 |
"64": "NON_PARTICIPATING",
|
| 223 |
+
"65": "NON_PARTICIPATING",
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "SUCCESS",
|
| 226 |
+
"68": "SUCCESS",
|
| 227 |
+
"69": "NON_PARTICIPATING",
|
| 228 |
"7": "NON_PARTICIPATING",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
"71": "NON_PARTICIPATING",
|
| 231 |
"72": "NON_PARTICIPATING",
|
| 232 |
"73": "NON_PARTICIPATING",
|
| 233 |
"74": "NON_PARTICIPATING",
|
| 234 |
+
"75": "SUCCESS",
|
| 235 |
"76": "NON_PARTICIPATING",
|
| 236 |
"77": "NON_PARTICIPATING",
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
"8": "NON_PARTICIPATING",
|
| 240 |
"80": "NON_PARTICIPATING",
|
| 241 |
+
"81": "SUCCESS",
|
| 242 |
+
"82": "SUCCESS",
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
+
"84": "SUCCESS",
|
| 245 |
+
"85": "SUCCESS",
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
"87": "NON_PARTICIPATING",
|
| 248 |
"88": "NON_PARTICIPATING",
|
| 249 |
"89": "NON_PARTICIPATING",
|
| 250 |
"9": "NON_PARTICIPATING",
|
| 251 |
+
"90": "NON_PARTICIPATING",
|
| 252 |
"91": "NON_PARTICIPATING",
|
| 253 |
"92": "NON_PARTICIPATING",
|
| 254 |
"93": "NON_PARTICIPATING",
|
| 255 |
"94": "NON_PARTICIPATING",
|
| 256 |
+
"95": "NON_PARTICIPATING",
|
| 257 |
"96": "NON_PARTICIPATING",
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
"98": "NON_PARTICIPATING",
|
| 260 |
+
"99": "SUCCESS"
|
| 261 |
},
|
| 262 |
"architectures": [
|
| 263 |
"GPTOptim"
|
|
|
|
| 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 |
"block_size": 1024,
|
| 272 |
"bos_token_id": 50256,
|
| 273 |
"embd_pdrop": 0.1,
|
| 274 |
"eos_token_id": 50256,
|
| 275 |
"initializer_range": 0.02,
|
| 276 |
+
"inner_step": 0,
|
| 277 |
"inner_steps": 0,
|
| 278 |
+
"last_allreduce_block": 5160840,
|
| 279 |
"layer_norm_epsilon": 1e-05,
|
| 280 |
"model_type": "gpt_optimized",
|
| 281 |
"n_embd": 1280,
|
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 sha256:2e1973401bfdcada6891ea5086b3b2b05facc87c1ff270537f55f7863b1e17ec
|
| 3 |
+
size 2688
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4040701744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d976cbb76f45821db2fa03ebd108d1deef8d2d4f5a2bfdb102fcf38713f86be
|
| 3 |
size 4040701744
|