Run 5. Outer Step 5. Inner Step 2.
Browse files- config.json +80 -89
- generation_config.json +7 -0
- inner_optimizer.pt +2 -2
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -1,21 +1,20 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
-
"activation_function": "gelu_new",
|
| 4 |
"all_reduce_scores": {
|
| 5 |
"0": "NON_PARTICIPATING",
|
| 6 |
-
"1": "
|
| 7 |
"10": "NON_PARTICIPATING",
|
| 8 |
"100": "NON_PARTICIPATING",
|
| 9 |
"101": "NON_PARTICIPATING",
|
| 10 |
"102": "NON_PARTICIPATING",
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
-
"104": "
|
| 13 |
"105": "NON_PARTICIPATING",
|
| 14 |
"106": "NON_PARTICIPATING",
|
| 15 |
"107": "NON_PARTICIPATING",
|
| 16 |
"108": "NON_PARTICIPATING",
|
| 17 |
"109": "NON_PARTICIPATING",
|
| 18 |
-
"11": "
|
| 19 |
"110": "NON_PARTICIPATING",
|
| 20 |
"111": "NON_PARTICIPATING",
|
| 21 |
"112": "NON_PARTICIPATING",
|
|
@@ -24,7 +23,7 @@
|
|
| 24 |
"115": "NON_PARTICIPATING",
|
| 25 |
"116": "NON_PARTICIPATING",
|
| 26 |
"117": "NON_PARTICIPATING",
|
| 27 |
-
"118": "
|
| 28 |
"119": "NON_PARTICIPATING",
|
| 29 |
"12": "SUCCESS",
|
| 30 |
"120": "NON_PARTICIPATING",
|
|
@@ -33,7 +32,7 @@
|
|
| 33 |
"123": "NON_PARTICIPATING",
|
| 34 |
"124": "NON_PARTICIPATING",
|
| 35 |
"125": "SUCCESS",
|
| 36 |
-
"126": "
|
| 37 |
"127": "NON_PARTICIPATING",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
|
@@ -57,7 +56,7 @@
|
|
| 57 |
"145": "NON_PARTICIPATING",
|
| 58 |
"146": "NON_PARTICIPATING",
|
| 59 |
"147": "NON_PARTICIPATING",
|
| 60 |
-
"148": "
|
| 61 |
"149": "NON_PARTICIPATING",
|
| 62 |
"15": "SUCCESS",
|
| 63 |
"150": "NON_PARTICIPATING",
|
|
@@ -70,15 +69,15 @@
|
|
| 70 |
"157": "NON_PARTICIPATING",
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
-
"16": "
|
| 74 |
"160": "NON_PARTICIPATING",
|
| 75 |
-
"161": "
|
| 76 |
-
"162": "
|
| 77 |
"163": "NON_PARTICIPATING",
|
| 78 |
"164": "NON_PARTICIPATING",
|
| 79 |
"165": "NON_PARTICIPATING",
|
| 80 |
"166": "NON_PARTICIPATING",
|
| 81 |
-
"167": "
|
| 82 |
"168": "NON_PARTICIPATING",
|
| 83 |
"169": "NON_PARTICIPATING",
|
| 84 |
"17": "SUCCESS",
|
|
@@ -86,17 +85,17 @@
|
|
| 86 |
"171": "NON_PARTICIPATING",
|
| 87 |
"172": "NON_PARTICIPATING",
|
| 88 |
"173": "NON_PARTICIPATING",
|
| 89 |
-
"174": "
|
| 90 |
"175": "NON_PARTICIPATING",
|
| 91 |
"176": "NON_PARTICIPATING",
|
| 92 |
-
"177": "
|
| 93 |
"178": "NON_PARTICIPATING",
|
| 94 |
"179": "NON_PARTICIPATING",
|
| 95 |
"18": "SUCCESS",
|
| 96 |
"180": "NON_PARTICIPATING",
|
| 97 |
"181": "NON_PARTICIPATING",
|
| 98 |
"182": "NON_PARTICIPATING",
|
| 99 |
-
"183": "
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
| 102 |
"186": "NON_PARTICIPATING",
|
|
@@ -106,7 +105,7 @@
|
|
| 106 |
"19": "SUCCESS",
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
-
"192": "
|
| 110 |
"193": "NON_PARTICIPATING",
|
| 111 |
"194": "NON_PARTICIPATING",
|
| 112 |
"195": "NON_PARTICIPATING",
|
|
@@ -121,11 +120,11 @@
|
|
| 121 |
"202": "NON_PARTICIPATING",
|
| 122 |
"203": "NON_PARTICIPATING",
|
| 123 |
"204": "NON_PARTICIPATING",
|
| 124 |
-
"205": "
|
| 125 |
"206": "NON_PARTICIPATING",
|
| 126 |
"207": "NON_PARTICIPATING",
|
| 127 |
"208": "NON_PARTICIPATING",
|
| 128 |
-
"209": "
|
| 129 |
"21": "SUCCESS",
|
| 130 |
"210": "NON_PARTICIPATING",
|
| 131 |
"211": "NON_PARTICIPATING",
|
|
@@ -133,22 +132,22 @@
|
|
| 133 |
"213": "NON_PARTICIPATING",
|
| 134 |
"214": "NON_PARTICIPATING",
|
| 135 |
"215": "NON_PARTICIPATING",
|
| 136 |
-
"216": "
|
| 137 |
-
"217": "
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
-
"219": "
|
| 140 |
"22": "SUCCESS",
|
| 141 |
"220": "NON_PARTICIPATING",
|
| 142 |
"221": "NON_PARTICIPATING",
|
| 143 |
"222": "NON_PARTICIPATING",
|
| 144 |
-
"223": "
|
| 145 |
"224": "NON_PARTICIPATING",
|
| 146 |
"225": "NON_PARTICIPATING",
|
| 147 |
"226": "NON_PARTICIPATING",
|
| 148 |
"227": "NON_PARTICIPATING",
|
| 149 |
"228": "NON_PARTICIPATING",
|
| 150 |
"229": "NON_PARTICIPATING",
|
| 151 |
-
"23": "
|
| 152 |
"230": "NON_PARTICIPATING",
|
| 153 |
"231": "NON_PARTICIPATING",
|
| 154 |
"232": "NON_PARTICIPATING",
|
|
@@ -167,54 +166,54 @@
|
|
| 167 |
"244": "NON_PARTICIPATING",
|
| 168 |
"245": "NON_PARTICIPATING",
|
| 169 |
"246": "NON_PARTICIPATING",
|
| 170 |
-
"247": "
|
| 171 |
"248": "NON_PARTICIPATING",
|
| 172 |
-
"249": "
|
| 173 |
"25": "SUCCESS",
|
| 174 |
-
"250": "
|
| 175 |
"251": "NON_PARTICIPATING",
|
| 176 |
"252": "NON_PARTICIPATING",
|
| 177 |
"253": "NON_PARTICIPATING",
|
| 178 |
"254": "NON_PARTICIPATING",
|
| 179 |
"255": "NON_PARTICIPATING",
|
| 180 |
-
"26": "
|
| 181 |
"27": "SUCCESS",
|
| 182 |
-
"28": "
|
| 183 |
"29": "SUCCESS",
|
| 184 |
-
"3": "
|
| 185 |
-
"30": "
|
| 186 |
"31": "SUCCESS",
|
| 187 |
-
"32": "
|
| 188 |
"33": "NON_PARTICIPATING",
|
| 189 |
"34": "SUCCESS",
|
| 190 |
"35": "SUCCESS",
|
| 191 |
-
"36": "
|
| 192 |
-
"37": "
|
| 193 |
"38": "SUCCESS",
|
| 194 |
-
"39": "
|
| 195 |
"4": "SUCCESS",
|
| 196 |
"40": "SUCCESS",
|
| 197 |
"41": "SUCCESS",
|
| 198 |
-
"42": "
|
| 199 |
-
"43": "
|
| 200 |
"44": "SUCCESS",
|
| 201 |
"45": "SUCCESS",
|
| 202 |
-
"46": "
|
| 203 |
"47": "SUCCESS",
|
| 204 |
"48": "NON_PARTICIPATING",
|
| 205 |
"49": "SUCCESS",
|
| 206 |
-
"5": "
|
| 207 |
-
"50": "
|
| 208 |
"51": "SUCCESS",
|
| 209 |
-
"52": "
|
| 210 |
-
"53": "
|
| 211 |
"54": "SUCCESS",
|
| 212 |
-
"55": "
|
| 213 |
"56": "SUCCESS",
|
| 214 |
-
"57": "
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
"59": "SUCCESS",
|
| 217 |
-
"6": "
|
| 218 |
"60": "SUCCESS",
|
| 219 |
"61": "SUCCESS",
|
| 220 |
"62": "SUCCESS",
|
|
@@ -223,8 +222,8 @@
|
|
| 223 |
"65": "SUCCESS",
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "SUCCESS",
|
| 226 |
-
"68": "
|
| 227 |
-
"69": "
|
| 228 |
"7": "SUCCESS",
|
| 229 |
"70": "SUCCESS",
|
| 230 |
"71": "SUCCESS",
|
|
@@ -232,71 +231,63 @@
|
|
| 232 |
"73": "SUCCESS",
|
| 233 |
"74": "SUCCESS",
|
| 234 |
"75": "SUCCESS",
|
| 235 |
-
"76": "
|
| 236 |
"77": "SUCCESS",
|
| 237 |
-
"78": "
|
| 238 |
-
"79": "
|
| 239 |
"8": "SUCCESS",
|
| 240 |
"80": "SUCCESS",
|
| 241 |
-
"81": "
|
| 242 |
"82": "SUCCESS",
|
| 243 |
-
"83": "
|
| 244 |
"84": "SUCCESS",
|
| 245 |
"85": "SUCCESS",
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
"87": "SUCCESS",
|
| 248 |
-
"88": "
|
| 249 |
"89": "SUCCESS",
|
| 250 |
-
"9": "
|
| 251 |
"90": "SUCCESS",
|
| 252 |
"91": "SUCCESS",
|
| 253 |
"92": "SUCCESS",
|
| 254 |
-
"93": "
|
| 255 |
-
"94": "
|
| 256 |
"95": "SUCCESS",
|
| 257 |
-
"96": "
|
| 258 |
"97": "SUCCESS",
|
| 259 |
"98": "SUCCESS",
|
| 260 |
"99": "SUCCESS"
|
| 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 |
],
|
| 274 |
-
"
|
| 275 |
-
"
|
| 276 |
-
"
|
| 277 |
-
"
|
| 278 |
"initializer_range": 0.02,
|
| 279 |
-
"inner_step":
|
| 280 |
-
"
|
| 281 |
-
"last_allreduce_block":
|
| 282 |
-
"
|
| 283 |
-
"
|
| 284 |
-
"
|
| 285 |
-
"
|
| 286 |
-
"
|
| 287 |
-
"
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
| 292 |
-
"
|
| 293 |
-
"summary_activation": null,
|
| 294 |
-
"summary_first_dropout": 0.1,
|
| 295 |
-
"summary_proj_to_labels": true,
|
| 296 |
-
"summary_type": "cls_index",
|
| 297 |
-
"summary_use_proj": true,
|
| 298 |
"torch_dtype": "float32",
|
| 299 |
"transformers_version": "4.39.3",
|
| 300 |
-
"use_cache":
|
| 301 |
-
"vocab_size":
|
| 302 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "distributed/llama-1b",
|
|
|
|
| 3 |
"all_reduce_scores": {
|
| 4 |
"0": "NON_PARTICIPATING",
|
| 5 |
+
"1": "SUCCESS",
|
| 6 |
"10": "NON_PARTICIPATING",
|
| 7 |
"100": "NON_PARTICIPATING",
|
| 8 |
"101": "NON_PARTICIPATING",
|
| 9 |
"102": "NON_PARTICIPATING",
|
| 10 |
"103": "NON_PARTICIPATING",
|
| 11 |
+
"104": "SUCCESS",
|
| 12 |
"105": "NON_PARTICIPATING",
|
| 13 |
"106": "NON_PARTICIPATING",
|
| 14 |
"107": "NON_PARTICIPATING",
|
| 15 |
"108": "NON_PARTICIPATING",
|
| 16 |
"109": "NON_PARTICIPATING",
|
| 17 |
+
"11": "SUCCESS",
|
| 18 |
"110": "NON_PARTICIPATING",
|
| 19 |
"111": "NON_PARTICIPATING",
|
| 20 |
"112": "NON_PARTICIPATING",
|
|
|
|
| 23 |
"115": "NON_PARTICIPATING",
|
| 24 |
"116": "NON_PARTICIPATING",
|
| 25 |
"117": "NON_PARTICIPATING",
|
| 26 |
+
"118": "SUCCESS",
|
| 27 |
"119": "NON_PARTICIPATING",
|
| 28 |
"12": "SUCCESS",
|
| 29 |
"120": "NON_PARTICIPATING",
|
|
|
|
| 32 |
"123": "NON_PARTICIPATING",
|
| 33 |
"124": "NON_PARTICIPATING",
|
| 34 |
"125": "SUCCESS",
|
| 35 |
+
"126": "SUCCESS",
|
| 36 |
"127": "NON_PARTICIPATING",
|
| 37 |
"128": "NON_PARTICIPATING",
|
| 38 |
"129": "NON_PARTICIPATING",
|
|
|
|
| 56 |
"145": "NON_PARTICIPATING",
|
| 57 |
"146": "NON_PARTICIPATING",
|
| 58 |
"147": "NON_PARTICIPATING",
|
| 59 |
+
"148": "SUCCESS",
|
| 60 |
"149": "NON_PARTICIPATING",
|
| 61 |
"15": "SUCCESS",
|
| 62 |
"150": "NON_PARTICIPATING",
|
|
|
|
| 69 |
"157": "NON_PARTICIPATING",
|
| 70 |
"158": "NON_PARTICIPATING",
|
| 71 |
"159": "NON_PARTICIPATING",
|
| 72 |
+
"16": "SUCCESS",
|
| 73 |
"160": "NON_PARTICIPATING",
|
| 74 |
+
"161": "NON_PARTICIPATING",
|
| 75 |
+
"162": "NON_PARTICIPATING",
|
| 76 |
"163": "NON_PARTICIPATING",
|
| 77 |
"164": "NON_PARTICIPATING",
|
| 78 |
"165": "NON_PARTICIPATING",
|
| 79 |
"166": "NON_PARTICIPATING",
|
| 80 |
+
"167": "SUCCESS",
|
| 81 |
"168": "NON_PARTICIPATING",
|
| 82 |
"169": "NON_PARTICIPATING",
|
| 83 |
"17": "SUCCESS",
|
|
|
|
| 85 |
"171": "NON_PARTICIPATING",
|
| 86 |
"172": "NON_PARTICIPATING",
|
| 87 |
"173": "NON_PARTICIPATING",
|
| 88 |
+
"174": "NON_PARTICIPATING",
|
| 89 |
"175": "NON_PARTICIPATING",
|
| 90 |
"176": "NON_PARTICIPATING",
|
| 91 |
+
"177": "SUCCESS",
|
| 92 |
"178": "NON_PARTICIPATING",
|
| 93 |
"179": "NON_PARTICIPATING",
|
| 94 |
"18": "SUCCESS",
|
| 95 |
"180": "NON_PARTICIPATING",
|
| 96 |
"181": "NON_PARTICIPATING",
|
| 97 |
"182": "NON_PARTICIPATING",
|
| 98 |
+
"183": "NON_PARTICIPATING",
|
| 99 |
"184": "NON_PARTICIPATING",
|
| 100 |
"185": "NON_PARTICIPATING",
|
| 101 |
"186": "NON_PARTICIPATING",
|
|
|
|
| 105 |
"19": "SUCCESS",
|
| 106 |
"190": "NON_PARTICIPATING",
|
| 107 |
"191": "NON_PARTICIPATING",
|
| 108 |
+
"192": "NON_PARTICIPATING",
|
| 109 |
"193": "NON_PARTICIPATING",
|
| 110 |
"194": "NON_PARTICIPATING",
|
| 111 |
"195": "NON_PARTICIPATING",
|
|
|
|
| 120 |
"202": "NON_PARTICIPATING",
|
| 121 |
"203": "NON_PARTICIPATING",
|
| 122 |
"204": "NON_PARTICIPATING",
|
| 123 |
+
"205": "NON_PARTICIPATING",
|
| 124 |
"206": "NON_PARTICIPATING",
|
| 125 |
"207": "NON_PARTICIPATING",
|
| 126 |
"208": "NON_PARTICIPATING",
|
| 127 |
+
"209": "SUCCESS",
|
| 128 |
"21": "SUCCESS",
|
| 129 |
"210": "NON_PARTICIPATING",
|
| 130 |
"211": "NON_PARTICIPATING",
|
|
|
|
| 132 |
"213": "NON_PARTICIPATING",
|
| 133 |
"214": "NON_PARTICIPATING",
|
| 134 |
"215": "NON_PARTICIPATING",
|
| 135 |
+
"216": "NON_PARTICIPATING",
|
| 136 |
+
"217": "NON_PARTICIPATING",
|
| 137 |
"218": "NON_PARTICIPATING",
|
| 138 |
+
"219": "NON_PARTICIPATING",
|
| 139 |
"22": "SUCCESS",
|
| 140 |
"220": "NON_PARTICIPATING",
|
| 141 |
"221": "NON_PARTICIPATING",
|
| 142 |
"222": "NON_PARTICIPATING",
|
| 143 |
+
"223": "SUCCESS",
|
| 144 |
"224": "NON_PARTICIPATING",
|
| 145 |
"225": "NON_PARTICIPATING",
|
| 146 |
"226": "NON_PARTICIPATING",
|
| 147 |
"227": "NON_PARTICIPATING",
|
| 148 |
"228": "NON_PARTICIPATING",
|
| 149 |
"229": "NON_PARTICIPATING",
|
| 150 |
+
"23": "SUCCESS",
|
| 151 |
"230": "NON_PARTICIPATING",
|
| 152 |
"231": "NON_PARTICIPATING",
|
| 153 |
"232": "NON_PARTICIPATING",
|
|
|
|
| 166 |
"244": "NON_PARTICIPATING",
|
| 167 |
"245": "NON_PARTICIPATING",
|
| 168 |
"246": "NON_PARTICIPATING",
|
| 169 |
+
"247": "NON_PARTICIPATING",
|
| 170 |
"248": "NON_PARTICIPATING",
|
| 171 |
+
"249": "SUCCESS",
|
| 172 |
"25": "SUCCESS",
|
| 173 |
+
"250": "SUCCESS",
|
| 174 |
"251": "NON_PARTICIPATING",
|
| 175 |
"252": "NON_PARTICIPATING",
|
| 176 |
"253": "NON_PARTICIPATING",
|
| 177 |
"254": "NON_PARTICIPATING",
|
| 178 |
"255": "NON_PARTICIPATING",
|
| 179 |
+
"26": "NON_PARTICIPATING",
|
| 180 |
"27": "SUCCESS",
|
| 181 |
+
"28": "SUCCESS",
|
| 182 |
"29": "SUCCESS",
|
| 183 |
+
"3": "SUCCESS",
|
| 184 |
+
"30": "SUCCESS",
|
| 185 |
"31": "SUCCESS",
|
| 186 |
+
"32": "SUCCESS",
|
| 187 |
"33": "NON_PARTICIPATING",
|
| 188 |
"34": "SUCCESS",
|
| 189 |
"35": "SUCCESS",
|
| 190 |
+
"36": "SUCCESS",
|
| 191 |
+
"37": "SUCCESS",
|
| 192 |
"38": "SUCCESS",
|
| 193 |
+
"39": "SUCCESS",
|
| 194 |
"4": "SUCCESS",
|
| 195 |
"40": "SUCCESS",
|
| 196 |
"41": "SUCCESS",
|
| 197 |
+
"42": "SUCCESS",
|
| 198 |
+
"43": "SUCCESS",
|
| 199 |
"44": "SUCCESS",
|
| 200 |
"45": "SUCCESS",
|
| 201 |
+
"46": "SUCCESS",
|
| 202 |
"47": "SUCCESS",
|
| 203 |
"48": "NON_PARTICIPATING",
|
| 204 |
"49": "SUCCESS",
|
| 205 |
+
"5": "SUCCESS",
|
| 206 |
+
"50": "SUCCESS",
|
| 207 |
"51": "SUCCESS",
|
| 208 |
+
"52": "NON_PARTICIPATING",
|
| 209 |
+
"53": "SUCCESS",
|
| 210 |
"54": "SUCCESS",
|
| 211 |
+
"55": "SUCCESS",
|
| 212 |
"56": "SUCCESS",
|
| 213 |
+
"57": "SUCCESS",
|
| 214 |
"58": "NON_PARTICIPATING",
|
| 215 |
"59": "SUCCESS",
|
| 216 |
+
"6": "SUCCESS",
|
| 217 |
"60": "SUCCESS",
|
| 218 |
"61": "SUCCESS",
|
| 219 |
"62": "SUCCESS",
|
|
|
|
| 222 |
"65": "SUCCESS",
|
| 223 |
"66": "NON_PARTICIPATING",
|
| 224 |
"67": "SUCCESS",
|
| 225 |
+
"68": "SUCCESS",
|
| 226 |
+
"69": "SUCCESS",
|
| 227 |
"7": "SUCCESS",
|
| 228 |
"70": "SUCCESS",
|
| 229 |
"71": "SUCCESS",
|
|
|
|
| 231 |
"73": "SUCCESS",
|
| 232 |
"74": "SUCCESS",
|
| 233 |
"75": "SUCCESS",
|
| 234 |
+
"76": "SUCCESS",
|
| 235 |
"77": "SUCCESS",
|
| 236 |
+
"78": "SUCCESS",
|
| 237 |
+
"79": "SUCCESS",
|
| 238 |
"8": "SUCCESS",
|
| 239 |
"80": "SUCCESS",
|
| 240 |
+
"81": "SUCCESS",
|
| 241 |
"82": "SUCCESS",
|
| 242 |
+
"83": "SUCCESS",
|
| 243 |
"84": "SUCCESS",
|
| 244 |
"85": "SUCCESS",
|
| 245 |
"86": "NON_PARTICIPATING",
|
| 246 |
"87": "SUCCESS",
|
| 247 |
+
"88": "SUCCESS",
|
| 248 |
"89": "SUCCESS",
|
| 249 |
+
"9": "NON_PARTICIPATING",
|
| 250 |
"90": "SUCCESS",
|
| 251 |
"91": "SUCCESS",
|
| 252 |
"92": "SUCCESS",
|
| 253 |
+
"93": "SUCCESS",
|
| 254 |
+
"94": "SUCCESS",
|
| 255 |
"95": "SUCCESS",
|
| 256 |
+
"96": "SUCCESS",
|
| 257 |
"97": "SUCCESS",
|
| 258 |
"98": "SUCCESS",
|
| 259 |
"99": "SUCCESS"
|
| 260 |
},
|
| 261 |
"architectures": [
|
| 262 |
+
"LlamaForCausalLM"
|
| 263 |
],
|
| 264 |
+
"attention_bias": false,
|
| 265 |
+
"attention_dropout": 0.0,
|
|
|
|
|
|
|
|
|
|
| 266 |
"block_list": [
|
| 267 |
+
5706753,
|
| 268 |
+
5706783
|
| 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": 2,
|
| 276 |
+
"intermediate_size": 5632,
|
| 277 |
+
"last_allreduce_block": 5706512,
|
| 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 |
+
}
|
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 sha256:9b2f716c04c0cba3f9e685730d4c55dcff56c8e5f46c8654043587768087b8f4
|
| 3 |
+
size 8800564012
|
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 sha256:0a8b6a63448eac9dab06bbe49e61ab0c65c0840cb52eaac91743fe2eb7edd164
|
| 3 |
+
size 4400216536
|