Run 4. Outer Step 83. Inner Step 2.
Browse files- config.json +51 -51
- inner_optimizer.pt +1 -1
- model.safetensors +1 -1
config.json
CHANGED
|
@@ -7,7 +7,7 @@
|
|
| 7 |
"10": "NON_PARTICIPATING",
|
| 8 |
"100": "NON_PARTICIPATING",
|
| 9 |
"101": "NON_PARTICIPATING",
|
| 10 |
-
"102": "
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
"104": "NON_PARTICIPATING",
|
| 13 |
"105": "NON_PARTICIPATING",
|
|
@@ -36,41 +36,41 @@
|
|
| 36 |
"126": "NON_PARTICIPATING",
|
| 37 |
"127": "NON_PARTICIPATING",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
-
"129": "
|
| 40 |
"13": "SUCCESS",
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
"131": "NON_PARTICIPATING",
|
| 43 |
"132": "NON_PARTICIPATING",
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
"134": "NON_PARTICIPATING",
|
| 46 |
-
"135": "
|
| 47 |
"136": "NON_PARTICIPATING",
|
| 48 |
"137": "NON_PARTICIPATING",
|
| 49 |
"138": "NON_PARTICIPATING",
|
| 50 |
"139": "NON_PARTICIPATING",
|
| 51 |
-
"14": "
|
| 52 |
"140": "NON_PARTICIPATING",
|
| 53 |
"141": "NON_PARTICIPATING",
|
| 54 |
"142": "NON_PARTICIPATING",
|
| 55 |
"143": "NON_PARTICIPATING",
|
| 56 |
"144": "NON_PARTICIPATING",
|
| 57 |
-
"145": "
|
| 58 |
"146": "NON_PARTICIPATING",
|
| 59 |
"147": "NON_PARTICIPATING",
|
| 60 |
"148": "NON_PARTICIPATING",
|
| 61 |
"149": "NON_PARTICIPATING",
|
| 62 |
-
"15": "
|
| 63 |
"150": "NON_PARTICIPATING",
|
| 64 |
"151": "NON_PARTICIPATING",
|
| 65 |
"152": "NON_PARTICIPATING",
|
| 66 |
"153": "NON_PARTICIPATING",
|
| 67 |
"154": "NON_PARTICIPATING",
|
| 68 |
"155": "NON_PARTICIPATING",
|
| 69 |
-
"156": "
|
| 70 |
"157": "NON_PARTICIPATING",
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
-
"16": "
|
| 74 |
"160": "NON_PARTICIPATING",
|
| 75 |
"161": "NON_PARTICIPATING",
|
| 76 |
"162": "NON_PARTICIPATING",
|
|
@@ -103,7 +103,7 @@
|
|
| 103 |
"187": "NON_PARTICIPATING",
|
| 104 |
"188": "NON_PARTICIPATING",
|
| 105 |
"189": "NON_PARTICIPATING",
|
| 106 |
-
"19": "
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
"192": "NON_PARTICIPATING",
|
|
@@ -114,8 +114,8 @@
|
|
| 114 |
"197": "NON_PARTICIPATING",
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
-
"2": "
|
| 118 |
-
"20": "
|
| 119 |
"200": "NON_PARTICIPATING",
|
| 120 |
"201": "NON_PARTICIPATING",
|
| 121 |
"202": "NON_PARTICIPATING",
|
|
@@ -129,26 +129,26 @@
|
|
| 129 |
"21": "SUCCESS",
|
| 130 |
"210": "NON_PARTICIPATING",
|
| 131 |
"211": "NON_PARTICIPATING",
|
| 132 |
-
"212": "
|
| 133 |
"213": "NON_PARTICIPATING",
|
| 134 |
"214": "NON_PARTICIPATING",
|
| 135 |
"215": "NON_PARTICIPATING",
|
| 136 |
"216": "NON_PARTICIPATING",
|
| 137 |
-
"217": "
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
"219": "NON_PARTICIPATING",
|
| 140 |
"22": "NON_PARTICIPATING",
|
| 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",
|
|
@@ -170,7 +170,7 @@
|
|
| 170 |
"247": "NON_PARTICIPATING",
|
| 171 |
"248": "NON_PARTICIPATING",
|
| 172 |
"249": "NON_PARTICIPATING",
|
| 173 |
-
"25": "
|
| 174 |
"250": "NON_PARTICIPATING",
|
| 175 |
"251": "NON_PARTICIPATING",
|
| 176 |
"252": "NON_PARTICIPATING",
|
|
@@ -179,7 +179,7 @@
|
|
| 179 |
"255": "NON_PARTICIPATING",
|
| 180 |
"26": "NON_PARTICIPATING",
|
| 181 |
"27": "SUCCESS",
|
| 182 |
-
"28": "
|
| 183 |
"29": "NON_PARTICIPATING",
|
| 184 |
"3": "NON_PARTICIPATING",
|
| 185 |
"30": "NON_PARTICIPATING",
|
|
@@ -189,74 +189,74 @@
|
|
| 189 |
"34": "SUCCESS",
|
| 190 |
"35": "NON_PARTICIPATING",
|
| 191 |
"36": "NON_PARTICIPATING",
|
| 192 |
-
"37": "
|
| 193 |
-
"38": "
|
| 194 |
"39": "SUCCESS",
|
| 195 |
"4": "NON_PARTICIPATING",
|
| 196 |
-
"40": "
|
| 197 |
-
"41": "
|
| 198 |
-
"42": "
|
| 199 |
"43": "NON_PARTICIPATING",
|
| 200 |
"44": "SUCCESS",
|
| 201 |
"45": "NON_PARTICIPATING",
|
| 202 |
"46": "NON_PARTICIPATING",
|
| 203 |
"47": "NON_PARTICIPATING",
|
| 204 |
"48": "NON_PARTICIPATING",
|
| 205 |
-
"49": "
|
| 206 |
-
"5": "
|
| 207 |
-
"50": "
|
| 208 |
-
"51": "
|
| 209 |
-
"52": "
|
| 210 |
-
"53": "
|
| 211 |
-
"54": "
|
| 212 |
"55": "NON_PARTICIPATING",
|
| 213 |
"56": "NON_PARTICIPATING",
|
| 214 |
"57": "NON_PARTICIPATING",
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
-
"59": "
|
| 217 |
-
"6": "
|
| 218 |
-
"60": "
|
| 219 |
"61": "NON_PARTICIPATING",
|
| 220 |
"62": "NON_PARTICIPATING",
|
| 221 |
-
"63": "
|
| 222 |
"64": "SUCCESS",
|
| 223 |
-
"65": "
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "NON_PARTICIPATING",
|
| 226 |
"68": "NON_PARTICIPATING",
|
| 227 |
"69": "NON_PARTICIPATING",
|
| 228 |
"7": "NON_PARTICIPATING",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
-
"71": "
|
| 231 |
"72": "NON_PARTICIPATING",
|
| 232 |
-
"73": "
|
| 233 |
"74": "NON_PARTICIPATING",
|
| 234 |
"75": "NON_PARTICIPATING",
|
| 235 |
"76": "NON_PARTICIPATING",
|
| 236 |
-
"77": "
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
-
"8": "
|
| 240 |
-
"80": "
|
| 241 |
-
"81": "
|
| 242 |
"82": "NON_PARTICIPATING",
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
"84": "NON_PARTICIPATING",
|
| 245 |
-
"85": "
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
-
"87": "
|
| 248 |
"88": "NON_PARTICIPATING",
|
| 249 |
"89": "NON_PARTICIPATING",
|
| 250 |
"9": "SUCCESS",
|
| 251 |
-
"90": "
|
| 252 |
-
"91": "
|
| 253 |
-
"92": "
|
| 254 |
"93": "NON_PARTICIPATING",
|
| 255 |
"94": "NON_PARTICIPATING",
|
| 256 |
-
"95": "
|
| 257 |
"96": "NON_PARTICIPATING",
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
-
"98": "
|
| 260 |
"99": "NON_PARTICIPATING"
|
| 261 |
},
|
| 262 |
"architectures": [
|
|
@@ -268,17 +268,17 @@
|
|
| 268 |
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
| 269 |
},
|
| 270 |
"block_list": [
|
| 271 |
-
|
| 272 |
-
|
| 273 |
],
|
| 274 |
"block_size": 1024,
|
| 275 |
"bos_token_id": 50256,
|
| 276 |
"embd_pdrop": 0.1,
|
| 277 |
"eos_token_id": 50256,
|
| 278 |
"initializer_range": 0.02,
|
| 279 |
-
"inner_step":
|
| 280 |
"inner_steps": 0,
|
| 281 |
-
"last_allreduce_block":
|
| 282 |
"layer_norm_epsilon": 1e-05,
|
| 283 |
"model_type": "gpt_optimized",
|
| 284 |
"n_embd": 1280,
|
|
|
|
| 7 |
"10": "NON_PARTICIPATING",
|
| 8 |
"100": "NON_PARTICIPATING",
|
| 9 |
"101": "NON_PARTICIPATING",
|
| 10 |
+
"102": "NON_PARTICIPATING",
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
"104": "NON_PARTICIPATING",
|
| 13 |
"105": "NON_PARTICIPATING",
|
|
|
|
| 36 |
"126": "NON_PARTICIPATING",
|
| 37 |
"127": "NON_PARTICIPATING",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
+
"129": "SUCCESS",
|
| 40 |
"13": "SUCCESS",
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
"131": "NON_PARTICIPATING",
|
| 43 |
"132": "NON_PARTICIPATING",
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
"134": "NON_PARTICIPATING",
|
| 46 |
+
"135": "SUCCESS",
|
| 47 |
"136": "NON_PARTICIPATING",
|
| 48 |
"137": "NON_PARTICIPATING",
|
| 49 |
"138": "NON_PARTICIPATING",
|
| 50 |
"139": "NON_PARTICIPATING",
|
| 51 |
+
"14": "NON_PARTICIPATING",
|
| 52 |
"140": "NON_PARTICIPATING",
|
| 53 |
"141": "NON_PARTICIPATING",
|
| 54 |
"142": "NON_PARTICIPATING",
|
| 55 |
"143": "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",
|
| 63 |
"150": "NON_PARTICIPATING",
|
| 64 |
"151": "NON_PARTICIPATING",
|
| 65 |
"152": "NON_PARTICIPATING",
|
| 66 |
"153": "NON_PARTICIPATING",
|
| 67 |
"154": "NON_PARTICIPATING",
|
| 68 |
"155": "NON_PARTICIPATING",
|
| 69 |
+
"156": "SUCCESS",
|
| 70 |
"157": "NON_PARTICIPATING",
|
| 71 |
"158": "NON_PARTICIPATING",
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
+
"16": "NON_PARTICIPATING",
|
| 74 |
"160": "NON_PARTICIPATING",
|
| 75 |
"161": "NON_PARTICIPATING",
|
| 76 |
"162": "NON_PARTICIPATING",
|
|
|
|
| 103 |
"187": "NON_PARTICIPATING",
|
| 104 |
"188": "NON_PARTICIPATING",
|
| 105 |
"189": "NON_PARTICIPATING",
|
| 106 |
+
"19": "SUCCESS",
|
| 107 |
"190": "NON_PARTICIPATING",
|
| 108 |
"191": "NON_PARTICIPATING",
|
| 109 |
"192": "NON_PARTICIPATING",
|
|
|
|
| 114 |
"197": "NON_PARTICIPATING",
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
+
"2": "NON_PARTICIPATING",
|
| 118 |
+
"20": "SUCCESS",
|
| 119 |
"200": "NON_PARTICIPATING",
|
| 120 |
"201": "NON_PARTICIPATING",
|
| 121 |
"202": "NON_PARTICIPATING",
|
|
|
|
| 129 |
"21": "SUCCESS",
|
| 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": "SUCCESS",
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
"219": "NON_PARTICIPATING",
|
| 140 |
"22": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 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",
|
|
|
|
| 170 |
"247": "NON_PARTICIPATING",
|
| 171 |
"248": "NON_PARTICIPATING",
|
| 172 |
"249": "NON_PARTICIPATING",
|
| 173 |
+
"25": "SUCCESS",
|
| 174 |
"250": "NON_PARTICIPATING",
|
| 175 |
"251": "NON_PARTICIPATING",
|
| 176 |
"252": "NON_PARTICIPATING",
|
|
|
|
| 179 |
"255": "NON_PARTICIPATING",
|
| 180 |
"26": "NON_PARTICIPATING",
|
| 181 |
"27": "SUCCESS",
|
| 182 |
+
"28": "SUCCESS",
|
| 183 |
"29": "NON_PARTICIPATING",
|
| 184 |
"3": "NON_PARTICIPATING",
|
| 185 |
"30": "NON_PARTICIPATING",
|
|
|
|
| 189 |
"34": "SUCCESS",
|
| 190 |
"35": "NON_PARTICIPATING",
|
| 191 |
"36": "NON_PARTICIPATING",
|
| 192 |
+
"37": "NON_PARTICIPATING",
|
| 193 |
+
"38": "SUCCESS",
|
| 194 |
"39": "SUCCESS",
|
| 195 |
"4": "NON_PARTICIPATING",
|
| 196 |
+
"40": "SUCCESS",
|
| 197 |
+
"41": "NON_PARTICIPATING",
|
| 198 |
+
"42": "SUCCESS",
|
| 199 |
"43": "NON_PARTICIPATING",
|
| 200 |
"44": "SUCCESS",
|
| 201 |
"45": "NON_PARTICIPATING",
|
| 202 |
"46": "NON_PARTICIPATING",
|
| 203 |
"47": "NON_PARTICIPATING",
|
| 204 |
"48": "NON_PARTICIPATING",
|
| 205 |
+
"49": "NON_PARTICIPATING",
|
| 206 |
+
"5": "SUCCESS",
|
| 207 |
+
"50": "NON_PARTICIPATING",
|
| 208 |
+
"51": "SUCCESS",
|
| 209 |
+
"52": "SUCCESS",
|
| 210 |
+
"53": "NON_PARTICIPATING",
|
| 211 |
+
"54": "NON_PARTICIPATING",
|
| 212 |
"55": "NON_PARTICIPATING",
|
| 213 |
"56": "NON_PARTICIPATING",
|
| 214 |
"57": "NON_PARTICIPATING",
|
| 215 |
"58": "NON_PARTICIPATING",
|
| 216 |
+
"59": "SUCCESS",
|
| 217 |
+
"6": "NON_PARTICIPATING",
|
| 218 |
+
"60": "SUCCESS",
|
| 219 |
"61": "NON_PARTICIPATING",
|
| 220 |
"62": "NON_PARTICIPATING",
|
| 221 |
+
"63": "SUCCESS",
|
| 222 |
"64": "SUCCESS",
|
| 223 |
+
"65": "NON_PARTICIPATING",
|
| 224 |
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "NON_PARTICIPATING",
|
| 226 |
"68": "NON_PARTICIPATING",
|
| 227 |
"69": "NON_PARTICIPATING",
|
| 228 |
"7": "NON_PARTICIPATING",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
+
"71": "SUCCESS",
|
| 231 |
"72": "NON_PARTICIPATING",
|
| 232 |
+
"73": "NON_PARTICIPATING",
|
| 233 |
"74": "NON_PARTICIPATING",
|
| 234 |
"75": "NON_PARTICIPATING",
|
| 235 |
"76": "NON_PARTICIPATING",
|
| 236 |
+
"77": "SUCCESS",
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
+
"8": "SUCCESS",
|
| 240 |
+
"80": "NON_PARTICIPATING",
|
| 241 |
+
"81": "NON_PARTICIPATING",
|
| 242 |
"82": "NON_PARTICIPATING",
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
"84": "NON_PARTICIPATING",
|
| 245 |
+
"85": "SUCCESS",
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
+
"87": "NON_PARTICIPATING",
|
| 248 |
"88": "NON_PARTICIPATING",
|
| 249 |
"89": "NON_PARTICIPATING",
|
| 250 |
"9": "SUCCESS",
|
| 251 |
+
"90": "NON_PARTICIPATING",
|
| 252 |
+
"91": "SUCCESS",
|
| 253 |
+
"92": "NON_PARTICIPATING",
|
| 254 |
"93": "NON_PARTICIPATING",
|
| 255 |
"94": "NON_PARTICIPATING",
|
| 256 |
+
"95": "SUCCESS",
|
| 257 |
"96": "NON_PARTICIPATING",
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
+
"98": "NON_PARTICIPATING",
|
| 260 |
"99": "NON_PARTICIPATING"
|
| 261 |
},
|
| 262 |
"architectures": [
|
|
|
|
| 268 |
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
| 269 |
},
|
| 270 |
"block_list": [
|
| 271 |
+
5672604,
|
| 272 |
+
5672633
|
| 273 |
],
|
| 274 |
"block_size": 1024,
|
| 275 |
"bos_token_id": 50256,
|
| 276 |
"embd_pdrop": 0.1,
|
| 277 |
"eos_token_id": 50256,
|
| 278 |
"initializer_range": 0.02,
|
| 279 |
+
"inner_step": 2,
|
| 280 |
"inner_steps": 0,
|
| 281 |
+
"last_allreduce_block": 5672513,
|
| 282 |
"layer_norm_epsilon": 1e-05,
|
| 283 |
"model_type": "gpt_optimized",
|
| 284 |
"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 8081782026
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f5954e182ec9a9743a3385c419628998c5beb2b4088b11560f8708902137169
|
| 3 |
size 8081782026
|
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:c3bf475d86c62e5df3e6674c90229f83837544e92a20f8d128947fc3cba92d6b
|
| 3 |
size 4040701744
|