e3eb67b9bd63b710aa1e13f823137d2b5ec5d1f622a5162a486b7bbc74ec66a6
Browse files- config.json +51 -51
- model.safetensors +1 -1
- smash_config.json +1 -1
config.json
CHANGED
|
@@ -1,32 +1,32 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "/covalent/.cache/models/
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"all_reduce_scores": {
|
| 5 |
"0": "NON_PARTICIPATING",
|
| 6 |
"1": "NON_PARTICIPATING",
|
| 7 |
-
"10": "
|
| 8 |
"100": "NON_PARTICIPATING",
|
| 9 |
-
"101": "
|
| 10 |
"102": "NON_PARTICIPATING",
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
-
"104": "
|
| 13 |
-
"105": "
|
| 14 |
"106": "NON_PARTICIPATING",
|
| 15 |
-
"107": "
|
| 16 |
"108": "NON_PARTICIPATING",
|
| 17 |
"109": "NON_PARTICIPATING",
|
| 18 |
-
"11": "
|
| 19 |
"110": "NON_PARTICIPATING",
|
| 20 |
-
"111": "
|
| 21 |
"112": "NON_PARTICIPATING",
|
| 22 |
"113": "NON_PARTICIPATING",
|
| 23 |
"114": "NON_PARTICIPATING",
|
| 24 |
-
"115": "
|
| 25 |
-
"116": "
|
| 26 |
"117": "NON_PARTICIPATING",
|
| 27 |
"118": "NON_PARTICIPATING",
|
| 28 |
-
"119": "
|
| 29 |
-
"12": "
|
| 30 |
"120": "NON_PARTICIPATING",
|
| 31 |
"121": "NON_PARTICIPATING",
|
| 32 |
"122": "NON_PARTICIPATING",
|
|
@@ -37,22 +37,22 @@
|
|
| 37 |
"127": "NON_PARTICIPATING",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
| 40 |
-
"13": "
|
| 41 |
"130": "NON_PARTICIPATING",
|
| 42 |
"131": "NON_PARTICIPATING",
|
| 43 |
"132": "NON_PARTICIPATING",
|
| 44 |
"133": "NON_PARTICIPATING",
|
| 45 |
-
"134": "
|
| 46 |
"135": "NON_PARTICIPATING",
|
| 47 |
"136": "NON_PARTICIPATING",
|
| 48 |
"137": "NON_PARTICIPATING",
|
| 49 |
"138": "NON_PARTICIPATING",
|
| 50 |
"139": "NON_PARTICIPATING",
|
| 51 |
-
"14": "
|
| 52 |
"140": "NON_PARTICIPATING",
|
| 53 |
-
"141": "
|
| 54 |
-
"142": "
|
| 55 |
-
"143": "
|
| 56 |
"144": "NON_PARTICIPATING",
|
| 57 |
"145": "NON_PARTICIPATING",
|
| 58 |
"146": "NON_PARTICIPATING",
|
|
@@ -61,21 +61,21 @@
|
|
| 61 |
"149": "NON_PARTICIPATING",
|
| 62 |
"15": "NON_PARTICIPATING",
|
| 63 |
"150": "NON_PARTICIPATING",
|
| 64 |
-
"151": "
|
| 65 |
"152": "NON_PARTICIPATING",
|
| 66 |
"153": "NON_PARTICIPATING",
|
| 67 |
"154": "NON_PARTICIPATING",
|
| 68 |
"155": "NON_PARTICIPATING",
|
| 69 |
"156": "NON_PARTICIPATING",
|
| 70 |
"157": "NON_PARTICIPATING",
|
| 71 |
-
"158": "
|
| 72 |
"159": "NON_PARTICIPATING",
|
| 73 |
-
"16": "
|
| 74 |
"160": "NON_PARTICIPATING",
|
| 75 |
"161": "NON_PARTICIPATING",
|
| 76 |
"162": "NON_PARTICIPATING",
|
| 77 |
"163": "NON_PARTICIPATING",
|
| 78 |
-
"164": "
|
| 79 |
"165": "NON_PARTICIPATING",
|
| 80 |
"166": "NON_PARTICIPATING",
|
| 81 |
"167": "NON_PARTICIPATING",
|
|
@@ -85,36 +85,36 @@
|
|
| 85 |
"170": "NON_PARTICIPATING",
|
| 86 |
"171": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 98 |
-
"182": "
|
| 99 |
"183": "NON_PARTICIPATING",
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
| 102 |
"186": "NON_PARTICIPATING",
|
| 103 |
"187": "NON_PARTICIPATING",
|
| 104 |
"188": "NON_PARTICIPATING",
|
| 105 |
-
"189": "
|
| 106 |
"19": "NON_PARTICIPATING",
|
| 107 |
-
"190": "
|
| 108 |
-
"191": "
|
| 109 |
"192": "NON_PARTICIPATING",
|
| 110 |
-
"193": "
|
| 111 |
-
"194": "
|
| 112 |
"195": "NON_PARTICIPATING",
|
| 113 |
-
"196": "
|
| 114 |
"197": "NON_PARTICIPATING",
|
| 115 |
"198": "NON_PARTICIPATING",
|
| 116 |
"199": "NON_PARTICIPATING",
|
| 117 |
-
"2": "
|
| 118 |
"20": "NON_PARTICIPATING",
|
| 119 |
"200": "NON_PARTICIPATING",
|
| 120 |
"201": "NON_PARTICIPATING",
|
|
@@ -124,7 +124,7 @@
|
|
| 124 |
"205": "NON_PARTICIPATING",
|
| 125 |
"206": "NON_PARTICIPATING",
|
| 126 |
"207": "NON_PARTICIPATING",
|
| 127 |
-
"208": "
|
| 128 |
"209": "NON_PARTICIPATING",
|
| 129 |
"21": "NON_PARTICIPATING",
|
| 130 |
"210": "NON_PARTICIPATING",
|
|
@@ -133,7 +133,7 @@
|
|
| 133 |
"213": "NON_PARTICIPATING",
|
| 134 |
"214": "NON_PARTICIPATING",
|
| 135 |
"215": "NON_PARTICIPATING",
|
| 136 |
-
"216": "
|
| 137 |
"217": "NON_PARTICIPATING",
|
| 138 |
"218": "NON_PARTICIPATING",
|
| 139 |
"219": "NON_PARTICIPATING",
|
|
@@ -182,17 +182,17 @@
|
|
| 182 |
"28": "NON_PARTICIPATING",
|
| 183 |
"29": "NON_PARTICIPATING",
|
| 184 |
"3": "SUCCESS",
|
| 185 |
-
"30": "
|
| 186 |
"31": "NON_PARTICIPATING",
|
| 187 |
"32": "NON_PARTICIPATING",
|
| 188 |
"33": "NON_PARTICIPATING",
|
| 189 |
-
"34": "
|
| 190 |
"35": "NON_PARTICIPATING",
|
| 191 |
"36": "NON_PARTICIPATING",
|
| 192 |
"37": "NON_PARTICIPATING",
|
| 193 |
"38": "NON_PARTICIPATING",
|
| 194 |
"39": "NON_PARTICIPATING",
|
| 195 |
-
"4": "
|
| 196 |
"40": "NON_PARTICIPATING",
|
| 197 |
"41": "NON_PARTICIPATING",
|
| 198 |
"42": "NON_PARTICIPATING",
|
|
@@ -212,49 +212,49 @@
|
|
| 212 |
"55": "SUCCESS",
|
| 213 |
"56": "NON_PARTICIPATING",
|
| 214 |
"57": "NON_PARTICIPATING",
|
| 215 |
-
"58": "
|
| 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": "NON_PARTICIPATING",
|
| 224 |
-
"66": "
|
| 225 |
"67": "SUCCESS",
|
| 226 |
"68": "NON_PARTICIPATING",
|
| 227 |
"69": "NON_PARTICIPATING",
|
| 228 |
"7": "SUCCESS",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
"71": "NON_PARTICIPATING",
|
| 231 |
-
"72": "
|
| 232 |
"73": "NON_PARTICIPATING",
|
| 233 |
-
"74": "
|
| 234 |
"75": "NON_PARTICIPATING",
|
| 235 |
"76": "NON_PARTICIPATING",
|
| 236 |
"77": "NON_PARTICIPATING",
|
| 237 |
"78": "NON_PARTICIPATING",
|
| 238 |
"79": "NON_PARTICIPATING",
|
| 239 |
"8": "NON_PARTICIPATING",
|
| 240 |
-
"80": "
|
| 241 |
"81": "NON_PARTICIPATING",
|
| 242 |
-
"82": "
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
-
"84": "
|
| 245 |
"85": "NON_PARTICIPATING",
|
| 246 |
"86": "NON_PARTICIPATING",
|
| 247 |
"87": "NON_PARTICIPATING",
|
| 248 |
"88": "NON_PARTICIPATING",
|
| 249 |
"89": "NON_PARTICIPATING",
|
| 250 |
-
"9": "
|
| 251 |
"90": "NON_PARTICIPATING",
|
| 252 |
-
"91": "
|
| 253 |
"92": "NON_PARTICIPATING",
|
| 254 |
-
"93": "
|
| 255 |
-
"94": "
|
| 256 |
-
"95": "
|
| 257 |
-
"96": "
|
| 258 |
"97": "NON_PARTICIPATING",
|
| 259 |
"98": "NON_PARTICIPATING",
|
| 260 |
"99": "NON_PARTICIPATING"
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/covalent/.cache/models/tmpu6vuwt5l5tm7idek",
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"all_reduce_scores": {
|
| 5 |
"0": "NON_PARTICIPATING",
|
| 6 |
"1": "NON_PARTICIPATING",
|
| 7 |
+
"10": "SUCCESS",
|
| 8 |
"100": "NON_PARTICIPATING",
|
| 9 |
+
"101": "NON_PARTICIPATING",
|
| 10 |
"102": "NON_PARTICIPATING",
|
| 11 |
"103": "NON_PARTICIPATING",
|
| 12 |
+
"104": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 27 |
"118": "NON_PARTICIPATING",
|
| 28 |
+
"119": "NON_PARTICIPATING",
|
| 29 |
+
"12": "NON_PARTICIPATING",
|
| 30 |
"120": "NON_PARTICIPATING",
|
| 31 |
"121": "NON_PARTICIPATING",
|
| 32 |
"122": "NON_PARTICIPATING",
|
|
|
|
| 37 |
"127": "NON_PARTICIPATING",
|
| 38 |
"128": "NON_PARTICIPATING",
|
| 39 |
"129": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 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",
|
|
|
|
| 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": "NON_PARTICIPATING",
|
| 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",
|
| 77 |
"163": "NON_PARTICIPATING",
|
| 78 |
+
"164": "NON_PARTICIPATING",
|
| 79 |
"165": "NON_PARTICIPATING",
|
| 80 |
"166": "NON_PARTICIPATING",
|
| 81 |
"167": "NON_PARTICIPATING",
|
|
|
|
| 85 |
"170": "NON_PARTICIPATING",
|
| 86 |
"171": "NON_PARTICIPATING",
|
| 87 |
"172": "NON_PARTICIPATING",
|
| 88 |
+
"173": "NON_PARTICIPATING",
|
| 89 |
"174": "NON_PARTICIPATING",
|
| 90 |
"175": "NON_PARTICIPATING",
|
| 91 |
+
"176": "NON_PARTICIPATING",
|
| 92 |
"177": "NON_PARTICIPATING",
|
| 93 |
"178": "NON_PARTICIPATING",
|
| 94 |
"179": "NON_PARTICIPATING",
|
| 95 |
"18": "NON_PARTICIPATING",
|
| 96 |
"180": "NON_PARTICIPATING",
|
| 97 |
"181": "NON_PARTICIPATING",
|
| 98 |
+
"182": "NON_PARTICIPATING",
|
| 99 |
"183": "NON_PARTICIPATING",
|
| 100 |
"184": "NON_PARTICIPATING",
|
| 101 |
"185": "NON_PARTICIPATING",
|
| 102 |
"186": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 120 |
"201": "NON_PARTICIPATING",
|
|
|
|
| 124 |
"205": "NON_PARTICIPATING",
|
| 125 |
"206": "NON_PARTICIPATING",
|
| 126 |
"207": "NON_PARTICIPATING",
|
| 127 |
+
"208": "NON_PARTICIPATING",
|
| 128 |
"209": "NON_PARTICIPATING",
|
| 129 |
"21": "NON_PARTICIPATING",
|
| 130 |
"210": "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",
|
|
|
|
| 182 |
"28": "NON_PARTICIPATING",
|
| 183 |
"29": "NON_PARTICIPATING",
|
| 184 |
"3": "SUCCESS",
|
| 185 |
+
"30": "NON_PARTICIPATING",
|
| 186 |
"31": "NON_PARTICIPATING",
|
| 187 |
"32": "NON_PARTICIPATING",
|
| 188 |
"33": "NON_PARTICIPATING",
|
| 189 |
+
"34": "SUCCESS",
|
| 190 |
"35": "NON_PARTICIPATING",
|
| 191 |
"36": "NON_PARTICIPATING",
|
| 192 |
"37": "NON_PARTICIPATING",
|
| 193 |
"38": "NON_PARTICIPATING",
|
| 194 |
"39": "NON_PARTICIPATING",
|
| 195 |
+
"4": "NON_PARTICIPATING",
|
| 196 |
"40": "NON_PARTICIPATING",
|
| 197 |
"41": "NON_PARTICIPATING",
|
| 198 |
"42": "NON_PARTICIPATING",
|
|
|
|
| 212 |
"55": "SUCCESS",
|
| 213 |
"56": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 221 |
"63": "NON_PARTICIPATING",
|
| 222 |
"64": "NON_PARTICIPATING",
|
| 223 |
"65": "NON_PARTICIPATING",
|
| 224 |
+
"66": "NON_PARTICIPATING",
|
| 225 |
"67": "SUCCESS",
|
| 226 |
"68": "NON_PARTICIPATING",
|
| 227 |
"69": "NON_PARTICIPATING",
|
| 228 |
"7": "SUCCESS",
|
| 229 |
"70": "NON_PARTICIPATING",
|
| 230 |
"71": "NON_PARTICIPATING",
|
| 231 |
+
"72": "NON_PARTICIPATING",
|
| 232 |
"73": "NON_PARTICIPATING",
|
| 233 |
+
"74": "NON_PARTICIPATING",
|
| 234 |
"75": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING",
|
| 242 |
+
"82": "NON_PARTICIPATING",
|
| 243 |
"83": "NON_PARTICIPATING",
|
| 244 |
+
"84": "NON_PARTICIPATING",
|
| 245 |
"85": "NON_PARTICIPATING",
|
| 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": "NON_PARTICIPATING"
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1207575528
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2d49bcb8d682557c99d6a494c3886ea7f787c74c8a2884704f1d6ddf6f59a59
|
| 3 |
size 1207575528
|
smash_config.json
CHANGED
|
@@ -28,7 +28,7 @@
|
|
| 28 |
"quant_llm-int8_weight_bits": 8,
|
| 29 |
"max_batch_size": 1,
|
| 30 |
"device": "cuda",
|
| 31 |
-
"cache_dir": "/covalent/.cache/models/
|
| 32 |
"task": "",
|
| 33 |
"save_load_fn": "bitsandbytes",
|
| 34 |
"save_load_fn_args": {}
|
|
|
|
| 28 |
"quant_llm-int8_weight_bits": 8,
|
| 29 |
"max_batch_size": 1,
|
| 30 |
"device": "cuda",
|
| 31 |
+
"cache_dir": "/covalent/.cache/models/tmpu6vuwt5l",
|
| 32 |
"task": "",
|
| 33 |
"save_load_fn": "bitsandbytes",
|
| 34 |
"save_load_fn_args": {}
|