Mhair commited on
Commit
e9decbd
·
verified ·
1 Parent(s): 6d2bf51

Run 9. Outer Step 10. Inner Step 98.

Browse files
Files changed (4) hide show
  1. config.json +260 -16
  2. gradients.pt +1 -1
  3. inner_optimizer.pt +1 -1
  4. model.safetensors +1 -1
config.json CHANGED
@@ -1,18 +1,262 @@
1
  {
2
- "_name_or_path": "distributed/llama-1b",
3
  "all_reduce_scores": {
4
  "0": "NOT_ALIVE",
5
- "1": "NOT_ALIVE",
6
- "10": "FAIL",
7
- "11": "NOT_ALIVE",
8
- "2": "NOT_ALIVE",
9
- "3": "NOT_ALIVE",
10
- "4": "NOT_ALIVE",
11
- "5": "NOT_ALIVE",
12
- "6": "NOT_ALIVE",
13
- "7": "NOT_ALIVE",
14
- "8": "NOT_ALIVE",
15
- "9": "NOT_ALIVE"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  },
17
  "architectures": [
18
  "LlamaForCausalLM"
@@ -20,17 +264,17 @@
20
  "attention_bias": false,
21
  "attention_dropout": 0.0,
22
  "block_list": [
23
- 6454764,
24
- 6454794
25
  ],
26
  "bos_token_id": 1,
27
  "eos_token_id": 2,
28
  "hidden_act": "silu",
29
  "hidden_size": 2048,
30
  "initializer_range": 0.02,
31
- "inner_step": 9,
32
  "intermediate_size": 5632,
33
- "last_allreduce_block": 5428935,
34
  "max_position_embeddings": 2048,
35
  "mlp_bias": false,
36
  "model_type": "llama",
 
1
  {
2
+ "_name_or_path": "Mhair/KW",
3
  "all_reduce_scores": {
4
  "0": "NOT_ALIVE",
5
+ "1": "FAIL",
6
+ "10": "NOT_ALIVE",
7
+ "100": "NOT_ALIVE",
8
+ "101": "NOT_ALIVE",
9
+ "102": "NOT_ALIVE",
10
+ "103": "NON_PARTICIPATING",
11
+ "104": "NON_PARTICIPATING",
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": "NON_PARTICIPATING",
20
+ "112": "NOT_ALIVE",
21
+ "113": "NON_PARTICIPATING",
22
+ "114": "NON_PARTICIPATING",
23
+ "115": "NON_PARTICIPATING",
24
+ "116": "FAIL",
25
+ "117": "NON_PARTICIPATING",
26
+ "118": "NON_PARTICIPATING",
27
+ "119": "NON_PARTICIPATING",
28
+ "12": "NOT_ALIVE",
29
+ "120": "NON_PARTICIPATING",
30
+ "121": "NOT_ALIVE",
31
+ "122": "NON_PARTICIPATING",
32
+ "123": "NON_PARTICIPATING",
33
+ "124": "NON_PARTICIPATING",
34
+ "125": "NOT_ALIVE",
35
+ "126": "NON_PARTICIPATING",
36
+ "127": "FAIL",
37
+ "128": "FAIL",
38
+ "129": "NON_PARTICIPATING",
39
+ "13": "NON_PARTICIPATING",
40
+ "130": "NON_PARTICIPATING",
41
+ "131": "NOT_ALIVE",
42
+ "132": "NOT_ALIVE",
43
+ "133": "NOT_ALIVE",
44
+ "134": "NON_PARTICIPATING",
45
+ "135": "NON_PARTICIPATING",
46
+ "136": "NOT_ALIVE",
47
+ "137": "NOT_ALIVE",
48
+ "138": "NON_PARTICIPATING",
49
+ "139": "NOT_ALIVE",
50
+ "14": "NOT_ALIVE",
51
+ "140": "NOT_ALIVE",
52
+ "141": "NOT_ALIVE",
53
+ "142": "NON_PARTICIPATING",
54
+ "143": "NON_PARTICIPATING",
55
+ "144": "NON_PARTICIPATING",
56
+ "145": "NON_PARTICIPATING",
57
+ "146": "NON_PARTICIPATING",
58
+ "147": "FAIL",
59
+ "148": "FAIL",
60
+ "149": "FAIL",
61
+ "15": "NON_PARTICIPATING",
62
+ "150": "NON_PARTICIPATING",
63
+ "151": "NOT_ALIVE",
64
+ "152": "NOT_ALIVE",
65
+ "153": "NON_PARTICIPATING",
66
+ "154": "NON_PARTICIPATING",
67
+ "155": "FAIL",
68
+ "156": "FAIL",
69
+ "157": "FAIL",
70
+ "158": "NOT_ALIVE",
71
+ "159": "FAIL",
72
+ "16": "NON_PARTICIPATING",
73
+ "160": "NOT_ALIVE",
74
+ "161": "NOT_ALIVE",
75
+ "162": "NON_PARTICIPATING",
76
+ "163": "NOT_ALIVE",
77
+ "164": "NOT_ALIVE",
78
+ "165": "NON_PARTICIPATING",
79
+ "166": "NON_PARTICIPATING",
80
+ "167": "NON_PARTICIPATING",
81
+ "168": "NON_PARTICIPATING",
82
+ "169": "FAIL",
83
+ "17": "FAIL",
84
+ "170": "NON_PARTICIPATING",
85
+ "171": "NON_PARTICIPATING",
86
+ "172": "NOT_ALIVE",
87
+ "173": "FAIL",
88
+ "174": "NOT_ALIVE",
89
+ "175": "FAIL",
90
+ "176": "NOT_ALIVE",
91
+ "177": "NON_PARTICIPATING",
92
+ "178": "NON_PARTICIPATING",
93
+ "179": "NOT_ALIVE",
94
+ "18": "NOT_ALIVE",
95
+ "180": "NOT_ALIVE",
96
+ "181": "NON_PARTICIPATING",
97
+ "182": "NON_PARTICIPATING",
98
+ "183": "NON_PARTICIPATING",
99
+ "184": "FAIL",
100
+ "185": "NOT_ALIVE",
101
+ "186": "FAIL",
102
+ "187": "NOT_ALIVE",
103
+ "188": "FAIL",
104
+ "189": "NOT_ALIVE",
105
+ "19": "NON_PARTICIPATING",
106
+ "190": "NON_PARTICIPATING",
107
+ "191": "NOT_ALIVE",
108
+ "192": "NON_PARTICIPATING",
109
+ "193": "NON_PARTICIPATING",
110
+ "194": "FAIL",
111
+ "195": "NON_PARTICIPATING",
112
+ "196": "NON_PARTICIPATING",
113
+ "197": "FAIL",
114
+ "198": "FAIL",
115
+ "199": "FAIL",
116
+ "2": "FAIL",
117
+ "20": "FAIL",
118
+ "200": "NOT_ALIVE",
119
+ "201": "NOT_ALIVE",
120
+ "202": "NON_PARTICIPATING",
121
+ "203": "NON_PARTICIPATING",
122
+ "204": "NOT_ALIVE",
123
+ "205": "NOT_ALIVE",
124
+ "206": "FAIL",
125
+ "207": "NOT_ALIVE",
126
+ "208": "FAIL",
127
+ "209": "NON_PARTICIPATING",
128
+ "21": "FAIL",
129
+ "210": "FAIL",
130
+ "211": "NON_PARTICIPATING",
131
+ "212": "FAIL",
132
+ "213": "NON_PARTICIPATING",
133
+ "214": "NON_PARTICIPATING",
134
+ "215": "NOT_ALIVE",
135
+ "216": "FAIL",
136
+ "217": "FAIL",
137
+ "218": "FAIL",
138
+ "219": "NON_PARTICIPATING",
139
+ "22": "NON_PARTICIPATING",
140
+ "220": "FAIL",
141
+ "221": "FAIL",
142
+ "222": "FAIL",
143
+ "223": "FAIL",
144
+ "224": "NOT_ALIVE",
145
+ "225": "FAIL",
146
+ "226": "NON_PARTICIPATING",
147
+ "227": "FAIL",
148
+ "228": "NON_PARTICIPATING",
149
+ "229": "NOT_ALIVE",
150
+ "23": "NON_PARTICIPATING",
151
+ "230": "NOT_ALIVE",
152
+ "231": "NON_PARTICIPATING",
153
+ "232": "FAIL",
154
+ "233": "NON_PARTICIPATING",
155
+ "234": "FAIL",
156
+ "235": "NON_PARTICIPATING",
157
+ "236": "NOT_ALIVE",
158
+ "237": "FAIL",
159
+ "238": "NON_PARTICIPATING",
160
+ "239": "FAIL",
161
+ "24": "FAIL",
162
+ "240": "NON_PARTICIPATING",
163
+ "241": "FAIL",
164
+ "242": "NOT_ALIVE",
165
+ "243": "FAIL",
166
+ "244": "NOT_ALIVE",
167
+ "245": "NON_PARTICIPATING",
168
+ "246": "NON_PARTICIPATING",
169
+ "247": "FAIL",
170
+ "248": "NON_PARTICIPATING",
171
+ "249": "NON_PARTICIPATING",
172
+ "25": "SUCCESS",
173
+ "250": "NON_PARTICIPATING",
174
+ "251": "NON_PARTICIPATING",
175
+ "252": "NON_PARTICIPATING",
176
+ "253": "NOT_ALIVE",
177
+ "254": "NOT_ALIVE",
178
+ "255": "NON_PARTICIPATING",
179
+ "26": "NOT_ALIVE",
180
+ "27": "FAIL",
181
+ "28": "NON_PARTICIPATING",
182
+ "29": "NON_PARTICIPATING",
183
+ "3": "NON_PARTICIPATING",
184
+ "30": "NON_PARTICIPATING",
185
+ "31": "NOT_ALIVE",
186
+ "32": "NOT_ALIVE",
187
+ "33": "NOT_ALIVE",
188
+ "34": "NOT_ALIVE",
189
+ "35": "FAIL",
190
+ "36": "FAIL",
191
+ "37": "NOT_ALIVE",
192
+ "38": "NOT_ALIVE",
193
+ "39": "FAIL",
194
+ "4": "NON_PARTICIPATING",
195
+ "40": "NOT_ALIVE",
196
+ "41": "NOT_ALIVE",
197
+ "42": "NON_PARTICIPATING",
198
+ "43": "FAIL",
199
+ "44": "FAIL",
200
+ "45": "NON_PARTICIPATING",
201
+ "46": "NON_PARTICIPATING",
202
+ "47": "NON_PARTICIPATING",
203
+ "48": "NOT_ALIVE",
204
+ "49": "NON_PARTICIPATING",
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": "NON_PARTICIPATING",
212
+ "56": "NON_PARTICIPATING",
213
+ "57": "NON_PARTICIPATING",
214
+ "58": "NOT_ALIVE",
215
+ "59": "NON_PARTICIPATING",
216
+ "6": "NON_PARTICIPATING",
217
+ "60": "FAIL",
218
+ "61": "FAIL",
219
+ "62": "NON_PARTICIPATING",
220
+ "63": "NON_PARTICIPATING",
221
+ "64": "NON_PARTICIPATING",
222
+ "65": "NON_PARTICIPATING",
223
+ "66": "NON_PARTICIPATING",
224
+ "67": "NON_PARTICIPATING",
225
+ "68": "NOT_ALIVE",
226
+ "69": "NON_PARTICIPATING",
227
+ "7": "FAIL",
228
+ "70": "NON_PARTICIPATING",
229
+ "71": "NOT_ALIVE",
230
+ "72": "FAIL",
231
+ "73": "NON_PARTICIPATING",
232
+ "74": "NON_PARTICIPATING",
233
+ "75": "NON_PARTICIPATING",
234
+ "76": "NON_PARTICIPATING",
235
+ "77": "NOT_ALIVE",
236
+ "78": "NON_PARTICIPATING",
237
+ "79": "NON_PARTICIPATING",
238
+ "8": "NON_PARTICIPATING",
239
+ "80": "NOT_ALIVE",
240
+ "81": "NOT_ALIVE",
241
+ "82": "NOT_ALIVE",
242
+ "83": "NON_PARTICIPATING",
243
+ "84": "NOT_ALIVE",
244
+ "85": "NON_PARTICIPATING",
245
+ "86": "NON_PARTICIPATING",
246
+ "87": "FAIL",
247
+ "88": "NON_PARTICIPATING",
248
+ "89": "NON_PARTICIPATING",
249
+ "9": "NON_PARTICIPATING",
250
+ "90": "NON_PARTICIPATING",
251
+ "91": "NON_PARTICIPATING",
252
+ "92": "NOT_ALIVE",
253
+ "93": "NOT_ALIVE",
254
+ "94": "FAIL",
255
+ "95": "NOT_ALIVE",
256
+ "96": "NON_PARTICIPATING",
257
+ "97": "NON_PARTICIPATING",
258
+ "98": "NON_PARTICIPATING",
259
+ "99": "FAIL"
260
  },
261
  "architectures": [
262
  "LlamaForCausalLM"
 
264
  "attention_bias": false,
265
  "attention_dropout": 0.0,
266
  "block_list": [
267
+ 6454933,
268
+ 6454962
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": 98,
276
  "intermediate_size": 5632,
277
+ "last_allreduce_block": 6452845,
278
  "max_position_embeddings": 2048,
279
  "mlp_bias": false,
280
  "model_type": "llama",
gradients.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:96a14aa57f40a8c7fb6088fff54d2aa50ca396d3bcb72e510f5763a404e48b78
3
  size 88389626
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecf49fd6d9f26b38e83a7697352785e896976632d2c45062f81a6b1facf5e8ef
3
  size 88389626
inner_optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c817cdfc0a98bb53f4c20f368c8d1d76cd2ea7203cd54dcd5e5710fe4c513477
3
  size 4924926906
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d7eb4aa5e7080e9ccf6f34e2348b89b60644c59f6cd39dbba2d883a8cf429d8
3
  size 4924926906
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbfd11b7e117ed271baf7a014c70c9bdf862235f49841c0cb3c54b602e3b43a1
3
  size 4400216536
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:275305a3276071a0530a49877c520dd258f0c5d72fcea322e64352f8ca6d248d
3
  size 4400216536