ncaagcc commited on
Commit
1b53585
·
verified ·
1 Parent(s): acd209c

Run 9. Outer Step 10. Inner Step 92.

Browse files
Files changed (4) hide show
  1. config.json +259 -14
  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 +264,18 @@
20
  "attention_bias": false,
21
  "attention_dropout": 0.0,
22
  "block_list": [
23
- 6454806,
24
- 6454808
 
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": 13,
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": "ncaagcc/trt2",
3
  "all_reduce_scores": {
4
  "0": "NOT_ALIVE",
5
+ "1": "FAIL",
6
+ "10": "NON_PARTICIPATING",
7
+ "100": "NOT_ALIVE",
8
+ "101": "NOT_ALIVE",
9
+ "102": "NOT_ALIVE",
10
+ "103": "NON_PARTICIPATING",
11
+ "104": "FAIL",
12
+ "105": "NON_PARTICIPATING",
13
+ "106": "NON_PARTICIPATING",
14
+ "107": "NON_PARTICIPATING",
15
+ "108": "NOT_ALIVE",
16
+ "109": "FAIL",
17
+ "11": "NON_PARTICIPATING",
18
+ "110": "NOT_ALIVE",
19
+ "111": "FAIL",
20
+ "112": "NON_PARTICIPATING",
21
+ "113": "NOT_ALIVE",
22
+ "114": "NON_PARTICIPATING",
23
+ "115": "NON_PARTICIPATING",
24
+ "116": "NOT_ALIVE",
25
+ "117": "NON_PARTICIPATING",
26
+ "118": "NON_PARTICIPATING",
27
+ "119": "NOT_ALIVE",
28
+ "12": "NON_PARTICIPATING",
29
+ "120": "NON_PARTICIPATING",
30
+ "121": "NOT_ALIVE",
31
+ "122": "FAIL",
32
+ "123": "NON_PARTICIPATING",
33
+ "124": "FAIL",
34
+ "125": "NON_PARTICIPATING",
35
+ "126": "NON_PARTICIPATING",
36
+ "127": "FAIL",
37
+ "128": "NOT_ALIVE",
38
+ "129": "NON_PARTICIPATING",
39
+ "13": "FAIL",
40
+ "130": "NOT_ALIVE",
41
+ "131": "NOT_ALIVE",
42
+ "132": "NOT_ALIVE",
43
+ "133": "NON_PARTICIPATING",
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": "NOT_ALIVE",
56
+ "145": "NON_PARTICIPATING",
57
+ "146": "FAIL",
58
+ "147": "NON_PARTICIPATING",
59
+ "148": "NOT_ALIVE",
60
+ "149": "NON_PARTICIPATING",
61
+ "15": "NON_PARTICIPATING",
62
+ "150": "NON_PARTICIPATING",
63
+ "151": "NOT_ALIVE",
64
+ "152": "NOT_ALIVE",
65
+ "153": "FAIL",
66
+ "154": "NON_PARTICIPATING",
67
+ "155": "NOT_ALIVE",
68
+ "156": "FAIL",
69
+ "157": "FAIL",
70
+ "158": "NON_PARTICIPATING",
71
+ "159": "NON_PARTICIPATING",
72
+ "16": "NON_PARTICIPATING",
73
+ "160": "NOT_ALIVE",
74
+ "161": "NOT_ALIVE",
75
+ "162": "FAIL",
76
+ "163": "NOT_ALIVE",
77
+ "164": "NOT_ALIVE",
78
+ "165": "NON_PARTICIPATING",
79
+ "166": "NON_PARTICIPATING",
80
+ "167": "NOT_ALIVE",
81
+ "168": "NON_PARTICIPATING",
82
+ "169": "FAIL",
83
+ "17": "FAIL",
84
+ "170": "NON_PARTICIPATING",
85
+ "171": "NOT_ALIVE",
86
+ "172": "NOT_ALIVE",
87
+ "173": "FAIL",
88
+ "174": "NOT_ALIVE",
89
+ "175": "NON_PARTICIPATING",
90
+ "176": "NOT_ALIVE",
91
+ "177": "NON_PARTICIPATING",
92
+ "178": "NON_PARTICIPATING",
93
+ "179": "NON_PARTICIPATING",
94
+ "18": "FAIL",
95
+ "180": "FAIL",
96
+ "181": "FAIL",
97
+ "182": "FAIL",
98
+ "183": "NON_PARTICIPATING",
99
+ "184": "NON_PARTICIPATING",
100
+ "185": "NOT_ALIVE",
101
+ "186": "FAIL",
102
+ "187": "NOT_ALIVE",
103
+ "188": "NOT_ALIVE",
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": "NON_PARTICIPATING",
111
+ "195": "NON_PARTICIPATING",
112
+ "196": "NON_PARTICIPATING",
113
+ "197": "NON_PARTICIPATING",
114
+ "198": "FAIL",
115
+ "199": "FAIL",
116
+ "2": "NON_PARTICIPATING",
117
+ "20": "NON_PARTICIPATING",
118
+ "200": "NON_PARTICIPATING",
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": "NOT_ALIVE",
127
+ "209": "FAIL",
128
+ "21": "FAIL",
129
+ "210": "FAIL",
130
+ "211": "NON_PARTICIPATING",
131
+ "212": "NON_PARTICIPATING",
132
+ "213": "NOT_ALIVE",
133
+ "214": "FAIL",
134
+ "215": "NON_PARTICIPATING",
135
+ "216": "FAIL",
136
+ "217": "NON_PARTICIPATING",
137
+ "218": "NOT_ALIVE",
138
+ "219": "NON_PARTICIPATING",
139
+ "22": "NON_PARTICIPATING",
140
+ "220": "NOT_ALIVE",
141
+ "221": "NON_PARTICIPATING",
142
+ "222": "NOT_ALIVE",
143
+ "223": "NON_PARTICIPATING",
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": "NON_PARTICIPATING",
154
+ "233": "FAIL",
155
+ "234": "FAIL",
156
+ "235": "NON_PARTICIPATING",
157
+ "236": "FAIL",
158
+ "237": "NOT_ALIVE",
159
+ "238": "NON_PARTICIPATING",
160
+ "239": "NON_PARTICIPATING",
161
+ "24": "NON_PARTICIPATING",
162
+ "240": "NOT_ALIVE",
163
+ "241": "NON_PARTICIPATING",
164
+ "242": "NOT_ALIVE",
165
+ "243": "NON_PARTICIPATING",
166
+ "244": "NOT_ALIVE",
167
+ "245": "NOT_ALIVE",
168
+ "246": "FAIL",
169
+ "247": "FAIL",
170
+ "248": "NOT_ALIVE",
171
+ "249": "NON_PARTICIPATING",
172
+ "25": "SUCCESS",
173
+ "250": "NON_PARTICIPATING",
174
+ "251": "NON_PARTICIPATING",
175
+ "252": "NON_PARTICIPATING",
176
+ "253": "NON_PARTICIPATING",
177
+ "254": "NON_PARTICIPATING",
178
+ "255": "NON_PARTICIPATING",
179
+ "26": "FAIL",
180
+ "27": "FAIL",
181
+ "28": "NON_PARTICIPATING",
182
+ "29": "NON_PARTICIPATING",
183
+ "3": "NON_PARTICIPATING",
184
+ "30": "NON_PARTICIPATING",
185
+ "31": "FAIL",
186
+ "32": "NOT_ALIVE",
187
+ "33": "NOT_ALIVE",
188
+ "34": "NON_PARTICIPATING",
189
+ "35": "NON_PARTICIPATING",
190
+ "36": "NON_PARTICIPATING",
191
+ "37": "NOT_ALIVE",
192
+ "38": "FAIL",
193
+ "39": "NON_PARTICIPATING",
194
+ "4": "FAIL",
195
+ "40": "NOT_ALIVE",
196
+ "41": "NOT_ALIVE",
197
+ "42": "NON_PARTICIPATING",
198
+ "43": "NOT_ALIVE",
199
+ "44": "FAIL",
200
+ "45": "FAIL",
201
+ "46": "NON_PARTICIPATING",
202
+ "47": "FAIL",
203
+ "48": "NON_PARTICIPATING",
204
+ "49": "NON_PARTICIPATING",
205
+ "5": "NON_PARTICIPATING",
206
+ "50": "NON_PARTICIPATING",
207
+ "51": "NOT_ALIVE",
208
+ "52": "FAIL",
209
+ "53": "NON_PARTICIPATING",
210
+ "54": "NOT_ALIVE",
211
+ "55": "NON_PARTICIPATING",
212
+ "56": "FAIL",
213
+ "57": "NON_PARTICIPATING",
214
+ "58": "NOT_ALIVE",
215
+ "59": "NON_PARTICIPATING",
216
  "6": "NOT_ALIVE",
217
+ "60": "NON_PARTICIPATING",
218
+ "61": "FAIL",
219
+ "62": "NON_PARTICIPATING",
220
+ "63": "NON_PARTICIPATING",
221
+ "64": "NOT_ALIVE",
222
+ "65": "NON_PARTICIPATING",
223
+ "66": "NON_PARTICIPATING",
224
+ "67": "NON_PARTICIPATING",
225
+ "68": "NOT_ALIVE",
226
+ "69": "FAIL",
227
  "7": "NOT_ALIVE",
228
+ "70": "NON_PARTICIPATING",
229
+ "71": "NOT_ALIVE",
230
+ "72": "NON_PARTICIPATING",
231
+ "73": "FAIL",
232
+ "74": "FAIL",
233
+ "75": "FAIL",
234
+ "76": "FAIL",
235
+ "77": "NOT_ALIVE",
236
+ "78": "NOT_ALIVE",
237
+ "79": "NON_PARTICIPATING",
238
+ "8": "NON_PARTICIPATING",
239
+ "80": "FAIL",
240
+ "81": "NOT_ALIVE",
241
+ "82": "NON_PARTICIPATING",
242
+ "83": "NON_PARTICIPATING",
243
+ "84": "FAIL",
244
+ "85": "NON_PARTICIPATING",
245
+ "86": "NOT_ALIVE",
246
+ "87": "FAIL",
247
+ "88": "NON_PARTICIPATING",
248
+ "89": "NON_PARTICIPATING",
249
+ "9": "NOT_ALIVE",
250
+ "90": "NON_PARTICIPATING",
251
+ "91": "FAIL",
252
+ "92": "NON_PARTICIPATING",
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": "NON_PARTICIPATING"
260
  },
261
  "architectures": [
262
  "LlamaForCausalLM"
 
264
  "attention_bias": false,
265
  "attention_dropout": 0.0,
266
  "block_list": [
267
+ 6454946,
268
+ 6454952,
269
+ 6454962
270
  ],
271
  "bos_token_id": 1,
272
  "eos_token_id": 2,
273
  "hidden_act": "silu",
274
  "hidden_size": 2048,
275
  "initializer_range": 0.02,
276
+ "inner_step": 91,
277
  "intermediate_size": 5632,
278
+ "last_allreduce_block": 6454064,
279
  "max_position_embeddings": 2048,
280
  "mlp_bias": false,
281
  "model_type": "llama",
gradients.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ff21b5bd94f1ccb17eec936ad67937aba78ecfda2acd9a4a4407e897deae5613
3
  size 88390091
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d48d7d36fbf73322dc2be61dc3ed3d6ff4e1abbdf4817083ec830509e0b61085
3
  size 88390091
inner_optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:649715842359b95940bd652c839ca57aa8b31ee2040208a3bc6c976f56112fb6
3
  size 4924928111
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bc8583fc1e7499b93a3dadddf0f300941f1a5f250bbe7bf005734703cf16fe5
3
  size 4924928111
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:737cce817f0f3099d38400cda275d0720fe167d47ba4bc89931d350fe8d5f30a
3
  size 4400216536
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d6cf7b0c45d16f0b84be0bad0d1a2dfb22080b94678139da2c30973a5334dd5
3
  size 4400216536