MBA4215 commited on
Commit
4d0f000
·
verified ·
1 Parent(s): 99ea36f

Run 5. Outer Step 52. Inner Step 2.

Browse files
Files changed (3) hide show
  1. config.json +53 -53
  2. inner_optimizer.pt +1 -1
  3. model.safetensors +1 -1
config.json CHANGED
@@ -2,16 +2,16 @@
2
  "_name_or_path": "distributed/llama-1b",
3
  "all_reduce_scores": {
4
  "0": "NOT_ALIVE",
5
- "1": "SUCCESS",
6
  "10": "NOT_ALIVE",
7
  "100": "NOT_ALIVE",
8
- "101": "NON_PARTICIPATING",
9
  "102": "NOT_ALIVE",
10
  "103": "NON_PARTICIPATING",
11
  "104": "NOT_ALIVE",
12
  "105": "SUCCESS",
13
  "106": "NOT_ALIVE",
14
- "107": "NOT_ALIVE",
15
  "108": "NON_PARTICIPATING",
16
  "109": "NON_PARTICIPATING",
17
  "11": "NOT_ALIVE",
@@ -24,7 +24,7 @@
24
  "116": "NON_PARTICIPATING",
25
  "117": "NOT_ALIVE",
26
  "118": "SUCCESS",
27
- "119": "NOT_ALIVE",
28
  "12": "NON_PARTICIPATING",
29
  "120": "NOT_ALIVE",
30
  "121": "NOT_ALIVE",
@@ -33,26 +33,26 @@
33
  "124": "SUCCESS",
34
  "125": "NOT_ALIVE",
35
  "126": "NON_PARTICIPATING",
36
- "127": "NON_PARTICIPATING",
37
  "128": "NOT_ALIVE",
38
  "129": "SUCCESS",
39
  "13": "NOT_ALIVE",
40
- "130": "SUCCESS",
41
  "131": "NOT_ALIVE",
42
  "132": "NOT_ALIVE",
43
  "133": "SUCCESS",
44
- "134": "NOT_ALIVE",
45
  "135": "NON_PARTICIPATING",
46
- "136": "NOT_ALIVE",
47
  "137": "NOT_ALIVE",
48
  "138": "NOT_ALIVE",
49
- "139": "SUCCESS",
50
- "14": "NOT_ALIVE",
51
  "140": "NOT_ALIVE",
52
  "141": "SUCCESS",
53
- "142": "NOT_ALIVE",
54
  "143": "NON_PARTICIPATING",
55
- "144": "NON_PARTICIPATING",
56
  "145": "NOT_ALIVE",
57
  "146": "SUCCESS",
58
  "147": "NON_PARTICIPATING",
@@ -61,27 +61,27 @@
61
  "15": "NON_PARTICIPATING",
62
  "150": "SUCCESS",
63
  "151": "NOT_ALIVE",
64
- "152": "NON_PARTICIPATING",
65
  "153": "NON_PARTICIPATING",
66
  "154": "NOT_ALIVE",
67
  "155": "SUCCESS",
68
- "156": "NON_PARTICIPATING",
69
- "157": "SUCCESS",
70
  "158": "NOT_ALIVE",
71
  "159": "NOT_ALIVE",
72
  "16": "NON_PARTICIPATING",
73
  "160": "NON_PARTICIPATING",
74
  "161": "SUCCESS",
75
- "162": "NON_PARTICIPATING",
76
  "163": "NOT_ALIVE",
77
  "164": "SUCCESS",
78
  "165": "NON_PARTICIPATING",
79
  "166": "NON_PARTICIPATING",
80
- "167": "NOT_ALIVE",
81
  "168": "NOT_ALIVE",
82
  "169": "NOT_ALIVE",
83
  "17": "NOT_ALIVE",
84
- "170": "NOT_ALIVE",
85
  "171": "SUCCESS",
86
  "172": "NOT_ALIVE",
87
  "173": "NOT_ALIVE",
@@ -100,10 +100,10 @@
100
  "185": "NON_PARTICIPATING",
101
  "186": "NOT_ALIVE",
102
  "187": "SUCCESS",
103
- "188": "NON_PARTICIPATING",
104
  "189": "NOT_ALIVE",
105
  "19": "NOT_ALIVE",
106
- "190": "NOT_ALIVE",
107
  "191": "NOT_ALIVE",
108
  "192": "NON_PARTICIPATING",
109
  "193": "NOT_ALIVE",
@@ -112,54 +112,54 @@
112
  "196": "NOT_ALIVE",
113
  "197": "NOT_ALIVE",
114
  "198": "NOT_ALIVE",
115
- "199": "NON_PARTICIPATING",
116
  "2": "NOT_ALIVE",
117
  "20": "NOT_ALIVE",
118
  "200": "NOT_ALIVE",
119
  "201": "NON_PARTICIPATING",
120
  "202": "NON_PARTICIPATING",
121
  "203": "NON_PARTICIPATING",
122
- "204": "NON_PARTICIPATING",
123
  "205": "NOT_ALIVE",
124
- "206": "SUCCESS",
125
  "207": "NOT_ALIVE",
126
- "208": "NOT_ALIVE",
127
  "209": "SUCCESS",
128
  "21": "SUCCESS",
129
  "210": "SUCCESS",
130
- "211": "NOT_ALIVE",
131
  "212": "NOT_ALIVE",
132
  "213": "NOT_ALIVE",
133
- "214": "NON_PARTICIPATING",
134
  "215": "NON_PARTICIPATING",
135
  "216": "NOT_ALIVE",
136
- "217": "NON_PARTICIPATING",
137
  "218": "NON_PARTICIPATING",
138
  "219": "NOT_ALIVE",
139
- "22": "SUCCESS",
140
  "220": "SUCCESS",
141
- "221": "NON_PARTICIPATING",
142
- "222": "NON_PARTICIPATING",
143
- "223": "NON_PARTICIPATING",
144
  "224": "NOT_ALIVE",
145
  "225": "NOT_ALIVE",
146
  "226": "NOT_ALIVE",
147
  "227": "SUCCESS",
148
- "228": "NON_PARTICIPATING",
149
  "229": "NOT_ALIVE",
150
  "23": "NOT_ALIVE",
151
  "230": "NOT_ALIVE",
152
  "231": "NON_PARTICIPATING",
153
  "232": "NOT_ALIVE",
154
  "233": "NOT_ALIVE",
155
- "234": "NOT_ALIVE",
156
  "235": "NOT_ALIVE",
157
  "236": "NOT_ALIVE",
158
  "237": "NOT_ALIVE",
159
  "238": "NON_PARTICIPATING",
160
  "239": "NOT_ALIVE",
161
  "24": "SUCCESS",
162
- "240": "NON_PARTICIPATING",
163
  "241": "NOT_ALIVE",
164
  "242": "SUCCESS",
165
  "243": "NON_PARTICIPATING",
@@ -173,12 +173,12 @@
173
  "250": "NON_PARTICIPATING",
174
  "251": "NOT_ALIVE",
175
  "252": "NON_PARTICIPATING",
176
- "253": "NOT_ALIVE",
177
  "254": "NOT_ALIVE",
178
  "255": "NON_PARTICIPATING",
179
  "26": "SUCCESS",
180
  "27": "SUCCESS",
181
- "28": "SUCCESS",
182
  "29": "NOT_ALIVE",
183
  "3": "SUCCESS",
184
  "30": "NOT_ALIVE",
@@ -190,9 +190,9 @@
190
  "36": "NON_PARTICIPATING",
191
  "37": "NOT_ALIVE",
192
  "38": "SUCCESS",
193
- "39": "NOT_ALIVE",
194
  "4": "NOT_ALIVE",
195
- "40": "SUCCESS",
196
  "41": "SUCCESS",
197
  "42": "SUCCESS",
198
  "43": "SUCCESS",
@@ -203,14 +203,14 @@
203
  "48": "SUCCESS",
204
  "49": "NOT_ALIVE",
205
  "5": "NOT_ALIVE",
206
- "50": "NON_PARTICIPATING",
207
  "51": "NOT_ALIVE",
208
- "52": "NON_PARTICIPATING",
209
  "53": "NOT_ALIVE",
210
  "54": "SUCCESS",
211
  "55": "NOT_ALIVE",
212
  "56": "SUCCESS",
213
- "57": "NOT_ALIVE",
214
  "58": "SUCCESS",
215
  "59": "NOT_ALIVE",
216
  "6": "NOT_ALIVE",
@@ -218,21 +218,21 @@
218
  "61": "NOT_ALIVE",
219
  "62": "NOT_ALIVE",
220
  "63": "NOT_ALIVE",
221
- "64": "NOT_ALIVE",
222
  "65": "NOT_ALIVE",
223
- "66": "SUCCESS",
224
  "67": "NOT_ALIVE",
225
  "68": "NOT_ALIVE",
226
- "69": "NOT_ALIVE",
227
  "7": "SUCCESS",
228
  "70": "NOT_ALIVE",
229
- "71": "SUCCESS",
230
  "72": "NOT_ALIVE",
231
  "73": "SUCCESS",
232
  "74": "NOT_ALIVE",
233
  "75": "SUCCESS",
234
  "76": "SUCCESS",
235
- "77": "NON_PARTICIPATING",
236
  "78": "NOT_ALIVE",
237
  "79": "NOT_ALIVE",
238
  "8": "NOT_ALIVE",
@@ -248,15 +248,15 @@
248
  "89": "NOT_ALIVE",
249
  "9": "NOT_ALIVE",
250
  "90": "SUCCESS",
251
- "91": "NON_PARTICIPATING",
252
- "92": "SUCCESS",
253
  "93": "NON_PARTICIPATING",
254
  "94": "NON_PARTICIPATING",
255
  "95": "NOT_ALIVE",
256
  "96": "NON_PARTICIPATING",
257
  "97": "SUCCESS",
258
  "98": "SUCCESS",
259
- "99": "NON_PARTICIPATING"
260
  },
261
  "architectures": [
262
  "LlamaForCausalLM"
@@ -264,17 +264,17 @@
264
  "attention_bias": false,
265
  "attention_dropout": 0.0,
266
  "block_list": [
267
- 5889343,
268
- 5889374
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": 25,
276
  "intermediate_size": 5632,
277
- "last_allreduce_block": 5888030,
278
  "max_position_embeddings": 2048,
279
  "mlp_bias": false,
280
  "model_type": "llama",
 
2
  "_name_or_path": "distributed/llama-1b",
3
  "all_reduce_scores": {
4
  "0": "NOT_ALIVE",
5
+ "1": "NOT_ALIVE",
6
  "10": "NOT_ALIVE",
7
  "100": "NOT_ALIVE",
8
+ "101": "NOT_ALIVE",
9
  "102": "NOT_ALIVE",
10
  "103": "NON_PARTICIPATING",
11
  "104": "NOT_ALIVE",
12
  "105": "SUCCESS",
13
  "106": "NOT_ALIVE",
14
+ "107": "NON_PARTICIPATING",
15
  "108": "NON_PARTICIPATING",
16
  "109": "NON_PARTICIPATING",
17
  "11": "NOT_ALIVE",
 
24
  "116": "NON_PARTICIPATING",
25
  "117": "NOT_ALIVE",
26
  "118": "SUCCESS",
27
+ "119": "NON_PARTICIPATING",
28
  "12": "NON_PARTICIPATING",
29
  "120": "NOT_ALIVE",
30
  "121": "NOT_ALIVE",
 
33
  "124": "SUCCESS",
34
  "125": "NOT_ALIVE",
35
  "126": "NON_PARTICIPATING",
36
+ "127": "NOT_ALIVE",
37
  "128": "NOT_ALIVE",
38
  "129": "SUCCESS",
39
  "13": "NOT_ALIVE",
40
+ "130": "NOT_ALIVE",
41
  "131": "NOT_ALIVE",
42
  "132": "NOT_ALIVE",
43
  "133": "SUCCESS",
44
+ "134": "SUCCESS",
45
  "135": "NON_PARTICIPATING",
46
+ "136": "NON_PARTICIPATING",
47
  "137": "NOT_ALIVE",
48
  "138": "NOT_ALIVE",
49
+ "139": "NON_PARTICIPATING",
50
+ "14": "NON_PARTICIPATING",
51
  "140": "NOT_ALIVE",
52
  "141": "SUCCESS",
53
+ "142": "SUCCESS",
54
  "143": "NON_PARTICIPATING",
55
+ "144": "NOT_ALIVE",
56
  "145": "NOT_ALIVE",
57
  "146": "SUCCESS",
58
  "147": "NON_PARTICIPATING",
 
61
  "15": "NON_PARTICIPATING",
62
  "150": "SUCCESS",
63
  "151": "NOT_ALIVE",
64
+ "152": "NOT_ALIVE",
65
  "153": "NON_PARTICIPATING",
66
  "154": "NOT_ALIVE",
67
  "155": "SUCCESS",
68
+ "156": "NOT_ALIVE",
69
+ "157": "NON_PARTICIPATING",
70
  "158": "NOT_ALIVE",
71
  "159": "NOT_ALIVE",
72
  "16": "NON_PARTICIPATING",
73
  "160": "NON_PARTICIPATING",
74
  "161": "SUCCESS",
75
+ "162": "NOT_ALIVE",
76
  "163": "NOT_ALIVE",
77
  "164": "SUCCESS",
78
  "165": "NON_PARTICIPATING",
79
  "166": "NON_PARTICIPATING",
80
+ "167": "SUCCESS",
81
  "168": "NOT_ALIVE",
82
  "169": "NOT_ALIVE",
83
  "17": "NOT_ALIVE",
84
+ "170": "SUCCESS",
85
  "171": "SUCCESS",
86
  "172": "NOT_ALIVE",
87
  "173": "NOT_ALIVE",
 
100
  "185": "NON_PARTICIPATING",
101
  "186": "NOT_ALIVE",
102
  "187": "SUCCESS",
103
+ "188": "NOT_ALIVE",
104
  "189": "NOT_ALIVE",
105
  "19": "NOT_ALIVE",
106
+ "190": "NON_PARTICIPATING",
107
  "191": "NOT_ALIVE",
108
  "192": "NON_PARTICIPATING",
109
  "193": "NOT_ALIVE",
 
112
  "196": "NOT_ALIVE",
113
  "197": "NOT_ALIVE",
114
  "198": "NOT_ALIVE",
115
+ "199": "NOT_ALIVE",
116
  "2": "NOT_ALIVE",
117
  "20": "NOT_ALIVE",
118
  "200": "NOT_ALIVE",
119
  "201": "NON_PARTICIPATING",
120
  "202": "NON_PARTICIPATING",
121
  "203": "NON_PARTICIPATING",
122
+ "204": "NOT_ALIVE",
123
  "205": "NOT_ALIVE",
124
+ "206": "NOT_ALIVE",
125
  "207": "NOT_ALIVE",
126
+ "208": "NON_PARTICIPATING",
127
  "209": "SUCCESS",
128
  "21": "SUCCESS",
129
  "210": "SUCCESS",
130
+ "211": "NON_PARTICIPATING",
131
  "212": "NOT_ALIVE",
132
  "213": "NOT_ALIVE",
133
+ "214": "NOT_ALIVE",
134
  "215": "NON_PARTICIPATING",
135
  "216": "NOT_ALIVE",
136
+ "217": "NOT_ALIVE",
137
  "218": "NON_PARTICIPATING",
138
  "219": "NOT_ALIVE",
139
+ "22": "NON_PARTICIPATING",
140
  "220": "SUCCESS",
141
+ "221": "SUCCESS",
142
+ "222": "NOT_ALIVE",
143
+ "223": "NOT_ALIVE",
144
  "224": "NOT_ALIVE",
145
  "225": "NOT_ALIVE",
146
  "226": "NOT_ALIVE",
147
  "227": "SUCCESS",
148
+ "228": "NOT_ALIVE",
149
  "229": "NOT_ALIVE",
150
  "23": "NOT_ALIVE",
151
  "230": "NOT_ALIVE",
152
  "231": "NON_PARTICIPATING",
153
  "232": "NOT_ALIVE",
154
  "233": "NOT_ALIVE",
155
+ "234": "NON_PARTICIPATING",
156
  "235": "NOT_ALIVE",
157
  "236": "NOT_ALIVE",
158
  "237": "NOT_ALIVE",
159
  "238": "NON_PARTICIPATING",
160
  "239": "NOT_ALIVE",
161
  "24": "SUCCESS",
162
+ "240": "NOT_ALIVE",
163
  "241": "NOT_ALIVE",
164
  "242": "SUCCESS",
165
  "243": "NON_PARTICIPATING",
 
173
  "250": "NON_PARTICIPATING",
174
  "251": "NOT_ALIVE",
175
  "252": "NON_PARTICIPATING",
176
+ "253": "NON_PARTICIPATING",
177
  "254": "NOT_ALIVE",
178
  "255": "NON_PARTICIPATING",
179
  "26": "SUCCESS",
180
  "27": "SUCCESS",
181
+ "28": "NOT_ALIVE",
182
  "29": "NOT_ALIVE",
183
  "3": "SUCCESS",
184
  "30": "NOT_ALIVE",
 
190
  "36": "NON_PARTICIPATING",
191
  "37": "NOT_ALIVE",
192
  "38": "SUCCESS",
193
+ "39": "NON_PARTICIPATING",
194
  "4": "NOT_ALIVE",
195
+ "40": "NON_PARTICIPATING",
196
  "41": "SUCCESS",
197
  "42": "SUCCESS",
198
  "43": "SUCCESS",
 
203
  "48": "SUCCESS",
204
  "49": "NOT_ALIVE",
205
  "5": "NOT_ALIVE",
206
+ "50": "NOT_ALIVE",
207
  "51": "NOT_ALIVE",
208
+ "52": "NOT_ALIVE",
209
  "53": "NOT_ALIVE",
210
  "54": "SUCCESS",
211
  "55": "NOT_ALIVE",
212
  "56": "SUCCESS",
213
+ "57": "SUCCESS",
214
  "58": "SUCCESS",
215
  "59": "NOT_ALIVE",
216
  "6": "NOT_ALIVE",
 
218
  "61": "NOT_ALIVE",
219
  "62": "NOT_ALIVE",
220
  "63": "NOT_ALIVE",
221
+ "64": "NON_PARTICIPATING",
222
  "65": "NOT_ALIVE",
223
+ "66": "NOT_ALIVE",
224
  "67": "NOT_ALIVE",
225
  "68": "NOT_ALIVE",
226
+ "69": "SUCCESS",
227
  "7": "SUCCESS",
228
  "70": "NOT_ALIVE",
229
+ "71": "NON_PARTICIPATING",
230
  "72": "NOT_ALIVE",
231
  "73": "SUCCESS",
232
  "74": "NOT_ALIVE",
233
  "75": "SUCCESS",
234
  "76": "SUCCESS",
235
+ "77": "NOT_ALIVE",
236
  "78": "NOT_ALIVE",
237
  "79": "NOT_ALIVE",
238
  "8": "NOT_ALIVE",
 
248
  "89": "NOT_ALIVE",
249
  "9": "NOT_ALIVE",
250
  "90": "SUCCESS",
251
+ "91": "NOT_ALIVE",
252
+ "92": "NOT_ALIVE",
253
  "93": "NON_PARTICIPATING",
254
  "94": "NON_PARTICIPATING",
255
  "95": "NOT_ALIVE",
256
  "96": "NON_PARTICIPATING",
257
  "97": "SUCCESS",
258
  "98": "SUCCESS",
259
+ "99": "NOT_ALIVE"
260
  },
261
  "architectures": [
262
  "LlamaForCausalLM"
 
264
  "attention_bias": false,
265
  "attention_dropout": 0.0,
266
  "block_list": [
267
+ 5889484,
268
+ 5889514
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": 5889307,
278
  "max_position_embeddings": 2048,
279
  "mlp_bias": false,
280
  "model_type": "llama",
inner_optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a6114f9c2ce0e84d666d1b31f02c473ebbadbba0a7f9b77c806372ebcf640db
3
  size 8800564076
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51e7f3543ed9be6075ed8a0de5934025b1ceca282acb2c7cefcec9a744bdb694
3
  size 8800564076
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9f60ca3d15e5c164d74dd300acc9e6ff1dfeaa519364df1c613d2070cd961339
3
  size 4400216536
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1128c62b7e0d5a1af97b38d57468656ad797801ed1d302551a61bd2cc57f6b89
3
  size 4400216536