zeras141a commited on
Commit
d66e08c
·
verified ·
1 Parent(s): 884245f

Run 5. Outer Step 5. Inner Step 2.

Browse files
Files changed (4) hide show
  1. config.json +89 -98
  2. generation_config.json +7 -0
  3. inner_optimizer.pt +2 -2
  4. model.safetensors +2 -2
config.json CHANGED
@@ -1,6 +1,5 @@
1
  {
2
- "_name_or_path": "zeras141a/d9",
3
- "activation_function": "gelu_new",
4
  "all_reduce_scores": {
5
  "0": "NON_PARTICIPATING",
6
  "1": "SUCCESS",
@@ -9,7 +8,7 @@
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",
@@ -24,25 +23,25 @@
24
  "115": "NON_PARTICIPATING",
25
  "116": "NON_PARTICIPATING",
26
  "117": "NON_PARTICIPATING",
27
- "118": "NON_PARTICIPATING",
28
- "119": "SUCCESS",
29
- "12": "NON_PARTICIPATING",
30
  "120": "NON_PARTICIPATING",
31
  "121": "NON_PARTICIPATING",
32
  "122": "NON_PARTICIPATING",
33
  "123": "NON_PARTICIPATING",
34
  "124": "NON_PARTICIPATING",
35
- "125": "NON_PARTICIPATING",
36
- "126": "NON_PARTICIPATING",
37
  "127": "NON_PARTICIPATING",
38
  "128": "NON_PARTICIPATING",
39
  "129": "NON_PARTICIPATING",
40
- "13": "NON_PARTICIPATING",
41
  "130": "NON_PARTICIPATING",
42
  "131": "NON_PARTICIPATING",
43
  "132": "NON_PARTICIPATING",
44
  "133": "NON_PARTICIPATING",
45
- "134": "SUCCESS",
46
  "135": "NON_PARTICIPATING",
47
  "136": "NON_PARTICIPATING",
48
  "137": "NON_PARTICIPATING",
@@ -59,10 +58,10 @@
59
  "147": "NON_PARTICIPATING",
60
  "148": "SUCCESS",
61
  "149": "NON_PARTICIPATING",
62
- "15": "NON_PARTICIPATING",
63
  "150": "NON_PARTICIPATING",
64
  "151": "NON_PARTICIPATING",
65
- "152": "SUCCESS",
66
  "153": "NON_PARTICIPATING",
67
  "154": "NON_PARTICIPATING",
68
  "155": "NON_PARTICIPATING",
@@ -78,7 +77,7 @@
78
  "164": "NON_PARTICIPATING",
79
  "165": "NON_PARTICIPATING",
80
  "166": "NON_PARTICIPATING",
81
- "167": "NON_PARTICIPATING",
82
  "168": "NON_PARTICIPATING",
83
  "169": "NON_PARTICIPATING",
84
  "17": "SUCCESS",
@@ -89,7 +88,7 @@
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": "SUCCESS",
@@ -103,7 +102,7 @@
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",
@@ -115,7 +114,7 @@
115
  "198": "NON_PARTICIPATING",
116
  "199": "NON_PARTICIPATING",
117
  "2": "SUCCESS",
118
- "20": "NON_PARTICIPATING",
119
  "200": "NON_PARTICIPATING",
120
  "201": "NON_PARTICIPATING",
121
  "202": "NON_PARTICIPATING",
@@ -125,9 +124,9 @@
125
  "206": "NON_PARTICIPATING",
126
  "207": "NON_PARTICIPATING",
127
  "208": "NON_PARTICIPATING",
128
- "209": "NON_PARTICIPATING",
129
  "21": "SUCCESS",
130
- "210": "SUCCESS",
131
  "211": "NON_PARTICIPATING",
132
  "212": "NON_PARTICIPATING",
133
  "213": "NON_PARTICIPATING",
@@ -159,7 +158,7 @@
159
  "237": "NON_PARTICIPATING",
160
  "238": "NON_PARTICIPATING",
161
  "239": "NON_PARTICIPATING",
162
- "24": "NON_PARTICIPATING",
163
  "240": "NON_PARTICIPATING",
164
  "241": "NON_PARTICIPATING",
165
  "242": "NON_PARTICIPATING",
@@ -169,134 +168,126 @@
169
  "246": "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",
177
  "253": "NON_PARTICIPATING",
178
  "254": "NON_PARTICIPATING",
179
  "255": "NON_PARTICIPATING",
180
  "26": "NON_PARTICIPATING",
181
- "27": "NON_PARTICIPATING",
182
- "28": "NON_PARTICIPATING",
183
  "29": "SUCCESS",
184
  "3": "SUCCESS",
185
- "30": "NON_PARTICIPATING",
186
  "31": "SUCCESS",
187
  "32": "SUCCESS",
188
  "33": "NON_PARTICIPATING",
189
- "34": "NON_PARTICIPATING",
190
- "35": "NON_PARTICIPATING",
191
- "36": "NON_PARTICIPATING",
192
  "37": "SUCCESS",
193
- "38": "NON_PARTICIPATING",
194
  "39": "SUCCESS",
195
- "4": "NON_PARTICIPATING",
196
- "40": "NON_PARTICIPATING",
197
- "41": "NON_PARTICIPATING",
198
- "42": "NON_PARTICIPATING",
199
- "43": "NON_PARTICIPATING",
200
- "44": "NON_PARTICIPATING",
201
  "45": "SUCCESS",
202
  "46": "SUCCESS",
203
- "47": "NON_PARTICIPATING",
204
  "48": "NON_PARTICIPATING",
205
- "49": "NON_PARTICIPATING",
206
- "5": "NON_PARTICIPATING",
207
  "50": "SUCCESS",
208
- "51": "NON_PARTICIPATING",
209
  "52": "NON_PARTICIPATING",
210
  "53": "SUCCESS",
211
  "54": "SUCCESS",
212
- "55": "NON_PARTICIPATING",
213
- "56": "NON_PARTICIPATING",
214
  "57": "SUCCESS",
215
  "58": "NON_PARTICIPATING",
216
- "59": "NON_PARTICIPATING",
217
  "6": "SUCCESS",
218
- "60": "NON_PARTICIPATING",
219
  "61": "SUCCESS",
220
  "62": "SUCCESS",
221
- "63": "NON_PARTICIPATING",
222
- "64": "NON_PARTICIPATING",
223
- "65": "NON_PARTICIPATING",
224
  "66": "NON_PARTICIPATING",
225
  "67": "SUCCESS",
226
  "68": "SUCCESS",
227
  "69": "SUCCESS",
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": "SUCCESS",
236
- "77": "NON_PARTICIPATING",
237
  "78": "SUCCESS",
238
- "79": "NON_PARTICIPATING",
239
- "8": "NON_PARTICIPATING",
240
- "80": "NON_PARTICIPATING",
241
  "81": "SUCCESS",
242
- "82": "NON_PARTICIPATING",
243
- "83": "NON_PARTICIPATING",
244
  "84": "SUCCESS",
245
- "85": "NON_PARTICIPATING",
246
  "86": "NON_PARTICIPATING",
247
  "87": "SUCCESS",
248
- "88": "NON_PARTICIPATING",
249
  "89": "SUCCESS",
250
  "9": "NON_PARTICIPATING",
251
  "90": "SUCCESS",
252
- "91": "NON_PARTICIPATING",
253
- "92": "NON_PARTICIPATING",
254
  "93": "SUCCESS",
255
  "94": "SUCCESS",
256
- "95": "NON_PARTICIPATING",
257
- "96": "NON_PARTICIPATING",
258
- "97": "NON_PARTICIPATING",
259
  "98": "SUCCESS",
260
- "99": "NON_PARTICIPATING"
261
  },
262
  "architectures": [
263
- "GPTOptim"
264
  ],
265
- "attn_pdrop": 0.1,
266
- "auto_map": {
267
- "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
268
- "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
269
- },
270
  "block_list": [
271
- 5691787,
272
- 5691805
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": 3,
280
- "inner_steps": 0,
281
- "last_allreduce_block": 5690407,
282
- "layer_norm_epsilon": 1e-05,
283
- "model_type": "gpt_optimized",
284
- "n_embd": 1280,
285
- "n_head": 32,
286
- "n_inner": null,
287
- "n_layer": 48,
288
- "n_positions": 1024,
289
- "reorder_and_upcast_attn": false,
290
- "resid_pdrop": 0.1,
291
- "scale_attn_by_inverse_layer_idx": false,
292
- "scale_attn_weights": true,
293
- "summary_activation": null,
294
- "summary_first_dropout": 0.1,
295
- "summary_proj_to_labels": true,
296
- "summary_type": "cls_index",
297
- "summary_use_proj": true,
298
  "torch_dtype": "float32",
299
  "transformers_version": "4.39.3",
300
- "use_cache": true,
301
- "vocab_size": 50257
302
  }
 
1
  {
2
+ "_name_or_path": "distributed/llama-1b",
 
3
  "all_reduce_scores": {
4
  "0": "NON_PARTICIPATING",
5
  "1": "SUCCESS",
 
8
  "101": "NON_PARTICIPATING",
9
  "102": "NON_PARTICIPATING",
10
  "103": "NON_PARTICIPATING",
11
+ "104": "SUCCESS",
12
  "105": "NON_PARTICIPATING",
13
  "106": "NON_PARTICIPATING",
14
  "107": "NON_PARTICIPATING",
 
23
  "115": "NON_PARTICIPATING",
24
  "116": "NON_PARTICIPATING",
25
  "117": "NON_PARTICIPATING",
26
+ "118": "SUCCESS",
27
+ "119": "NON_PARTICIPATING",
28
+ "12": "SUCCESS",
29
  "120": "NON_PARTICIPATING",
30
  "121": "NON_PARTICIPATING",
31
  "122": "NON_PARTICIPATING",
32
  "123": "NON_PARTICIPATING",
33
  "124": "NON_PARTICIPATING",
34
+ "125": "SUCCESS",
35
+ "126": "SUCCESS",
36
  "127": "NON_PARTICIPATING",
37
  "128": "NON_PARTICIPATING",
38
  "129": "NON_PARTICIPATING",
39
+ "13": "SUCCESS",
40
  "130": "NON_PARTICIPATING",
41
  "131": "NON_PARTICIPATING",
42
  "132": "NON_PARTICIPATING",
43
  "133": "NON_PARTICIPATING",
44
+ "134": "NON_PARTICIPATING",
45
  "135": "NON_PARTICIPATING",
46
  "136": "NON_PARTICIPATING",
47
  "137": "NON_PARTICIPATING",
 
58
  "147": "NON_PARTICIPATING",
59
  "148": "SUCCESS",
60
  "149": "NON_PARTICIPATING",
61
+ "15": "SUCCESS",
62
  "150": "NON_PARTICIPATING",
63
  "151": "NON_PARTICIPATING",
64
+ "152": "NON_PARTICIPATING",
65
  "153": "NON_PARTICIPATING",
66
  "154": "NON_PARTICIPATING",
67
  "155": "NON_PARTICIPATING",
 
77
  "164": "NON_PARTICIPATING",
78
  "165": "NON_PARTICIPATING",
79
  "166": "NON_PARTICIPATING",
80
+ "167": "SUCCESS",
81
  "168": "NON_PARTICIPATING",
82
  "169": "NON_PARTICIPATING",
83
  "17": "SUCCESS",
 
88
  "174": "NON_PARTICIPATING",
89
  "175": "NON_PARTICIPATING",
90
  "176": "NON_PARTICIPATING",
91
+ "177": "SUCCESS",
92
  "178": "NON_PARTICIPATING",
93
  "179": "NON_PARTICIPATING",
94
  "18": "SUCCESS",
 
102
  "187": "NON_PARTICIPATING",
103
  "188": "NON_PARTICIPATING",
104
  "189": "NON_PARTICIPATING",
105
+ "19": "SUCCESS",
106
  "190": "NON_PARTICIPATING",
107
  "191": "NON_PARTICIPATING",
108
  "192": "NON_PARTICIPATING",
 
114
  "198": "NON_PARTICIPATING",
115
  "199": "NON_PARTICIPATING",
116
  "2": "SUCCESS",
117
+ "20": "SUCCESS",
118
  "200": "NON_PARTICIPATING",
119
  "201": "NON_PARTICIPATING",
120
  "202": "NON_PARTICIPATING",
 
124
  "206": "NON_PARTICIPATING",
125
  "207": "NON_PARTICIPATING",
126
  "208": "NON_PARTICIPATING",
127
+ "209": "SUCCESS",
128
  "21": "SUCCESS",
129
+ "210": "NON_PARTICIPATING",
130
  "211": "NON_PARTICIPATING",
131
  "212": "NON_PARTICIPATING",
132
  "213": "NON_PARTICIPATING",
 
158
  "237": "NON_PARTICIPATING",
159
  "238": "NON_PARTICIPATING",
160
  "239": "NON_PARTICIPATING",
161
+ "24": "SUCCESS",
162
  "240": "NON_PARTICIPATING",
163
  "241": "NON_PARTICIPATING",
164
  "242": "NON_PARTICIPATING",
 
168
  "246": "NON_PARTICIPATING",
169
  "247": "NON_PARTICIPATING",
170
  "248": "NON_PARTICIPATING",
171
+ "249": "SUCCESS",
172
  "25": "SUCCESS",
173
+ "250": "SUCCESS",
174
  "251": "NON_PARTICIPATING",
175
  "252": "NON_PARTICIPATING",
176
  "253": "NON_PARTICIPATING",
177
  "254": "NON_PARTICIPATING",
178
  "255": "NON_PARTICIPATING",
179
  "26": "NON_PARTICIPATING",
180
+ "27": "SUCCESS",
181
+ "28": "SUCCESS",
182
  "29": "SUCCESS",
183
  "3": "SUCCESS",
184
+ "30": "SUCCESS",
185
  "31": "SUCCESS",
186
  "32": "SUCCESS",
187
  "33": "NON_PARTICIPATING",
188
+ "34": "SUCCESS",
189
+ "35": "SUCCESS",
190
+ "36": "SUCCESS",
191
  "37": "SUCCESS",
192
+ "38": "SUCCESS",
193
  "39": "SUCCESS",
194
+ "4": "SUCCESS",
195
+ "40": "SUCCESS",
196
+ "41": "SUCCESS",
197
+ "42": "SUCCESS",
198
+ "43": "SUCCESS",
199
+ "44": "SUCCESS",
200
  "45": "SUCCESS",
201
  "46": "SUCCESS",
202
+ "47": "SUCCESS",
203
  "48": "NON_PARTICIPATING",
204
+ "49": "SUCCESS",
205
+ "5": "SUCCESS",
206
  "50": "SUCCESS",
207
+ "51": "SUCCESS",
208
  "52": "NON_PARTICIPATING",
209
  "53": "SUCCESS",
210
  "54": "SUCCESS",
211
+ "55": "SUCCESS",
212
+ "56": "SUCCESS",
213
  "57": "SUCCESS",
214
  "58": "NON_PARTICIPATING",
215
+ "59": "SUCCESS",
216
  "6": "SUCCESS",
217
+ "60": "SUCCESS",
218
  "61": "SUCCESS",
219
  "62": "SUCCESS",
220
+ "63": "SUCCESS",
221
+ "64": "SUCCESS",
222
+ "65": "SUCCESS",
223
  "66": "NON_PARTICIPATING",
224
  "67": "SUCCESS",
225
  "68": "SUCCESS",
226
  "69": "SUCCESS",
227
  "7": "SUCCESS",
228
+ "70": "SUCCESS",
229
+ "71": "SUCCESS",
230
+ "72": "SUCCESS",
231
+ "73": "SUCCESS",
232
+ "74": "SUCCESS",
233
+ "75": "SUCCESS",
234
  "76": "SUCCESS",
235
+ "77": "SUCCESS",
236
  "78": "SUCCESS",
237
+ "79": "SUCCESS",
238
+ "8": "SUCCESS",
239
+ "80": "SUCCESS",
240
  "81": "SUCCESS",
241
+ "82": "SUCCESS",
242
+ "83": "SUCCESS",
243
  "84": "SUCCESS",
244
+ "85": "SUCCESS",
245
  "86": "NON_PARTICIPATING",
246
  "87": "SUCCESS",
247
+ "88": "SUCCESS",
248
  "89": "SUCCESS",
249
  "9": "NON_PARTICIPATING",
250
  "90": "SUCCESS",
251
+ "91": "SUCCESS",
252
+ "92": "SUCCESS",
253
  "93": "SUCCESS",
254
  "94": "SUCCESS",
255
+ "95": "SUCCESS",
256
+ "96": "SUCCESS",
257
+ "97": "SUCCESS",
258
  "98": "SUCCESS",
259
+ "99": "SUCCESS"
260
  },
261
  "architectures": [
262
+ "LlamaForCausalLM"
263
  ],
264
+ "attention_bias": false,
265
+ "attention_dropout": 0.0,
 
 
 
266
  "block_list": [
267
+ 5706753,
268
+ 5706782
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": 5706512,
278
+ "max_position_embeddings": 2048,
279
+ "mlp_bias": false,
280
+ "model_type": "llama",
281
+ "num_attention_heads": 32,
282
+ "num_hidden_layers": 22,
283
+ "num_key_value_heads": 4,
284
+ "pretraining_tp": 1,
285
+ "rms_norm_eps": 1e-05,
286
+ "rope_scaling": null,
287
+ "rope_theta": 10000.0,
288
+ "tie_word_embeddings": false,
 
 
 
 
 
289
  "torch_dtype": "float32",
290
  "transformers_version": "4.39.3",
291
+ "use_cache": false,
292
+ "vocab_size": 32000
293
  }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.39.3",
6
+ "use_cache": false
7
+ }
inner_optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c981d91f66ab6aa9bcfd1432f68531b1b4cfe45034f17bd5e5fa4a42d35a8909
3
- size 8081782026
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90b865efe30c6aca264c64ca57dab63b184dc76f47f77ffff69bcf43d453630c
3
+ size 8800564012
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:50d6536a94379497362845e9c88048593790231c867288c724abac46ed22217e
3
- size 4040701744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20c8276e7113ea3f0fe57620fec96608762b75ae40c1d9bc0f85afe3638c173d
3
+ size 4400216536