kmfoda commited on
Commit
13a4c18
·
verified ·
1 Parent(s): 40fe251

Run 6. Outer Step 6. Inner Step 4.

Browse files
Files changed (5) hide show
  1. config.json +219 -228
  2. generation_config.json +7 -0
  3. gradients.pt +3 -0
  4. inner_optimizer.pt +2 -2
  5. model.safetensors +2 -2
config.json CHANGED
@@ -1,302 +1,293 @@
1
  {
2
- "_name_or_path": "distributed/optimized-gpt2-1b",
3
- "activation_function": "gelu_new",
4
  "all_reduce_scores": {
5
- "0": "NON_PARTICIPATING",
6
- "1": "SUCCESS",
7
- "10": "NON_PARTICIPATING",
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": "SUCCESS",
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": "SUCCESS",
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": "SUCCESS",
39
  "129": "NON_PARTICIPATING",
40
- "13": "SUCCESS",
41
- "130": "NON_PARTICIPATING",
42
- "131": "NON_PARTICIPATING",
43
- "132": "NON_PARTICIPATING",
44
- "133": "SUCCESS",
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": "SUCCESS",
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",
59
- "147": "NON_PARTICIPATING",
60
- "148": "NON_PARTICIPATING",
61
- "149": "NON_PARTICIPATING",
62
  "15": "SUCCESS",
63
- "150": "SUCCESS",
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": "SUCCESS",
74
- "160": "NON_PARTICIPATING",
75
- "161": "SUCCESS",
76
- "162": "NON_PARTICIPATING",
77
- "163": "NON_PARTICIPATING",
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",
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": "SUCCESS",
96
- "180": "NON_PARTICIPATING",
97
- "181": "NON_PARTICIPATING",
98
- "182": "NON_PARTICIPATING",
99
  "183": "NON_PARTICIPATING",
100
- "184": "SUCCESS",
101
- "185": "NON_PARTICIPATING",
102
- "186": "NON_PARTICIPATING",
103
  "187": "NON_PARTICIPATING",
104
- "188": "NON_PARTICIPATING",
105
- "189": "NON_PARTICIPATING",
106
- "19": "SUCCESS",
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": "SUCCESS",
114
  "197": "NON_PARTICIPATING",
115
- "198": "NON_PARTICIPATING",
116
- "199": "NON_PARTICIPATING",
117
- "2": "SUCCESS",
118
- "20": "SUCCESS",
119
  "200": "NON_PARTICIPATING",
120
  "201": "NON_PARTICIPATING",
121
  "202": "NON_PARTICIPATING",
122
- "203": "NON_PARTICIPATING",
123
- "204": "SUCCESS",
124
- "205": "NON_PARTICIPATING",
125
- "206": "NON_PARTICIPATING",
126
- "207": "NON_PARTICIPATING",
127
  "208": "NON_PARTICIPATING",
128
  "209": "NON_PARTICIPATING",
129
  "21": "SUCCESS",
130
  "210": "NON_PARTICIPATING",
131
- "211": "NON_PARTICIPATING",
132
- "212": "SUCCESS",
133
- "213": "NON_PARTICIPATING",
134
- "214": "NON_PARTICIPATING",
135
- "215": "NON_PARTICIPATING",
136
  "216": "SUCCESS",
137
- "217": "SUCCESS",
138
- "218": "NON_PARTICIPATING",
139
  "219": "NON_PARTICIPATING",
140
- "22": "SUCCESS",
141
  "220": "NON_PARTICIPATING",
142
- "221": "SUCCESS",
143
- "222": "NON_PARTICIPATING",
144
- "223": "NON_PARTICIPATING",
145
- "224": "NON_PARTICIPATING",
146
- "225": "NON_PARTICIPATING",
147
- "226": "NON_PARTICIPATING",
148
- "227": "SUCCESS",
149
  "228": "NON_PARTICIPATING",
150
- "229": "NON_PARTICIPATING",
151
  "23": "NON_PARTICIPATING",
152
- "230": "NON_PARTICIPATING",
153
- "231": "NON_PARTICIPATING",
154
- "232": "NON_PARTICIPATING",
155
- "233": "SUCCESS",
156
- "234": "NON_PARTICIPATING",
157
- "235": "NON_PARTICIPATING",
158
- "236": "NON_PARTICIPATING",
159
- "237": "NON_PARTICIPATING",
160
- "238": "NON_PARTICIPATING",
161
- "239": "NON_PARTICIPATING",
162
- "24": "SUCCESS",
163
- "240": "NON_PARTICIPATING",
164
  "241": "NON_PARTICIPATING",
165
- "242": "NON_PARTICIPATING",
166
- "243": "NON_PARTICIPATING",
167
- "244": "NON_PARTICIPATING",
168
- "245": "NON_PARTICIPATING",
169
- "246": "NON_PARTICIPATING",
170
- "247": "NON_PARTICIPATING",
171
  "248": "NON_PARTICIPATING",
172
  "249": "NON_PARTICIPATING",
173
  "25": "SUCCESS",
174
- "250": "SUCCESS",
175
- "251": "NON_PARTICIPATING",
176
- "252": "NON_PARTICIPATING",
177
- "253": "NON_PARTICIPATING",
178
- "254": "NON_PARTICIPATING",
179
  "255": "NON_PARTICIPATING",
180
- "26": "SUCCESS",
181
- "27": "SUCCESS",
182
- "28": "SUCCESS",
183
- "29": "SUCCESS",
184
  "3": "NON_PARTICIPATING",
185
  "30": "NON_PARTICIPATING",
186
- "31": "SUCCESS",
187
  "32": "NON_PARTICIPATING",
188
- "33": "NON_PARTICIPATING",
189
- "34": "SUCCESS",
190
- "35": "SUCCESS",
191
  "36": "NON_PARTICIPATING",
192
- "37": "NON_PARTICIPATING",
193
- "38": "SUCCESS",
194
  "39": "NON_PARTICIPATING",
195
- "4": "SUCCESS",
196
- "40": "SUCCESS",
197
- "41": "SUCCESS",
198
- "42": "SUCCESS",
199
- "43": "NON_PARTICIPATING",
200
- "44": "SUCCESS",
201
- "45": "SUCCESS",
202
- "46": "SUCCESS",
203
- "47": "SUCCESS",
204
- "48": "NON_PARTICIPATING",
205
- "49": "SUCCESS",
206
- "5": "SUCCESS",
207
- "50": "NON_PARTICIPATING",
208
- "51": "SUCCESS",
209
- "52": "SUCCESS",
210
  "53": "NON_PARTICIPATING",
211
- "54": "SUCCESS",
212
- "55": "SUCCESS",
213
- "56": "SUCCESS",
214
- "57": "NON_PARTICIPATING",
215
  "58": "NON_PARTICIPATING",
216
- "59": "SUCCESS",
217
- "6": "NON_PARTICIPATING",
218
- "60": "SUCCESS",
219
- "61": "SUCCESS",
220
- "62": "SUCCESS",
221
- "63": "SUCCESS",
222
- "64": "SUCCESS",
223
- "65": "SUCCESS",
224
- "66": "SUCCESS",
225
- "67": "SUCCESS",
226
- "68": "NON_PARTICIPATING",
227
  "69": "NON_PARTICIPATING",
228
- "7": "NON_PARTICIPATING",
229
- "70": "SUCCESS",
230
  "71": "SUCCESS",
231
- "72": "SUCCESS",
232
- "73": "SUCCESS",
233
- "74": "SUCCESS",
234
- "75": "SUCCESS",
235
- "76": "NON_PARTICIPATING",
236
- "77": "SUCCESS",
237
  "78": "NON_PARTICIPATING",
238
- "79": "NON_PARTICIPATING",
239
- "8": "SUCCESS",
240
  "80": "SUCCESS",
241
- "81": "NON_PARTICIPATING",
242
- "82": "SUCCESS",
243
- "83": "SUCCESS",
244
- "84": "SUCCESS",
245
- "85": "SUCCESS",
246
  "86": "NON_PARTICIPATING",
247
- "87": "SUCCESS",
248
- "88": "SUCCESS",
249
- "89": "SUCCESS",
250
- "9": "SUCCESS",
251
- "90": "SUCCESS",
252
- "91": "SUCCESS",
253
- "92": "NON_PARTICIPATING",
254
  "93": "NON_PARTICIPATING",
255
- "94": "SUCCESS",
256
- "95": "SUCCESS",
257
- "96": "NON_PARTICIPATING",
258
- "97": "SUCCESS",
259
- "98": "SUCCESS",
260
  "99": "SUCCESS"
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
- 5682602,
272
- 5682631
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": 2,
280
- "inner_steps": 0,
281
- "last_allreduce_block": 5682191,
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": "NOT_ALIVE",
5
+ "1": "NON_PARTICIPATING",
6
+ "10": "NOT_ALIVE",
7
+ "100": "NOT_ALIVE",
8
+ "101": "NOT_ALIVE",
9
+ "102": "NOT_ALIVE",
10
+ "103": "NOT_ALIVE",
11
+ "104": "NOT_ALIVE",
12
+ "105": "NOT_ALIVE",
13
+ "106": "NOT_ALIVE",
14
+ "107": "NOT_ALIVE",
15
  "108": "NON_PARTICIPATING",
16
  "109": "NON_PARTICIPATING",
17
+ "11": "NOT_ALIVE",
18
+ "110": "NOT_ALIVE",
19
+ "111": "SUCCESS",
20
+ "112": "NOT_ALIVE",
21
  "113": "NON_PARTICIPATING",
22
  "114": "NON_PARTICIPATING",
23
  "115": "NON_PARTICIPATING",
24
  "116": "NON_PARTICIPATING",
25
  "117": "NON_PARTICIPATING",
26
+ "118": "NOT_ALIVE",
27
+ "119": "NOT_ALIVE",
28
+ "12": "NOT_ALIVE",
29
+ "120": "NOT_ALIVE",
30
  "121": "NON_PARTICIPATING",
31
  "122": "NON_PARTICIPATING",
32
  "123": "NON_PARTICIPATING",
33
+ "124": "NOT_ALIVE",
34
+ "125": "NOT_ALIVE",
35
  "126": "NON_PARTICIPATING",
36
+ "127": "NOT_ALIVE",
37
+ "128": "NOT_ALIVE",
38
  "129": "NON_PARTICIPATING",
39
+ "13": "NON_PARTICIPATING",
40
+ "130": "SUCCESS",
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": "SUCCESS",
52
+ "141": "NOT_ALIVE",
53
+ "142": "NOT_ALIVE",
54
+ "143": "NOT_ALIVE",
55
  "144": "NON_PARTICIPATING",
56
+ "145": "NOT_ALIVE",
57
  "146": "NON_PARTICIPATING",
58
+ "147": "SUCCESS",
59
+ "148": "SUCCESS",
60
+ "149": "NOT_ALIVE",
61
  "15": "SUCCESS",
62
+ "150": "NOT_ALIVE",
63
+ "151": "NOT_ALIVE",
64
+ "152": "NOT_ALIVE",
65
+ "153": "NOT_ALIVE",
66
  "154": "NON_PARTICIPATING",
67
  "155": "NON_PARTICIPATING",
68
+ "156": "NOT_ALIVE",
69
  "157": "NON_PARTICIPATING",
70
+ "158": "NOT_ALIVE",
71
+ "159": "NOT_ALIVE",
72
+ "16": "NOT_ALIVE",
73
+ "160": "NOT_ALIVE",
74
+ "161": "NOT_ALIVE",
75
+ "162": "SUCCESS",
76
+ "163": "NOT_ALIVE",
77
  "164": "NON_PARTICIPATING",
78
  "165": "NON_PARTICIPATING",
79
  "166": "NON_PARTICIPATING",
80
  "167": "NON_PARTICIPATING",
81
+ "168": "NOT_ALIVE",
82
+ "169": "NOT_ALIVE",
83
+ "17": "NOT_ALIVE",
84
+ "170": "NOT_ALIVE",
85
+ "171": "NOT_ALIVE",
86
+ "172": "NOT_ALIVE",
87
  "173": "NON_PARTICIPATING",
88
+ "174": "NOT_ALIVE",
89
+ "175": "NOT_ALIVE",
90
+ "176": "NOT_ALIVE",
91
+ "177": "NOT_ALIVE",
92
+ "178": "NOT_ALIVE",
93
+ "179": "NOT_ALIVE",
94
+ "18": "NOT_ALIVE",
95
+ "180": "NOT_ALIVE",
96
+ "181": "NOT_ALIVE",
97
+ "182": "NOT_ALIVE",
98
  "183": "NON_PARTICIPATING",
99
+ "184": "NOT_ALIVE",
100
+ "185": "NOT_ALIVE",
101
+ "186": "NOT_ALIVE",
102
  "187": "NON_PARTICIPATING",
103
+ "188": "SUCCESS",
104
+ "189": "NOT_ALIVE",
105
+ "19": "NOT_ALIVE",
106
+ "190": "NOT_ALIVE",
107
  "191": "NON_PARTICIPATING",
108
  "192": "NON_PARTICIPATING",
109
  "193": "NON_PARTICIPATING",
110
+ "194": "NOT_ALIVE",
111
+ "195": "NOT_ALIVE",
112
+ "196": "NON_PARTICIPATING",
113
  "197": "NON_PARTICIPATING",
114
+ "198": "NOT_ALIVE",
115
+ "199": "NOT_ALIVE",
116
+ "2": "NON_PARTICIPATING",
117
+ "20": "NOT_ALIVE",
118
  "200": "NON_PARTICIPATING",
119
  "201": "NON_PARTICIPATING",
120
  "202": "NON_PARTICIPATING",
121
+ "203": "NOT_ALIVE",
122
+ "204": "NOT_ALIVE",
123
+ "205": "NOT_ALIVE",
124
+ "206": "NOT_ALIVE",
125
+ "207": "NOT_ALIVE",
126
  "208": "NON_PARTICIPATING",
127
  "209": "NON_PARTICIPATING",
128
  "21": "SUCCESS",
129
  "210": "NON_PARTICIPATING",
130
+ "211": "NOT_ALIVE",
131
+ "212": "NON_PARTICIPATING",
132
+ "213": "NOT_ALIVE",
133
+ "214": "NOT_ALIVE",
134
+ "215": "NOT_ALIVE",
135
  "216": "SUCCESS",
136
+ "217": "NOT_ALIVE",
137
+ "218": "NOT_ALIVE",
138
  "219": "NON_PARTICIPATING",
139
+ "22": "NOT_ALIVE",
140
  "220": "NON_PARTICIPATING",
141
+ "221": "NOT_ALIVE",
142
+ "222": "NOT_ALIVE",
143
+ "223": "NOT_ALIVE",
144
+ "224": "NOT_ALIVE",
145
+ "225": "NOT_ALIVE",
146
+ "226": "NOT_ALIVE",
147
+ "227": "NOT_ALIVE",
148
  "228": "NON_PARTICIPATING",
149
+ "229": "NOT_ALIVE",
150
  "23": "NON_PARTICIPATING",
151
+ "230": "SUCCESS",
152
+ "231": "NOT_ALIVE",
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": "NOT_ALIVE",
160
+ "239": "NOT_ALIVE",
161
+ "24": "NOT_ALIVE",
162
+ "240": "NOT_ALIVE",
163
  "241": "NON_PARTICIPATING",
164
+ "242": "SUCCESS",
165
+ "243": "NOT_ALIVE",
166
+ "244": "SUCCESS",
167
+ "245": "NOT_ALIVE",
168
+ "246": "NOT_ALIVE",
169
+ "247": "SUCCESS",
170
  "248": "NON_PARTICIPATING",
171
  "249": "NON_PARTICIPATING",
172
  "25": "SUCCESS",
173
+ "250": "NON_PARTICIPATING",
174
+ "251": "NOT_ALIVE",
175
+ "252": "NOT_ALIVE",
176
+ "253": "NOT_ALIVE",
177
+ "254": "NOT_ALIVE",
178
  "255": "NON_PARTICIPATING",
179
+ "26": "NON_PARTICIPATING",
180
+ "27": "NOT_ALIVE",
181
+ "28": "NOT_ALIVE",
182
+ "29": "NOT_ALIVE",
183
  "3": "NON_PARTICIPATING",
184
  "30": "NON_PARTICIPATING",
185
+ "31": "NON_PARTICIPATING",
186
  "32": "NON_PARTICIPATING",
187
+ "33": "NOT_ALIVE",
188
+ "34": "NON_PARTICIPATING",
189
+ "35": "NON_PARTICIPATING",
190
  "36": "NON_PARTICIPATING",
191
+ "37": "NOT_ALIVE",
192
+ "38": "NOT_ALIVE",
193
  "39": "NON_PARTICIPATING",
194
+ "4": "NOT_ALIVE",
195
+ "40": "NON_PARTICIPATING",
196
+ "41": "NOT_ALIVE",
197
+ "42": "NON_PARTICIPATING",
198
+ "43": "NOT_ALIVE",
199
+ "44": "NOT_ALIVE",
200
+ "45": "NOT_ALIVE",
201
+ "46": "NON_PARTICIPATING",
202
+ "47": "NON_PARTICIPATING",
203
+ "48": "NOT_ALIVE",
204
+ "49": "NOT_ALIVE",
205
+ "5": "NOT_ALIVE",
206
+ "50": "SUCCESS",
207
+ "51": "NOT_ALIVE",
208
+ "52": "NOT_ALIVE",
209
  "53": "NON_PARTICIPATING",
210
+ "54": "NOT_ALIVE",
211
+ "55": "NOT_ALIVE",
212
+ "56": "NON_PARTICIPATING",
213
+ "57": "NOT_ALIVE",
214
  "58": "NON_PARTICIPATING",
215
+ "59": "NON_PARTICIPATING",
216
+ "6": "NOT_ALIVE",
217
+ "60": "NOT_ALIVE",
218
+ "61": "NON_PARTICIPATING",
219
+ "62": "NOT_ALIVE",
220
+ "63": "NON_PARTICIPATING",
221
+ "64": "NOT_ALIVE",
222
+ "65": "NOT_ALIVE",
223
+ "66": "NOT_ALIVE",
224
+ "67": "NOT_ALIVE",
225
+ "68": "NOT_ALIVE",
226
  "69": "NON_PARTICIPATING",
227
+ "7": "NOT_ALIVE",
228
+ "70": "NON_PARTICIPATING",
229
  "71": "SUCCESS",
230
+ "72": "NON_PARTICIPATING",
231
+ "73": "NON_PARTICIPATING",
232
+ "74": "NON_PARTICIPATING",
233
+ "75": "NON_PARTICIPATING",
234
+ "76": "NOT_ALIVE",
235
+ "77": "NON_PARTICIPATING",
236
  "78": "NON_PARTICIPATING",
237
+ "79": "NOT_ALIVE",
238
+ "8": "NOT_ALIVE",
239
  "80": "SUCCESS",
240
+ "81": "NOT_ALIVE",
241
+ "82": "NON_PARTICIPATING",
242
+ "83": "NON_PARTICIPATING",
243
+ "84": "NOT_ALIVE",
244
+ "85": "NOT_ALIVE",
245
  "86": "NON_PARTICIPATING",
246
+ "87": "NOT_ALIVE",
247
+ "88": "NOT_ALIVE",
248
+ "89": "NOT_ALIVE",
249
+ "9": "NON_PARTICIPATING",
250
+ "90": "NOT_ALIVE",
251
+ "91": "NOT_ALIVE",
252
+ "92": "NOT_ALIVE",
253
  "93": "NON_PARTICIPATING",
254
+ "94": "NOT_ALIVE",
255
+ "95": "NOT_ALIVE",
256
+ "96": "NOT_ALIVE",
257
+ "97": "NON_PARTICIPATING",
258
+ "98": "NOT_ALIVE",
259
  "99": "SUCCESS"
260
  },
261
  "architectures": [
262
+ "LlamaForCausalLM"
263
  ],
264
+ "attention_bias": false,
265
+ "attention_dropout": 0.0,
 
 
 
266
  "block_list": [
267
+ 6125799,
268
+ 6125811
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": 3,
276
+ "intermediate_size": 5632,
277
+ "last_allreduce_block": 6124020,
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
+ }
gradients.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddc48849c184c4df3230c941cb023655eeb581273780f85e866b3cab07a1e306
3
+ size 88389626
inner_optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9072bd436f79bc2da3d3832c82a9b4576e313090a7fe0cc574f415ec5fd67046
3
- size 8081782026
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:493fa5755ca780721c64e856269497896f65324582f70021e3a38ba1936ce43f
3
+ size 8800564076
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e0ac60b6c65aad771927155adb6db3596b7658a38ad0ff3891a996e8df264e5e
3
- size 4040701744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d331c46ae2d35ad06832d2dd8e08962d36c9f5f06dd10d0dd4517e98cb3ab47
3
+ size 4400216536