Run 3. Outer Step 7. Inner Step 63.
Browse files- config.json +52 -52
- inner_optimizer.pt +1 -1
- model.safetensors +1 -1
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"activation_function": "gelu_new",
|
4 |
"all_reduce_scores": {
|
5 |
"0": "NON_PARTICIPATING",
|
@@ -7,14 +7,14 @@
|
|
7 |
"10": "NON_PARTICIPATING",
|
8 |
"100": "NON_PARTICIPATING",
|
9 |
"101": "NON_PARTICIPATING",
|
10 |
-
"102": "
|
11 |
"103": "NON_PARTICIPATING",
|
12 |
"104": "NON_PARTICIPATING",
|
13 |
"105": "NON_PARTICIPATING",
|
14 |
"106": "NON_PARTICIPATING",
|
15 |
"107": "NON_PARTICIPATING",
|
16 |
"108": "NON_PARTICIPATING",
|
17 |
-
"109": "
|
18 |
"11": "NON_PARTICIPATING",
|
19 |
"110": "NON_PARTICIPATING",
|
20 |
"111": "NON_PARTICIPATING",
|
@@ -23,7 +23,7 @@
|
|
23 |
"114": "NON_PARTICIPATING",
|
24 |
"115": "NON_PARTICIPATING",
|
25 |
"116": "NON_PARTICIPATING",
|
26 |
-
"117": "
|
27 |
"118": "NON_PARTICIPATING",
|
28 |
"119": "NON_PARTICIPATING",
|
29 |
"12": "NON_PARTICIPATING",
|
@@ -33,7 +33,7 @@
|
|
33 |
"123": "NON_PARTICIPATING",
|
34 |
"124": "NON_PARTICIPATING",
|
35 |
"125": "NON_PARTICIPATING",
|
36 |
-
"126": "
|
37 |
"127": "NON_PARTICIPATING",
|
38 |
"128": "NON_PARTICIPATING",
|
39 |
"129": "NON_PARTICIPATING",
|
@@ -54,7 +54,7 @@
|
|
54 |
"142": "NON_PARTICIPATING",
|
55 |
"143": "NON_PARTICIPATING",
|
56 |
"144": "NON_PARTICIPATING",
|
57 |
-
"145": "
|
58 |
"146": "NON_PARTICIPATING",
|
59 |
"147": "NON_PARTICIPATING",
|
60 |
"148": "NON_PARTICIPATING",
|
@@ -64,7 +64,7 @@
|
|
64 |
"151": "NON_PARTICIPATING",
|
65 |
"152": "NON_PARTICIPATING",
|
66 |
"153": "NON_PARTICIPATING",
|
67 |
-
"154": "
|
68 |
"155": "NON_PARTICIPATING",
|
69 |
"156": "NON_PARTICIPATING",
|
70 |
"157": "NON_PARTICIPATING",
|
@@ -75,13 +75,13 @@
|
|
75 |
"161": "NON_PARTICIPATING",
|
76 |
"162": "NON_PARTICIPATING",
|
77 |
"163": "NON_PARTICIPATING",
|
78 |
-
"164": "
|
79 |
-
"165": "
|
80 |
"166": "NON_PARTICIPATING",
|
81 |
"167": "NON_PARTICIPATING",
|
82 |
"168": "NON_PARTICIPATING",
|
83 |
"169": "NON_PARTICIPATING",
|
84 |
-
"17": "
|
85 |
"170": "NON_PARTICIPATING",
|
86 |
"171": "NON_PARTICIPATING",
|
87 |
"172": "NON_PARTICIPATING",
|
@@ -89,7 +89,7 @@
|
|
89 |
"174": "NON_PARTICIPATING",
|
90 |
"175": "NON_PARTICIPATING",
|
91 |
"176": "NON_PARTICIPATING",
|
92 |
-
"177": "
|
93 |
"178": "NON_PARTICIPATING",
|
94 |
"179": "NON_PARTICIPATING",
|
95 |
"18": "NON_PARTICIPATING",
|
@@ -100,23 +100,23 @@
|
|
100 |
"184": "NON_PARTICIPATING",
|
101 |
"185": "NON_PARTICIPATING",
|
102 |
"186": "NON_PARTICIPATING",
|
103 |
-
"187": "
|
104 |
"188": "NON_PARTICIPATING",
|
105 |
"189": "NON_PARTICIPATING",
|
106 |
"19": "NON_PARTICIPATING",
|
107 |
"190": "NON_PARTICIPATING",
|
108 |
"191": "NON_PARTICIPATING",
|
109 |
-
"192": "
|
110 |
-
"193": "
|
111 |
"194": "NON_PARTICIPATING",
|
112 |
"195": "NON_PARTICIPATING",
|
113 |
"196": "NON_PARTICIPATING",
|
114 |
"197": "NON_PARTICIPATING",
|
115 |
"198": "NON_PARTICIPATING",
|
116 |
"199": "NON_PARTICIPATING",
|
117 |
-
"2": "
|
118 |
-
"20": "
|
119 |
-
"200": "
|
120 |
"201": "NON_PARTICIPATING",
|
121 |
"202": "NON_PARTICIPATING",
|
122 |
"203": "NON_PARTICIPATING",
|
@@ -135,27 +135,27 @@
|
|
135 |
"215": "NON_PARTICIPATING",
|
136 |
"216": "NON_PARTICIPATING",
|
137 |
"217": "NON_PARTICIPATING",
|
138 |
-
"218": "
|
139 |
"219": "NON_PARTICIPATING",
|
140 |
-
"22": "
|
141 |
"220": "NON_PARTICIPATING",
|
142 |
-
"221": "
|
143 |
"222": "NON_PARTICIPATING",
|
144 |
-
"223": "
|
145 |
"224": "NON_PARTICIPATING",
|
146 |
-
"225": "
|
147 |
-
"226": "
|
148 |
"227": "NON_PARTICIPATING",
|
149 |
"228": "NON_PARTICIPATING",
|
150 |
"229": "NON_PARTICIPATING",
|
151 |
-
"23": "
|
152 |
"230": "NON_PARTICIPATING",
|
153 |
"231": "NON_PARTICIPATING",
|
154 |
"232": "NON_PARTICIPATING",
|
155 |
"233": "NON_PARTICIPATING",
|
156 |
"234": "NON_PARTICIPATING",
|
157 |
-
"235": "
|
158 |
-
"236": "
|
159 |
"237": "NON_PARTICIPATING",
|
160 |
"238": "NON_PARTICIPATING",
|
161 |
"239": "NON_PARTICIPATING",
|
@@ -169,14 +169,14 @@
|
|
169 |
"246": "NON_PARTICIPATING",
|
170 |
"247": "NON_PARTICIPATING",
|
171 |
"248": "NON_PARTICIPATING",
|
172 |
-
"249": "
|
173 |
"25": "SUCCESS",
|
174 |
-
"250": "
|
175 |
"251": "NON_PARTICIPATING",
|
176 |
"252": "NON_PARTICIPATING",
|
177 |
"253": "NON_PARTICIPATING",
|
178 |
"254": "NON_PARTICIPATING",
|
179 |
-
"255": "
|
180 |
"26": "NON_PARTICIPATING",
|
181 |
"27": "NON_PARTICIPATING",
|
182 |
"28": "NON_PARTICIPATING",
|
@@ -184,20 +184,20 @@
|
|
184 |
"3": "NON_PARTICIPATING",
|
185 |
"30": "NON_PARTICIPATING",
|
186 |
"31": "NON_PARTICIPATING",
|
187 |
-
"32": "
|
188 |
"33": "NON_PARTICIPATING",
|
189 |
"34": "NON_PARTICIPATING",
|
190 |
-
"35": "
|
191 |
-
"36": "
|
192 |
-
"37": "
|
193 |
"38": "NON_PARTICIPATING",
|
194 |
"39": "SUCCESS",
|
195 |
"4": "NON_PARTICIPATING",
|
196 |
"40": "NON_PARTICIPATING",
|
197 |
"41": "NON_PARTICIPATING",
|
198 |
-
"42": "
|
199 |
-
"43": "
|
200 |
-
"44": "
|
201 |
"45": "NON_PARTICIPATING",
|
202 |
"46": "NON_PARTICIPATING",
|
203 |
"47": "NON_PARTICIPATING",
|
@@ -206,7 +206,7 @@
|
|
206 |
"5": "NON_PARTICIPATING",
|
207 |
"50": "SUCCESS",
|
208 |
"51": "NON_PARTICIPATING",
|
209 |
-
"52": "
|
210 |
"53": "NON_PARTICIPATING",
|
211 |
"54": "NON_PARTICIPATING",
|
212 |
"55": "NON_PARTICIPATING",
|
@@ -222,12 +222,12 @@
|
|
222 |
"64": "NON_PARTICIPATING",
|
223 |
"65": "NON_PARTICIPATING",
|
224 |
"66": "NON_PARTICIPATING",
|
225 |
-
"67": "
|
226 |
"68": "NON_PARTICIPATING",
|
227 |
"69": "NON_PARTICIPATING",
|
228 |
"7": "NON_PARTICIPATING",
|
229 |
"70": "NON_PARTICIPATING",
|
230 |
-
"71": "
|
231 |
"72": "NON_PARTICIPATING",
|
232 |
"73": "NON_PARTICIPATING",
|
233 |
"74": "NON_PARTICIPATING",
|
@@ -237,11 +237,11 @@
|
|
237 |
"78": "NON_PARTICIPATING",
|
238 |
"79": "NON_PARTICIPATING",
|
239 |
"8": "NON_PARTICIPATING",
|
240 |
-
"80": "
|
241 |
-
"81": "
|
242 |
"82": "NON_PARTICIPATING",
|
243 |
"83": "NON_PARTICIPATING",
|
244 |
-
"84": "
|
245 |
"85": "NON_PARTICIPATING",
|
246 |
"86": "NON_PARTICIPATING",
|
247 |
"87": "NON_PARTICIPATING",
|
@@ -250,13 +250,13 @@
|
|
250 |
"9": "NON_PARTICIPATING",
|
251 |
"90": "NON_PARTICIPATING",
|
252 |
"91": "NON_PARTICIPATING",
|
253 |
-
"92": "
|
254 |
"93": "NON_PARTICIPATING",
|
255 |
"94": "NON_PARTICIPATING",
|
256 |
"95": "NON_PARTICIPATING",
|
257 |
-
"96": "
|
258 |
"97": "NON_PARTICIPATING",
|
259 |
-
"98": "
|
260 |
"99": "NON_PARTICIPATING"
|
261 |
},
|
262 |
"architectures": [
|
@@ -268,20 +268,20 @@
|
|
268 |
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
269 |
},
|
270 |
"block_list": [
|
271 |
-
|
272 |
-
|
273 |
-
|
274 |
-
|
275 |
-
|
276 |
],
|
277 |
"block_size": 1024,
|
278 |
"bos_token_id": 50256,
|
279 |
"embd_pdrop": 0.1,
|
280 |
"eos_token_id": 50256,
|
281 |
"initializer_range": 0.02,
|
282 |
-
"inner_step":
|
283 |
"inner_steps": 0,
|
284 |
-
"last_allreduce_block":
|
285 |
"layer_norm_epsilon": 1e-05,
|
286 |
"model_type": "gpt_optimized",
|
287 |
"n_embd": 1280,
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "distributed/optimized-gpt2-1b",
|
3 |
"activation_function": "gelu_new",
|
4 |
"all_reduce_scores": {
|
5 |
"0": "NON_PARTICIPATING",
|
|
|
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": "NON_PARTICIPATING",
|
16 |
"108": "NON_PARTICIPATING",
|
17 |
+
"109": "NON_PARTICIPATING",
|
18 |
"11": "NON_PARTICIPATING",
|
19 |
"110": "NON_PARTICIPATING",
|
20 |
"111": "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": "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",
|
|
|
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",
|
|
|
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",
|
|
|
75 |
"161": "NON_PARTICIPATING",
|
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",
|
|
|
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": "NON_PARTICIPATING",
|
|
|
100 |
"184": "NON_PARTICIPATING",
|
101 |
"185": "NON_PARTICIPATING",
|
102 |
"186": "NON_PARTICIPATING",
|
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",
|
110 |
+
"193": "NON_PARTICIPATING",
|
111 |
"194": "NON_PARTICIPATING",
|
112 |
"195": "NON_PARTICIPATING",
|
113 |
"196": "NON_PARTICIPATING",
|
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",
|
|
|
135 |
"215": "NON_PARTICIPATING",
|
136 |
"216": "NON_PARTICIPATING",
|
137 |
"217": "NON_PARTICIPATING",
|
138 |
+
"218": "NON_PARTICIPATING",
|
139 |
"219": "NON_PARTICIPATING",
|
140 |
+
"22": "FAIL",
|
141 |
"220": "NON_PARTICIPATING",
|
142 |
+
"221": "NON_PARTICIPATING",
|
143 |
"222": "NON_PARTICIPATING",
|
144 |
+
"223": "NON_PARTICIPATING",
|
145 |
"224": "NON_PARTICIPATING",
|
146 |
+
"225": "NON_PARTICIPATING",
|
147 |
+
"226": "NON_PARTICIPATING",
|
148 |
"227": "NON_PARTICIPATING",
|
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": "NON_PARTICIPATING",
|
156 |
"234": "NON_PARTICIPATING",
|
157 |
+
"235": "NON_PARTICIPATING",
|
158 |
+
"236": "NON_PARTICIPATING",
|
159 |
"237": "NON_PARTICIPATING",
|
160 |
"238": "NON_PARTICIPATING",
|
161 |
"239": "NON_PARTICIPATING",
|
|
|
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",
|
|
|
184 |
"3": "NON_PARTICIPATING",
|
185 |
"30": "NON_PARTICIPATING",
|
186 |
"31": "NON_PARTICIPATING",
|
187 |
+
"32": "NON_PARTICIPATING",
|
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": "FAIL",
|
200 |
+
"44": "SUCCESS",
|
201 |
"45": "NON_PARTICIPATING",
|
202 |
"46": "NON_PARTICIPATING",
|
203 |
"47": "NON_PARTICIPATING",
|
|
|
206 |
"5": "NON_PARTICIPATING",
|
207 |
"50": "SUCCESS",
|
208 |
"51": "NON_PARTICIPATING",
|
209 |
+
"52": "NON_PARTICIPATING",
|
210 |
"53": "NON_PARTICIPATING",
|
211 |
"54": "NON_PARTICIPATING",
|
212 |
"55": "NON_PARTICIPATING",
|
|
|
222 |
"64": "NON_PARTICIPATING",
|
223 |
"65": "NON_PARTICIPATING",
|
224 |
"66": "NON_PARTICIPATING",
|
225 |
+
"67": "FAIL",
|
226 |
"68": "NON_PARTICIPATING",
|
227 |
"69": "NON_PARTICIPATING",
|
228 |
"7": "NON_PARTICIPATING",
|
229 |
"70": "NON_PARTICIPATING",
|
230 |
+
"71": "SUCCESS",
|
231 |
"72": "NON_PARTICIPATING",
|
232 |
"73": "NON_PARTICIPATING",
|
233 |
"74": "NON_PARTICIPATING",
|
|
|
237 |
"78": "NON_PARTICIPATING",
|
238 |
"79": "NON_PARTICIPATING",
|
239 |
"8": "NON_PARTICIPATING",
|
240 |
+
"80": "FAIL",
|
241 |
+
"81": "NON_PARTICIPATING",
|
242 |
"82": "NON_PARTICIPATING",
|
243 |
"83": "NON_PARTICIPATING",
|
244 |
+
"84": "SUCCESS",
|
245 |
"85": "NON_PARTICIPATING",
|
246 |
"86": "NON_PARTICIPATING",
|
247 |
"87": "NON_PARTICIPATING",
|
|
|
250 |
"9": "NON_PARTICIPATING",
|
251 |
"90": "NON_PARTICIPATING",
|
252 |
"91": "NON_PARTICIPATING",
|
253 |
+
"92": "NON_PARTICIPATING",
|
254 |
"93": "NON_PARTICIPATING",
|
255 |
"94": "NON_PARTICIPATING",
|
256 |
"95": "NON_PARTICIPATING",
|
257 |
+
"96": "SUCCESS",
|
258 |
"97": "NON_PARTICIPATING",
|
259 |
+
"98": "FAIL",
|
260 |
"99": "NON_PARTICIPATING"
|
261 |
},
|
262 |
"architectures": [
|
|
|
268 |
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
269 |
},
|
270 |
"block_list": [
|
271 |
+
5394975,
|
272 |
+
5394980,
|
273 |
+
5394984,
|
274 |
+
5394988,
|
275 |
+
5394993
|
276 |
],
|
277 |
"block_size": 1024,
|
278 |
"bos_token_id": 50256,
|
279 |
"embd_pdrop": 0.1,
|
280 |
"eos_token_id": 50256,
|
281 |
"initializer_range": 0.02,
|
282 |
+
"inner_step": 63,
|
283 |
"inner_steps": 0,
|
284 |
+
"last_allreduce_block": 5392742,
|
285 |
"layer_norm_epsilon": 1e-05,
|
286 |
"model_type": "gpt_optimized",
|
287 |
"n_embd": 1280,
|
inner_optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 8081782026
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bf2d34027febdf33dff39ff7c9f9f79c8ee8d062168946379817036cd64ad988
|
3 |
size 8081782026
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4040701744
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4c2a9d55aee651626cc05e59ecb42509a80e9ee2f1b8352bf866c94ead072ded
|
3 |
size 4040701744
|