Outer Step 0. Inner Step 2. Batch Size 176
Browse files- config.json +84 -84
- model.safetensors +1 -1
config.json
CHANGED
@@ -1,35 +1,35 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"activation_function": "gelu_new",
|
4 |
"all_reduce_scores": {
|
5 |
"0": "NON_PARTICIPATING",
|
6 |
-
"1": "
|
7 |
-
"10": "
|
8 |
"100": "NON_PARTICIPATING",
|
9 |
"101": "NON_PARTICIPATING",
|
10 |
-
"102": "
|
11 |
"103": "NON_PARTICIPATING",
|
12 |
"104": "NON_PARTICIPATING",
|
13 |
-
"105": "
|
14 |
-
"106": "
|
15 |
"107": "NON_PARTICIPATING",
|
16 |
-
"108": "
|
17 |
-
"109": "
|
18 |
"11": "NON_PARTICIPATING",
|
19 |
-
"110": "
|
20 |
"111": "NON_PARTICIPATING",
|
21 |
"112": "NON_PARTICIPATING",
|
22 |
"113": "NON_PARTICIPATING",
|
23 |
"114": "NON_PARTICIPATING",
|
24 |
-
"115": "
|
25 |
"116": "NON_PARTICIPATING",
|
26 |
"117": "NON_PARTICIPATING",
|
27 |
-
"118": "
|
28 |
"119": "NON_PARTICIPATING",
|
29 |
-
"12": "
|
30 |
"120": "NON_PARTICIPATING",
|
31 |
"121": "SUCCESS",
|
32 |
-
"122": "
|
33 |
"123": "NON_PARTICIPATING",
|
34 |
"124": "NON_PARTICIPATING",
|
35 |
"125": "NON_PARTICIPATING",
|
@@ -53,10 +53,10 @@
|
|
53 |
"141": "NON_PARTICIPATING",
|
54 |
"142": "NON_PARTICIPATING",
|
55 |
"143": "NON_PARTICIPATING",
|
56 |
-
"144": "
|
57 |
-
"145": "
|
58 |
-
"146": "
|
59 |
-
"147": "
|
60 |
"148": "NON_PARTICIPATING",
|
61 |
"149": "NON_PARTICIPATING",
|
62 |
"15": "NON_PARTICIPATING",
|
@@ -66,10 +66,10 @@
|
|
66 |
"153": "NON_PARTICIPATING",
|
67 |
"154": "NON_PARTICIPATING",
|
68 |
"155": "NON_PARTICIPATING",
|
69 |
-
"156": "
|
70 |
-
"157": "
|
71 |
"158": "NON_PARTICIPATING",
|
72 |
-
"159": "
|
73 |
"16": "NON_PARTICIPATING",
|
74 |
"160": "NON_PARTICIPATING",
|
75 |
"161": "NON_PARTICIPATING",
|
@@ -85,61 +85,61 @@
|
|
85 |
"170": "NON_PARTICIPATING",
|
86 |
"171": "NON_PARTICIPATING",
|
87 |
"172": "NON_PARTICIPATING",
|
88 |
-
"173": "
|
89 |
-
"174": "
|
90 |
"175": "NON_PARTICIPATING",
|
91 |
"176": "NON_PARTICIPATING",
|
92 |
"177": "NON_PARTICIPATING",
|
93 |
"178": "NON_PARTICIPATING",
|
94 |
"179": "NON_PARTICIPATING",
|
95 |
"18": "NON_PARTICIPATING",
|
96 |
-
"180": "
|
97 |
-
"181": "
|
98 |
"182": "NON_PARTICIPATING",
|
99 |
"183": "NON_PARTICIPATING",
|
100 |
"184": "NON_PARTICIPATING",
|
101 |
"185": "NON_PARTICIPATING",
|
102 |
-
"186": "
|
103 |
-
"187": "
|
104 |
-
"188": "
|
105 |
"189": "NON_PARTICIPATING",
|
106 |
"19": "NON_PARTICIPATING",
|
107 |
"190": "NON_PARTICIPATING",
|
108 |
"191": "NON_PARTICIPATING",
|
109 |
"192": "NON_PARTICIPATING",
|
110 |
-
"193": "
|
111 |
"194": "NON_PARTICIPATING",
|
112 |
-
"195": "
|
113 |
-
"196": "
|
114 |
"197": "NON_PARTICIPATING",
|
115 |
-
"198": "
|
116 |
"199": "NON_PARTICIPATING",
|
117 |
"2": "NON_PARTICIPATING",
|
118 |
"20": "NON_PARTICIPATING",
|
119 |
"200": "NON_PARTICIPATING",
|
120 |
"201": "NON_PARTICIPATING",
|
121 |
"202": "NON_PARTICIPATING",
|
122 |
-
"203": "
|
123 |
"204": "NON_PARTICIPATING",
|
124 |
"205": "NON_PARTICIPATING",
|
125 |
"206": "NON_PARTICIPATING",
|
126 |
"207": "NON_PARTICIPATING",
|
127 |
"208": "NON_PARTICIPATING",
|
128 |
"209": "NON_PARTICIPATING",
|
129 |
-
"21": "
|
130 |
"210": "NON_PARTICIPATING",
|
131 |
"211": "NON_PARTICIPATING",
|
132 |
"212": "NON_PARTICIPATING",
|
133 |
-
"213": "
|
134 |
"214": "NON_PARTICIPATING",
|
135 |
"215": "NON_PARTICIPATING",
|
136 |
"216": "NON_PARTICIPATING",
|
137 |
"217": "NON_PARTICIPATING",
|
138 |
"218": "NON_PARTICIPATING",
|
139 |
-
"219": "
|
140 |
"22": "NON_PARTICIPATING",
|
141 |
"220": "NON_PARTICIPATING",
|
142 |
-
"221": "
|
143 |
"222": "NON_PARTICIPATING",
|
144 |
"223": "NON_PARTICIPATING",
|
145 |
"224": "NON_PARTICIPATING",
|
@@ -149,32 +149,32 @@
|
|
149 |
"228": "NON_PARTICIPATING",
|
150 |
"229": "NON_PARTICIPATING",
|
151 |
"23": "NON_PARTICIPATING",
|
152 |
-
"230": "
|
153 |
"231": "NON_PARTICIPATING",
|
154 |
-
"232": "
|
155 |
"233": "NON_PARTICIPATING",
|
156 |
"234": "SUCCESS",
|
157 |
"235": "SUCCESS",
|
158 |
-
"236": "
|
159 |
"237": "SUCCESS",
|
160 |
"238": "NON_PARTICIPATING",
|
161 |
"239": "NON_PARTICIPATING",
|
162 |
"24": "NON_PARTICIPATING",
|
163 |
"240": "NON_PARTICIPATING",
|
164 |
-
"241": "
|
165 |
-
"242": "
|
166 |
"243": "NON_PARTICIPATING",
|
167 |
"244": "NON_PARTICIPATING",
|
168 |
"245": "NON_PARTICIPATING",
|
169 |
-
"246": "
|
170 |
"247": "NON_PARTICIPATING",
|
171 |
"248": "NON_PARTICIPATING",
|
172 |
"249": "NON_PARTICIPATING",
|
173 |
-
"25": "
|
174 |
"250": "NON_PARTICIPATING",
|
175 |
-
"251": "
|
176 |
-
"252": "
|
177 |
-
"253": "
|
178 |
"254": "NON_PARTICIPATING",
|
179 |
"255": "NON_PARTICIPATING",
|
180 |
"26": "NON_PARTICIPATING",
|
@@ -182,8 +182,8 @@
|
|
182 |
"28": "NON_PARTICIPATING",
|
183 |
"29": "NON_PARTICIPATING",
|
184 |
"3": "NON_PARTICIPATING",
|
185 |
-
"30": "
|
186 |
-
"31": "
|
187 |
"32": "NON_PARTICIPATING",
|
188 |
"33": "NON_PARTICIPATING",
|
189 |
"34": "NON_PARTICIPATING",
|
@@ -192,72 +192,72 @@
|
|
192 |
"37": "NON_PARTICIPATING",
|
193 |
"38": "NON_PARTICIPATING",
|
194 |
"39": "NON_PARTICIPATING",
|
195 |
-
"4": "
|
196 |
-
"40": "
|
197 |
"41": "NON_PARTICIPATING",
|
198 |
"42": "NON_PARTICIPATING",
|
199 |
"43": "NON_PARTICIPATING",
|
200 |
-
"44": "
|
201 |
"45": "NON_PARTICIPATING",
|
202 |
"46": "NON_PARTICIPATING",
|
203 |
"47": "SUCCESS",
|
204 |
"48": "NON_PARTICIPATING",
|
205 |
"49": "NON_PARTICIPATING",
|
206 |
"5": "NON_PARTICIPATING",
|
207 |
-
"50": "
|
208 |
"51": "SUCCESS",
|
209 |
"52": "NON_PARTICIPATING",
|
210 |
"53": "NON_PARTICIPATING",
|
211 |
"54": "NON_PARTICIPATING",
|
212 |
"55": "NON_PARTICIPATING",
|
213 |
-
"56": "
|
214 |
"57": "NON_PARTICIPATING",
|
215 |
-
"58": "
|
216 |
"59": "NON_PARTICIPATING",
|
217 |
"6": "NON_PARTICIPATING",
|
218 |
"60": "NON_PARTICIPATING",
|
219 |
"61": "NON_PARTICIPATING",
|
220 |
-
"62": "
|
221 |
-
"63": "
|
222 |
-
"64": "
|
223 |
"65": "SUCCESS",
|
224 |
"66": "NON_PARTICIPATING",
|
225 |
"67": "SUCCESS",
|
226 |
-
"68": "
|
227 |
"69": "SUCCESS",
|
228 |
"7": "NON_PARTICIPATING",
|
229 |
"70": "NON_PARTICIPATING",
|
230 |
"71": "NON_PARTICIPATING",
|
231 |
-
"72": "
|
232 |
"73": "NON_PARTICIPATING",
|
233 |
-
"74": "
|
234 |
"75": "NON_PARTICIPATING",
|
235 |
-
"76": "
|
236 |
"77": "NON_PARTICIPATING",
|
237 |
-
"78": "
|
238 |
-
"79": "
|
239 |
"8": "NON_PARTICIPATING",
|
240 |
-
"80": "
|
241 |
-
"81": "
|
242 |
"82": "NON_PARTICIPATING",
|
243 |
-
"83": "
|
244 |
-
"84": "
|
245 |
-
"85": "
|
246 |
-
"86": "
|
247 |
-
"87": "
|
248 |
-
"88": "
|
249 |
-
"89": "
|
250 |
"9": "NON_PARTICIPATING",
|
251 |
"90": "SUCCESS",
|
252 |
"91": "NON_PARTICIPATING",
|
253 |
"92": "NON_PARTICIPATING",
|
254 |
-
"93": "
|
255 |
-
"94": "
|
256 |
"95": "SUCCESS",
|
257 |
-
"96": "
|
258 |
-
"97": "
|
259 |
"98": "NON_PARTICIPATING",
|
260 |
-
"99": "
|
261 |
},
|
262 |
"architectures": [
|
263 |
"GPTOptim"
|
@@ -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",
|
6 |
+
"1": "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": "SUCCESS",
|
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": "SUCCESS",
|
32 |
+
"122": "NON_PARTICIPATING",
|
33 |
"123": "NON_PARTICIPATING",
|
34 |
"124": "NON_PARTICIPATING",
|
35 |
"125": "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": "SUCCESS",
|
60 |
"148": "NON_PARTICIPATING",
|
61 |
"149": "NON_PARTICIPATING",
|
62 |
"15": "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": "NON_PARTICIPATING",
|
74 |
"160": "NON_PARTICIPATING",
|
75 |
"161": "NON_PARTICIPATING",
|
|
|
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": "NON_PARTICIPATING",
|
96 |
+
"180": "NON_PARTICIPATING",
|
97 |
+
"181": "NON_PARTICIPATING",
|
98 |
"182": "NON_PARTICIPATING",
|
99 |
"183": "NON_PARTICIPATING",
|
100 |
"184": "NON_PARTICIPATING",
|
101 |
"185": "NON_PARTICIPATING",
|
102 |
+
"186": "NON_PARTICIPATING",
|
103 |
+
"187": "SUCCESS",
|
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": "SUCCESS",
|
114 |
"197": "NON_PARTICIPATING",
|
115 |
+
"198": "NON_PARTICIPATING",
|
116 |
"199": "NON_PARTICIPATING",
|
117 |
"2": "NON_PARTICIPATING",
|
118 |
"20": "NON_PARTICIPATING",
|
119 |
"200": "NON_PARTICIPATING",
|
120 |
"201": "NON_PARTICIPATING",
|
121 |
"202": "NON_PARTICIPATING",
|
122 |
+
"203": "NON_PARTICIPATING",
|
123 |
"204": "NON_PARTICIPATING",
|
124 |
"205": "NON_PARTICIPATING",
|
125 |
"206": "NON_PARTICIPATING",
|
126 |
"207": "NON_PARTICIPATING",
|
127 |
"208": "NON_PARTICIPATING",
|
128 |
"209": "NON_PARTICIPATING",
|
129 |
+
"21": "NON_PARTICIPATING",
|
130 |
"210": "NON_PARTICIPATING",
|
131 |
"211": "NON_PARTICIPATING",
|
132 |
"212": "NON_PARTICIPATING",
|
133 |
+
"213": "SUCCESS",
|
134 |
"214": "NON_PARTICIPATING",
|
135 |
"215": "NON_PARTICIPATING",
|
136 |
"216": "NON_PARTICIPATING",
|
137 |
"217": "NON_PARTICIPATING",
|
138 |
"218": "NON_PARTICIPATING",
|
139 |
+
"219": "NON_PARTICIPATING",
|
140 |
"22": "NON_PARTICIPATING",
|
141 |
"220": "NON_PARTICIPATING",
|
142 |
+
"221": "NON_PARTICIPATING",
|
143 |
"222": "NON_PARTICIPATING",
|
144 |
"223": "NON_PARTICIPATING",
|
145 |
"224": "NON_PARTICIPATING",
|
|
|
149 |
"228": "NON_PARTICIPATING",
|
150 |
"229": "NON_PARTICIPATING",
|
151 |
"23": "NON_PARTICIPATING",
|
152 |
+
"230": "NON_PARTICIPATING",
|
153 |
"231": "NON_PARTICIPATING",
|
154 |
+
"232": "SUCCESS",
|
155 |
"233": "NON_PARTICIPATING",
|
156 |
"234": "SUCCESS",
|
157 |
"235": "SUCCESS",
|
158 |
+
"236": "NON_PARTICIPATING",
|
159 |
"237": "SUCCESS",
|
160 |
"238": "NON_PARTICIPATING",
|
161 |
"239": "NON_PARTICIPATING",
|
162 |
"24": "NON_PARTICIPATING",
|
163 |
"240": "NON_PARTICIPATING",
|
164 |
+
"241": "SUCCESS",
|
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": "NON_PARTICIPATING",
|
175 |
+
"251": "NON_PARTICIPATING",
|
176 |
+
"252": "NON_PARTICIPATING",
|
177 |
+
"253": "SUCCESS",
|
178 |
"254": "NON_PARTICIPATING",
|
179 |
"255": "NON_PARTICIPATING",
|
180 |
"26": "NON_PARTICIPATING",
|
|
|
182 |
"28": "NON_PARTICIPATING",
|
183 |
"29": "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",
|
|
|
192 |
"37": "NON_PARTICIPATING",
|
193 |
"38": "NON_PARTICIPATING",
|
194 |
"39": "NON_PARTICIPATING",
|
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": "NON_PARTICIPATING",
|
202 |
"46": "NON_PARTICIPATING",
|
203 |
"47": "SUCCESS",
|
204 |
"48": "NON_PARTICIPATING",
|
205 |
"49": "NON_PARTICIPATING",
|
206 |
"5": "NON_PARTICIPATING",
|
207 |
+
"50": "NON_PARTICIPATING",
|
208 |
"51": "SUCCESS",
|
209 |
"52": "NON_PARTICIPATING",
|
210 |
"53": "NON_PARTICIPATING",
|
211 |
"54": "NON_PARTICIPATING",
|
212 |
"55": "NON_PARTICIPATING",
|
213 |
+
"56": "NON_PARTICIPATING",
|
214 |
"57": "NON_PARTICIPATING",
|
215 |
+
"58": "NON_PARTICIPATING",
|
216 |
"59": "NON_PARTICIPATING",
|
217 |
"6": "NON_PARTICIPATING",
|
218 |
"60": "NON_PARTICIPATING",
|
219 |
"61": "NON_PARTICIPATING",
|
220 |
+
"62": "NON_PARTICIPATING",
|
221 |
+
"63": "NON_PARTICIPATING",
|
222 |
+
"64": "NON_PARTICIPATING",
|
223 |
"65": "SUCCESS",
|
224 |
"66": "NON_PARTICIPATING",
|
225 |
"67": "SUCCESS",
|
226 |
+
"68": "NON_PARTICIPATING",
|
227 |
"69": "SUCCESS",
|
228 |
"7": "NON_PARTICIPATING",
|
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": "NON_PARTICIPATING",
|
236 |
"77": "NON_PARTICIPATING",
|
237 |
+
"78": "NON_PARTICIPATING",
|
238 |
+
"79": "NON_PARTICIPATING",
|
239 |
"8": "NON_PARTICIPATING",
|
240 |
+
"80": "NON_PARTICIPATING",
|
241 |
+
"81": "NON_PARTICIPATING",
|
242 |
"82": "NON_PARTICIPATING",
|
243 |
+
"83": "NON_PARTICIPATING",
|
244 |
+
"84": "NON_PARTICIPATING",
|
245 |
+
"85": "NON_PARTICIPATING",
|
246 |
+
"86": "NON_PARTICIPATING",
|
247 |
+
"87": "NON_PARTICIPATING",
|
248 |
+
"88": "NON_PARTICIPATING",
|
249 |
+
"89": "NON_PARTICIPATING",
|
250 |
"9": "NON_PARTICIPATING",
|
251 |
"90": "SUCCESS",
|
252 |
"91": "NON_PARTICIPATING",
|
253 |
"92": "NON_PARTICIPATING",
|
254 |
+
"93": "NON_PARTICIPATING",
|
255 |
+
"94": "NON_PARTICIPATING",
|
256 |
"95": "SUCCESS",
|
257 |
+
"96": "NON_PARTICIPATING",
|
258 |
+
"97": "NON_PARTICIPATING",
|
259 |
"98": "NON_PARTICIPATING",
|
260 |
+
"99": "NON_PARTICIPATING"
|
261 |
},
|
262 |
"architectures": [
|
263 |
"GPTOptim"
|
|
|
268 |
"AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
|
269 |
},
|
270 |
"block_list": [
|
271 |
+
5129440,
|
272 |
+
5129451,
|
273 |
+
5129456,
|
274 |
+
5129465,
|
275 |
+
5129474
|
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": 2,
|
283 |
"inner_steps": 0,
|
284 |
+
"last_allreduce_block": 5072405,
|
285 |
"layer_norm_epsilon": 1e-05,
|
286 |
"model_type": "gpt_optimized",
|
287 |
"n_embd": 1280,
|
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 sha256:18bef1305a524e82d2ff19fbef3592aef223e870c5d9de4d2ea37ad3ec59f283
|
3 |
size 4040701744
|