Chauhan
commited on
Commit
•
30a49b0
1
Parent(s):
57cb960
End of training
Browse files- all_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +442 -92
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 4.
|
7 |
-
"train_steps_per_second": 4.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"train_loss": 0.22781734634577194,
|
4 |
+
"train_runtime": 7496.2711,
|
5 |
+
"train_samples": 3519,
|
6 |
+
"train_samples_per_second": 4.694,
|
7 |
+
"train_steps_per_second": 4.694
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples":
|
6 |
-
"train_samples_per_second": 4.
|
7 |
-
"train_steps_per_second": 4.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"train_loss": 0.22781734634577194,
|
4 |
+
"train_runtime": 7496.2711,
|
5 |
+
"train_samples": 3519,
|
6 |
+
"train_samples_per_second": 4.694,
|
7 |
+
"train_steps_per_second": 4.694
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,169 +1,519 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
-
"epoch": 0.
|
13 |
-
"grad_norm":
|
14 |
-
"learning_rate": 4.
|
15 |
-
"loss":
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
-
"epoch":
|
20 |
-
"grad_norm":
|
21 |
-
"learning_rate": 4.
|
22 |
-
"loss":
|
23 |
"step": 1000
|
24 |
},
|
25 |
{
|
26 |
-
"epoch":
|
27 |
-
"grad_norm":
|
28 |
-
"learning_rate": 4.
|
29 |
-
"loss":
|
30 |
"step": 1500
|
31 |
},
|
32 |
{
|
33 |
-
"epoch":
|
34 |
-
"grad_norm":
|
35 |
-
"learning_rate":
|
36 |
-
"loss": 0.
|
37 |
"step": 2000
|
38 |
},
|
39 |
{
|
40 |
-
"epoch":
|
41 |
-
"grad_norm":
|
42 |
-
"learning_rate":
|
43 |
-
"loss": 0.
|
44 |
"step": 2500
|
45 |
},
|
46 |
{
|
47 |
-
"epoch":
|
48 |
-
"grad_norm":
|
49 |
-
"learning_rate":
|
50 |
-
"loss": 0.
|
51 |
"step": 3000
|
52 |
},
|
53 |
{
|
54 |
-
"epoch":
|
55 |
-
"grad_norm":
|
56 |
-
"learning_rate":
|
57 |
-
"loss": 0.
|
58 |
"step": 3500
|
59 |
},
|
60 |
{
|
61 |
-
"epoch":
|
62 |
-
"grad_norm":
|
63 |
-
"learning_rate":
|
64 |
-
"loss": 0.
|
65 |
"step": 4000
|
66 |
},
|
67 |
{
|
68 |
-
"epoch":
|
69 |
-
"grad_norm":
|
70 |
-
"learning_rate":
|
71 |
-
"loss": 0.
|
72 |
"step": 4500
|
73 |
},
|
74 |
{
|
75 |
-
"epoch":
|
76 |
-
"grad_norm":
|
77 |
-
"learning_rate":
|
78 |
-
"loss": 0.
|
79 |
"step": 5000
|
80 |
},
|
81 |
{
|
82 |
-
"epoch":
|
83 |
-
"grad_norm":
|
84 |
-
"learning_rate":
|
85 |
-
"loss": 0.
|
86 |
"step": 5500
|
87 |
},
|
88 |
{
|
89 |
-
"epoch":
|
90 |
-
"grad_norm":
|
91 |
-
"learning_rate":
|
92 |
-
"loss": 0.
|
93 |
"step": 6000
|
94 |
},
|
95 |
{
|
96 |
-
"epoch":
|
97 |
-
"grad_norm":
|
98 |
-
"learning_rate":
|
99 |
-
"loss": 0.
|
100 |
"step": 6500
|
101 |
},
|
102 |
{
|
103 |
-
"epoch":
|
104 |
-
"grad_norm":
|
105 |
-
"learning_rate":
|
106 |
-
"loss": 0.
|
107 |
"step": 7000
|
108 |
},
|
109 |
{
|
110 |
-
"epoch":
|
111 |
-
"grad_norm":
|
112 |
-
"learning_rate":
|
113 |
-
"loss": 0.
|
114 |
"step": 7500
|
115 |
},
|
116 |
{
|
117 |
-
"epoch":
|
118 |
-
"grad_norm":
|
119 |
-
"learning_rate":
|
120 |
-
"loss": 0.
|
121 |
"step": 8000
|
122 |
},
|
123 |
{
|
124 |
-
"epoch":
|
125 |
-
"grad_norm":
|
126 |
-
"learning_rate":
|
127 |
-
"loss": 0.
|
128 |
"step": 8500
|
129 |
},
|
130 |
{
|
131 |
-
"epoch":
|
132 |
-
"grad_norm":
|
133 |
-
"learning_rate":
|
134 |
-
"loss": 0.
|
135 |
"step": 9000
|
136 |
},
|
137 |
{
|
138 |
-
"epoch":
|
139 |
-
"grad_norm":
|
140 |
-
"learning_rate":
|
141 |
-
"loss": 0.
|
142 |
"step": 9500
|
143 |
},
|
144 |
{
|
145 |
-
"epoch":
|
146 |
-
"grad_norm":
|
147 |
-
"learning_rate":
|
148 |
-
"loss": 0.
|
149 |
"step": 10000
|
150 |
},
|
151 |
{
|
152 |
-
"epoch":
|
153 |
-
"
|
154 |
-
"
|
155 |
-
"
|
156 |
-
"
|
157 |
-
|
158 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
159 |
}
|
160 |
],
|
161 |
"logging_steps": 500,
|
162 |
-
"max_steps":
|
163 |
"num_input_tokens_seen": 0,
|
164 |
-
"num_train_epochs":
|
165 |
"save_steps": 1000,
|
166 |
-
"total_flos":
|
167 |
"train_batch_size": 1,
|
168 |
"trial_name": null,
|
169 |
"trial_params": null
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 10.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 35190,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"epoch": 0.14,
|
13 |
+
"grad_norm": 2.325218677520752,
|
14 |
+
"learning_rate": 4.92895709008241e-05,
|
15 |
+
"loss": 0.8164,
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
+
"epoch": 0.28,
|
20 |
+
"grad_norm": 2.017394781112671,
|
21 |
+
"learning_rate": 4.85791418016482e-05,
|
22 |
+
"loss": 0.6693,
|
23 |
"step": 1000
|
24 |
},
|
25 |
{
|
26 |
+
"epoch": 0.43,
|
27 |
+
"grad_norm": 1.5872315168380737,
|
28 |
+
"learning_rate": 4.78687127024723e-05,
|
29 |
+
"loss": 0.6128,
|
30 |
"step": 1500
|
31 |
},
|
32 |
{
|
33 |
+
"epoch": 0.57,
|
34 |
+
"grad_norm": 1.8339617252349854,
|
35 |
+
"learning_rate": 4.71582836032964e-05,
|
36 |
+
"loss": 0.5742,
|
37 |
"step": 2000
|
38 |
},
|
39 |
{
|
40 |
+
"epoch": 0.71,
|
41 |
+
"grad_norm": 1.8466053009033203,
|
42 |
+
"learning_rate": 4.644785450412049e-05,
|
43 |
+
"loss": 0.5523,
|
44 |
"step": 2500
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 0.85,
|
48 |
+
"grad_norm": 1.4517784118652344,
|
49 |
+
"learning_rate": 4.573742540494459e-05,
|
50 |
+
"loss": 0.5346,
|
51 |
"step": 3000
|
52 |
},
|
53 |
{
|
54 |
+
"epoch": 0.99,
|
55 |
+
"grad_norm": 1.3925143480300903,
|
56 |
+
"learning_rate": 4.5026996305768686e-05,
|
57 |
+
"loss": 0.5028,
|
58 |
"step": 3500
|
59 |
},
|
60 |
{
|
61 |
+
"epoch": 1.14,
|
62 |
+
"grad_norm": 1.220083236694336,
|
63 |
+
"learning_rate": 4.4316567206592784e-05,
|
64 |
+
"loss": 0.4355,
|
65 |
"step": 4000
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 1.28,
|
69 |
+
"grad_norm": 1.5840811729431152,
|
70 |
+
"learning_rate": 4.360613810741688e-05,
|
71 |
+
"loss": 0.4228,
|
72 |
"step": 4500
|
73 |
},
|
74 |
{
|
75 |
+
"epoch": 1.42,
|
76 |
+
"grad_norm": 1.6422966718673706,
|
77 |
+
"learning_rate": 4.289570900824098e-05,
|
78 |
+
"loss": 0.4009,
|
79 |
"step": 5000
|
80 |
},
|
81 |
{
|
82 |
+
"epoch": 1.56,
|
83 |
+
"grad_norm": 1.7338138818740845,
|
84 |
+
"learning_rate": 4.218527990906508e-05,
|
85 |
+
"loss": 0.3916,
|
86 |
"step": 5500
|
87 |
},
|
88 |
{
|
89 |
+
"epoch": 1.71,
|
90 |
+
"grad_norm": 1.0937657356262207,
|
91 |
+
"learning_rate": 4.147485080988918e-05,
|
92 |
+
"loss": 0.3851,
|
93 |
"step": 6000
|
94 |
},
|
95 |
{
|
96 |
+
"epoch": 1.85,
|
97 |
+
"grad_norm": 1.1090480089187622,
|
98 |
+
"learning_rate": 4.076442171071328e-05,
|
99 |
+
"loss": 0.3672,
|
100 |
"step": 6500
|
101 |
},
|
102 |
{
|
103 |
+
"epoch": 1.99,
|
104 |
+
"grad_norm": 1.5349233150482178,
|
105 |
+
"learning_rate": 4.005399261153737e-05,
|
106 |
+
"loss": 0.3659,
|
107 |
"step": 7000
|
108 |
},
|
109 |
{
|
110 |
+
"epoch": 2.13,
|
111 |
+
"grad_norm": 1.6046255826950073,
|
112 |
+
"learning_rate": 3.934356351236147e-05,
|
113 |
+
"loss": 0.3117,
|
114 |
"step": 7500
|
115 |
},
|
116 |
{
|
117 |
+
"epoch": 2.27,
|
118 |
+
"grad_norm": 1.3127567768096924,
|
119 |
+
"learning_rate": 3.8633134413185566e-05,
|
120 |
+
"loss": 0.3043,
|
121 |
"step": 8000
|
122 |
},
|
123 |
{
|
124 |
+
"epoch": 2.42,
|
125 |
+
"grad_norm": 1.3286867141723633,
|
126 |
+
"learning_rate": 3.7922705314009665e-05,
|
127 |
+
"loss": 0.3029,
|
128 |
"step": 8500
|
129 |
},
|
130 |
{
|
131 |
+
"epoch": 2.56,
|
132 |
+
"grad_norm": 0.8038182258605957,
|
133 |
+
"learning_rate": 3.721227621483376e-05,
|
134 |
+
"loss": 0.2934,
|
135 |
"step": 9000
|
136 |
},
|
137 |
{
|
138 |
+
"epoch": 2.7,
|
139 |
+
"grad_norm": 1.097979187965393,
|
140 |
+
"learning_rate": 3.650184711565786e-05,
|
141 |
+
"loss": 0.2822,
|
142 |
"step": 9500
|
143 |
},
|
144 |
{
|
145 |
+
"epoch": 2.84,
|
146 |
+
"grad_norm": 1.42654287815094,
|
147 |
+
"learning_rate": 3.579141801648196e-05,
|
148 |
+
"loss": 0.2768,
|
149 |
"step": 10000
|
150 |
},
|
151 |
{
|
152 |
+
"epoch": 2.98,
|
153 |
+
"grad_norm": 1.3376920223236084,
|
154 |
+
"learning_rate": 3.508098891730606e-05,
|
155 |
+
"loss": 0.2733,
|
156 |
+
"step": 10500
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 3.13,
|
160 |
+
"grad_norm": 1.3901501893997192,
|
161 |
+
"learning_rate": 3.437055981813016e-05,
|
162 |
+
"loss": 0.2358,
|
163 |
+
"step": 11000
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 3.27,
|
167 |
+
"grad_norm": 0.9437574148178101,
|
168 |
+
"learning_rate": 3.366013071895425e-05,
|
169 |
+
"loss": 0.2347,
|
170 |
+
"step": 11500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 3.41,
|
174 |
+
"grad_norm": 1.7849817276000977,
|
175 |
+
"learning_rate": 3.294970161977835e-05,
|
176 |
+
"loss": 0.2282,
|
177 |
+
"step": 12000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 3.55,
|
181 |
+
"grad_norm": 1.3517547845840454,
|
182 |
+
"learning_rate": 3.2239272520602446e-05,
|
183 |
+
"loss": 0.2196,
|
184 |
+
"step": 12500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 3.69,
|
188 |
+
"grad_norm": 1.3112218379974365,
|
189 |
+
"learning_rate": 3.152884342142654e-05,
|
190 |
+
"loss": 0.2196,
|
191 |
+
"step": 13000
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 3.84,
|
195 |
+
"grad_norm": 1.532402515411377,
|
196 |
+
"learning_rate": 3.081841432225064e-05,
|
197 |
+
"loss": 0.222,
|
198 |
+
"step": 13500
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 3.98,
|
202 |
+
"grad_norm": 1.0918519496917725,
|
203 |
+
"learning_rate": 3.0107985223074735e-05,
|
204 |
+
"loss": 0.217,
|
205 |
+
"step": 14000
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 4.12,
|
209 |
+
"grad_norm": 1.1216357946395874,
|
210 |
+
"learning_rate": 2.9397556123898834e-05,
|
211 |
+
"loss": 0.1902,
|
212 |
+
"step": 14500
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 4.26,
|
216 |
+
"grad_norm": 0.9181265234947205,
|
217 |
+
"learning_rate": 2.8687127024722932e-05,
|
218 |
+
"loss": 0.1854,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 4.4,
|
223 |
+
"grad_norm": 1.3160160779953003,
|
224 |
+
"learning_rate": 2.797669792554703e-05,
|
225 |
+
"loss": 0.1784,
|
226 |
+
"step": 15500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 4.55,
|
230 |
+
"grad_norm": 1.198168396949768,
|
231 |
+
"learning_rate": 2.7266268826371126e-05,
|
232 |
+
"loss": 0.1812,
|
233 |
+
"step": 16000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 4.69,
|
237 |
+
"grad_norm": 1.515939474105835,
|
238 |
+
"learning_rate": 2.6555839727195225e-05,
|
239 |
+
"loss": 0.1824,
|
240 |
+
"step": 16500
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 4.83,
|
244 |
+
"grad_norm": 1.1988515853881836,
|
245 |
+
"learning_rate": 2.5845410628019323e-05,
|
246 |
+
"loss": 0.178,
|
247 |
+
"step": 17000
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 4.97,
|
251 |
+
"grad_norm": 0.8168569803237915,
|
252 |
+
"learning_rate": 2.5134981528843422e-05,
|
253 |
+
"loss": 0.1758,
|
254 |
+
"step": 17500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 5.12,
|
258 |
+
"grad_norm": 1.039239764213562,
|
259 |
+
"learning_rate": 2.442455242966752e-05,
|
260 |
+
"loss": 0.1517,
|
261 |
+
"step": 18000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 5.26,
|
265 |
+
"grad_norm": 0.9721378087997437,
|
266 |
+
"learning_rate": 2.371412333049162e-05,
|
267 |
+
"loss": 0.1452,
|
268 |
+
"step": 18500
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 5.4,
|
272 |
+
"grad_norm": 1.0437672138214111,
|
273 |
+
"learning_rate": 2.3003694231315714e-05,
|
274 |
+
"loss": 0.1513,
|
275 |
+
"step": 19000
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 5.54,
|
279 |
+
"grad_norm": 1.620819091796875,
|
280 |
+
"learning_rate": 2.2293265132139813e-05,
|
281 |
+
"loss": 0.1539,
|
282 |
+
"step": 19500
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 5.68,
|
286 |
+
"grad_norm": 1.2683171033859253,
|
287 |
+
"learning_rate": 2.158283603296391e-05,
|
288 |
+
"loss": 0.152,
|
289 |
+
"step": 20000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 5.83,
|
293 |
+
"grad_norm": 1.0495145320892334,
|
294 |
+
"learning_rate": 2.0872406933788007e-05,
|
295 |
+
"loss": 0.1435,
|
296 |
+
"step": 20500
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 5.97,
|
300 |
+
"grad_norm": 1.5612547397613525,
|
301 |
+
"learning_rate": 2.0161977834612105e-05,
|
302 |
+
"loss": 0.1471,
|
303 |
+
"step": 21000
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 6.11,
|
307 |
+
"grad_norm": 1.3628000020980835,
|
308 |
+
"learning_rate": 1.9451548735436204e-05,
|
309 |
+
"loss": 0.1349,
|
310 |
+
"step": 21500
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 6.25,
|
314 |
+
"grad_norm": 0.9959810376167297,
|
315 |
+
"learning_rate": 1.8741119636260302e-05,
|
316 |
+
"loss": 0.1229,
|
317 |
+
"step": 22000
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 6.39,
|
321 |
+
"grad_norm": 0.9821630120277405,
|
322 |
+
"learning_rate": 1.80306905370844e-05,
|
323 |
+
"loss": 0.1288,
|
324 |
+
"step": 22500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 6.54,
|
328 |
+
"grad_norm": 1.5605982542037964,
|
329 |
+
"learning_rate": 1.7320261437908496e-05,
|
330 |
+
"loss": 0.1285,
|
331 |
+
"step": 23000
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 6.68,
|
335 |
+
"grad_norm": 1.138551115989685,
|
336 |
+
"learning_rate": 1.6609832338732595e-05,
|
337 |
+
"loss": 0.128,
|
338 |
+
"step": 23500
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 6.82,
|
342 |
+
"grad_norm": 0.9666039347648621,
|
343 |
+
"learning_rate": 1.5899403239556693e-05,
|
344 |
+
"loss": 0.127,
|
345 |
+
"step": 24000
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 6.96,
|
349 |
+
"grad_norm": 0.7725051641464233,
|
350 |
+
"learning_rate": 1.518897414038079e-05,
|
351 |
+
"loss": 0.1268,
|
352 |
+
"step": 24500
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 7.1,
|
356 |
+
"grad_norm": 0.8455806970596313,
|
357 |
+
"learning_rate": 1.4478545041204889e-05,
|
358 |
+
"loss": 0.1077,
|
359 |
+
"step": 25000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 7.25,
|
363 |
+
"grad_norm": 0.9111607074737549,
|
364 |
+
"learning_rate": 1.3768115942028985e-05,
|
365 |
+
"loss": 0.1129,
|
366 |
+
"step": 25500
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 7.39,
|
370 |
+
"grad_norm": 1.432874083518982,
|
371 |
+
"learning_rate": 1.3057686842853084e-05,
|
372 |
+
"loss": 0.1105,
|
373 |
+
"step": 26000
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 7.53,
|
377 |
+
"grad_norm": 1.1109156608581543,
|
378 |
+
"learning_rate": 1.2347257743677183e-05,
|
379 |
+
"loss": 0.1151,
|
380 |
+
"step": 26500
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 7.67,
|
384 |
+
"grad_norm": 0.8510033488273621,
|
385 |
+
"learning_rate": 1.163682864450128e-05,
|
386 |
+
"loss": 0.1113,
|
387 |
+
"step": 27000
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 7.81,
|
391 |
+
"grad_norm": 0.6120481491088867,
|
392 |
+
"learning_rate": 1.0926399545325378e-05,
|
393 |
+
"loss": 0.1116,
|
394 |
+
"step": 27500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 7.96,
|
398 |
+
"grad_norm": 0.8336113691329956,
|
399 |
+
"learning_rate": 1.0215970446149475e-05,
|
400 |
+
"loss": 0.112,
|
401 |
+
"step": 28000
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 8.1,
|
405 |
+
"grad_norm": 1.1527196168899536,
|
406 |
+
"learning_rate": 9.505541346973572e-06,
|
407 |
+
"loss": 0.0994,
|
408 |
+
"step": 28500
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 8.24,
|
412 |
+
"grad_norm": 0.9409528970718384,
|
413 |
+
"learning_rate": 8.79511224779767e-06,
|
414 |
+
"loss": 0.0999,
|
415 |
+
"step": 29000
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 8.38,
|
419 |
+
"grad_norm": 1.298326849937439,
|
420 |
+
"learning_rate": 8.084683148621767e-06,
|
421 |
+
"loss": 0.1007,
|
422 |
+
"step": 29500
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 8.53,
|
426 |
+
"grad_norm": 1.2974679470062256,
|
427 |
+
"learning_rate": 7.374254049445865e-06,
|
428 |
+
"loss": 0.1007,
|
429 |
+
"step": 30000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 8.67,
|
433 |
+
"grad_norm": 1.1797147989273071,
|
434 |
+
"learning_rate": 6.6638249502699635e-06,
|
435 |
+
"loss": 0.1012,
|
436 |
+
"step": 30500
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 8.81,
|
440 |
+
"grad_norm": 1.3059791326522827,
|
441 |
+
"learning_rate": 5.953395851094061e-06,
|
442 |
+
"loss": 0.0987,
|
443 |
+
"step": 31000
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 8.95,
|
447 |
+
"grad_norm": 0.8831413388252258,
|
448 |
+
"learning_rate": 5.242966751918159e-06,
|
449 |
+
"loss": 0.0988,
|
450 |
+
"step": 31500
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 9.09,
|
454 |
+
"grad_norm": 0.7082082629203796,
|
455 |
+
"learning_rate": 4.532537652742257e-06,
|
456 |
+
"loss": 0.0961,
|
457 |
+
"step": 32000
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 9.24,
|
461 |
+
"grad_norm": 0.956742525100708,
|
462 |
+
"learning_rate": 3.822108553566354e-06,
|
463 |
+
"loss": 0.0911,
|
464 |
+
"step": 32500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 9.38,
|
468 |
+
"grad_norm": 0.6049565672874451,
|
469 |
+
"learning_rate": 3.111679454390452e-06,
|
470 |
+
"loss": 0.0907,
|
471 |
+
"step": 33000
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 9.52,
|
475 |
+
"grad_norm": 1.5750799179077148,
|
476 |
+
"learning_rate": 2.40125035521455e-06,
|
477 |
+
"loss": 0.0957,
|
478 |
+
"step": 33500
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 9.66,
|
482 |
+
"grad_norm": 1.0990025997161865,
|
483 |
+
"learning_rate": 1.6908212560386474e-06,
|
484 |
+
"loss": 0.0913,
|
485 |
+
"step": 34000
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 9.8,
|
489 |
+
"grad_norm": 1.086700677871704,
|
490 |
+
"learning_rate": 9.80392156862745e-07,
|
491 |
+
"loss": 0.0937,
|
492 |
+
"step": 34500
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 9.95,
|
496 |
+
"grad_norm": 0.9267112612724304,
|
497 |
+
"learning_rate": 2.6996305768684286e-07,
|
498 |
+
"loss": 0.0941,
|
499 |
+
"step": 35000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 10.0,
|
503 |
+
"step": 35190,
|
504 |
+
"total_flos": 1.838973321216e+16,
|
505 |
+
"train_loss": 0.22781734634577194,
|
506 |
+
"train_runtime": 7496.2711,
|
507 |
+
"train_samples_per_second": 4.694,
|
508 |
+
"train_steps_per_second": 4.694
|
509 |
}
|
510 |
],
|
511 |
"logging_steps": 500,
|
512 |
+
"max_steps": 35190,
|
513 |
"num_input_tokens_seen": 0,
|
514 |
+
"num_train_epochs": 10,
|
515 |
"save_steps": 1000,
|
516 |
+
"total_flos": 1.838973321216e+16,
|
517 |
"train_batch_size": 1,
|
518 |
"trial_name": null,
|
519 |
"trial_params": null
|