Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +733 -0
trainer_state.json
ADDED
@@ -0,0 +1,733 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 5000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"grad_norm": 0.3745611011981964,
|
14 |
+
"learning_rate": 5e-06,
|
15 |
+
"loss": 2.5464,
|
16 |
+
"step": 50
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"grad_norm": 0.6350923180580139,
|
21 |
+
"learning_rate": 1e-05,
|
22 |
+
"loss": 2.5921,
|
23 |
+
"step": 100
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.03,
|
27 |
+
"grad_norm": 1.1115145683288574,
|
28 |
+
"learning_rate": 1.5e-05,
|
29 |
+
"loss": 2.5699,
|
30 |
+
"step": 150
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.04,
|
34 |
+
"grad_norm": 1.1577266454696655,
|
35 |
+
"learning_rate": 2e-05,
|
36 |
+
"loss": 2.5473,
|
37 |
+
"step": 200
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.05,
|
41 |
+
"grad_norm": 3.4771063327789307,
|
42 |
+
"learning_rate": 2.5e-05,
|
43 |
+
"loss": 2.5239,
|
44 |
+
"step": 250
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.06,
|
48 |
+
"grad_norm": 1.695399522781372,
|
49 |
+
"learning_rate": 3e-05,
|
50 |
+
"loss": 2.4753,
|
51 |
+
"step": 300
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.07,
|
55 |
+
"grad_norm": 1.5696115493774414,
|
56 |
+
"learning_rate": 3.5e-05,
|
57 |
+
"loss": 2.4721,
|
58 |
+
"step": 350
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.08,
|
62 |
+
"grad_norm": 1.8710404634475708,
|
63 |
+
"learning_rate": 4e-05,
|
64 |
+
"loss": 2.3697,
|
65 |
+
"step": 400
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.09,
|
69 |
+
"grad_norm": 2.613884210586548,
|
70 |
+
"learning_rate": 4.5e-05,
|
71 |
+
"loss": 2.2966,
|
72 |
+
"step": 450
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.1,
|
76 |
+
"grad_norm": 2.416072368621826,
|
77 |
+
"learning_rate": 5e-05,
|
78 |
+
"loss": 2.4018,
|
79 |
+
"step": 500
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.11,
|
83 |
+
"grad_norm": 1.746003270149231,
|
84 |
+
"learning_rate": 4.99847706754774e-05,
|
85 |
+
"loss": 2.2825,
|
86 |
+
"step": 550
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.12,
|
90 |
+
"grad_norm": 2.941234827041626,
|
91 |
+
"learning_rate": 4.993910125649561e-05,
|
92 |
+
"loss": 2.473,
|
93 |
+
"step": 600
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.13,
|
97 |
+
"grad_norm": 2.632983922958374,
|
98 |
+
"learning_rate": 4.9863047384206835e-05,
|
99 |
+
"loss": 2.2963,
|
100 |
+
"step": 650
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.14,
|
104 |
+
"grad_norm": 5.2639946937561035,
|
105 |
+
"learning_rate": 4.975670171853926e-05,
|
106 |
+
"loss": 2.2385,
|
107 |
+
"step": 700
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.15,
|
111 |
+
"grad_norm": 1.5031816959381104,
|
112 |
+
"learning_rate": 4.962019382530521e-05,
|
113 |
+
"loss": 2.2266,
|
114 |
+
"step": 750
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.16,
|
118 |
+
"grad_norm": 1.3354930877685547,
|
119 |
+
"learning_rate": 4.9453690018345144e-05,
|
120 |
+
"loss": 2.2189,
|
121 |
+
"step": 800
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.17,
|
125 |
+
"grad_norm": 2.2369980812072754,
|
126 |
+
"learning_rate": 4.925739315689991e-05,
|
127 |
+
"loss": 2.2916,
|
128 |
+
"step": 850
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.18,
|
132 |
+
"grad_norm": 1.5904576778411865,
|
133 |
+
"learning_rate": 4.9031542398457974e-05,
|
134 |
+
"loss": 2.1422,
|
135 |
+
"step": 900
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.19,
|
139 |
+
"grad_norm": 3.333570957183838,
|
140 |
+
"learning_rate": 4.877641290737884e-05,
|
141 |
+
"loss": 2.3032,
|
142 |
+
"step": 950
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.2,
|
146 |
+
"grad_norm": 2.157676935195923,
|
147 |
+
"learning_rate": 4.849231551964771e-05,
|
148 |
+
"loss": 2.1895,
|
149 |
+
"step": 1000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.21,
|
153 |
+
"grad_norm": 8.637140274047852,
|
154 |
+
"learning_rate": 4.817959636416969e-05,
|
155 |
+
"loss": 2.3375,
|
156 |
+
"step": 1050
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.22,
|
160 |
+
"grad_norm": 2.910135507583618,
|
161 |
+
"learning_rate": 4.783863644106502e-05,
|
162 |
+
"loss": 2.261,
|
163 |
+
"step": 1100
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.23,
|
167 |
+
"grad_norm": 1.8060033321380615,
|
168 |
+
"learning_rate": 4.7469851157479177e-05,
|
169 |
+
"loss": 2.198,
|
170 |
+
"step": 1150
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.24,
|
174 |
+
"grad_norm": 1.5828884840011597,
|
175 |
+
"learning_rate": 4.707368982147318e-05,
|
176 |
+
"loss": 2.1015,
|
177 |
+
"step": 1200
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.25,
|
181 |
+
"grad_norm": 2.2127833366394043,
|
182 |
+
"learning_rate": 4.665063509461097e-05,
|
183 |
+
"loss": 2.2895,
|
184 |
+
"step": 1250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.26,
|
188 |
+
"grad_norm": 1.5719356536865234,
|
189 |
+
"learning_rate": 4.620120240391065e-05,
|
190 |
+
"loss": 2.3689,
|
191 |
+
"step": 1300
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.27,
|
195 |
+
"grad_norm": 2.7176144123077393,
|
196 |
+
"learning_rate": 4.572593931387604e-05,
|
197 |
+
"loss": 2.3083,
|
198 |
+
"step": 1350
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.28,
|
202 |
+
"grad_norm": 2.401768684387207,
|
203 |
+
"learning_rate": 4.522542485937369e-05,
|
204 |
+
"loss": 2.1516,
|
205 |
+
"step": 1400
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.29,
|
209 |
+
"grad_norm": 2.553105354309082,
|
210 |
+
"learning_rate": 4.4700268840168045e-05,
|
211 |
+
"loss": 2.3064,
|
212 |
+
"step": 1450
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.3,
|
216 |
+
"grad_norm": 2.074903964996338,
|
217 |
+
"learning_rate": 4.415111107797445e-05,
|
218 |
+
"loss": 2.1008,
|
219 |
+
"step": 1500
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.31,
|
223 |
+
"grad_norm": 2.7607364654541016,
|
224 |
+
"learning_rate": 4.357862063693486e-05,
|
225 |
+
"loss": 2.1576,
|
226 |
+
"step": 1550
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.32,
|
230 |
+
"grad_norm": 3.275789976119995,
|
231 |
+
"learning_rate": 4.2983495008466276e-05,
|
232 |
+
"loss": 2.2486,
|
233 |
+
"step": 1600
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.33,
|
237 |
+
"grad_norm": 3.713785409927368,
|
238 |
+
"learning_rate": 4.2366459261474933e-05,
|
239 |
+
"loss": 2.1851,
|
240 |
+
"step": 1650
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.34,
|
244 |
+
"grad_norm": 2.8428258895874023,
|
245 |
+
"learning_rate": 4.172826515897146e-05,
|
246 |
+
"loss": 2.2279,
|
247 |
+
"step": 1700
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.35,
|
251 |
+
"grad_norm": 4.787942409515381,
|
252 |
+
"learning_rate": 4.1069690242163484e-05,
|
253 |
+
"loss": 2.1374,
|
254 |
+
"step": 1750
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.36,
|
258 |
+
"grad_norm": 4.18039083480835,
|
259 |
+
"learning_rate": 4.039153688314145e-05,
|
260 |
+
"loss": 2.2694,
|
261 |
+
"step": 1800
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.37,
|
265 |
+
"grad_norm": 1.7699658870697021,
|
266 |
+
"learning_rate": 3.969463130731183e-05,
|
267 |
+
"loss": 2.2034,
|
268 |
+
"step": 1850
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.38,
|
272 |
+
"grad_norm": 1.5621023178100586,
|
273 |
+
"learning_rate": 3.897982258676867e-05,
|
274 |
+
"loss": 2.1479,
|
275 |
+
"step": 1900
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.39,
|
279 |
+
"grad_norm": 2.7159836292266846,
|
280 |
+
"learning_rate": 3.824798160583012e-05,
|
281 |
+
"loss": 2.066,
|
282 |
+
"step": 1950
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.4,
|
286 |
+
"grad_norm": 2.071038246154785,
|
287 |
+
"learning_rate": 3.7500000000000003e-05,
|
288 |
+
"loss": 2.3053,
|
289 |
+
"step": 2000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.41,
|
293 |
+
"grad_norm": 2.285961866378784,
|
294 |
+
"learning_rate": 3.673678906964727e-05,
|
295 |
+
"loss": 2.3655,
|
296 |
+
"step": 2050
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.42,
|
300 |
+
"grad_norm": 2.2050671577453613,
|
301 |
+
"learning_rate": 3.5959278669726935e-05,
|
302 |
+
"loss": 2.386,
|
303 |
+
"step": 2100
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.43,
|
307 |
+
"grad_norm": 1.8379002809524536,
|
308 |
+
"learning_rate": 3.516841607689501e-05,
|
309 |
+
"loss": 2.2381,
|
310 |
+
"step": 2150
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.44,
|
314 |
+
"grad_norm": 3.3484580516815186,
|
315 |
+
"learning_rate": 3.436516483539781e-05,
|
316 |
+
"loss": 2.2748,
|
317 |
+
"step": 2200
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.45,
|
321 |
+
"grad_norm": 2.6434378623962402,
|
322 |
+
"learning_rate": 3.355050358314172e-05,
|
323 |
+
"loss": 2.1108,
|
324 |
+
"step": 2250
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.46,
|
328 |
+
"grad_norm": 2.832470417022705,
|
329 |
+
"learning_rate": 3.272542485937369e-05,
|
330 |
+
"loss": 2.2859,
|
331 |
+
"step": 2300
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.47,
|
335 |
+
"grad_norm": 2.5317020416259766,
|
336 |
+
"learning_rate": 3.1890933895424976e-05,
|
337 |
+
"loss": 2.1841,
|
338 |
+
"step": 2350
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.48,
|
342 |
+
"grad_norm": 2.41705060005188,
|
343 |
+
"learning_rate": 3.104804738999169e-05,
|
344 |
+
"loss": 2.2283,
|
345 |
+
"step": 2400
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.49,
|
349 |
+
"grad_norm": 2.1975114345550537,
|
350 |
+
"learning_rate": 3.0197792270443982e-05,
|
351 |
+
"loss": 2.1952,
|
352 |
+
"step": 2450
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.5,
|
356 |
+
"grad_norm": 3.486776828765869,
|
357 |
+
"learning_rate": 2.9341204441673266e-05,
|
358 |
+
"loss": 2.289,
|
359 |
+
"step": 2500
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.51,
|
363 |
+
"grad_norm": 2.697965145111084,
|
364 |
+
"learning_rate": 2.8479327524001636e-05,
|
365 |
+
"loss": 2.2576,
|
366 |
+
"step": 2550
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.52,
|
370 |
+
"grad_norm": 1.713478684425354,
|
371 |
+
"learning_rate": 2.761321158169134e-05,
|
372 |
+
"loss": 2.1399,
|
373 |
+
"step": 2600
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.53,
|
377 |
+
"grad_norm": 1.4894109964370728,
|
378 |
+
"learning_rate": 2.674391184360313e-05,
|
379 |
+
"loss": 2.0986,
|
380 |
+
"step": 2650
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.54,
|
384 |
+
"grad_norm": 2.0305914878845215,
|
385 |
+
"learning_rate": 2.587248741756253e-05,
|
386 |
+
"loss": 2.1002,
|
387 |
+
"step": 2700
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.55,
|
391 |
+
"grad_norm": 2.2819197177886963,
|
392 |
+
"learning_rate": 2.5e-05,
|
393 |
+
"loss": 2.2805,
|
394 |
+
"step": 2750
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.56,
|
398 |
+
"grad_norm": 2.552293539047241,
|
399 |
+
"learning_rate": 2.4127512582437485e-05,
|
400 |
+
"loss": 2.0637,
|
401 |
+
"step": 2800
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.57,
|
405 |
+
"grad_norm": 1.8868838548660278,
|
406 |
+
"learning_rate": 2.3256088156396868e-05,
|
407 |
+
"loss": 2.2437,
|
408 |
+
"step": 2850
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.58,
|
412 |
+
"grad_norm": 2.3621270656585693,
|
413 |
+
"learning_rate": 2.238678841830867e-05,
|
414 |
+
"loss": 2.1867,
|
415 |
+
"step": 2900
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.59,
|
419 |
+
"grad_norm": 2.550295829772949,
|
420 |
+
"learning_rate": 2.1520672475998373e-05,
|
421 |
+
"loss": 2.2582,
|
422 |
+
"step": 2950
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.6,
|
426 |
+
"grad_norm": 2.509620428085327,
|
427 |
+
"learning_rate": 2.0658795558326743e-05,
|
428 |
+
"loss": 2.144,
|
429 |
+
"step": 3000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.61,
|
433 |
+
"grad_norm": 3.188652276992798,
|
434 |
+
"learning_rate": 1.980220772955602e-05,
|
435 |
+
"loss": 2.1568,
|
436 |
+
"step": 3050
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.62,
|
440 |
+
"grad_norm": 2.8035781383514404,
|
441 |
+
"learning_rate": 1.895195261000831e-05,
|
442 |
+
"loss": 2.1571,
|
443 |
+
"step": 3100
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.63,
|
447 |
+
"grad_norm": 2.5277318954467773,
|
448 |
+
"learning_rate": 1.8109066104575023e-05,
|
449 |
+
"loss": 2.1844,
|
450 |
+
"step": 3150
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.64,
|
454 |
+
"grad_norm": 5.469127655029297,
|
455 |
+
"learning_rate": 1.7274575140626318e-05,
|
456 |
+
"loss": 2.1792,
|
457 |
+
"step": 3200
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.65,
|
461 |
+
"grad_norm": 2.5568430423736572,
|
462 |
+
"learning_rate": 1.6449496416858284e-05,
|
463 |
+
"loss": 2.2361,
|
464 |
+
"step": 3250
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.66,
|
468 |
+
"grad_norm": 3.626446008682251,
|
469 |
+
"learning_rate": 1.56348351646022e-05,
|
470 |
+
"loss": 2.1619,
|
471 |
+
"step": 3300
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.67,
|
475 |
+
"grad_norm": 2.9157896041870117,
|
476 |
+
"learning_rate": 1.4831583923104999e-05,
|
477 |
+
"loss": 2.1609,
|
478 |
+
"step": 3350
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.68,
|
482 |
+
"grad_norm": 2.5856120586395264,
|
483 |
+
"learning_rate": 1.4040721330273062e-05,
|
484 |
+
"loss": 2.1897,
|
485 |
+
"step": 3400
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.69,
|
489 |
+
"grad_norm": 3.538245677947998,
|
490 |
+
"learning_rate": 1.3263210930352737e-05,
|
491 |
+
"loss": 2.0751,
|
492 |
+
"step": 3450
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.7,
|
496 |
+
"grad_norm": 1.6384202241897583,
|
497 |
+
"learning_rate": 1.2500000000000006e-05,
|
498 |
+
"loss": 2.2173,
|
499 |
+
"step": 3500
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.71,
|
503 |
+
"grad_norm": 2.493316888809204,
|
504 |
+
"learning_rate": 1.175201839416988e-05,
|
505 |
+
"loss": 2.1041,
|
506 |
+
"step": 3550
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.72,
|
510 |
+
"grad_norm": 3.1629319190979004,
|
511 |
+
"learning_rate": 1.1020177413231334e-05,
|
512 |
+
"loss": 2.2025,
|
513 |
+
"step": 3600
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.73,
|
517 |
+
"grad_norm": 2.7992451190948486,
|
518 |
+
"learning_rate": 1.0305368692688174e-05,
|
519 |
+
"loss": 2.1657,
|
520 |
+
"step": 3650
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.74,
|
524 |
+
"grad_norm": 5.4246907234191895,
|
525 |
+
"learning_rate": 9.608463116858542e-06,
|
526 |
+
"loss": 2.1475,
|
527 |
+
"step": 3700
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.75,
|
531 |
+
"grad_norm": 3.659865617752075,
|
532 |
+
"learning_rate": 8.930309757836517e-06,
|
533 |
+
"loss": 2.1799,
|
534 |
+
"step": 3750
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.76,
|
538 |
+
"grad_norm": 2.3712964057922363,
|
539 |
+
"learning_rate": 8.271734841028553e-06,
|
540 |
+
"loss": 2.0888,
|
541 |
+
"step": 3800
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.77,
|
545 |
+
"grad_norm": 1.6899696588516235,
|
546 |
+
"learning_rate": 7.633540738525066e-06,
|
547 |
+
"loss": 2.1238,
|
548 |
+
"step": 3850
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.78,
|
552 |
+
"grad_norm": 1.6770470142364502,
|
553 |
+
"learning_rate": 7.016504991533726e-06,
|
554 |
+
"loss": 2.0053,
|
555 |
+
"step": 3900
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.79,
|
559 |
+
"grad_norm": 2.5809788703918457,
|
560 |
+
"learning_rate": 6.421379363065142e-06,
|
561 |
+
"loss": 2.1939,
|
562 |
+
"step": 3950
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.8,
|
566 |
+
"grad_norm": 2.4868805408477783,
|
567 |
+
"learning_rate": 5.848888922025553e-06,
|
568 |
+
"loss": 2.2744,
|
569 |
+
"step": 4000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.81,
|
573 |
+
"grad_norm": 2.573330879211426,
|
574 |
+
"learning_rate": 5.299731159831953e-06,
|
575 |
+
"loss": 2.2459,
|
576 |
+
"step": 4050
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.82,
|
580 |
+
"grad_norm": 3.3828341960906982,
|
581 |
+
"learning_rate": 4.7745751406263165e-06,
|
582 |
+
"loss": 2.1207,
|
583 |
+
"step": 4100
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.83,
|
587 |
+
"grad_norm": 3.077352285385132,
|
588 |
+
"learning_rate": 4.274060686123959e-06,
|
589 |
+
"loss": 2.2318,
|
590 |
+
"step": 4150
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.84,
|
594 |
+
"grad_norm": 2.399592399597168,
|
595 |
+
"learning_rate": 3.798797596089351e-06,
|
596 |
+
"loss": 2.0679,
|
597 |
+
"step": 4200
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.85,
|
601 |
+
"grad_norm": 2.191920042037964,
|
602 |
+
"learning_rate": 3.3493649053890326e-06,
|
603 |
+
"loss": 2.1493,
|
604 |
+
"step": 4250
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.86,
|
608 |
+
"grad_norm": 1.8285448551177979,
|
609 |
+
"learning_rate": 2.9263101785268254e-06,
|
610 |
+
"loss": 2.2512,
|
611 |
+
"step": 4300
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.87,
|
615 |
+
"grad_norm": 3.347266912460327,
|
616 |
+
"learning_rate": 2.5301488425208296e-06,
|
617 |
+
"loss": 2.1985,
|
618 |
+
"step": 4350
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.88,
|
622 |
+
"grad_norm": 1.9266479015350342,
|
623 |
+
"learning_rate": 2.1613635589349756e-06,
|
624 |
+
"loss": 2.2455,
|
625 |
+
"step": 4400
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.89,
|
629 |
+
"grad_norm": 2.067202091217041,
|
630 |
+
"learning_rate": 1.8204036358303173e-06,
|
631 |
+
"loss": 2.1952,
|
632 |
+
"step": 4450
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.9,
|
636 |
+
"grad_norm": 2.765047788619995,
|
637 |
+
"learning_rate": 1.5076844803522922e-06,
|
638 |
+
"loss": 2.0983,
|
639 |
+
"step": 4500
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.91,
|
643 |
+
"grad_norm": 3.322967767715454,
|
644 |
+
"learning_rate": 1.2235870926211619e-06,
|
645 |
+
"loss": 2.1133,
|
646 |
+
"step": 4550
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.92,
|
650 |
+
"grad_norm": 2.262035608291626,
|
651 |
+
"learning_rate": 9.684576015420278e-07,
|
652 |
+
"loss": 2.1524,
|
653 |
+
"step": 4600
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.93,
|
657 |
+
"grad_norm": 2.373628854751587,
|
658 |
+
"learning_rate": 7.426068431000882e-07,
|
659 |
+
"loss": 2.2132,
|
660 |
+
"step": 4650
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.94,
|
664 |
+
"grad_norm": 2.6327779293060303,
|
665 |
+
"learning_rate": 5.463099816548579e-07,
|
666 |
+
"loss": 2.1586,
|
667 |
+
"step": 4700
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.95,
|
671 |
+
"grad_norm": 1.8346914052963257,
|
672 |
+
"learning_rate": 3.7980617469479953e-07,
|
673 |
+
"loss": 2.2988,
|
674 |
+
"step": 4750
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.96,
|
678 |
+
"grad_norm": 2.114342451095581,
|
679 |
+
"learning_rate": 2.4329828146074095e-07,
|
680 |
+
"loss": 2.2557,
|
681 |
+
"step": 4800
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.97,
|
685 |
+
"grad_norm": 2.1969785690307617,
|
686 |
+
"learning_rate": 1.3695261579316777e-07,
|
687 |
+
"loss": 2.1417,
|
688 |
+
"step": 4850
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.98,
|
692 |
+
"grad_norm": 1.7341632843017578,
|
693 |
+
"learning_rate": 6.089874350439506e-08,
|
694 |
+
"loss": 2.1895,
|
695 |
+
"step": 4900
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.99,
|
699 |
+
"grad_norm": 2.1261510848999023,
|
700 |
+
"learning_rate": 1.522932452260595e-08,
|
701 |
+
"loss": 2.1405,
|
702 |
+
"step": 4950
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 1.0,
|
706 |
+
"grad_norm": 2.2943620681762695,
|
707 |
+
"learning_rate": 0.0,
|
708 |
+
"loss": 2.1998,
|
709 |
+
"step": 5000
|
710 |
+
}
|
711 |
+
],
|
712 |
+
"logging_steps": 50,
|
713 |
+
"max_steps": 5000,
|
714 |
+
"num_input_tokens_seen": 0,
|
715 |
+
"num_train_epochs": 1,
|
716 |
+
"save_steps": 500,
|
717 |
+
"stateful_callbacks": {
|
718 |
+
"TrainerControl": {
|
719 |
+
"args": {
|
720 |
+
"should_epoch_stop": false,
|
721 |
+
"should_evaluate": false,
|
722 |
+
"should_log": false,
|
723 |
+
"should_save": true,
|
724 |
+
"should_training_stop": true
|
725 |
+
},
|
726 |
+
"attributes": {}
|
727 |
+
}
|
728 |
+
},
|
729 |
+
"total_flos": 7.314072519030374e+16,
|
730 |
+
"train_batch_size": 2,
|
731 |
+
"trial_name": null,
|
732 |
+
"trial_params": null
|
733 |
+
}
|