Falcon3-10B-Instruct


Find this model in the Falcon model summary

Falcon3-10B-Instruct Model Set Plots

Falcon Compared to Base Model Plots

Falcon3-10B-Instruct Model Selected Details
id layer_type N M Q alpha D alpha-hat num_spikes warning
1 dense 23040 3072 7.5 3.949894 0.022737 -3.509183 76
2 dense 23040 3072 7.5 2.217955 0.025861 -0.795543 789
3 dense 23040 3072 7.5 2.218784 0.025544 -1.113255 761
4 dense 3072 1024 3.0 1.594416 0.016757 -0.888957 203 over-trained
5 dense 3072 3072 1.0 2.128297 0.022344 -1.693396 88
6 dense 3072 3072 1.0 1.529318 0.028412 0.437775 1013 over-trained
7 dense 3072 1024 3.0 2.166983 0.023369 -2.763700 157
8 dense 23040 3072 7.5 3.622127 0.020237 -3.393369 151
9 dense 23040 3072 7.5 2.494432 0.021715 -0.905935 619
10 dense 23040 3072 7.5 2.441410 0.027650 -0.964790 644
11 dense 3072 1024 3.0 2.301683 0.018764 -3.171796 164
12 dense 3072 3072 1.0 2.568620 0.019061 -2.402091 150
13 dense 3072 3072 1.0 2.288077 0.016819 -2.688666 230
14 dense 3072 1024 3.0 2.570249 0.023714 -3.628573 118
15 dense 23040 3072 7.5 3.314011 0.016942 -3.117900 241
16 dense 23040 3072 7.5 2.864976 0.007904 -0.964619 484
17 dense 23040 3072 7.5 2.827609 0.010069 -1.026520 404
18 dense 3072 1024 3.0 2.316329 0.014750 -3.618389 156
19 dense 3072 3072 1.0 2.688425 0.018010 -3.078092 98
20 dense 3072 3072 1.0 2.371422 0.021725 -3.067055 237
21 dense 3072 1024 3.0 2.728130 0.021808 -4.070718 107
22 dense 23040 3072 7.5 3.080582 0.027296 -2.593945 331
23 dense 23040 3072 7.5 2.945323 0.011124 -0.924775 524
24 dense 23040 3072 7.5 2.784027 0.018817 -1.172650 365
25 dense 3072 1024 3.0 2.279472 0.014645 -3.587790 160
26 dense 3072 3072 1.0 2.659396 0.016527 -3.111033 135
27 dense 3072 3072 1.0 2.351967 0.016715 -2.898416 177
28 dense 3072 1024 3.0 2.779401 0.021188 -4.496507 112
29 dense 23040 3072 7.5 3.183438 0.025202 -2.523783 290
30 dense 23040 3072 7.5 3.092287 0.020097 -1.001707 676
31 dense 23040 3072 7.5 2.914759 0.010752 -1.317155 181
32 dense 3072 1024 3.0 2.297522 0.020042 -3.575919 183
33 dense 3072 1024 3.0 2.604983 0.025323 -3.958461 114
34 dense 3072 3072 1.0 2.441859 0.025883 -2.783201 133
35 dense 3072 3072 1.0 2.355989 0.020258 -2.907204 194
36 dense 23040 3072 7.5 2.912029 0.017800 -0.837946 631
37 dense 3072 3072 1.0 2.698994 0.029903 -3.483221 148
38 dense 23040 3072 7.5 2.807229 0.012738 -1.116638 305
39 dense 3072 1024 3.0 2.376427 0.016276 -3.686006 162
40 dense 23040 3072 7.5 3.374609 0.017764 -2.796394 189
41 dense 3072 1024 3.0 2.788646 0.023209 -4.614759 104
42 dense 3072 3072 1.0 2.397638 0.015538 -2.871010 165
43 dense 3072 3072 1.0 2.343618 0.027348 -2.790611 178
44 dense 3072 3072 1.0 3.034733 0.020863 -3.901858 76
45 dense 23040 3072 7.5 2.847915 0.014798 -0.824407 670
46 dense 23040 3072 7.5 3.462274 0.020247 -3.006496 204
47 dense 3072 1024 3.0 2.307432 0.019390 -3.720789 131
48 dense 23040 3072 7.5 2.801556 0.008977 -1.007251 326
49 dense 3072 1024 3.0 2.843595 0.028069 -4.834879 100
50 dense 3072 1024 3.0 2.365780 0.020371 -3.775663 143
51 dense 23040 3072 7.5 2.786676 0.010419 -0.677528 519
52 dense 23040 3072 7.5 3.371675 0.018090 -2.558081 134
53 dense 3072 3072 1.0 2.390366 0.024252 -2.935779 160
54 dense 3072 3072 1.0 2.927124 0.028157 -3.684683 77
55 dense 3072 1024 3.0 2.807957 0.031189 -4.820930 94
56 dense 23040 3072 7.5 2.744093 0.008337 -0.844206 296
57 dense 23040 3072 7.5 2.744235 0.006999 -0.523949 413
58 dense 23040 3072 7.5 2.672390 0.010986 -0.699040 307
59 dense 3072 1024 3.0 2.342510 0.015912 -3.091164 133
60 dense 3072 3072 1.0 2.864366 0.025902 -3.388296 77
61 dense 3072 3072 1.0 2.386825 0.021806 -2.774573 189
62 dense 3072 1024 3.0 2.759544 0.025038 -4.263689 76
63 dense 23040 3072 7.5 3.220519 0.011687 -2.646307 184
64 dense 3072 1024 3.0 2.799983 0.028263 -4.909292 82
65 dense 3072 3072 1.0 2.619574 0.028127 -3.347406 142
66 dense 23040 3072 7.5 3.126106 0.010334 -2.366297 206
67 dense 23040 3072 7.5 2.690887 0.008622 -0.395413 495
68 dense 23040 3072 7.5 2.651932 0.006496 -0.572648 280
69 dense 3072 3072 1.0 2.354319 0.023585 -2.611659 160
70 dense 3072 1024 3.0 2.371040 0.029135 -3.719691 136
71 dense 23040 3072 7.5 3.188167 0.014431 -2.470164 193
72 dense 23040 3072 7.5 2.712317 0.009359 -0.337607 423
73 dense 23040 3072 7.5 2.690893 0.009413 -0.612186 204
74 dense 3072 1024 3.0 2.342382 0.023456 -3.635752 133
75 dense 3072 3072 1.0 2.655432 0.037069 -3.339665 121
76 dense 3072 3072 1.0 2.305262 0.027169 -2.365248 208
77 dense 3072 1024 3.0 2.818133 0.027881 -4.751083 96
78 dense 23040 3072 7.5 3.134020 0.011925 -2.269799 206
79 dense 23040 3072 7.5 2.631477 0.010236 -0.141651 530
80 dense 23040 3072 7.5 2.609444 0.007080 -0.349498 331
81 dense 3072 1024 3.0 2.433097 0.024941 -3.934827 123
82 dense 3072 3072 1.0 2.967347 0.035153 -3.735784 67
83 dense 3072 3072 1.0 2.371634 0.023540 -2.693253 194
84 dense 3072 1024 3.0 2.817030 0.031481 -4.687094 80
85 dense 23040 3072 7.5 3.217772 0.017078 -2.530454 210
86 dense 23040 3072 7.5 2.629853 0.007136 -0.199087 572
87 dense 23040 3072 7.5 2.663044 0.007194 -0.453748 229
88 dense 3072 1024 3.0 2.409431 0.026193 -3.924524 148
89 dense 3072 3072 1.0 2.812796 0.026914 -3.556646 91
90 dense 3072 3072 1.0 2.425720 0.026003 -2.662797 164
91 dense 3072 1024 3.0 2.778981 0.033287 -4.786758 104
92 dense 23040 3072 7.5 3.248368 0.023098 -2.555211 163
93 dense 23040 3072 7.5 2.590812 0.008534 -0.154392 607
94 dense 23040 3072 7.5 2.600042 0.009300 -0.356190 345
95 dense 3072 1024 3.0 2.349746 0.025754 -3.427291 126
96 dense 3072 3072 1.0 2.899619 0.027963 -3.580046 73
97 dense 3072 1024 3.0 2.717625 0.023773 -4.106731 108
98 dense 3072 3072 1.0 2.371421 0.023906 -2.446692 195
99 dense 3072 3072 1.0 2.443449 0.025013 -2.728274 185
100 dense 3072 3072 1.0 3.032658 0.037564 -3.955849 72
101 dense 23040 3072 7.5 2.603520 0.007800 -0.212817 576
102 dense 23040 3072 7.5 3.284415 0.024241 -2.656631 188
103 dense 3072 1024 3.0 2.455684 0.027327 -3.876515 127
104 dense 23040 3072 7.5 2.628422 0.007495 -0.470051 317
105 dense 3072 1024 3.0 2.790464 0.022088 -4.398895 86
106 dense 23040 3072 7.5 3.406133 0.029185 -2.894212 127
107 dense 23040 3072 7.5 2.584560 0.007597 -0.228386 532
108 dense 23040 3072 7.5 2.622485 0.007856 -0.431619 307
109 dense 3072 1024 3.0 2.533641 0.029595 -3.953437 103
110 dense 3072 3072 1.0 2.439332 0.080042 -3.272511 367
111 dense 3072 3072 1.0 2.405771 0.027308 -2.651840 196
112 dense 3072 1024 3.0 3.363199 0.034444 -5.866242 42
113 dense 23040 3072 7.5 3.401657 0.029029 -2.892180 146
114 dense 23040 3072 7.5 2.639842 0.006926 -0.413186 509
115 dense 23040 3072 7.5 2.682980 0.008348 -0.664016 271
116 dense 3072 1024 3.0 2.542419 0.027023 -4.131435 117
117 dense 3072 3072 1.0 3.171498 0.035619 -4.222057 66
118 dense 3072 3072 1.0 2.397194 0.026474 -2.718790 240
119 dense 3072 1024 3.0 3.047268 0.026287 -5.068474 86
120 dense 23040 3072 7.5 3.318653 0.029374 -2.516495 174
121 dense 23040 3072 7.5 2.646458 0.008472 -0.390327 497
122 dense 23040 3072 7.5 2.682112 0.009726 -0.581073 310
123 dense 3072 1024 3.0 2.401264 0.037065 -3.973284 139
124 dense 3072 3072 1.0 3.099043 0.033821 -4.170155 76
125 dense 3072 3072 1.0 2.389854 0.031388 -2.449512 143
126 dense 3072 1024 3.0 3.093572 0.039139 -5.692681 93
127 dense 23040 3072 7.5 3.703088 0.024595 -3.147252 112
128 dense 23040 3072 7.5 2.724173 0.008418 -0.495700 443
129 dense 3072 1024 3.0 3.260422 0.029610 -5.561218 71
130 dense 3072 1024 3.0 2.402468 0.032978 -3.665249 147
131 dense 3072 3072 1.0 2.243654 0.083463 -2.798820 417
132 dense 23040 3072 7.5 2.766543 0.010473 -0.843468 307
133 dense 3072 3072 1.0 2.414547 0.027450 -2.734415 159
134 dense 3072 3072 1.0 2.913033 0.080983 -4.260345 189
135 dense 3072 3072 1.0 2.454523 0.026591 -2.649584 165
136 dense 3072 1024 3.0 2.672054 0.033872 -4.353297 105
137 dense 23040 3072 7.5 2.723361 0.008530 -0.642819 390
138 dense 23040 3072 7.5 2.679069 0.007359 -0.455552 491
139 dense 23040 3072 7.5 3.617740 0.034577 -3.378695 178
140 dense 3072 1024 3.0 3.654536 0.037896 -6.875762 56
141 dense 3072 1024 3.0 2.680002 0.027634 -4.280733 104
142 dense 23040 3072 7.5 2.858137 0.009868 -0.918784 281
143 dense 23040 3072 7.5 2.761346 0.008405 -0.573233 480
144 dense 23040 3072 7.5 3.989606 0.027028 -3.600159 124
145 dense 3072 3072 1.0 2.413287 0.027059 -2.557134 234
146 dense 3072 1024 3.0 3.474254 0.037442 -6.521697 68
147 dense 3072 3072 1.0 2.903082 0.091193 -4.357425 235
148 dense 3072 3072 1.0 2.242812 0.081694 -3.189199 490
149 dense 23040 3072 7.5 4.107682 0.032007 -4.015814 84
150 dense 3072 1024 3.0 2.795203 0.024986 -4.572364 72
151 dense 23040 3072 7.5 2.780398 0.009985 -0.707780 410
152 dense 23040 3072 7.5 2.719992 0.007917 -0.503555 467
153 dense 3072 1024 3.0 3.241606 0.040754 -5.736386 67
154 dense 3072 3072 1.0 2.535254 0.021547 -2.825753 135
155 dense 3072 3072 1.0 2.478684 0.022795 -2.689492 134
156 dense 3072 3072 1.0 3.149910 0.024171 -3.420402 71
157 dense 23040 3072 7.5 2.806176 0.008047 -0.670633 493
158 dense 23040 3072 7.5 4.000060 0.032364 -3.752945 146
159 dense 3072 1024 3.0 2.652613 0.020645 -4.038403 84
160 dense 23040 3072 7.5 2.905742 0.010553 -1.078204 331
161 dense 3072 1024 3.0 2.930730 0.022879 -4.638937 80
162 dense 23040 3072 7.5 2.774218 0.010301 -0.827362 433
163 dense 3072 3072 1.0 3.425241 0.039579 -4.820672 53
164 dense 3072 1024 3.0 3.206606 0.033703 -5.736699 86
165 dense 23040 3072 7.5 4.290491 0.032710 -4.154298 60
166 dense 23040 3072 7.5 2.719389 0.009192 -0.583004 507
167 dense 3072 1024 3.0 2.560625 0.029340 -4.061976 89
168 dense 3072 3072 1.0 2.422231 0.028735 -2.698251 166
169 dense 23040 3072 7.5 2.745502 0.011343 -0.703282 594
170 dense 3072 1024 3.0 2.675372 0.034666 -4.244533 72
171 dense 3072 3072 1.0 2.440857 0.026243 -2.827835 165
172 dense 3072 1024 3.0 3.103973 0.036815 -5.569516 100
173 dense 23040 3072 7.5 4.323864 0.022931 -4.101137 85
174 dense 23040 3072 7.5 2.839223 0.009494 -1.116875 349
175 dense 3072 3072 1.0 3.257576 0.041387 -4.364593 91
176 dense 23040 3072 7.5 2.661933 0.010674 -0.631188 412
177 dense 23040 3072 7.5 2.704197 0.013709 -0.788453 410
178 dense 23040 3072 7.5 4.265300 0.021056 -4.312157 88
179 dense 3072 3072 1.0 3.391047 0.030089 -4.277027 94
180 dense 3072 1024 3.0 2.615777 0.026190 -4.007410 95
181 dense 3072 1024 3.0 3.069733 0.026538 -5.434940 87
182 dense 3072 3072 1.0 2.495820 0.030235 -2.958953 131
183 dense 23040 3072 7.5 2.839303 0.015002 -1.056776 274
184 dense 3072 3072 1.0 4.115695 0.031032 -5.883421 60
185 dense 23040 3072 7.5 2.716140 0.011036 -0.731340 369
186 dense 23040 3072 7.5 4.331889 0.016587 -4.262079 121
187 dense 3072 1024 3.0 2.663189 0.023192 -4.232162 121
188 dense 3072 3072 1.0 2.462801 0.021041 -2.999029 181
189 dense 3072 1024 3.0 3.420756 0.038016 -6.130760 51
190 dense 3072 3072 1.0 2.541958 0.023631 -2.793228 137
191 dense 3072 3072 1.0 3.942544 0.049169 -5.857265 59
192 dense 3072 1024 3.0 2.655102 0.037704 -4.223913 142
193 dense 3072 1024 3.0 3.226891 0.036651 -5.738681 85
194 dense 23040 3072 7.5 4.204335 0.019821 -4.544848 119
195 dense 23040 3072 7.5 2.608740 0.008314 -0.508538 451
196 dense 23040 3072 7.5 2.667140 0.009667 -0.647186 379
197 dense 3072 1024 3.0 2.840911 0.036039 -4.474329 70
198 dense 3072 3072 1.0 3.819712 0.020325 -5.624860 74
199 dense 3072 1024 3.0 3.030850 0.020205 -5.088884 94
200 dense 3072 3072 1.0 2.508768 0.017920 -2.767264 164
201 dense 23040 3072 7.5 2.759121 0.010767 -0.933903 402
202 dense 23040 3072 7.5 4.323840 0.019491 -4.430759 131
203 dense 23040 3072 7.5 2.656309 0.007321 -0.625047 486
204 dense 23040 3072 7.5 2.587624 0.006740 -0.442248 606
205 dense 3072 1024 3.0 2.686124 0.035186 -4.254699 96
206 dense 23040 3072 7.5 2.650988 0.005559 -0.554676 521
207 dense 3072 3072 1.0 3.473218 0.035406 -4.860383 54
208 dense 23040 3072 7.5 4.205406 0.022829 -4.381159 132
209 dense 3072 3072 1.0 2.367337 0.022778 -2.497168 216
210 dense 3072 1024 3.0 2.863503 0.035058 -4.912606 123
211 dense 23040 3072 7.5 2.632873 0.007414 -0.467997 686
212 dense 23040 3072 7.5 4.508830 0.024262 -4.873987 118
213 dense 3072 1024 3.0 2.591540 0.025740 -4.129891 117
214 dense 23040 3072 7.5 2.751359 0.007308 -0.769750 466
215 dense 3072 3072 1.0 2.420775 0.020199 -2.656383 167
216 dense 3072 1024 3.0 2.549404 0.020508 -3.819330 147
217 dense 3072 3072 1.0 2.873059 0.022111 -3.196886 128
218 dense 23040 3072 7.5 2.714335 0.008467 -0.570293 558
219 dense 3072 3072 1.0 2.875937 0.017749 -3.354314 114
220 dense 3072 1024 3.0 2.988583 0.031645 -5.191214 92
221 dense 23040 3072 7.5 4.376892 0.020629 -4.690985 127
222 dense 23040 3072 7.5 2.628461 0.007657 -0.407412 629
223 dense 3072 1024 3.0 2.534855 0.018784 -3.630130 112
224 dense 3072 3072 1.0 2.447579 0.018270 -2.689536 218
225 dense 23040 3072 7.5 2.651026 0.008803 -0.427695 660
226 dense 3072 1024 3.0 2.417491 0.017560 -2.818806 134
227 dense 3072 3072 1.0 2.411684 0.018641 -2.586424 174
228 dense 3072 1024 3.0 2.868628 0.032221 -4.842477 105
229 dense 23040 3072 7.5 4.497092 0.015661 -4.751191 132
230 dense 23040 3072 7.5 2.779108 0.007632 -0.791487 470
231 dense 3072 3072 1.0 2.947310 0.021258 -3.273108 116
232 dense 3072 3072 1.0 3.405583 0.030911 -4.483439 101
233 dense 23040 3072 7.5 2.798323 0.008506 -0.832962 505
234 dense 23040 3072 7.5 2.667711 0.010047 -0.521205 658
235 dense 23040 3072 7.5 4.404984 0.018375 -4.619220 127
236 dense 3072 1024 3.0 2.688799 0.019797 -4.189840 119
237 dense 3072 1024 3.0 2.859723 0.030341 -4.755060 145
238 dense 3072 3072 1.0 2.500440 0.018580 -2.859522 166
239 dense 23040 3072 7.5 2.757686 0.010659 -0.750150 551
240 dense 3072 1024 3.0 2.409558 0.028312 -3.519978 151
241 dense 23040 3072 7.5 2.636048 0.010292 -0.483263 704
242 dense 3072 3072 1.0 2.734629 0.017292 -3.445384 136
243 dense 23040 3072 7.5 4.019895 0.018523 -4.158497 209
244 dense 3072 3072 1.0 2.343252 0.015792 -2.265632 203
245 dense 3072 1024 3.0 2.547375 0.024043 -3.513344 145
246 dense 23040 3072 7.5 3.680752 0.015473 -3.775617 319
247 dense 23040 3072 7.5 2.606811 0.008831 -0.417173 679
248 dense 23040 3072 7.5 2.732224 0.009040 -0.678690 549
249 dense 3072 1024 3.0 2.792293 0.028112 -4.582159 88
250 dense 3072 3072 1.0 4.070912 0.042770 -6.137448 48
251 dense 3072 3072 1.0 2.446963 0.022232 -2.605288 186
252 dense 3072 1024 3.0 2.808258 0.021219 -4.281290 116
253 dense 23040 3072 7.5 3.568636 0.016090 -3.673485 272
254 dense 3072 1024 3.0 2.672336 0.017691 -3.767086 188
255 dense 3072 3072 1.0 2.358732 0.019112 -2.230987 238
256 dense 3072 3072 1.0 2.331469 0.082648 -3.303236 551
257 dense 23040 3072 7.5 2.719900 0.012935 -0.573753 749
258 dense 23040 3072 7.5 2.604642 0.015026 -0.347186 914
259 dense 3072 1024 3.0 2.715743 0.020372 -4.207056 101
260 dense 3072 3072 1.0 2.921670 0.021829 -3.295386 161
261 dense 23040 3072 7.5 3.713777 0.030190 -3.324335 142
262 dense 3072 3072 1.0 2.350296 0.017432 -2.482955 279
263 dense 3072 1024 3.0 2.557802 0.023957 -4.071125 122
264 dense 3072 1024 3.0 2.679500 0.014798 -4.083492 151
265 dense 23040 3072 7.5 2.536120 0.013333 -0.504404 134
266 dense 23040 3072 7.5 2.452749 0.017107 -0.307458 187
267 dense 3072 3072 1.0 2.957495 0.019111 -3.576538 126
268 dense 3072 1024 3.0 2.753531 0.014268 -4.137627 153
269 dense 23040 3072 7.5 3.670335 0.035107 -1.500461 152
270 dense 23040 3072 7.5 2.631671 0.017377 -0.276566 958
271 dense 23040 3072 7.5 2.750803 0.015182 -0.504740 787
272 dense 3072 1024 3.0 2.606204 0.024541 -4.180106 96
273 dense 3072 3072 1.0 2.399797 0.016886 -2.484608 209
274 dense 3072 1024 3.0 2.429032 0.018688 -2.444177 208
275 dense 23040 3072 7.5 2.911920 0.013683 -0.057166 491
276 dense 3072 3072 1.0 2.510297 0.011566 -1.488887 163
277 dense 23040 3072 7.5 3.364591 0.025175 -0.846246 162
278 dense 3072 3072 1.0 2.261733 0.024288 -1.810518 290
279 dense 3072 1024 3.0 2.393219 0.020453 -3.343518 113
280 dense 23040 3072 7.5 2.788095 0.010086 -0.166480 661