Qwen2.5-1.5B-Instruct


Find this model in the Qwen2.5-small model summary

Qwen2.5-1.5B-Instruct Model Set Plots

Qwen2.5-small Compared to Base Model Plots

Qwen2.5-1.5B-Instruct Model Selected Details
id layer_type N M Q alpha D alpha-hat num_spikes warning
1 dense 8960 1536 5.833333 6.192732 0.053414 10.346296 337 under-trained
2 dense 8960 1536 5.833333 3.988269 0.014106 8.540256 176
3 dense 8960 1536 5.833333 11.216515 0.074838 15.578239 61 under-trained
4 dense 1536 256 6.000000 6.601371 0.053442 8.175576 35 under-trained
5 dense 1536 1536 1.000000 3.857758 0.078567 4.199012 380
6 dense 1536 1536 1.000000 2.491033 0.016290 7.162338 187
7 dense 1536 256 6.000000 9.249598 0.117407 0.807286 54 under-trained
8 dense 1536 256 6.000000 8.659670 0.057693 1.346619 38 under-trained
9 dense 1536 1536 1.000000 2.946589 0.039966 4.751012 87
10 dense 1536 1536 1.000000 3.799894 0.031738 4.510958 129
11 dense 8960 1536 5.833333 5.766564 0.020838 11.512075 173
12 dense 8960 1536 5.833333 4.295077 0.017742 9.836010 255
13 dense 8960 1536 5.833333 4.188555 0.038459 7.440409 44
14 dense 1536 256 6.000000 1.948153 0.138447 2.265253 180 over-trained
15 dense 1536 256 6.000000 5.640992 0.104665 1.722965 72
16 dense 8960 1536 5.833333 6.425356 0.037866 10.678067 187 under-trained
17 dense 1536 256 6.000000 2.809418 0.073513 2.514817 110
18 dense 1536 1536 1.000000 3.982889 0.057416 4.401463 128
19 dense 1536 1536 1.000000 2.463721 0.044057 3.413705 240
20 dense 8960 1536 5.833333 3.791164 0.014593 8.803422 346
21 dense 8960 1536 5.833333 5.177057 0.026783 8.574793 52
22 dense 8960 1536 5.833333 6.661750 0.012078 10.975939 124 under-trained
23 dense 8960 1536 5.833333 3.807793 0.021621 8.607070 395
24 dense 8960 1536 5.833333 11.687328 0.111947 14.980680 150 under-trained
25 dense 1536 256 6.000000 4.633157 0.122331 4.279891 77
26 dense 1536 1536 1.000000 4.689202 0.073950 5.028190 115
27 dense 1536 1536 1.000000 3.221272 0.037350 4.581188 121
28 dense 1536 256 6.000000 6.763702 0.052068 2.697367 46 under-trained
29 dense 8960 1536 5.833333 13.868327 0.048306 18.761084 75 under-trained
30 dense 8960 1536 5.833333 4.198176 0.043489 7.517608 385
31 dense 8960 1536 5.833333 4.162315 0.011687 9.223583 286
32 dense 1536 256 6.000000 5.732522 0.141886 4.100718 63
33 dense 1536 1536 1.000000 4.991207 0.031678 5.431063 89
34 dense 1536 1536 1.000000 4.631432 0.050596 6.066159 55
35 dense 1536 256 6.000000 12.222344 0.047042 4.711628 31 under-trained
36 dense 1536 1536 1.000000 3.399360 0.052377 4.696224 98
37 dense 1536 256 6.000000 5.607214 0.134396 3.884405 54
38 dense 8960 1536 5.833333 15.307873 0.064693 19.818747 76 under-trained
39 dense 8960 1536 5.833333 5.206945 0.029760 8.469914 233
40 dense 1536 256 6.000000 5.100932 0.087394 2.921472 73
41 dense 8960 1536 5.833333 3.699872 0.018507 9.005504 409
42 dense 1536 1536 1.000000 2.230894 0.105703 2.976226 346
43 dense 1536 1536 1.000000 2.490843 0.078598 3.396442 190
44 dense 8960 1536 5.833333 3.872802 0.020060 8.484631 375
45 dense 8960 1536 5.833333 10.010717 0.118530 14.129108 209 under-trained
46 dense 1536 256 6.000000 3.359665 0.118421 3.170409 78
47 dense 1536 1536 1.000000 5.200839 0.022721 6.194827 74
48 dense 8960 1536 5.833333 4.944765 0.063119 8.517043 348
49 dense 1536 256 6.000000 8.301245 0.051683 4.037567 30 under-trained
50 dense 1536 256 6.000000 9.330274 0.054402 4.280315 37 under-trained
51 dense 1536 1536 1.000000 4.598839 0.061739 6.011422 101
52 dense 1536 256 6.000000 2.097507 0.149502 1.830708 159
53 dense 1536 1536 1.000000 2.300247 0.089473 3.046359 329
54 dense 8960 1536 5.833333 3.988181 0.016082 8.459346 309
55 dense 8960 1536 5.833333 6.006131 0.017519 10.636528 103 under-trained
56 dense 8960 1536 5.833333 10.912916 0.047292 16.559290 105 under-trained
57 dense 1536 256 6.000000 7.069447 0.088015 2.744517 65 under-trained
58 dense 1536 1536 1.000000 2.342271 0.080442 3.149462 217
59 dense 8960 1536 5.833333 5.102557 0.046228 10.102247 198
60 dense 8960 1536 5.833333 4.134582 0.014998 8.758208 263
61 dense 8960 1536 5.833333 11.177652 0.023147 15.806928 60 under-trained
62 dense 1536 256 6.000000 3.003304 0.059474 2.579017 73
63 dense 1536 1536 1.000000 4.653224 0.045670 4.577975 76
64 dense 1536 1536 1.000000 2.194472 0.085262 2.836288 290
65 dense 8960 1536 5.833333 3.786920 0.072466 8.053799 471
66 dense 8960 1536 5.833333 4.151295 0.016712 8.445251 228
67 dense 8960 1536 5.833333 9.015780 0.019240 13.012854 81 under-trained
68 dense 1536 256 6.000000 7.474136 0.053709 6.141252 30 under-trained
69 dense 1536 1536 1.000000 3.681284 0.028524 4.153183 135
70 dense 1536 256 6.000000 6.536903 0.038494 3.163638 53 under-trained
71 dense 8960 1536 5.833333 3.642787 0.022077 7.547328 296
72 dense 1536 256 6.000000 11.068272 0.036770 4.138400 35 under-trained
73 dense 1536 1536 1.000000 2.286889 0.075278 3.030090 280
74 dense 1536 1536 1.000000 4.720976 0.119084 6.068936 134
75 dense 1536 256 6.000000 5.349618 0.077302 4.411973 28
76 dense 8960 1536 5.833333 7.660972 0.027495 10.951116 104 under-trained
77 dense 8960 1536 5.833333 3.717588 0.057197 8.118905 448
78 dense 1536 256 6.000000 8.965046 0.041249 2.988253 40 under-trained
79 dense 1536 1536 1.000000 3.717166 0.047234 5.094646 188
80 dense 1536 256 6.000000 4.253726 0.130903 4.158571 55
81 dense 8960 1536 5.833333 6.292470 0.039819 8.807026 138 under-trained
82 dense 8960 1536 5.833333 3.452131 0.038413 7.488068 298
83 dense 8960 1536 5.833333 5.040999 0.016244 10.865299 199
84 dense 1536 1536 1.000000 1.888479 0.102641 2.559120 490 over-trained
85 dense 1536 256 6.000000 5.940743 0.117987 2.228796 74
86 dense 1536 1536 1.000000 2.042567 0.099624 2.703976 370
87 dense 1536 1536 1.000000 3.663782 0.056835 4.413034 93
88 dense 1536 256 6.000000 3.696612 0.144186 2.904817 62
89 dense 8960 1536 5.833333 6.138574 0.031224 8.564404 157 under-trained
90 dense 8960 1536 5.833333 4.061060 0.024548 8.469624 170
91 dense 8960 1536 5.833333 4.559149 0.020936 9.606735 171
92 dense 8960 1536 5.833333 4.226422 0.020800 8.605664 167
93 dense 1536 256 6.000000 5.002470 0.074783 2.065032 91
94 dense 1536 1536 1.000000 3.508407 0.066514 4.916390 69
95 dense 1536 1536 1.000000 4.182174 0.032282 5.054072 120
96 dense 8960 1536 5.833333 4.142703 0.031644 8.487528 231
97 dense 8960 1536 5.833333 6.120409 0.029749 8.720710 146 under-trained
98 dense 1536 256 6.000000 5.300690 0.075661 4.990967 25
99 dense 1536 1536 1.000000 3.070896 0.104453 3.527150 168
100 dense 1536 256 6.000000 1.749499 0.121859 1.627044 140 over-trained
101 dense 1536 256 6.000000 6.135047 0.093281 2.390938 46 under-trained
102 dense 1536 1536 1.000000 1.923431 0.079839 2.749048 318 over-trained
103 dense 8960 1536 5.833333 3.510089 0.075406 7.377644 427
104 dense 8960 1536 5.833333 4.271436 0.016885 8.563691 143
105 dense 8960 1536 5.833333 5.807736 0.019748 8.421993 153
106 dense 1536 1536 1.000000 3.527058 0.044948 8.014304 27
107 dense 1536 256 6.000000 9.188657 0.122880 3.087044 74 under-trained
108 dense 8960 1536 5.833333 6.537858 0.048491 9.416969 69 under-trained
109 dense 8960 1536 5.833333 4.046365 0.033967 8.135159 135
110 dense 8960 1536 5.833333 4.875848 0.012841 10.508988 154
111 dense 1536 256 6.000000 3.850850 0.079676 3.816533 23
112 dense 1536 1536 1.000000 5.295649 0.102790 6.596783 87
113 dense 8960 1536 5.833333 4.343250 0.020733 8.904405 204
114 dense 8960 1536 5.833333 4.332672 0.041877 8.449353 136
115 dense 8960 1536 5.833333 7.312091 0.043358 10.202291 83 under-trained
116 dense 1536 1536 1.000000 4.793134 0.062272 6.492009 44
117 dense 1536 1536 1.000000 2.066577 0.071377 2.963435 280
118 dense 1536 256 6.000000 7.331428 0.127659 2.524294 65 under-trained
119 dense 1536 256 6.000000 2.082208 0.153395 1.702594 115
120 dense 8960 1536 5.833333 4.349535 0.015914 8.149673 185
121 dense 8960 1536 5.833333 4.291488 0.032393 8.466334 149
122 dense 8960 1536 5.833333 6.249738 0.046397 8.951590 158 under-trained
123 dense 1536 256 6.000000 4.211558 0.058996 2.912169 40
124 dense 1536 1536 1.000000 2.946031 0.085670 3.728661 174
125 dense 1536 1536 1.000000 1.968943 0.102475 2.511965 424 over-trained
126 dense 1536 256 6.000000 5.534511 0.121136 3.343237 46
127 dense 8960 1536 5.833333 5.271300 0.031121 10.276681 30
128 dense 8960 1536 5.833333 6.829580 0.033925 10.100236 104 under-trained
129 dense 1536 256 6.000000 6.711048 0.094003 4.948870 17 under-trained
130 dense 1536 1536 1.000000 2.911070 0.038801 3.451668 149
131 dense 1536 1536 1.000000 2.149709 0.080178 2.821936 293
132 dense 8960 1536 5.833333 4.145997 0.019223 7.388965 243
133 dense 1536 256 6.000000 6.582742 0.110474 2.693819 76 under-trained
134 dense 8960 1536 5.833333 5.822343 0.029026 10.817062 172
135 dense 8960 1536 5.833333 4.610228 0.031049 9.324222 108
136 dense 8960 1536 5.833333 6.666224 0.022174 12.088863 138 under-trained
137 dense 1536 1536 1.000000 7.986119 0.058497 9.752339 43 under-trained
138 dense 1536 1536 1.000000 2.135236 0.071866 3.185385 258
139 dense 1536 256 6.000000 8.081471 0.076485 3.296480 42 under-trained
140 dense 1536 256 6.000000 2.031798 0.099713 1.266101 121
141 dense 1536 256 6.000000 5.745552 0.114150 3.202652 107
142 dense 8960 1536 5.833333 7.178888 0.024819 12.139773 126 under-trained
143 dense 8960 1536 5.833333 5.131118 0.018190 10.478010 123
144 dense 8960 1536 5.833333 8.218023 0.030217 12.486075 85 under-trained
145 dense 1536 256 6.000000 1.878093 0.145534 1.449685 138 over-trained
146 dense 1536 1536 1.000000 5.154539 0.024396 7.526251 81
147 dense 1536 1536 1.000000 1.872556 0.091615 2.497908 442 over-trained
148 dense 8960 1536 5.833333 5.442999 0.015276 11.488695 129
149 dense 8960 1536 5.833333 5.885151 0.032589 9.738623 28
150 dense 8960 1536 5.833333 8.841463 0.027298 13.505339 96 under-trained
151 dense 1536 256 6.000000 5.834123 0.058242 4.475806 15
152 dense 1536 1536 1.000000 3.268522 0.052785 3.984235 130
153 dense 1536 1536 1.000000 2.091116 0.082548 2.808717 358
154 dense 1536 256 6.000000 4.988411 0.070037 3.364337 70
155 dense 1536 1536 1.000000 2.387682 0.083307 3.392633 259
156 dense 8960 1536 5.833333 8.823411 0.047814 14.090040 121 under-trained
157 dense 1536 1536 1.000000 4.751384 0.108364 5.686731 120
158 dense 1536 256 6.000000 2.562504 0.093931 1.954115 95
159 dense 1536 256 6.000000 8.818753 0.136663 4.531508 58 under-trained
160 dense 8960 1536 5.833333 6.025584 0.029758 12.763737 144 under-trained
161 dense 8960 1536 5.833333 7.168961 0.058317 12.039945 214 under-trained
162 dense 1536 256 6.000000 2.819453 0.098939 1.848671 198
163 dense 1536 1536 1.000000 5.010596 0.040776 7.078160 78
164 dense 1536 256 6.000000 2.670785 0.066310 2.266284 44
165 dense 8960 1536 5.833333 8.429790 0.041717 13.912143 104 under-trained
166 dense 8960 1536 5.833333 5.269237 0.024233 11.309271 149
167 dense 8960 1536 5.833333 10.859550 0.074042 17.533613 119 under-trained
168 dense 1536 1536 1.000000 2.261389 0.062543 3.286702 219
169 dense 1536 256 6.000000 6.479276 0.052193 4.468302 59 under-trained
170 dense 1536 1536 1.000000 2.528988 0.063315 3.744746 231
171 dense 1536 1536 1.000000 8.120524 0.047558 10.181946 52 under-trained
172 dense 1536 256 6.000000 4.171582 0.054418 3.403425 32
173 dense 8960 1536 5.833333 8.931620 0.057484 14.679871 114 under-trained
174 dense 8960 1536 5.833333 5.906404 0.026772 12.384088 135
175 dense 8960 1536 5.833333 11.296101 0.048852 17.401210 90 under-trained
176 dense 1536 256 6.000000 6.021168 0.035875 4.901322 42 under-trained
177 dense 8960 1536 5.833333 7.045234 0.061599 10.716275 187 under-trained
178 dense 8960 1536 5.833333 5.936938 0.024073 11.820454 135
179 dense 8960 1536 5.833333 9.431707 0.052813 15.379787 106 under-trained
180 dense 1536 256 6.000000 1.999358 0.060977 1.449005 155 over-trained
181 dense 1536 1536 1.000000 3.583733 0.016739 5.704252 159
182 dense 1536 1536 1.000000 3.008343 0.027258 4.704169 116
183 dense 1536 256 6.000000 6.119400 0.053731 6.727109 37 under-trained
184 dense 1536 1536 1.000000 3.041233 0.065331 4.264973 70
185 dense 1536 1536 1.000000 6.541357 0.059740 11.305734 91 under-trained
186 dense 1536 256 6.000000 1.816577 0.074641 1.534134 139 over-trained
187 dense 8960 1536 5.833333 6.768009 0.054838 15.718645 182 under-trained
188 dense 8960 1536 5.833333 4.737632 0.015891 10.142019 209
189 dense 8960 1536 5.833333 12.682194 0.035168 18.843943 48 under-trained
190 dense 1536 256 6.000000 6.032333 0.118819 6.288753 84 under-trained
191 dense 1536 1536 1.000000 2.187914 0.060869 3.056463 216
192 dense 1536 1536 1.000000 2.884349 0.046297 5.053284 227
193 dense 1536 256 6.000000 1.886137 0.051696 1.467891 124 over-trained
194 dense 8960 1536 5.833333 5.459954 0.063053 11.643075 289
195 dense 8960 1536 5.833333 4.602602 0.017919 8.044780 187
196 dense 8960 1536 5.833333 4.534571 0.034116 10.790930 344