layer_id int64 0 223 | name stringlengths 26 32 | D float64 0.03 0.21 | M int64 1.02k 4.1k | N int64 4.1k 14.3k | Q float64 1 4 | alpha float64 2.55 30.7 | alpha_weighted float64 -100.7 -4.79 | entropy float64 1.07 1.57 | has_esd bool 1
class | lambda_max float32 0 0.01 | layer_type stringclasses 1
value | log_alpha_norm float64 -100.63 -4.73 | log_norm float32 -1.66 -0.94 | log_spectral_norm float32 -3.31 -1.88 | matrix_rank int64 64 64 | norm float32 0.02 0.11 | num_evals int64 1.02k 4.1k | num_pl_spikes int64 5 64 | rank_loss int64 960 4.03k | rf int64 1 1 | sigma float64 0.35 9.03 | spectral_norm float32 0 0.01 | stable_rank float32 4.91 54.6 | status stringclasses 1
value | sv_max float64 0.02 0.11 | sv_min float64 0 0 | warning stringclasses 2
values | weak_rank_loss int64 960 4.03k | xmax float64 0 0.01 | xmin float64 0 0 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
100 | model.layers.14.mlp.up_proj | 0.088045 | 4,096 | 14,336 | 3.5 | 4.728654 | -10.57703 | 1.558624 | true | 0.005797 | dense | -10.503501 | -0.974476 | -2.236795 | 64 | 0.106053 | 4,096 | 10 | 4,032 | 1 | 1.179104 | 0.005797 | 18.294415 | success | 0.076138 | 0.000001 | 4,032 | 0.005797 | 0.001769 | |
101 | model.layers.14.self_attn.k_proj | 0.052885 | 1,024 | 4,096 | 4 | 7.301511 | -19.055762 | 1.129169 | true | 0.002456 | dense | -19.051874 | -1.337155 | -2.609838 | 64 | 0.046009 | 1,024 | 62 | 960 | 1 | 0.800293 | 0.002456 | 18.736265 | success | 0.049554 | 0.000001 | under-trained | 960 | 0.002456 | 0.000601 |
102 | model.layers.14.self_attn.o_proj | 0.081484 | 4,096 | 4,096 | 1 | 6.163327 | -16.709728 | 1.564028 | true | 0.001945 | dense | -16.529209 | -1.267445 | -2.711154 | 64 | 0.05402 | 4,096 | 14 | 4,032 | 1 | 1.379957 | 0.001945 | 27.778484 | success | 0.044098 | 0 | under-trained | 4,032 | 0.001945 | 0.000851 |
103 | model.layers.14.self_attn.q_proj | 0.078813 | 4,096 | 4,096 | 1 | 9.171611 | -23.894232 | 1.561724 | true | 0.002482 | dense | -23.888121 | -1.267009 | -2.605238 | 64 | 0.054074 | 4,096 | 64 | 4,032 | 1 | 1.021451 | 0.002482 | 21.788572 | success | 0.049817 | 0 | under-trained | 4,032 | 0.002482 | 0.000732 |
104 | model.layers.14.self_attn.v_proj | 0.052027 | 1,024 | 4,096 | 4 | 15.531955 | -48.761606 | 1.136586 | true | 0.000725 | dense | -48.656321 | -1.501851 | -3.139438 | 64 | 0.031488 | 1,024 | 26 | 960 | 1 | 2.849951 | 0.000725 | 43.409729 | success | 0.026933 | 0.000001 | under-trained | 960 | 0.000725 | 0.000495 |
105 | model.layers.15.mlp.down_proj | 0.03222 | 4,096 | 14,336 | 3.5 | 20.178456 | -54.219032 | 1.567939 | true | 0.002056 | dense | -54.08938 | -1.004997 | -2.686976 | 64 | 0.098856 | 4,096 | 62 | 4,032 | 1 | 2.435666 | 0.002056 | 48.08157 | success | 0.045343 | 0.000001 | under-trained | 4,032 | 0.002056 | 0.001466 |
106 | model.layers.15.mlp.gate_proj | 0.071743 | 4,096 | 14,336 | 3.5 | 5.233762 | -11.880146 | 1.559517 | true | 0.005371 | dense | -11.838327 | -1.002592 | -2.269906 | 64 | 0.099405 | 4,096 | 13 | 4,032 | 1 | 1.174234 | 0.005371 | 18.506042 | success | 0.07329 | 0.000001 | 4,032 | 0.005371 | 0.00155 | |
107 | model.layers.15.mlp.up_proj | 0.066682 | 4,096 | 14,336 | 3.5 | 5.879578 | -13.534449 | 1.561189 | true | 0.00499 | dense | -13.495264 | -0.986517 | -2.301942 | 64 | 0.103153 | 4,096 | 16 | 4,032 | 1 | 1.219894 | 0.00499 | 20.674023 | success | 0.070636 | 0.000001 | 4,032 | 0.00499 | 0.001588 | |
108 | model.layers.15.self_attn.k_proj | 0.099247 | 1,024 | 4,096 | 4 | 9.423515 | -28.762983 | 1.136285 | true | 0.000887 | dense | -27.768991 | -1.35326 | -3.052256 | 64 | 0.044334 | 1,024 | 50 | 960 | 1 | 1.191265 | 0.000887 | 50.003071 | success | 0.029776 | 0.000001 | under-trained | 960 | 0.000887 | 0.000633 |
109 | model.layers.15.self_attn.o_proj | 0.032321 | 4,096 | 4,096 | 1 | 15.634389 | -46.124784 | 1.567676 | true | 0.001121 | dense | -45.919358 | -1.286214 | -2.950213 | 64 | 0.051735 | 4,096 | 34 | 4,032 | 1 | 2.509777 | 0.001121 | 46.13166 | success | 0.033488 | 0 | under-trained | 4,032 | 0.001121 | 0.00079 |
110 | model.layers.15.self_attn.q_proj | 0.064558 | 4,096 | 4,096 | 1 | 12.589234 | -35.11913 | 1.566411 | true | 0.001623 | dense | -35.114409 | -1.288 | -2.789616 | 64 | 0.051523 | 4,096 | 53 | 4,032 | 1 | 1.591904 | 0.001623 | 31.740629 | success | 0.04029 | 0 | under-trained | 4,032 | 0.001623 | 0.000746 |
111 | model.layers.15.self_attn.v_proj | 0.064377 | 1,024 | 4,096 | 4 | 28.481666 | -92.912677 | 1.137139 | true | 0.000547 | dense | -92.497945 | -1.528765 | -3.262192 | 64 | 0.029596 | 1,024 | 16 | 960 | 1 | 6.870417 | 0.000547 | 54.128735 | success | 0.023383 | 0.000001 | under-trained | 960 | 0.000547 | 0.000482 |
112 | model.layers.16.mlp.down_proj | 0.087122 | 4,096 | 14,336 | 3.5 | 18.278263 | -47.557774 | 1.567344 | true | 0.002501 | dense | -47.528136 | -1.00071 | -2.601876 | 64 | 0.099837 | 4,096 | 64 | 4,032 | 1 | 2.159783 | 0.002501 | 39.917759 | success | 0.050011 | 0.000001 | under-trained | 4,032 | 0.002501 | 0.001466 |
113 | model.layers.16.mlp.gate_proj | 0.110417 | 4,096 | 14,336 | 3.5 | 4.460473 | -9.956797 | 1.55842 | true | 0.005858 | dense | -9.879789 | -0.988953 | -2.232229 | 64 | 0.102576 | 4,096 | 8 | 4,032 | 1 | 1.223462 | 0.005858 | 17.509598 | success | 0.076539 | 0.000001 | 4,032 | 0.005858 | 0.001726 | |
114 | model.layers.16.mlp.up_proj | 0.107535 | 4,096 | 14,336 | 3.5 | 4.569904 | -10.342811 | 1.559859 | true | 0.005455 | dense | -10.244776 | -0.983598 | -2.263245 | 64 | 0.103849 | 4,096 | 8 | 4,032 | 1 | 1.262152 | 0.005455 | 19.039099 | success | 0.073855 | 0.000001 | 4,032 | 0.005455 | 0.001755 | |
115 | model.layers.16.self_attn.k_proj | 0.095658 | 1,024 | 4,096 | 4 | 28.090519 | -86.373939 | 1.136644 | true | 0.000842 | dense | -86.024576 | -1.364159 | -3.074843 | 64 | 0.043236 | 1,024 | 9 | 960 | 1 | 9.030173 | 0.000842 | 51.367046 | success | 0.029012 | 0.000001 | under-trained | 960 | 0.000842 | 0.000754 |
116 | model.layers.16.self_attn.o_proj | 0.030218 | 4,096 | 4,096 | 1 | 14.659859 | -42.945266 | 1.567561 | true | 0.001176 | dense | -42.756544 | -1.27997 | -2.929446 | 64 | 0.052484 | 4,096 | 45 | 4,032 | 1 | 2.036292 | 0.001176 | 44.614502 | success | 0.034299 | 0 | under-trained | 4,032 | 0.001176 | 0.000781 |
117 | model.layers.16.self_attn.q_proj | 0.066623 | 4,096 | 4,096 | 1 | 12.333208 | -33.995883 | 1.565982 | true | 0.001752 | dense | -33.993732 | -1.286641 | -2.756451 | 64 | 0.051684 | 4,096 | 55 | 4,032 | 1 | 1.528169 | 0.001752 | 29.499187 | success | 0.041858 | 0 | under-trained | 4,032 | 0.001752 | 0.000744 |
118 | model.layers.16.self_attn.v_proj | 0.086352 | 1,024 | 4,096 | 4 | 27.09401 | -87.411861 | 1.137041 | true | 0.000594 | dense | -87.27872 | -1.519833 | -3.226243 | 64 | 0.030211 | 1,024 | 15 | 960 | 1 | 6.737444 | 0.000594 | 50.863968 | success | 0.024371 | 0.000001 | under-trained | 960 | 0.000594 | 0.000497 |
119 | model.layers.17.mlp.down_proj | 0.057253 | 4,096 | 14,336 | 3.5 | 19.359662 | -51.849015 | 1.567815 | true | 0.002098 | dense | -51.711726 | -1.007425 | -2.678198 | 64 | 0.098305 | 4,096 | 64 | 4,032 | 1 | 2.294958 | 0.002098 | 46.856842 | success | 0.045804 | 0.000001 | under-trained | 4,032 | 0.002098 | 0.001451 |
120 | model.layers.17.mlp.gate_proj | 0.087451 | 4,096 | 14,336 | 3.5 | 5.048414 | -11.485839 | 1.560231 | true | 0.005307 | dense | -11.425627 | -0.99271 | -2.275138 | 64 | 0.101693 | 4,096 | 11 | 4,032 | 1 | 1.220643 | 0.005307 | 19.161421 | success | 0.07285 | 0.000001 | 4,032 | 0.005307 | 0.001603 | |
121 | model.layers.17.mlp.up_proj | 0.106461 | 4,096 | 14,336 | 3.5 | 4.94852 | -11.495758 | 1.561953 | true | 0.004753 | dense | -11.37885 | -0.988145 | -2.32307 | 64 | 0.102767 | 4,096 | 8 | 4,032 | 1 | 1.396013 | 0.004753 | 21.623453 | success | 0.068939 | 0.000001 | 4,032 | 0.004753 | 0.001731 | |
122 | model.layers.17.self_attn.k_proj | 0.079491 | 1,024 | 4,096 | 4 | 7.654828 | -19.952351 | 1.126326 | true | 0.002475 | dense | -19.939089 | -1.35584 | -2.606506 | 64 | 0.044072 | 1,024 | 63 | 960 | 1 | 0.838429 | 0.002475 | 17.810081 | success | 0.049745 | 0.000001 | under-trained | 960 | 0.002475 | 0.000574 |
123 | model.layers.17.self_attn.o_proj | 0.028133 | 4,096 | 4,096 | 1 | 12.521953 | -36.428269 | 1.567143 | true | 0.001233 | dense | -36.314292 | -1.304693 | -2.909152 | 64 | 0.04958 | 4,096 | 59 | 4,032 | 1 | 1.500031 | 0.001233 | 40.221554 | success | 0.035109 | 0 | under-trained | 4,032 | 0.001233 | 0.000712 |
124 | model.layers.17.self_attn.q_proj | 0.083729 | 4,096 | 4,096 | 1 | 9.104787 | -22.625788 | 1.557598 | true | 0.003273 | dense | -22.625483 | -1.267234 | -2.485043 | 64 | 0.054046 | 4,096 | 64 | 4,032 | 1 | 1.013098 | 0.003273 | 16.512367 | success | 0.057211 | 0 | under-trained | 4,032 | 0.003273 | 0.000723 |
125 | model.layers.17.self_attn.v_proj | 0.070676 | 1,024 | 4,096 | 4 | 25.021776 | -81.282501 | 1.137071 | true | 0.000564 | dense | -80.976471 | -1.530491 | -3.248471 | 64 | 0.029479 | 1,024 | 17 | 960 | 1 | 5.826136 | 0.000564 | 52.237156 | success | 0.023756 | 0.000001 | under-trained | 960 | 0.000564 | 0.000479 |
126 | model.layers.18.mlp.down_proj | 0.050305 | 4,096 | 14,336 | 3.5 | 22.059051 | -59.651624 | 1.567991 | true | 0.001976 | dense | -59.524687 | -1.015775 | -2.704179 | 64 | 0.096433 | 4,096 | 62 | 4,032 | 1 | 2.674502 | 0.001976 | 48.798252 | success | 0.044454 | 0.000001 | under-trained | 4,032 | 0.001976 | 0.001437 |
127 | model.layers.18.mlp.gate_proj | 0.101667 | 4,096 | 14,336 | 3.5 | 4.72942 | -10.654039 | 1.559201 | true | 0.005588 | dense | -10.596874 | -1.00335 | -2.252716 | 64 | 0.099232 | 4,096 | 9 | 4,032 | 1 | 1.24314 | 0.005588 | 17.756847 | success | 0.074755 | 0.000001 | 4,032 | 0.005588 | 0.001601 | |
128 | model.layers.18.mlp.up_proj | 0.093648 | 4,096 | 14,336 | 3.5 | 5.078744 | -11.618822 | 1.560661 | true | 0.005155 | dense | -11.560412 | -1.001033 | -2.287735 | 64 | 0.099762 | 4,096 | 10 | 4,032 | 1 | 1.289812 | 0.005155 | 19.350937 | success | 0.071801 | 0.000001 | 4,032 | 0.005155 | 0.001593 | |
129 | model.layers.18.self_attn.k_proj | 0.084797 | 1,024 | 4,096 | 4 | 9.731679 | -29.269051 | 1.135682 | true | 0.000983 | dense | -28.591311 | -1.352156 | -3.007606 | 64 | 0.044447 | 1,024 | 34 | 960 | 1 | 1.497471 | 0.000983 | 45.232372 | success | 0.031347 | 0.000001 | under-trained | 960 | 0.000983 | 0.000673 |
130 | model.layers.18.self_attn.o_proj | 0.053086 | 4,096 | 4,096 | 1 | 12.688044 | -36.572721 | 1.566637 | true | 0.001311 | dense | -36.512068 | -1.321364 | -2.882455 | 64 | 0.047713 | 4,096 | 64 | 4,032 | 1 | 1.461006 | 0.001311 | 36.399193 | success | 0.036205 | 0 | under-trained | 4,032 | 0.001311 | 0.00068 |
131 | model.layers.18.self_attn.q_proj | 0.044601 | 4,096 | 4,096 | 1 | 11.852522 | -32.606474 | 1.565888 | true | 0.001774 | dense | -32.602863 | -1.2781 | -2.751016 | 64 | 0.052711 | 4,096 | 43 | 4,032 | 1 | 1.654994 | 0.001774 | 29.71092 | success | 0.04212 | 0 | under-trained | 4,032 | 0.001774 | 0.000775 |
132 | model.layers.18.self_attn.v_proj | 0.08094 | 1,024 | 4,096 | 4 | 25.83335 | -83.407907 | 1.137018 | true | 0.000591 | dense | -83.302488 | -1.530775 | -3.228691 | 64 | 0.029459 | 1,024 | 11 | 960 | 1 | 7.487537 | 0.000591 | 49.878788 | success | 0.024303 | 0.000001 | under-trained | 960 | 0.000591 | 0.000491 |
133 | model.layers.19.mlp.down_proj | 0.046149 | 4,096 | 14,336 | 3.5 | 24.274677 | -66.061206 | 1.568086 | true | 0.001899 | dense | -65.986934 | -1.028575 | -2.721404 | 64 | 0.093632 | 4,096 | 63 | 4,032 | 1 | 2.932334 | 0.001899 | 49.298027 | success | 0.043581 | 0.000001 | under-trained | 4,032 | 0.001899 | 0.001401 |
134 | model.layers.19.mlp.gate_proj | 0.123848 | 4,096 | 14,336 | 3.5 | 4.70368 | -10.993695 | 1.561626 | true | 0.0046 | dense | -10.883689 | -1.026132 | -2.337254 | 64 | 0.09416 | 4,096 | 7 | 4,032 | 1 | 1.39986 | 0.0046 | 20.470171 | success | 0.067822 | 0.000001 | 4,032 | 0.0046 | 0.001578 | |
135 | model.layers.19.mlp.up_proj | 0.098325 | 4,096 | 14,336 | 3.5 | 5.139143 | -12.166721 | 1.562589 | true | 0.004291 | dense | -12.069168 | -1.027631 | -2.367461 | 64 | 0.093836 | 4,096 | 9 | 4,032 | 1 | 1.379714 | 0.004291 | 21.869057 | success | 0.065504 | 0.000001 | 4,032 | 0.004291 | 0.001497 | |
136 | model.layers.19.self_attn.k_proj | 0.051625 | 1,024 | 4,096 | 4 | 10.897674 | -32.478873 | 1.135455 | true | 0.001046 | dense | -32.330556 | -1.402143 | -2.980349 | 64 | 0.039615 | 1,024 | 22 | 960 | 1 | 2.110191 | 0.001046 | 37.862217 | success | 0.032346 | 0.000001 | under-trained | 960 | 0.001046 | 0.000633 |
137 | model.layers.19.self_attn.o_proj | 0.04253 | 4,096 | 4,096 | 1 | 17.472803 | -52.45571 | 1.567787 | true | 0.000995 | dense | -52.225657 | -1.323864 | -3.002135 | 64 | 0.047439 | 4,096 | 33 | 4,032 | 1 | 2.867547 | 0.000995 | 47.672775 | success | 0.031545 | 0 | under-trained | 4,032 | 0.000995 | 0.000728 |
138 | model.layers.19.self_attn.q_proj | 0.04941 | 4,096 | 4,096 | 1 | 10.202023 | -28.205337 | 1.565039 | true | 0.001719 | dense | -28.201599 | -1.339994 | -2.764681 | 64 | 0.045709 | 4,096 | 64 | 4,032 | 1 | 1.150253 | 0.001719 | 26.588095 | success | 0.041463 | 0 | under-trained | 4,032 | 0.001719 | 0.000634 |
139 | model.layers.19.self_attn.v_proj | 0.064489 | 1,024 | 4,096 | 4 | 22.368846 | -71.606298 | 1.136987 | true | 0.000629 | dense | -71.518224 | -1.51859 | -3.201162 | 64 | 0.030298 | 1,024 | 19 | 960 | 1 | 4.90235 | 0.000629 | 48.147243 | success | 0.025085 | 0.000001 | under-trained | 960 | 0.000629 | 0.000488 |
140 | model.layers.20.mlp.down_proj | 0.066527 | 4,096 | 14,336 | 3.5 | 26.106228 | -71.931461 | 1.56813 | true | 0.001757 | dense | -71.776794 | -1.045039 | -2.755337 | 64 | 0.090149 | 4,096 | 63 | 4,032 | 1 | 3.163087 | 0.001757 | 51.321369 | success | 0.041911 | 0.000001 | under-trained | 4,032 | 0.001757 | 0.001353 |
141 | model.layers.20.mlp.gate_proj | 0.112275 | 4,096 | 14,336 | 3.5 | 5.347478 | -12.838731 | 1.563282 | true | 0.003973 | dense | -12.746096 | -1.045397 | -2.400895 | 64 | 0.090075 | 4,096 | 8 | 4,032 | 1 | 1.537066 | 0.003973 | 22.672419 | success | 0.063031 | 0.000001 | 4,032 | 0.003973 | 0.001456 | |
142 | model.layers.20.mlp.up_proj | 0.096243 | 4,096 | 14,336 | 3.5 | 5.597637 | -13.746404 | 1.564401 | true | 0.003501 | dense | -13.614599 | -1.052082 | -2.455751 | 64 | 0.088699 | 4,096 | 9 | 4,032 | 1 | 1.532546 | 0.003501 | 25.332001 | success | 0.059173 | 0.000001 | 4,032 | 0.003501 | 0.001402 | |
143 | model.layers.20.self_attn.k_proj | 0.056116 | 1,024 | 4,096 | 4 | 11.830419 | -36.589825 | 1.136153 | true | 0.000807 | dense | -36.31897 | -1.462648 | -3.09286 | 64 | 0.034463 | 1,024 | 32 | 960 | 1 | 1.914566 | 0.000807 | 42.678757 | success | 0.028416 | 0.000001 | under-trained | 960 | 0.000807 | 0.000528 |
144 | model.layers.20.self_attn.o_proj | 0.069663 | 4,096 | 4,096 | 1 | 28.4237 | -87.707927 | 1.568112 | true | 0.000821 | dense | -87.527496 | -1.362754 | -3.085732 | 64 | 0.043376 | 4,096 | 22 | 4,032 | 1 | 5.846752 | 0.000821 | 52.841881 | success | 0.028651 | 0 | under-trained | 4,032 | 0.000821 | 0.000688 |
145 | model.layers.20.self_attn.q_proj | 0.05119 | 4,096 | 4,096 | 1 | 10.494094 | -29.652063 | 1.565325 | true | 0.001494 | dense | -29.646998 | -1.3821 | -2.825595 | 64 | 0.041486 | 4,096 | 64 | 4,032 | 1 | 1.186762 | 0.001494 | 27.764874 | success | 0.038655 | 0 | under-trained | 4,032 | 0.001494 | 0.000577 |
146 | model.layers.20.self_attn.v_proj | 0.093938 | 1,024 | 4,096 | 4 | 24.501967 | -79.827178 | 1.137067 | true | 0.000552 | dense | -79.417424 | -1.53325 | -3.257991 | 64 | 0.029292 | 1,024 | 19 | 960 | 1 | 5.391721 | 0.000552 | 53.056709 | success | 0.023497 | 0.000001 | under-trained | 960 | 0.000552 | 0.000474 |
147 | model.layers.21.mlp.down_proj | 0.064974 | 4,096 | 14,336 | 3.5 | 26.468458 | -73.706019 | 1.568107 | true | 0.001642 | dense | -73.33047 | -1.065387 | -2.784674 | 64 | 0.086023 | 4,096 | 63 | 4,032 | 1 | 3.208724 | 0.001642 | 52.394669 | success | 0.040519 | 0.000001 | under-trained | 4,032 | 0.001642 | 0.001292 |
148 | model.layers.21.mlp.gate_proj | 0.115016 | 4,096 | 14,336 | 3.5 | 6.141163 | -15.168297 | 1.564783 | true | 0.003389 | dense | -15.086632 | -1.064503 | -2.469939 | 64 | 0.086198 | 4,096 | 8 | 4,032 | 1 | 1.817676 | 0.003389 | 25.435242 | success | 0.058214 | 0.000001 | under-trained | 4,032 | 0.003389 | 0.001383 |
149 | model.layers.21.mlp.up_proj | 0.106059 | 4,096 | 14,336 | 3.5 | 6.094915 | -15.4517 | 1.565676 | true | 0.002916 | dense | -15.282092 | -1.076371 | -2.535179 | 64 | 0.083874 | 4,096 | 8 | 4,032 | 1 | 1.801325 | 0.002916 | 28.761208 | success | 0.054002 | 0.000001 | under-trained | 4,032 | 0.002916 | 0.001339 |
150 | model.layers.21.self_attn.k_proj | 0.049746 | 1,024 | 4,096 | 4 | 7.629897 | -21.971864 | 1.132671 | true | 0.001319 | dense | -21.927961 | -1.469238 | -2.879707 | 64 | 0.033944 | 1,024 | 60 | 960 | 1 | 0.855916 | 0.001319 | 25.731724 | success | 0.03632 | 0.000001 | under-trained | 960 | 0.001319 | 0.000453 |
151 | model.layers.21.self_attn.o_proj | 0.061163 | 4,096 | 4,096 | 1 | 22.63699 | -70.924932 | 1.568127 | true | 0.000736 | dense | -70.323526 | -1.395994 | -3.133143 | 64 | 0.04018 | 4,096 | 42 | 4,032 | 1 | 3.33866 | 0.000736 | 54.594528 | success | 0.027129 | 0 | under-trained | 4,032 | 0.000736 | 0.000612 |
152 | model.layers.21.self_attn.q_proj | 0.053863 | 4,096 | 4,096 | 1 | 9.267359 | -25.456937 | 1.562925 | true | 0.001791 | dense | -25.45525 | -1.401946 | -2.746946 | 64 | 0.039633 | 4,096 | 64 | 4,032 | 1 | 1.03342 | 0.001791 | 22.130991 | success | 0.042318 | 0 | under-trained | 4,032 | 0.001791 | 0.000539 |
153 | model.layers.21.self_attn.v_proj | 0.08334 | 1,024 | 4,096 | 4 | 30.733462 | -100.703772 | 1.137111 | true | 0.000529 | dense | -100.628457 | -1.568105 | -3.276682 | 64 | 0.027033 | 1,024 | 11 | 960 | 1 | 8.964976 | 0.000529 | 51.118313 | success | 0.022996 | 0.000001 | under-trained | 960 | 0.000529 | 0.000449 |
154 | model.layers.22.mlp.down_proj | 0.092055 | 4,096 | 14,336 | 3.5 | 26.495596 | -73.430247 | 1.568081 | true | 0.001693 | dense | -73.376784 | -1.081067 | -2.771413 | 64 | 0.082972 | 4,096 | 62 | 4,032 | 1 | 3.237944 | 0.001693 | 49.01701 | success | 0.041143 | 0.000001 | under-trained | 4,032 | 0.001693 | 0.001247 |
155 | model.layers.22.mlp.gate_proj | 0.115706 | 4,096 | 14,336 | 3.5 | 6.272838 | -15.50362 | 1.564791 | true | 0.003376 | dense | -15.429582 | -1.06528 | -2.471548 | 64 | 0.086044 | 4,096 | 8 | 4,032 | 1 | 1.86423 | 0.003376 | 25.484016 | success | 0.058107 | 0.000001 | under-trained | 4,032 | 0.003376 | 0.00139 |
156 | model.layers.22.mlp.up_proj | 0.109543 | 4,096 | 14,336 | 3.5 | 6.262851 | -15.84857 | 1.565468 | true | 0.002947 | dense | -15.714697 | -1.082538 | -2.530568 | 64 | 0.082692 | 4,096 | 8 | 4,032 | 1 | 1.860699 | 0.002947 | 28.056255 | success | 0.05429 | 0.000001 | under-trained | 4,032 | 0.002947 | 0.001335 |
157 | model.layers.22.self_attn.k_proj | 0.061843 | 1,024 | 4,096 | 4 | 7.847612 | -24.359356 | 1.134993 | true | 0.000787 | dense | -23.882569 | -1.512321 | -3.104047 | 64 | 0.030738 | 1,024 | 44 | 960 | 1 | 1.032316 | 0.000787 | 39.059383 | success | 0.028053 | 0.000001 | under-trained | 960 | 0.000787 | 0.000436 |
158 | model.layers.22.self_attn.o_proj | 0.047603 | 4,096 | 4,096 | 1 | 15.036358 | -46.856894 | 1.56769 | true | 0.000765 | dense | -46.407214 | -1.432106 | -3.11624 | 64 | 0.036974 | 4,096 | 63 | 4,032 | 1 | 1.768415 | 0.000765 | 48.320675 | success | 0.027662 | 0 | under-trained | 4,032 | 0.000765 | 0.000537 |
159 | model.layers.22.self_attn.q_proj | 0.046817 | 4,096 | 4,096 | 1 | 9.643863 | -27.110748 | 1.563508 | true | 0.001545 | dense | -27.109173 | -1.447843 | -2.811192 | 64 | 0.035658 | 4,096 | 49 | 4,032 | 1 | 1.234838 | 0.001545 | 23.08602 | success | 0.039301 | 0 | under-trained | 4,032 | 0.001545 | 0.000505 |
160 | model.layers.22.self_attn.v_proj | 0.058297 | 1,024 | 4,096 | 4 | 24.13966 | -79.221554 | 1.137047 | true | 0.000523 | dense | -79.091265 | -1.58329 | -3.281801 | 64 | 0.026104 | 1,024 | 17 | 960 | 1 | 5.612192 | 0.000523 | 49.947212 | success | 0.022861 | 0.000001 | under-trained | 960 | 0.000523 | 0.000423 |
161 | model.layers.23.mlp.down_proj | 0.080725 | 4,096 | 14,336 | 3.5 | 26.884795 | -73.691522 | 1.567962 | true | 0.001815 | dense | -73.689571 | -1.094295 | -2.741011 | 64 | 0.080483 | 4,096 | 62 | 4,032 | 1 | 3.287372 | 0.001815 | 44.331905 | success | 0.042608 | 0.000001 | under-trained | 4,032 | 0.001815 | 0.001209 |
162 | model.layers.23.mlp.gate_proj | 0.105507 | 4,096 | 14,336 | 3.5 | 6.553426 | -16.146519 | 1.564583 | true | 0.003437 | dense | -16.098287 | -1.069908 | -2.463829 | 64 | 0.085132 | 4,096 | 9 | 4,032 | 1 | 1.851142 | 0.003437 | 24.769682 | success | 0.058625 | 0.000001 | under-trained | 4,032 | 0.003437 | 0.001351 |
163 | model.layers.23.mlp.up_proj | 0.098312 | 4,096 | 14,336 | 3.5 | 6.451926 | -16.419561 | 1.565579 | true | 0.002852 | dense | -16.298551 | -1.0934 | -2.544909 | 64 | 0.080649 | 4,096 | 9 | 4,032 | 1 | 1.817309 | 0.002852 | 28.281919 | success | 0.053401 | 0.000001 | under-trained | 4,032 | 0.002852 | 0.001282 |
164 | model.layers.23.self_attn.k_proj | 0.044231 | 1,024 | 4,096 | 4 | 8.685878 | -26.818006 | 1.135234 | true | 0.000817 | dense | -26.599642 | -1.528306 | -3.087541 | 64 | 0.029627 | 1,024 | 50 | 960 | 1 | 1.086947 | 0.000817 | 36.243912 | success | 0.028591 | 0.000001 | under-trained | 960 | 0.000817 | 0.000417 |
165 | model.layers.23.self_attn.o_proj | 0.086024 | 4,096 | 4,096 | 1 | 13.179125 | -38.548754 | 1.566138 | true | 0.001189 | dense | -38.540363 | -1.420885 | -2.924986 | 64 | 0.037942 | 4,096 | 64 | 4,032 | 1 | 1.522391 | 0.001189 | 31.922808 | success | 0.034475 | 0 | under-trained | 4,032 | 0.001189 | 0.000542 |
166 | model.layers.23.self_attn.q_proj | 0.047519 | 4,096 | 4,096 | 1 | 10.634295 | -31.235869 | 1.565653 | true | 0.001155 | dense | -31.228127 | -1.471754 | -2.937277 | 64 | 0.033748 | 4,096 | 48 | 4,032 | 1 | 1.390591 | 0.001155 | 29.209417 | success | 0.033991 | 0 | under-trained | 4,032 | 0.001155 | 0.000486 |
167 | model.layers.23.self_attn.v_proj | 0.04169 | 1,024 | 4,096 | 4 | 17.404055 | -56.238253 | 1.136861 | true | 0.000587 | dense | -56.055635 | -1.56089 | -3.23133 | 64 | 0.027486 | 1,024 | 29 | 960 | 1 | 3.046157 | 0.000587 | 46.821026 | success | 0.024229 | 0.000001 | under-trained | 960 | 0.000587 | 0.000428 |
168 | model.layers.24.mlp.down_proj | 0.09546 | 4,096 | 14,336 | 3.5 | 24.953419 | -67.179076 | 1.567588 | true | 0.002032 | dense | -67.178959 | -1.106459 | -2.692179 | 64 | 0.07826 | 4,096 | 64 | 4,032 | 1 | 2.994177 | 0.002032 | 38.523045 | success | 0.045072 | 0.000001 | under-trained | 4,032 | 0.002032 | 0.001169 |
169 | model.layers.24.mlp.gate_proj | 0.125072 | 4,096 | 14,336 | 3.5 | 15.395967 | -37.963343 | 1.564485 | true | 0.003421 | dense | -37.963322 | -1.075045 | -2.465798 | 64 | 0.084131 | 4,096 | 64 | 4,032 | 1 | 1.799496 | 0.003421 | 24.589685 | success | 0.058493 | 0.000001 | under-trained | 4,032 | 0.003421 | 0.001211 |
170 | model.layers.24.mlp.up_proj | 0.095826 | 4,096 | 14,336 | 3.5 | 6.089533 | -15.509972 | 1.565375 | true | 0.002838 | dense | -15.365067 | -1.10238 | -2.546989 | 64 | 0.078999 | 4,096 | 9 | 4,032 | 1 | 1.696511 | 0.002838 | 27.836147 | success | 0.053273 | 0.000001 | under-trained | 4,032 | 0.002838 | 0.001248 |
171 | model.layers.24.self_attn.k_proj | 0.047288 | 1,024 | 4,096 | 4 | 7.493427 | -22.751314 | 1.133444 | true | 0.00092 | dense | -22.595868 | -1.558617 | -3.036169 | 64 | 0.02763 | 1,024 | 51 | 960 | 1 | 0.909262 | 0.00092 | 30.029797 | success | 0.030333 | 0.000001 | under-trained | 960 | 0.00092 | 0.000378 |
172 | model.layers.24.self_attn.o_proj | 0.071055 | 4,096 | 4,096 | 1 | 16.48228 | -49.786721 | 1.567127 | true | 0.000954 | dense | -49.785523 | -1.479156 | -3.020621 | 64 | 0.033178 | 4,096 | 64 | 4,032 | 1 | 1.935285 | 0.000954 | 34.79084 | success | 0.030881 | 0 | under-trained | 4,032 | 0.000954 | 0.000484 |
173 | model.layers.24.self_attn.q_proj | 0.036041 | 4,096 | 4,096 | 1 | 9.925885 | -28.910437 | 1.564274 | true | 0.001223 | dense | -28.908021 | -1.519081 | -2.912631 | 64 | 0.030263 | 4,096 | 56 | 4,032 | 1 | 1.192772 | 0.001223 | 24.74852 | success | 0.034969 | 0 | under-trained | 4,032 | 0.001223 | 0.000424 |
174 | model.layers.24.self_attn.v_proj | 0.058889 | 1,024 | 4,096 | 4 | 23.276687 | -76.282564 | 1.13697 | true | 0.000528 | dense | -76.228967 | -1.599989 | -3.277209 | 64 | 0.02512 | 1,024 | 18 | 960 | 1 | 5.250666 | 0.000528 | 47.557606 | success | 0.022982 | 0.000001 | under-trained | 960 | 0.000528 | 0.000407 |
175 | model.layers.25.mlp.down_proj | 0.12544 | 4,096 | 14,336 | 3.5 | 22.601148 | -60.042548 | 1.567144 | true | 0.002205 | dense | -60.04247 | -1.113156 | -2.656615 | 64 | 0.077063 | 4,096 | 64 | 4,032 | 1 | 2.700143 | 0.002205 | 34.95097 | success | 0.046956 | 0.000001 | under-trained | 4,032 | 0.002205 | 0.001145 |
176 | model.layers.25.mlp.gate_proj | 0.12068 | 4,096 | 14,336 | 3.5 | 13.940288 | -34.545454 | 1.564761 | true | 0.003326 | dense | -34.545329 | -1.073043 | -2.478102 | 64 | 0.084519 | 4,096 | 64 | 4,032 | 1 | 1.617536 | 0.003326 | 25.413172 | success | 0.05767 | 0.000001 | under-trained | 4,032 | 0.003326 | 0.001208 |
177 | model.layers.25.mlp.up_proj | 0.09619 | 4,096 | 14,336 | 3.5 | 5.860057 | -14.98204 | 1.565308 | true | 0.002776 | dense | -14.800526 | -1.108379 | -2.556637 | 64 | 0.077915 | 4,096 | 9 | 4,032 | 1 | 1.620019 | 0.002776 | 28.071037 | success | 0.052684 | 0.000001 | 4,032 | 0.002776 | 0.001226 | |
178 | model.layers.25.self_attn.k_proj | 0.086496 | 1,024 | 4,096 | 4 | 7.71323 | -25.242188 | 1.135881 | true | 0.000534 | dense | -24.216169 | -1.590238 | -3.272583 | 64 | 0.02569 | 1,024 | 63 | 960 | 1 | 0.845788 | 0.000534 | 48.122189 | success | 0.023105 | 0.000001 | under-trained | 960 | 0.000534 | 0.000344 |
179 | model.layers.25.self_attn.o_proj | 0.083508 | 4,096 | 4,096 | 1 | 13.930974 | -42.769201 | 1.566749 | true | 0.000851 | dense | -42.640661 | -1.498674 | -3.07008 | 64 | 0.031719 | 4,096 | 63 | 4,032 | 1 | 1.62915 | 0.000851 | 37.273968 | success | 0.029172 | 0 | under-trained | 4,032 | 0.000851 | 0.000457 |
180 | model.layers.25.self_attn.q_proj | 0.039704 | 4,096 | 4,096 | 1 | 9.001914 | -26.694807 | 1.564694 | true | 0.001083 | dense | -26.675192 | -1.524629 | -2.965459 | 64 | 0.029879 | 4,096 | 51 | 4,032 | 1 | 1.120492 | 0.001083 | 27.594992 | success | 0.032906 | 0 | under-trained | 4,032 | 0.001083 | 0.000419 |
181 | model.layers.25.self_attn.v_proj | 0.079096 | 1,024 | 4,096 | 4 | 23.327133 | -76.709015 | 1.136906 | true | 0.000515 | dense | -76.58582 | -1.598524 | -3.288403 | 64 | 0.025204 | 1,024 | 11 | 960 | 1 | 6.731884 | 0.000515 | 48.964264 | success | 0.022688 | 0.000001 | under-trained | 960 | 0.000515 | 0.000422 |
182 | model.layers.26.mlp.down_proj | 0.134245 | 4,096 | 14,336 | 3.5 | 21.974989 | -57.246944 | 1.566523 | true | 0.002483 | dense | -57.246938 | -1.118854 | -2.605095 | 64 | 0.076058 | 4,096 | 64 | 4,032 | 1 | 2.621874 | 0.002483 | 30.636671 | success | 0.049826 | 0.000001 | under-trained | 4,032 | 0.002483 | 0.001126 |
183 | model.layers.26.mlp.gate_proj | 0.145916 | 4,096 | 14,336 | 3.5 | 14.525749 | -35.792439 | 1.564134 | true | 0.003435 | dense | -35.791913 | -1.068163 | -2.464068 | 64 | 0.085475 | 4,096 | 64 | 4,032 | 1 | 1.690719 | 0.003435 | 24.883167 | success | 0.058609 | 0.000001 | under-trained | 4,032 | 0.003435 | 0.001223 |
184 | model.layers.26.mlp.up_proj | 0.108684 | 4,096 | 14,336 | 3.5 | 6.368245 | -16.239063 | 1.564809 | true | 0.002818 | dense | -16.093494 | -1.103885 | -2.550006 | 64 | 0.078725 | 4,096 | 13 | 4,032 | 1 | 1.488883 | 0.002818 | 27.93321 | success | 0.053088 | 0.000001 | under-trained | 4,032 | 0.002818 | 0.001206 |
185 | model.layers.26.self_attn.k_proj | 0.034856 | 1,024 | 4,096 | 4 | 7.456504 | -22.91698 | 1.132999 | true | 0.000844 | dense | -22.644402 | -1.587041 | -3.073421 | 64 | 0.02588 | 1,024 | 49 | 960 | 1 | 0.922358 | 0.000844 | 30.646473 | success | 0.02906 | 0.000001 | under-trained | 960 | 0.000844 | 0.000356 |
186 | model.layers.26.self_attn.o_proj | 0.167697 | 4,096 | 4,096 | 1 | 10.922671 | -31.43949 | 1.562381 | true | 0.001323 | dense | -31.402265 | -1.469525 | -2.87837 | 64 | 0.033922 | 4,096 | 64 | 4,032 | 1 | 1.240334 | 0.001323 | 25.635735 | success | 0.036376 | 0 | under-trained | 4,032 | 0.001323 | 0.00047 |
187 | model.layers.26.self_attn.q_proj | 0.042271 | 4,096 | 4,096 | 1 | 7.985482 | -22.489588 | 1.560908 | true | 0.001526 | dense | -22.486146 | -1.516497 | -2.816309 | 64 | 0.030444 | 4,096 | 64 | 4,032 | 1 | 0.873185 | 0.001526 | 19.943983 | success | 0.03907 | 0 | under-trained | 4,032 | 0.001526 | 0.000403 |
188 | model.layers.26.self_attn.v_proj | 0.058746 | 1,024 | 4,096 | 4 | 24.701824 | -81.647791 | 1.137018 | true | 0.000495 | dense | -81.5235 | -1.607704 | -3.305335 | 64 | 0.024677 | 1,024 | 15 | 960 | 1 | 6.119785 | 0.000495 | 49.846031 | success | 0.02225 | 0.000001 | under-trained | 960 | 0.000495 | 0.000404 |
189 | model.layers.27.mlp.down_proj | 0.169859 | 4,096 | 14,336 | 3.5 | 19.350975 | -49.017357 | 1.565103 | true | 0.00293 | dense | -49.017354 | -1.122905 | -2.533069 | 64 | 0.075352 | 4,096 | 64 | 4,032 | 1 | 2.293872 | 0.00293 | 25.71369 | success | 0.054133 | 0.000001 | under-trained | 4,032 | 0.00293 | 0.001103 |
190 | model.layers.27.mlp.gate_proj | 0.154579 | 4,096 | 14,336 | 3.5 | 4.657099 | -11.249424 | 1.562604 | true | 0.003841 | dense | -11.058984 | -1.059703 | -2.415543 | 64 | 0.087156 | 4,096 | 7 | 4,032 | 1 | 1.382253 | 0.003841 | 22.690315 | success | 0.061977 | 0.000001 | 4,032 | 0.003841 | 0.001419 | |
191 | model.layers.27.mlp.up_proj | 0.139529 | 4,096 | 14,336 | 3.5 | 5.798713 | -14.547875 | 1.563372 | true | 0.003099 | dense | -14.383105 | -1.09817 | -2.508811 | 64 | 0.079768 | 4,096 | 14 | 4,032 | 1 | 1.28251 | 0.003099 | 25.741913 | success | 0.055667 | 0.000001 | 4,032 | 0.003099 | 0.001208 | |
192 | model.layers.27.self_attn.k_proj | 0.053247 | 1,024 | 4,096 | 4 | 7.088601 | -20.734525 | 1.129563 | true | 0.001188 | dense | -20.46086 | -1.529256 | -2.925052 | 64 | 0.029563 | 1,024 | 63 | 960 | 1 | 0.767092 | 0.001188 | 24.876902 | success | 0.034473 | 0.000001 | under-trained | 960 | 0.001188 | 0.000383 |
193 | model.layers.27.self_attn.o_proj | 0.101908 | 4,096 | 4,096 | 1 | 12.312543 | -37.227076 | 1.566002 | true | 0.000947 | dense | -37.146804 | -1.485103 | -3.023508 | 64 | 0.032726 | 4,096 | 64 | 4,032 | 1 | 1.414068 | 0.000947 | 34.546589 | success | 0.030778 | 0 | under-trained | 4,032 | 0.000947 | 0.000464 |
194 | model.layers.27.self_attn.q_proj | 0.042044 | 4,096 | 4,096 | 1 | 7.92628 | -21.679954 | 1.559018 | true | 0.00184 | dense | -21.678854 | -1.490915 | -2.735199 | 64 | 0.032291 | 4,096 | 64 | 4,032 | 1 | 0.865785 | 0.00184 | 17.550272 | success | 0.042894 | 0 | under-trained | 4,032 | 0.00184 | 0.000425 |
195 | model.layers.27.self_attn.v_proj | 0.063521 | 1,024 | 4,096 | 4 | 16.373393 | -53.042834 | 1.136855 | true | 0.000576 | dense | -52.951057 | -1.590571 | -3.239575 | 64 | 0.02567 | 1,024 | 38 | 960 | 1 | 2.493894 | 0.000576 | 44.566044 | success | 0.024 | 0.000001 | under-trained | 960 | 0.000576 | 0.00039 |
196 | model.layers.28.mlp.down_proj | 0.17393 | 4,096 | 14,336 | 3.5 | 4.513984 | -11.095919 | 1.563425 | true | 0.003482 | dense | -10.931005 | -1.115081 | -2.458121 | 64 | 0.076722 | 4,096 | 5 | 4,032 | 1 | 1.571502 | 0.003482 | 22.03126 | success | 0.059012 | 0.000001 | 4,032 | 0.003482 | 0.001234 | |
197 | model.layers.28.mlp.gate_proj | 0.174762 | 4,096 | 14,336 | 3.5 | 12.366044 | -29.875316 | 1.561714 | true | 0.003838 | dense | -29.865175 | -1.061048 | -2.415915 | 64 | 0.086886 | 4,096 | 64 | 4,032 | 1 | 1.420755 | 0.003838 | 22.639526 | success | 0.06195 | 0.000001 | under-trained | 4,032 | 0.003838 | 0.001217 |
198 | model.layers.28.mlp.up_proj | 0.132646 | 4,096 | 14,336 | 3.5 | 5.04215 | -12.611106 | 1.562622 | true | 0.003154 | dense | -12.314449 | -1.088936 | -2.501137 | 64 | 0.081482 | 4,096 | 12 | 4,032 | 1 | 1.166868 | 0.003154 | 25.834534 | success | 0.056161 | 0.000001 | 4,032 | 0.003154 | 0.001248 | |
199 | model.layers.28.self_attn.k_proj | 0.065966 | 1,024 | 4,096 | 4 | 7.460837 | -22.041873 | 1.131685 | true | 0.001111 | dense | -21.921246 | -1.543238 | -2.954343 | 64 | 0.028626 | 1,024 | 64 | 960 | 1 | 0.807605 | 0.001111 | 25.769484 | success | 0.033329 | 0.000001 | under-trained | 960 | 0.001111 | 0.000376 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.