| layer_type | N | M | Q | alpha | D | alpha-hat | log_SN | % Rand | num_traps | num_fingers | rank_loss | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| layer_id | ||||||||||||
| 3 | EMBEDDING | 768 | 514 | 1.49 | 1.44 | 0.09 | 4.60 | 3.20 | 27.65 | 1 | 1 | 1 |
| 12 | DENSE | 768 | 768 | 1.00 | 3.58 | 0.04 | 7.48 | 2.09 | 89.63 | 0 | 0 | 1 |
| 13 | DENSE | 768 | 768 | 1.00 | 1.93 | 0.09 | 4.50 | 2.33 | 76.24 | 0 | 1 | 1 |
| 14 | DENSE | 768 | 768 | 1.00 | 2.08 | 0.12 | 2.57 | 1.23 | 82.23 | 0 | 3 | 2 |
| 17 | DENSE | 768 | 768 | 1.00 | 2.85 | 0.10 | 4.77 | 1.67 | 83.01 | 1 | 3 | 2 |
| 21 | DENSE | 3072 | 768 | 4.00 | 2.53 | 0.04 | 7.94 | 3.14 | 70.53 | 1 | 0 | 0 |
| 24 | DENSE | 3072 | 768 | 4.00 | 6.28 | 0.06 | 11.59 | 1.85 | 92.51 | 0 | 0 | 0 |
| 30 | DENSE | 768 | 768 | 1.00 | 2.49 | 0.04 | 5.70 | 2.28 | 81.60 | 0 | 0 | 2 |
| 31 | DENSE | 768 | 768 | 1.00 | 2.60 | 0.06 | 5.30 | 2.04 | 80.64 | 0 | 0 | 2 |
| 32 | DENSE | 768 | 768 | 1.00 | 4.88 | 0.07 | 5.31 | 1.09 | 85.41 | 0 | 0 | 1 |
| 35 | DENSE | 768 | 768 | 1.00 | 4.74 | 0.06 | 4.73 | 1.00 | 89.31 | 0 | 0 | 2 |
| 39 | DENSE | 3072 | 768 | 4.00 | 2.62 | 0.03 | 8.08 | 3.09 | 79.71 | 1 | 0 | 0 |
| 42 | DENSE | 3072 | 768 | 4.00 | 5.09 | 0.02 | 9.01 | 1.77 | 89.11 | 0 | 0 | 0 |
| 48 | DENSE | 768 | 768 | 1.00 | 2.46 | 0.04 | 4.82 | 1.96 | 81.14 | 0 | 0 | 2 |
| 49 | DENSE | 768 | 768 | 1.00 | 2.65 | 0.04 | 5.26 | 1.98 | 81.19 | 0 | 0 | 2 |
| 50 | DENSE | 768 | 768 | 1.00 | 4.19 | 0.06 | 4.37 | 1.04 | 88.10 | 0 | 0 | 1 |
| 53 | DENSE | 768 | 768 | 1.00 | 3.33 | 0.03 | 4.49 | 1.35 | 83.44 | 0 | 0 | 3 |
| 57 | DENSE | 3072 | 768 | 4.00 | 2.67 | 0.03 | 8.09 | 3.03 | 80.72 | 1 | 0 | 0 |
| 60 | DENSE | 3072 | 768 | 4.00 | 3.26 | 0.04 | 5.90 | 1.81 | 82.68 | 0 | 0 | 0 |
| 66 | DENSE | 768 | 768 | 1.00 | 3.86 | 0.07 | 6.17 | 1.60 | 85.45 | 0 | 0 | 2 |
| 67 | DENSE | 768 | 768 | 1.00 | 3.78 | 0.05 | 5.88 | 1.56 | 88.39 | 0 | 0 | 2 |
| 68 | DENSE | 768 | 768 | 1.00 | 5.03 | 0.10 | 5.82 | 1.16 | 87.26 | 0 | 0 | 1 |
| 71 | DENSE | 768 | 768 | 1.00 | 2.48 | 0.04 | 3.49 | 1.41 | 80.46 | 0 | 0 | 1 |
| 75 | DENSE | 3072 | 768 | 4.00 | 3.22 | 0.03 | 10.09 | 3.14 | 81.50 | 1 | 0 | 0 |
| 78 | DENSE | 3072 | 768 | 4.00 | 4.77 | 0.03 | 10.40 | 2.18 | 83.01 | 0 | 0 | 0 |
| 84 | DENSE | 768 | 768 | 1.00 | 3.18 | 0.07 | 5.37 | 1.69 | 86.87 | 0 | 0 | 1 |
| 85 | DENSE | 768 | 768 | 1.00 | 3.31 | 0.05 | 5.58 | 1.68 | 87.48 | 0 | 0 | 1 |
| 86 | DENSE | 768 | 768 | 1.00 | 4.87 | 0.09 | 6.01 | 1.23 | 85.69 | 0 | 0 | 0 |
| 89 | DENSE | 768 | 768 | 1.00 | 2.07 | 0.06 | 3.66 | 1.77 | 74.93 | 0 | 0 | 3 |
| 93 | DENSE | 3072 | 768 | 4.00 | 3.59 | 0.04 | 11.75 | 3.27 | 80.12 | 1 | 0 | 0 |
| 96 | DENSE | 3072 | 768 | 4.00 | 4.56 | 0.03 | 10.96 | 2.40 | 80.24 | 0 | 0 | 0 |
| 102 | DENSE | 768 | 768 | 1.00 | 4.09 | 0.09 | 5.83 | 1.43 | 86.79 | 0 | 0 | 3 |
| 103 | DENSE | 768 | 768 | 1.00 | 3.34 | 0.04 | 5.43 | 1.62 | 84.94 | 0 | 0 | 4 |
| 104 | DENSE | 768 | 768 | 1.00 | 5.07 | 0.11 | 6.47 | 1.28 | 87.82 | 0 | 1 | 1 |
| 107 | DENSE | 768 | 768 | 1.00 | 1.92 | 0.08 | 3.05 | 1.59 | 76.75 | 0 | 0 | 2 |
| 111 | DENSE | 3072 | 768 | 4.00 | 3.75 | 0.04 | 12.24 | 3.27 | 80.06 | 1 | 0 | 0 |
| 114 | DENSE | 3072 | 768 | 4.00 | 4.04 | 0.03 | 9.77 | 2.42 | 76.19 | 0 | 0 | 0 |
| 120 | DENSE | 768 | 768 | 1.00 | 3.44 | 0.08 | 6.67 | 1.94 | 85.57 | 0 | 0 | 3 |
| 121 | DENSE | 768 | 768 | 1.00 | 3.35 | 0.07 | 5.24 | 1.56 | 86.31 | 0 | 0 | 4 |
| 122 | DENSE | 768 | 768 | 1.00 | 2.71 | 0.11 | 3.77 | 1.39 | 85.27 | 0 | 1 | 2 |
| 125 | DENSE | 768 | 768 | 1.00 | 2.00 | 0.07 | 3.27 | 1.64 | 75.55 | 0 | 0 | 2 |
| 129 | DENSE | 3072 | 768 | 4.00 | 3.62 | 0.05 | 11.68 | 3.23 | 80.14 | 1 | 0 | 0 |
| 132 | DENSE | 3072 | 768 | 4.00 | 3.56 | 0.03 | 9.40 | 2.64 | 73.93 | 0 | 0 | 0 |
| 138 | DENSE | 768 | 768 | 1.00 | 3.87 | 0.05 | 6.79 | 1.75 | 85.10 | 0 | 0 | 3 |
| 139 | DENSE | 768 | 768 | 1.00 | 3.54 | 0.06 | 5.64 | 1.59 | 83.54 | 0 | 0 | 4 |
| 140 | DENSE | 768 | 768 | 1.00 | 1.73 | 0.11 | 1.98 | 1.14 | 89.95 | 0 | 0 | 2 |
| 143 | DENSE | 768 | 768 | 1.00 | 2.14 | 0.07 | 3.56 | 1.66 | 80.01 | 0 | 0 | 2 |
| 147 | DENSE | 3072 | 768 | 4.00 | 3.58 | 0.05 | 11.24 | 3.14 | 81.26 | 1 | 0 | 0 |
| 150 | DENSE | 3072 | 768 | 4.00 | 3.77 | 0.03 | 9.70 | 2.57 | 75.25 | 0 | 0 | 0 |
| 156 | DENSE | 768 | 768 | 1.00 | 3.30 | 0.12 | 6.10 | 1.85 | 83.45 | 0 | 4 | 3 |
| 157 | DENSE | 768 | 768 | 1.00 | 4.24 | 0.08 | 6.54 | 1.54 | 83.35 | 0 | 0 | 4 |
| 158 | DENSE | 768 | 768 | 1.00 | 3.48 | 0.13 | 3.91 | 1.13 | 90.93 | 0 | 4 | 4 |
| 161 | DENSE | 768 | 768 | 1.00 | 1.94 | 0.05 | 3.31 | 1.71 | 71.71 | 0 | 0 | 1 |
| 165 | DENSE | 3072 | 768 | 4.00 | 3.61 | 0.04 | 11.02 | 3.06 | 81.24 | 1 | 0 | 0 |
| 168 | DENSE | 3072 | 768 | 4.00 | 3.61 | 0.02 | 8.90 | 2.47 | 76.94 | 0 | 0 | 0 |
| 174 | DENSE | 768 | 768 | 1.00 | 3.36 | 0.10 | 7.07 | 2.10 | 83.64 | 0 | 1 | 3 |
| 175 | DENSE | 768 | 768 | 1.00 | 3.82 | 0.07 | 6.17 | 1.62 | 84.06 | 0 | 0 | 4 |
| 176 | DENSE | 768 | 768 | 1.00 | 5.80 | 0.13 | 6.43 | 1.11 | 93.04 | 0 | 6 | 3 |
| 179 | DENSE | 768 | 768 | 1.00 | 2.67 | 0.09 | 4.41 | 1.65 | 85.08 | 0 | 1 | 1 |
| 183 | DENSE | 3072 | 768 | 4.00 | 3.54 | 0.05 | 10.49 | 2.96 | 82.81 | 1 | 0 | 0 |
| 186 | DENSE | 3072 | 768 | 4.00 | 5.24 | 0.06 | 12.78 | 2.44 | 82.71 | 0 | 0 | 0 |
| 192 | DENSE | 768 | 768 | 1.00 | 3.37 | 0.09 | 7.50 | 2.23 | 81.90 | 0 | 0 | 1 |
| 193 | DENSE | 768 | 768 | 1.00 | 2.73 | 0.10 | 4.31 | 1.58 | 84.44 | 0 | 1 | 3 |
| 194 | DENSE | 768 | 768 | 1.00 | 6.86 | 0.05 | 7.94 | 1.16 | 95.63 | 0 | 0 | 3 |
| 197 | DENSE | 768 | 768 | 1.00 | 5.25 | 0.07 | 9.13 | 1.74 | 85.83 | 0 | 0 | 1 |
| 201 | DENSE | 3072 | 768 | 4.00 | 3.50 | 0.02 | 9.33 | 2.66 | 83.45 | 1 | 0 | 0 |
| 204 | DENSE | 3072 | 768 | 4.00 | 4.69 | 0.08 | 12.43 | 2.65 | 79.06 | 0 | 0 | 0 |
| 210 | DENSE | 768 | 768 | 1.00 | 2.46 | 0.08 | 5.00 | 2.03 | 82.68 | 0 | 0 | 2 |
| 211 | DENSE | 768 | 768 | 1.00 | 2.98 | 0.09 | 5.12 | 1.72 | 87.99 | 0 | 1 | 2 |
| 212 | DENSE | 768 | 768 | 1.00 | 8.50 | 0.13 | 7.59 | 0.89 | 97.11 | 0 | 1 | 3 |
| 215 | DENSE | 768 | 768 | 1.00 | 4.22 | 0.08 | 7.85 | 1.86 | 84.36 | 2 | 0 | 1 |
| 219 | DENSE | 3072 | 768 | 4.00 | 2.67 | 0.02 | 6.32 | 2.37 | 79.35 | 1 | 0 | 0 |
| 222 | DENSE | 3072 | 768 | 4.00 | 3.71 | 0.09 | 10.35 | 2.79 | 70.00 | 4 | 0 | 0 |
| 226 | DENSE | 768 | 768 | 1.00 | 7.22 | 0.14 | 0.61 | 0.08 | 99.80 | 0 | 7 | 2 |



























































































































































































































































































































































































































































