| 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 | 1.50 | 0.10 | 4.29 | 2.86 | 27.70 | 0 | 0 | 1 |
| 12 | DENSE | 768 | 768 | 1.00 | 3.35 | 0.03 | 7.06 | 2.11 | 86.78 | 0 | 0 | 1 |
| 13 | DENSE | 768 | 768 | 1.00 | 2.75 | 0.07 | 5.94 | 2.16 | 80.44 | 0 | 0 | 1 |
| 14 | DENSE | 768 | 768 | 1.00 | 2.78 | 0.09 | 3.02 | 1.09 | 86.78 | 0 | 0 | 1 |
| 17 | DENSE | 768 | 768 | 1.00 | 2.72 | 0.08 | 4.48 | 1.65 | 83.32 | 0 | 1 | 1 |
| 21 | DENSE | 3072 | 768 | 4.00 | 2.80 | 0.02 | 7.46 | 2.66 | 81.35 | 1 | 0 | 0 |
| 24 | DENSE | 3072 | 768 | 4.00 | 3.31 | 0.03 | 6.56 | 1.98 | 84.67 | 0 | 0 | 0 |
| 30 | DENSE | 768 | 768 | 1.00 | 3.76 | 0.05 | 8.30 | 2.21 | 82.67 | 0 | 0 | 2 |
| 31 | DENSE | 768 | 768 | 1.00 | 2.91 | 0.04 | 5.19 | 1.78 | 83.24 | 0 | 0 | 1 |
| 32 | DENSE | 768 | 768 | 1.00 | 2.33 | 0.10 | 3.28 | 1.41 | 84.84 | 0 | 1 | 0 |
| 35 | DENSE | 768 | 768 | 1.00 | 3.36 | 0.08 | 4.45 | 1.32 | 86.58 | 0 | 0 | 2 |
| 39 | DENSE | 3072 | 768 | 4.00 | 3.58 | 0.02 | 10.05 | 2.81 | 85.33 | 1 | 0 | 0 |
| 42 | DENSE | 3072 | 768 | 4.00 | 3.27 | 0.07 | 6.60 | 2.02 | 83.28 | 0 | 1 | 0 |
| 48 | DENSE | 768 | 768 | 1.00 | 3.92 | 0.04 | 8.85 | 2.25 | 83.24 | 0 | 0 | 1 |
| 49 | DENSE | 768 | 768 | 1.00 | 2.87 | 0.09 | 4.84 | 1.69 | 86.10 | 0 | 3 | 2 |
| 50 | DENSE | 768 | 768 | 1.00 | 4.15 | 0.11 | 6.57 | 1.58 | 83.93 | 0 | 1 | 1 |
| 53 | DENSE | 768 | 768 | 1.00 | 3.59 | 0.08 | 4.64 | 1.29 | 86.50 | 0 | 0 | 1 |
| 57 | DENSE | 3072 | 768 | 4.00 | 3.69 | 0.02 | 9.97 | 2.70 | 84.30 | 1 | 0 | 0 |
| 60 | DENSE | 3072 | 768 | 4.00 | 3.23 | 0.07 | 6.56 | 2.03 | 82.79 | 0 | 1 | 0 |
| 66 | DENSE | 768 | 768 | 1.00 | 4.19 | 0.08 | 7.32 | 1.75 | 85.63 | 0 | 0 | 1 |
| 67 | DENSE | 768 | 768 | 1.00 | 3.72 | 0.05 | 6.46 | 1.74 | 83.44 | 0 | 0 | 2 |
| 68 | DENSE | 768 | 768 | 1.00 | 2.65 | 0.10 | 3.43 | 1.29 | 86.43 | 0 | 1 | 1 |
| 71 | DENSE | 768 | 768 | 1.00 | 5.06 | 0.05 | 6.63 | 1.31 | 89.73 | 0 | 0 | 1 |
| 75 | DENSE | 3072 | 768 | 4.00 | 3.56 | 0.04 | 9.53 | 2.68 | 84.05 | 1 | 0 | 0 |
| 78 | DENSE | 3072 | 768 | 4.00 | 3.63 | 0.02 | 8.08 | 2.23 | 81.48 | 0 | 0 | 0 |
| 84 | DENSE | 768 | 768 | 1.00 | 3.54 | 0.08 | 7.26 | 2.05 | 79.88 | 0 | 0 | 2 |
| 85 | DENSE | 768 | 768 | 1.00 | 3.26 | 0.09 | 5.47 | 1.68 | 81.09 | 0 | 0 | 2 |
| 86 | DENSE | 768 | 768 | 1.00 | 3.87 | 0.10 | 6.13 | 1.59 | 92.20 | 0 | 1 | 2 |
| 89 | DENSE | 768 | 768 | 1.00 | 3.48 | 0.06 | 4.87 | 1.40 | 92.88 | 0 | 0 | 1 |
| 93 | DENSE | 3072 | 768 | 4.00 | 3.39 | 0.03 | 9.21 | 2.72 | 83.49 | 1 | 0 | 0 |
| 96 | DENSE | 3072 | 768 | 4.00 | 4.43 | 0.04 | 9.36 | 2.12 | 87.36 | 0 | 0 | 0 |
| 102 | DENSE | 768 | 768 | 1.00 | 2.29 | 0.11 | 4.92 | 2.15 | 80.09 | 0 | 1 | 1 |
| 103 | DENSE | 768 | 768 | 1.00 | 2.61 | 0.10 | 4.17 | 1.60 | 83.35 | 0 | 1 | 2 |
| 104 | DENSE | 768 | 768 | 1.00 | 7.08 | 0.09 | 9.95 | 1.41 | 94.71 | 0 | 0 | 2 |
| 107 | DENSE | 768 | 768 | 1.00 | 4.90 | 0.10 | 7.08 | 1.44 | 92.84 | 0 | 0 | 1 |
| 111 | DENSE | 3072 | 768 | 4.00 | 3.34 | 0.02 | 8.77 | 2.62 | 82.85 | 1 | 0 | 0 |
| 114 | DENSE | 3072 | 768 | 4.00 | 4.69 | 0.08 | 10.82 | 2.31 | 82.21 | 0 | 0 | 0 |
| 118 | DENSE | 768 | 768 | 1.00 | 9.58 | 0.14 | 0.81 | 0.08 | 99.82 | 0 | 1 | 1 |



































































































































































































































