xlnet-base-cased
Find this model in the BertXLNet model summary
xlnet-base-cased Model Summary Plots
xlnet-base-cased Model Selected Details
layer_type
N
M
Q
alpha
D
alpha-hat
log_SN
% Rand
num_traps
num_fingers
rank_loss
layer_id
1
EMBEDDING
32000
768
41.67
2.85
0.03
10.90
3.83
71.70
0
0
0
9
DENSE
3072
768
4.00
2.33
0.04
5.62
2.42
74.84
0
0
0
10
DENSE
3072
768
4.00
3.19
0.02
6.94
2.17
81.95
7
0
0
20
DENSE
3072
768
4.00
2.84
0.02
6.77
2.39
78.37
0
0
0
21
DENSE
3072
768
4.00
3.39
0.04
6.84
2.02
84.23
4
0
0
31
DENSE
3072
768
4.00
3.04
0.02
7.44
2.44
79.05
0
0
0
32
DENSE
3072
768
4.00
3.99
0.05
7.16
1.79
86.25
1
0
0
42
DENSE
3072
768
4.00
3.32
0.02
8.47
2.55
78.18
0
0
0
43
DENSE
3072
768
4.00
4.37
0.05
9.55
2.18
87.51
8
0
0
53
DENSE
3072
768
4.00
3.68
0.03
8.57
2.33
82.21
0
0
0
54
DENSE
3072
768
4.00
4.29
0.03
9.11
2.12
87.90
3
0
0
64
DENSE
3072
768
4.00
3.98
0.03
9.02
2.27
82.75
0
0
0
65
DENSE
3072
768
4.00
4.56
0.03
8.73
1.91
87.30
3
0
0
75
DENSE
3072
768
4.00
3.57
0.04
8.17
2.29
82.96
0
0
0
76
DENSE
3072
768
4.00
3.27
0.10
5.91
1.81
88.13
5
7
0
86
DENSE
3072
768
4.00
3.76
0.03
8.55
2.27
83.34
0
0
0
87
DENSE
3072
768
4.00
3.95
0.03
8.94
2.26
86.53
5
0
0
97
DENSE
3072
768
4.00
3.57
0.02
8.88
2.49
82.24
0
0
0
98
DENSE
3072
768
4.00
3.63
0.04
9.85
2.71
81.03
9
0
0
108
DENSE
3072
768
4.00
3.81
0.02
9.36
2.46
82.89
0
0
0
109
DENSE
3072
768
4.00
3.54
0.02
9.48
2.67
81.95
6
0
0
119
DENSE
3072
768
4.00
3.61
0.02
8.72
2.41
83.00
0
0
0
120
DENSE
3072
768
4.00
3.80
0.02
8.89
2.34
84.51
14
0
0
130
DENSE
3072
768
4.00
3.39
0.02
7.18
2.12
84.16
0
0
0
131
DENSE
3072
768
4.00
3.43
0.06
7.70
2.25
87.94
23
0
0
xlnet-base-cased Layer Plots
Layer 1
Layer=1 | N=32000 | M=768 | Q=41.67 | alpha=2.85 | D_ks=0.03 | alpha-hat=10.90 | num traps=0
Layer 9
Layer=9 | N=3072 | M=768 | Q=4.00 | alpha=2.33 | D_ks=0.04 | alpha-hat=5.62 | num traps=0
Layer 10
Layer=10 | N=3072 | M=768 | Q=4.00 | alpha=3.19 | D_ks=0.02 | alpha-hat=6.94 | num traps=7
Layer 20
Layer=20 | N=3072 | M=768 | Q=4.00 | alpha=2.84 | D_ks=0.02 | alpha-hat=6.77 | num traps=0
Layer 21
Layer=21 | N=3072 | M=768 | Q=4.00 | alpha=3.39 | D_ks=0.04 | alpha-hat=6.84 | num traps=4
Layer 31
Layer=31 | N=3072 | M=768 | Q=4.00 | alpha=3.04 | D_ks=0.02 | alpha-hat=7.44 | num traps=0
Layer 32
Layer=32 | N=3072 | M=768 | Q=4.00 | alpha=3.99 | D_ks=0.05 | alpha-hat=7.16 | num traps=1
Layer 42
Layer=42 | N=3072 | M=768 | Q=4.00 | alpha=3.32 | D_ks=0.02 | alpha-hat=8.47 | num traps=0
Layer 43
Layer=43 | N=3072 | M=768 | Q=4.00 | alpha=4.37 | D_ks=0.05 | alpha-hat=9.55 | num traps=8
Layer 53
Layer=53 | N=3072 | M=768 | Q=4.00 | alpha=3.68 | D_ks=0.03 | alpha-hat=8.57 | num traps=0
Layer 54
Layer=54 | N=3072 | M=768 | Q=4.00 | alpha=4.29 | D_ks=0.03 | alpha-hat=9.11 | num traps=3
Layer 64
Layer=64 | N=3072 | M=768 | Q=4.00 | alpha=3.98 | D_ks=0.03 | alpha-hat=9.02 | num traps=0
Layer 65
Layer=65 | N=3072 | M=768 | Q=4.00 | alpha=4.56 | D_ks=0.03 | alpha-hat=8.73 | num traps=3
Layer 75
Layer=75 | N=3072 | M=768 | Q=4.00 | alpha=3.57 | D_ks=0.04 | alpha-hat=8.17 | num traps=0
Layer 76
Layer=76 | N=3072 | M=768 | Q=4.00 | alpha=3.27 | D_ks=0.10 | alpha-hat=5.91 | num traps=5
Layer 86
Layer=86 | N=3072 | M=768 | Q=4.00 | alpha=3.76 | D_ks=0.03 | alpha-hat=8.55 | num traps=0
Layer 87
Layer=87 | N=3072 | M=768 | Q=4.00 | alpha=3.95 | D_ks=0.03 | alpha-hat=8.94 | num traps=5
Layer 97
Layer=97 | N=3072 | M=768 | Q=4.00 | alpha=3.57 | D_ks=0.02 | alpha-hat=8.88 | num traps=0
Layer 98
Layer=98 | N=3072 | M=768 | Q=4.00 | alpha=3.63 | D_ks=0.04 | alpha-hat=9.85 | num traps=9
Layer 108
Layer=108 | N=3072 | M=768 | Q=4.00 | alpha=3.81 | D_ks=0.02 | alpha-hat=9.36 | num traps=0
Layer 109
Layer=109 | N=3072 | M=768 | Q=4.00 | alpha=3.54 | D_ks=0.02 | alpha-hat=9.48 | num traps=6
Layer 119
Layer=119 | N=3072 | M=768 | Q=4.00 | alpha=3.61 | D_ks=0.02 | alpha-hat=8.72 | num traps=0
Layer 120
Layer=120 | N=3072 | M=768 | Q=4.00 | alpha=3.80 | D_ks=0.02 | alpha-hat=8.89 | num traps=14
Layer 130
Layer=130 | N=3072 | M=768 | Q=4.00 | alpha=3.39 | D_ks=0.02 | alpha-hat=7.18 | num traps=0
Layer 131
Layer=131 | N=3072 | M=768 | Q=4.00 | alpha=3.43 | D_ks=0.06 | alpha-hat=7.70 | num traps=23
xlnet-base-cased Model Summary Plots
xlnet-base-cased Model Selected Details
| layer_type | N | M | Q | alpha | D | alpha-hat | log_SN | % Rand | num_traps | num_fingers | rank_loss | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| layer_id | ||||||||||||
| 1 | EMBEDDING | 32000 | 768 | 41.67 | 2.85 | 0.03 | 10.90 | 3.83 | 71.70 | 0 | 0 | 0 |
| 9 | DENSE | 3072 | 768 | 4.00 | 2.33 | 0.04 | 5.62 | 2.42 | 74.84 | 0 | 0 | 0 |
| 10 | DENSE | 3072 | 768 | 4.00 | 3.19 | 0.02 | 6.94 | 2.17 | 81.95 | 7 | 0 | 0 |
| 20 | DENSE | 3072 | 768 | 4.00 | 2.84 | 0.02 | 6.77 | 2.39 | 78.37 | 0 | 0 | 0 |
| 21 | DENSE | 3072 | 768 | 4.00 | 3.39 | 0.04 | 6.84 | 2.02 | 84.23 | 4 | 0 | 0 |
| 31 | DENSE | 3072 | 768 | 4.00 | 3.04 | 0.02 | 7.44 | 2.44 | 79.05 | 0 | 0 | 0 |
| 32 | DENSE | 3072 | 768 | 4.00 | 3.99 | 0.05 | 7.16 | 1.79 | 86.25 | 1 | 0 | 0 |
| 42 | DENSE | 3072 | 768 | 4.00 | 3.32 | 0.02 | 8.47 | 2.55 | 78.18 | 0 | 0 | 0 |
| 43 | DENSE | 3072 | 768 | 4.00 | 4.37 | 0.05 | 9.55 | 2.18 | 87.51 | 8 | 0 | 0 |
| 53 | DENSE | 3072 | 768 | 4.00 | 3.68 | 0.03 | 8.57 | 2.33 | 82.21 | 0 | 0 | 0 |
| 54 | DENSE | 3072 | 768 | 4.00 | 4.29 | 0.03 | 9.11 | 2.12 | 87.90 | 3 | 0 | 0 |
| 64 | DENSE | 3072 | 768 | 4.00 | 3.98 | 0.03 | 9.02 | 2.27 | 82.75 | 0 | 0 | 0 |
| 65 | DENSE | 3072 | 768 | 4.00 | 4.56 | 0.03 | 8.73 | 1.91 | 87.30 | 3 | 0 | 0 |
| 75 | DENSE | 3072 | 768 | 4.00 | 3.57 | 0.04 | 8.17 | 2.29 | 82.96 | 0 | 0 | 0 |
| 76 | DENSE | 3072 | 768 | 4.00 | 3.27 | 0.10 | 5.91 | 1.81 | 88.13 | 5 | 7 | 0 |
| 86 | DENSE | 3072 | 768 | 4.00 | 3.76 | 0.03 | 8.55 | 2.27 | 83.34 | 0 | 0 | 0 |
| 87 | DENSE | 3072 | 768 | 4.00 | 3.95 | 0.03 | 8.94 | 2.26 | 86.53 | 5 | 0 | 0 |
| 97 | DENSE | 3072 | 768 | 4.00 | 3.57 | 0.02 | 8.88 | 2.49 | 82.24 | 0 | 0 | 0 |
| 98 | DENSE | 3072 | 768 | 4.00 | 3.63 | 0.04 | 9.85 | 2.71 | 81.03 | 9 | 0 | 0 |
| 108 | DENSE | 3072 | 768 | 4.00 | 3.81 | 0.02 | 9.36 | 2.46 | 82.89 | 0 | 0 | 0 |
| 109 | DENSE | 3072 | 768 | 4.00 | 3.54 | 0.02 | 9.48 | 2.67 | 81.95 | 6 | 0 | 0 |
| 119 | DENSE | 3072 | 768 | 4.00 | 3.61 | 0.02 | 8.72 | 2.41 | 83.00 | 0 | 0 | 0 |
| 120 | DENSE | 3072 | 768 | 4.00 | 3.80 | 0.02 | 8.89 | 2.34 | 84.51 | 14 | 0 | 0 |
| 130 | DENSE | 3072 | 768 | 4.00 | 3.39 | 0.02 | 7.18 | 2.12 | 84.16 | 0 | 0 | 0 |
| 131 | DENSE | 3072 | 768 | 4.00 | 3.43 | 0.06 | 7.70 | 2.25 | 87.94 | 23 | 0 | 0 |
xlnet-base-cased Layer Plots
Layer 1
Layer=1 | N=32000 | M=768 | Q=41.67 | alpha=2.85 | D_ks=0.03 | alpha-hat=10.90 | num traps=0
Layer 9
Layer=9 | N=3072 | M=768 | Q=4.00 | alpha=2.33 | D_ks=0.04 | alpha-hat=5.62 | num traps=0
Layer 10
Layer=10 | N=3072 | M=768 | Q=4.00 | alpha=3.19 | D_ks=0.02 | alpha-hat=6.94 | num traps=7
Layer 20
Layer=20 | N=3072 | M=768 | Q=4.00 | alpha=2.84 | D_ks=0.02 | alpha-hat=6.77 | num traps=0
Layer 21
Layer=21 | N=3072 | M=768 | Q=4.00 | alpha=3.39 | D_ks=0.04 | alpha-hat=6.84 | num traps=4
Layer 31
Layer=31 | N=3072 | M=768 | Q=4.00 | alpha=3.04 | D_ks=0.02 | alpha-hat=7.44 | num traps=0
Layer 32
Layer=32 | N=3072 | M=768 | Q=4.00 | alpha=3.99 | D_ks=0.05 | alpha-hat=7.16 | num traps=1
Layer 42
Layer=42 | N=3072 | M=768 | Q=4.00 | alpha=3.32 | D_ks=0.02 | alpha-hat=8.47 | num traps=0
Layer 43
Layer=43 | N=3072 | M=768 | Q=4.00 | alpha=4.37 | D_ks=0.05 | alpha-hat=9.55 | num traps=8
Layer 53
Layer=53 | N=3072 | M=768 | Q=4.00 | alpha=3.68 | D_ks=0.03 | alpha-hat=8.57 | num traps=0
Layer 54
Layer=54 | N=3072 | M=768 | Q=4.00 | alpha=4.29 | D_ks=0.03 | alpha-hat=9.11 | num traps=3
Layer 64
Layer=64 | N=3072 | M=768 | Q=4.00 | alpha=3.98 | D_ks=0.03 | alpha-hat=9.02 | num traps=0
Layer 65
Layer=65 | N=3072 | M=768 | Q=4.00 | alpha=4.56 | D_ks=0.03 | alpha-hat=8.73 | num traps=3
Layer 75
Layer=75 | N=3072 | M=768 | Q=4.00 | alpha=3.57 | D_ks=0.04 | alpha-hat=8.17 | num traps=0
Layer 76
Layer=76 | N=3072 | M=768 | Q=4.00 | alpha=3.27 | D_ks=0.10 | alpha-hat=5.91 | num traps=5
Layer 86
Layer=86 | N=3072 | M=768 | Q=4.00 | alpha=3.76 | D_ks=0.03 | alpha-hat=8.55 | num traps=0
Layer 87
Layer=87 | N=3072 | M=768 | Q=4.00 | alpha=3.95 | D_ks=0.03 | alpha-hat=8.94 | num traps=5
Layer 97
Layer=97 | N=3072 | M=768 | Q=4.00 | alpha=3.57 | D_ks=0.02 | alpha-hat=8.88 | num traps=0
Layer 98
Layer=98 | N=3072 | M=768 | Q=4.00 | alpha=3.63 | D_ks=0.04 | alpha-hat=9.85 | num traps=9
Layer 108
Layer=108 | N=3072 | M=768 | Q=4.00 | alpha=3.81 | D_ks=0.02 | alpha-hat=9.36 | num traps=0
Layer 109
Layer=109 | N=3072 | M=768 | Q=4.00 | alpha=3.54 | D_ks=0.02 | alpha-hat=9.48 | num traps=6
Layer 119
Layer=119 | N=3072 | M=768 | Q=4.00 | alpha=3.61 | D_ks=0.02 | alpha-hat=8.72 | num traps=0
Layer 120
Layer=120 | N=3072 | M=768 | Q=4.00 | alpha=3.80 | D_ks=0.02 | alpha-hat=8.89 | num traps=14
Layer 130
Layer=130 | N=3072 | M=768 | Q=4.00 | alpha=3.39 | D_ks=0.02 | alpha-hat=7.18 | num traps=0
Layer 131
Layer=131 | N=3072 | M=768 | Q=4.00 | alpha=3.43 | D_ks=0.06 | alpha-hat=7.70 | num traps=23