dolly-v2-3b


Find this model in the Dolly model summary

Model source: https://huggingface.co/databricks/dolly-v2-3b


dolly-v2-3b Model Summary Plots




dolly-v2-3b Model Selected Details
  layer_type N M Q alpha D alpha-hat log_SN rank_loss
layer_id                  
4 DENSE 7680 2560 3.00 3.44 0.02 7.96 2.32 0
5 DENSE 2560 2560 1.00 2.98 0.02 4.88 1.64 5
6 DENSE 10240 2560 4.00 8.11 0.09 16.14 1.99 0
7 DENSE 10240 2560 4.00 8.44 0.05 13.85 1.64 0
10 DENSE 7680 2560 3.00 3.49 0.02 7.10 2.03 0
11 DENSE 2560 2560 1.00 4.22 0.03 4.60 1.09 3
12 DENSE 10240 2560 4.00 2.84 0.02 6.75 2.38 0
13 DENSE 10240 2560 4.00 5.60 0.05 9.99 1.78 0
16 DENSE 7680 2560 3.00 3.69 0.03 7.24 1.96 0
17 DENSE 2560 2560 1.00 4.39 0.03 5.86 1.34 4
18 DENSE 10240 2560 4.00 2.77 0.01 6.46 2.33 0
19 DENSE 10240 2560 4.00 3.89 0.02 7.34 1.89 0
22 DENSE 7680 2560 3.00 3.89 0.02 7.91 2.04 0
23 DENSE 2560 2560 1.00 4.72 0.05 5.38 1.14 3
24 DENSE 10240 2560 4.00 3.26 0.01 7.18 2.20 0
25 DENSE 10240 2560 4.00 3.23 0.01 5.31 1.64 0
28 DENSE 7680 2560 3.00 3.77 0.03 7.74 2.05 0
29 DENSE 2560 2560 1.00 4.25 0.04 5.14 1.21 3
30 DENSE 10240 2560 4.00 3.33 0.01 7.73 2.32 0
31 DENSE 10240 2560 4.00 3.18 0.02 5.95 1.87 0
34 DENSE 7680 2560 3.00 4.00 0.02 8.20 2.05 0
35 DENSE 2560 2560 1.00 6.88 0.05 7.15 1.04 2
36 DENSE 10240 2560 4.00 3.45 0.01 7.62 2.21 0
37 DENSE 10240 2560 4.00 3.71 0.02 6.19 1.67 0
40 DENSE 7680 2560 3.00 3.26 0.03 6.53 2.00 0
41 DENSE 2560 2560 1.00 4.22 0.06 4.47 1.06 2
42 DENSE 10240 2560 4.00 3.08 0.01 6.91 2.24 0
43 DENSE 10240 2560 4.00 3.34 0.04 5.41 1.62 0
46 DENSE 7680 2560 3.00 3.54 0.03 7.41 2.09 0
47 DENSE 2560 2560 1.00 4.44 0.06 4.95 1.11 3
48 DENSE 10240 2560 4.00 3.05 0.02 6.88 2.26 0
49 DENSE 10240 2560 4.00 3.10 0.03 5.22 1.68 0
52 DENSE 7680 2560 3.00 3.92 0.03 8.48 2.17 0
53 DENSE 2560 2560 1.00 6.55 0.06 7.80 1.19 3
54 DENSE 10240 2560 4.00 3.15 0.02 7.06 2.24 0
55 DENSE 10240 2560 4.00 3.16 0.03 5.10 1.62 0
58 DENSE 7680 2560 3.00 3.17 0.03 6.74 2.12 0
59 DENSE 2560 2560 1.00 3.16 0.06 3.76 1.19 3
60 DENSE 10240 2560 4.00 3.16 0.03 7.06 2.24 0
61 DENSE 10240 2560 4.00 3.01 0.03 4.76 1.58 0
64 DENSE 7680 2560 3.00 3.18 0.04 6.88 2.17 0
65 DENSE 2560 2560 1.00 3.47 0.05 3.87 1.12 3
66 DENSE 10240 2560 4.00 3.19 0.03 7.03 2.20 0
67 DENSE 10240 2560 4.00 3.19 0.03 4.63 1.45 0
70 DENSE 7680 2560 3.00 3.15 0.04 6.55 2.08 0
71 DENSE 2560 2560 1.00 3.42 0.05 3.95 1.15 3
72 DENSE 10240 2560 4.00 3.20 0.03 6.99 2.19 0
73 DENSE 10240 2560 4.00 3.32 0.03 4.84 1.46 0
76 DENSE 7680 2560 3.00 3.23 0.03 6.76 2.09 0
77 DENSE 2560 2560 1.00 3.83 0.07 4.07 1.06 2
78 DENSE 10240 2560 4.00 3.26 0.02 7.29 2.24 0
79 DENSE 10240 2560 4.00 3.30 0.03 4.89 1.48 0
82 DENSE 7680 2560 3.00 3.30 0.03 6.83 2.07 0
83 DENSE 2560 2560 1.00 3.73 0.04 3.88 1.04 3
84 DENSE 10240 2560 4.00 3.66 0.03 8.21 2.24 0
85 DENSE 10240 2560 4.00 3.80 0.02 5.40 1.42 0
88 DENSE 7680 2560 3.00 3.19 0.01 8.07 2.53 0
89 DENSE 2560 2560 1.00 4.01 0.05 4.01 1.00 3
90 DENSE 10240 2560 4.00 3.74 0.03 8.40 2.25 0
91 DENSE 10240 2560 4.00 4.34 0.02 6.20 1.43 0
94 DENSE 7680 2560 3.00 3.25 0.02 6.96 2.14 0
95 DENSE 2560 2560 1.00 14.00 0.06 11.74 0.84 3
96 DENSE 10240 2560 4.00 3.54 0.03 7.96 2.25 0
97 DENSE 10240 2560 4.00 3.58 0.03 5.49 1.53 0
100 DENSE 7680 2560 3.00 3.49 0.01 7.59 2.18 0
101 DENSE 2560 2560 1.00 4.26 0.05 4.33 1.02 3
102 DENSE 10240 2560 4.00 4.18 0.04 9.46 2.26 0
103 DENSE 10240 2560 4.00 3.89 0.02 5.75 1.48 0
106 DENSE 7680 2560 3.00 3.08 0.01 7.54 2.45 0
107 DENSE 2560 2560 1.00 8.07 0.02 7.71 0.96 2
108 DENSE 10240 2560 4.00 3.56 0.04 8.08 2.27 0
109 DENSE 10240 2560 4.00 4.51 0.02 6.75 1.50 0
112 DENSE 7680 2560 3.00 3.41 0.03 8.19 2.40 0
113 DENSE 2560 2560 1.00 8.60 0.06 9.67 1.12 2
114 DENSE 10240 2560 4.00 3.66 0.04 8.22 2.24 0
115 DENSE 10240 2560 4.00 4.42 0.02 6.90 1.56 0
118 DENSE 7680 2560 3.00 3.26 0.03 8.41 2.58 0
119 DENSE 2560 2560 1.00 7.75 0.04 8.63 1.11 2
120 DENSE 10240 2560 4.00 3.25 0.04 7.22 2.22 0
121 DENSE 10240 2560 4.00 4.24 0.01 7.08 1.67 0
124 DENSE 7680 2560 3.00 2.94 0.04 8.03 2.73 0
125 DENSE 2560 2560 1.00 6.44 0.03 7.77 1.21 2
126 DENSE 10240 2560 4.00 4.37 0.04 9.55 2.19 0
127 DENSE 10240 2560 4.00 4.65 0.01 8.40 1.81 0
130 DENSE 7680 2560 3.00 2.84 0.04 8.11 2.85 0
131 DENSE 2560 2560 1.00 5.70 0.04 7.45 1.31 2
132 DENSE 10240 2560 4.00 3.22 0.05 6.90 2.15 0
133 DENSE 10240 2560 4.00 5.21 0.01 9.89 1.90 0
136 DENSE 7680 2560 3.00 2.27 0.03 6.68 2.94 0
137 DENSE 2560 2560 1.00 5.41 0.03 8.10 1.50 2
138 DENSE 10240 2560 4.00 4.11 0.04 8.76 2.13 0
139 DENSE 10240 2560 4.00 5.44 0.02 10.52 1.93 0
142 DENSE 7680 2560 3.00 2.45 0.04 7.30 2.98 0
143 DENSE 2560 2560 1.00 5.75 0.02 7.66 1.33 2
144 DENSE 10240 2560 4.00 3.45 0.05 7.29 2.11 0
145 DENSE 10240 2560 4.00 6.16 0.03 10.78 1.75 0
148 DENSE 7680 2560 3.00 2.44 0.04 7.62 3.13 0
149 DENSE 2560 2560 1.00 6.82 0.03 7.96 1.17 2
150 DENSE 10240 2560 4.00 5.01 0.04 10.51 2.10 0
151 DENSE 10240 2560 4.00 6.62 0.03 10.50 1.59 0
154 DENSE 7680 2560 3.00 2.43 0.04 7.53 3.09 0
155 DENSE 2560 2560 1.00 6.92 0.01 6.68 0.97 3
156 DENSE 10240 2560 4.00 4.75 0.04 9.95 2.10 0
157 DENSE 10240 2560 4.00 7.44 0.02 10.91 1.47 0
160 DENSE 7680 2560 3.00 1.97 0.03 6.10 3.10 0
161 DENSE 2560 2560 1.00 7.83 0.03 5.99 0.77 3
162 DENSE 10240 2560 4.00 4.96 0.03 10.32 2.08 0
163 DENSE 10240 2560 4.00 6.95 0.02 9.72 1.40 0
166 DENSE 7680 2560 3.00 2.31 0.04 7.40 3.20 0
167 DENSE 2560 2560 1.00 7.53 0.03 5.68 0.75 3
168 DENSE 10240 2560 4.00 4.52 0.03 9.32 2.06 0
169 DENSE 10240 2560 4.00 7.29 0.02 10.30 1.41 0
172 DENSE 7680 2560 3.00 2.36 0.04 7.19 3.05 0
173 DENSE 2560 2560 1.00 10.74 0.03 7.98 0.74 4
174 DENSE 10240 2560 4.00 4.24 0.03 8.74 2.06 0
175 DENSE 10240 2560 4.00 6.73 0.01 10.18 1.51 0
178 DENSE 7680 2560 3.00 2.36 0.04 7.33 3.10 0
179 DENSE 2560 2560 1.00 8.17 0.03 6.49 0.80 3
180 DENSE 10240 2560 4.00 4.67 0.02 9.59 2.06 0
181 DENSE 10240 2560 4.00 5.70 0.01 9.54 1.67 0
184 DENSE 7680 2560 3.00 2.29 0.03 7.35 3.21 0
185 DENSE 2560 2560 1.00 5.07 0.04 7.00 1.38 3
186 DENSE 10240 2560 4.00 4.42 0.01 9.14 2.07 0
187 DENSE 10240 2560 4.00 4.80 0.01 10.88 2.27 0
190 DENSE 7680 2560 3.00 2.34 0.04 7.10 3.04 0
191 DENSE 2560 2560 1.00 4.56 0.04 6.12 1.34 3
192 DENSE 10240 2560 4.00 3.91 0.03 8.62 2.21 0
193 DENSE 10240 2560 4.00 3.76 0.02 8.52 2.27 0