Hugging Face's logo

AbstractPhil
/
aleph-splat-0

English
attention
mixture-of-experts
softmax-free
linear-attention
geometric-deep-learning
research-log
aleph
Model card Files
xet
 Community
AbstractPhil commited on
Commit
d8a37b7
·
verified ·
1 Parent(s): 0debcab

V1b cell R64_K0_s0_D16

Browse files
Files changed (1) hide show
  1. battery/v1b_recall_army/R64_K0_s0_D16.json +64 -64
battery/v1b_recall_army/R64_K0_s0_D16.json CHANGED
@@ -4,168 +4,168 @@
4
  "seed": 0,
5
  "steps": 8000,
6
  "arm": "sdpa_tri",
7
- "final_acc": 0.937,
8
- "final_bpb": 0.5567,
9
  "curve": [
10
  {
11
  "step": 250,
12
- "ans_acc": 0.003,
13
- "ans_bpb": 8.0183
14
  },
15
  {
16
  "step": 500,
17
- "ans_acc": 0.0041,
18
- "ans_bpb": 8.0882
19
  },
20
  {
21
  "step": 750,
22
- "ans_acc": 0.0135,
23
- "ans_bpb": 8.025
24
  },
25
  {
26
  "step": 1000,
27
  "ans_acc": 0.0316,
28
- "ans_bpb": 7.9055
29
  },
30
  {
31
  "step": 1250,
32
- "ans_acc": 0.0648,
33
- "ans_bpb": 7.5767
34
  },
35
  {
36
  "step": 1500,
37
- "ans_acc": 0.4381,
38
- "ans_bpb": 4.6237
39
  },
40
  {
41
  "step": 1750,
42
- "ans_acc": 0.8976,
43
- "ans_bpb": 1.0868
44
  },
45
  {
46
  "step": 2000,
47
- "ans_acc": 0.9222,
48
- "ans_bpb": 0.7677
49
  },
50
  {
51
  "step": 2250,
52
- "ans_acc": 0.9373,
53
- "ans_bpb": 0.6132
54
  },
55
  {
56
  "step": 2500,
57
- "ans_acc": 0.9418,
58
- "ans_bpb": 0.552
59
  },
60
  {
61
  "step": 2750,
62
- "ans_acc": 0.9353,
63
- "ans_bpb": 0.5836
64
  },
65
  {
66
  "step": 3000,
67
- "ans_acc": 0.9355,
68
- "ans_bpb": 0.5893
69
  },
70
  {
71
  "step": 3250,
72
- "ans_acc": 0.9334,
73
- "ans_bpb": 0.603
74
  },
75
  {
76
  "step": 3500,
77
- "ans_acc": 0.9409,
78
- "ans_bpb": 0.5369
79
  },
80
  {
81
  "step": 3750,
82
- "ans_acc": 0.9368,
83
- "ans_bpb": 0.5655
84
  },
85
  {
86
  "step": 4000,
87
- "ans_acc": 0.9417,
88
- "ans_bpb": 0.5332
89
  },
90
  {
91
  "step": 4250,
92
- "ans_acc": 0.9399,
93
- "ans_bpb": 0.5309
94
  },
95
  {
96
  "step": 4500,
97
- "ans_acc": 0.9362,
98
- "ans_bpb": 0.5565
99
  },
100
  {
101
  "step": 4750,
102
- "ans_acc": 0.9363,
103
- "ans_bpb": 0.5617
104
  },
105
  {
106
  "step": 5000,
107
- "ans_acc": 0.9378,
108
- "ans_bpb": 0.5326
109
  },
110
  {
111
  "step": 5250,
112
- "ans_acc": 0.9364,
113
- "ans_bpb": 0.5511
114
  },
115
  {
116
  "step": 5500,
117
  "ans_acc": 0.9379,
118
- "ans_bpb": 0.5453
119
  },
120
  {
121
  "step": 5750,
122
- "ans_acc": 0.9396,
123
- "ans_bpb": 0.5282
124
  },
125
  {
126
  "step": 6000,
127
- "ans_acc": 0.939,
128
- "ans_bpb": 0.5411
129
  },
130
  {
131
  "step": 6250,
132
- "ans_acc": 0.9419,
133
- "ans_bpb": 0.5186
134
  },
135
  {
136
  "step": 6500,
137
- "ans_acc": 0.9401,
138
- "ans_bpb": 0.5325
139
  },
140
  {
141
  "step": 6750,
142
- "ans_acc": 0.9377,
143
- "ans_bpb": 0.5623
144
  },
145
  {
146
  "step": 7000,
147
- "ans_acc": 0.9414,
148
- "ans_bpb": 0.5132
149
  },
150
  {
151
  "step": 7250,
152
- "ans_acc": 0.9393,
153
- "ans_bpb": 0.5558
154
  },
155
  {
156
  "step": 7500,
157
- "ans_acc": 0.9391,
158
- "ans_bpb": 0.5537
159
  },
160
  {
161
  "step": 7750,
162
- "ans_acc": 0.9393,
163
- "ans_bpb": 0.5527
164
  },
165
  {
166
  "step": 8000,
167
- "ans_acc": 0.9373,
168
- "ans_bpb": 0.5535
169
  }
170
  ],
171
  "census": null,
 
4
  "seed": 0,
5
  "steps": 8000,
6
  "arm": "sdpa_tri",
7
+ "final_acc": 0.9364,
8
+ "final_bpb": 0.5657,
9
  "curve": [
10
  {
11
  "step": 250,
12
+ "ans_acc": 0.0038,
13
+ "ans_bpb": 8.0251
14
  },
15
  {
16
  "step": 500,
17
+ "ans_acc": 0.0071,
18
+ "ans_bpb": 8.0984
19
  },
20
  {
21
  "step": 750,
22
+ "ans_acc": 0.0186,
23
+ "ans_bpb": 8.01
24
  },
25
  {
26
  "step": 1000,
27
  "ans_acc": 0.0316,
28
+ "ans_bpb": 7.8478
29
  },
30
  {
31
  "step": 1250,
32
+ "ans_acc": 0.0909,
33
+ "ans_bpb": 7.2213
34
  },
35
  {
36
  "step": 1500,
37
+ "ans_acc": 0.8377,
38
+ "ans_bpb": 1.8879
39
  },
40
  {
41
  "step": 1750,
42
+ "ans_acc": 0.9153,
43
+ "ans_bpb": 0.8723
44
  },
45
  {
46
  "step": 2000,
47
+ "ans_acc": 0.9252,
48
+ "ans_bpb": 0.7051
49
  },
50
  {
51
  "step": 2250,
52
+ "ans_acc": 0.9399,
53
+ "ans_bpb": 0.5817
54
  },
55
  {
56
  "step": 2500,
57
+ "ans_acc": 0.942,
58
+ "ans_bpb": 0.5455
59
  },
60
  {
61
  "step": 2750,
62
+ "ans_acc": 0.9362,
63
+ "ans_bpb": 0.5837
64
  },
65
  {
66
  "step": 3000,
67
+ "ans_acc": 0.9338,
68
+ "ans_bpb": 0.5739
69
  },
70
  {
71
  "step": 3250,
72
+ "ans_acc": 0.9313,
73
+ "ans_bpb": 0.5975
74
  },
75
  {
76
  "step": 3500,
77
+ "ans_acc": 0.9413,
78
+ "ans_bpb": 0.5483
79
  },
80
  {
81
  "step": 3750,
82
+ "ans_acc": 0.9394,
83
+ "ans_bpb": 0.5354
84
  },
85
  {
86
  "step": 4000,
87
+ "ans_acc": 0.9392,
88
+ "ans_bpb": 0.5557
89
  },
90
  {
91
  "step": 4250,
92
+ "ans_acc": 0.9414,
93
+ "ans_bpb": 0.524
94
  },
95
  {
96
  "step": 4500,
97
+ "ans_acc": 0.9377,
98
+ "ans_bpb": 0.5612
99
  },
100
  {
101
  "step": 4750,
102
+ "ans_acc": 0.9337,
103
+ "ans_bpb": 0.5816
104
  },
105
  {
106
  "step": 5000,
107
+ "ans_acc": 0.9408,
108
+ "ans_bpb": 0.5195
109
  },
110
  {
111
  "step": 5250,
112
+ "ans_acc": 0.9373,
113
+ "ans_bpb": 0.5529
114
  },
115
  {
116
  "step": 5500,
117
  "ans_acc": 0.9379,
118
+ "ans_bpb": 0.5593
119
  },
120
  {
121
  "step": 5750,
122
+ "ans_acc": 0.9405,
123
+ "ans_bpb": 0.5333
124
  },
125
  {
126
  "step": 6000,
127
+ "ans_acc": 0.9412,
128
+ "ans_bpb": 0.5291
129
  },
130
  {
131
  "step": 6250,
132
+ "ans_acc": 0.9411,
133
+ "ans_bpb": 0.5479
134
  },
135
  {
136
  "step": 6500,
137
+ "ans_acc": 0.942,
138
+ "ans_bpb": 0.5115
139
  },
140
  {
141
  "step": 6750,
142
+ "ans_acc": 0.9374,
143
+ "ans_bpb": 0.5528
144
  },
145
  {
146
  "step": 7000,
147
+ "ans_acc": 0.941,
148
+ "ans_bpb": 0.5133
149
  },
150
  {
151
  "step": 7250,
152
+ "ans_acc": 0.9402,
153
+ "ans_bpb": 0.5279
154
  },
155
  {
156
  "step": 7500,
157
+ "ans_acc": 0.9383,
158
+ "ans_bpb": 0.5543
159
  },
160
  {
161
  "step": 7750,
162
+ "ans_acc": 0.9387,
163
+ "ans_bpb": 0.5627
164
  },
165
  {
166
  "step": 8000,
167
+ "ans_acc": 0.9353,
168
+ "ans_bpb": 0.5808
169
  }
170
  ],
171
  "census": null,