AbstractPhil commited on
Commit
e00a51a
·
verified ·
1 Parent(s): beecc35

bed: alhena-both_s1_L24000_ctx4096_frdyn_cap_mode_sz32-128-256-512_r4_n30_o0_askany_gag50_ahdigit1_scalerep weights

Browse files
bed/weights/alhena-both_s1_L24000_ctx4096_frdyn_cap_mode_sz32-128-256-512_r4_n30_o0_askany_gag50_ahdigit1_scalerep/addr_head.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b097fe6416f90ab237b3316371024a7ec9b886887e86b127194d8b443376478c
3
+ size 16584
bed/weights/alhena-both_s1_L24000_ctx4096_frdyn_cap_mode_sz32-128-256-512_r4_n30_o0_askany_gag50_ahdigit1_scalerep/config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "bed-alhena-both",
3
+ "d_model": 256,
4
+ "n_layers": 6,
5
+ "n_heads": 8,
6
+ "context": 4096,
7
+ "vocab_size": 256,
8
+ "tokenizer": "byte-trigram",
9
+ "hub_layers": [],
10
+ "hub_K": 32,
11
+ "hub_D": 16,
12
+ "tau": 0.1,
13
+ "bank_experts": 7,
14
+ "bank_ff": 256,
15
+ "head_K": 64,
16
+ "head_D": 16,
17
+ "head_mode": "",
18
+ "head_dial_band": 4.0,
19
+ "head_dial_rho": 0.015625,
20
+ "gate_init": -3.0,
21
+ "tie_embeddings": false,
22
+ "hub_chunk": 128,
23
+ "hub_const": 1,
24
+ "hub_ckpt": 0,
25
+ "fusion": null,
26
+ "qk_norm": "",
27
+ "attn_fp32": false,
28
+ "attn_kernel": "sdpa",
29
+ "lookup_arm_sites": [],
30
+ "lookup_arm_depth": 1,
31
+ "attn_rotary": true,
32
+ "hub_rotary": false,
33
+ "rotary_base": 10000.0,
34
+ "hub_rotary_frac": 1.0,
35
+ "hub_dual": false,
36
+ "pos_table": true,
37
+ "bank_byte_table": "add",
38
+ "bank_last_fix": "",
39
+ "fix_rho": 0.0625,
40
+ "fix_warm": 200,
41
+ "fix_clamp": [
42
+ 0.25,
43
+ 4.0
44
+ ],
45
+ "canon": ""
46
+ }
bed/weights/alhena-both_s1_L24000_ctx4096_frdyn_cap_mode_sz32-128-256-512_r4_n30_o0_askany_gag50_ahdigit1_scalerep/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3f07f54d18686ef79a5794f2abfbb0c8cafbf6a3e34610bbeff6245d4280a44
3
+ size 36916028
bed/weights/alhena-both_s1_L24000_ctx4096_frdyn_cap_mode_sz32-128-256-512_r4_n30_o0_askany_gag50_ahdigit1_scalerep/scale.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "grid": "ag",
3
+ "grid_p": 0.5,
4
+ "n_ag": 4,
5
+ "rotary": {
6
+ "form": "anchored",
7
+ "pairs": 16,
8
+ "rel_pairs": 6,
9
+ "content_pairs": 2,
10
+ "anchor_freqs": 4,
11
+ "levels": 2,
12
+ "ntk_scale": 4.0,
13
+ "base": 10000.0,
14
+ "null": 8,
15
+ "w": [
16
+ 1.0,
17
+ 0.16327561755730105,
18
+ 0.026658927288718035,
19
+ 0.004352752816480621,
20
+ 0.000710698404185155,
21
+ 0.00011603972084031943,
22
+ 0.0,
23
+ 0.0,
24
+ 0.6981317007977318,
25
+ 1.3962634015954636,
26
+ 2.0943951023931953,
27
+ 2.792526803190927,
28
+ 0.6981317007977318,
29
+ 1.3962634015954636,
30
+ 2.0943951023931953,
31
+ 2.792526803190927
32
+ ],
33
+ "kind": [
34
+ 0,
35
+ 0,
36
+ 0,
37
+ 0,
38
+ 0,
39
+ 0,
40
+ 1,
41
+ 1,
42
+ 2,
43
+ 2,
44
+ 2,
45
+ 2,
46
+ 3,
47
+ 3,
48
+ 3,
49
+ 3
50
+ ]
51
+ },
52
+ "head": "digit",
53
+ "head_lam": 1.0,
54
+ "layout": {
55
+ "slot": 128,
56
+ "comps": 24,
57
+ "radix": 8,
58
+ "levels": 2
59
+ }
60
+ }