AdamF92 commited on
Commit
d987458
·
verified ·
1 Parent(s): d7a262e

Epoch 0 - Val loss 1.4337

Browse files
Files changed (2) hide show
  1. config.json +23 -0
  2. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "att_dropout": 0.0,
3
+ "att_heads": 16,
4
+ "embed_dim": 1024,
5
+ "head_dim": 128,
6
+ "interlayer_att_dropout": 0.0,
7
+ "interlayer_att_heads": 16,
8
+ "interlayer_kv_heads": 8,
9
+ "interlayer_qk_norm": true,
10
+ "kv_heads": 8,
11
+ "kv_input_dim": 1024,
12
+ "num_groups": 2,
13
+ "num_layers": 14,
14
+ "residual_gate_init": 3.0,
15
+ "residual_gate_type": "elementwise",
16
+ "rope_base": 1000000,
17
+ "seq_len": 8192,
18
+ "stm_size": 4096,
19
+ "use_flash_attention": false,
20
+ "use_gqa": true,
21
+ "use_interlayer_gqa": true,
22
+ "use_qk_norm": true
23
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85b23367f6004fd1bf8bbbcfb695ff490b5dbba38ca5a45bba7bf5299c077988
3
+ size 471288288