Bc-AI commited on
Commit
e58ab4a
·
1 Parent(s): b061eed

Final model step=4240 val=2.2820

Browse files
Files changed (1) hide show
  1. final/config.json +45 -0
final/config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 50304,
3
+ "d_model": 4096,
4
+ "n_layers": 56,
5
+ "n_heads": 32,
6
+ "n_kv_heads": 8,
7
+ "max_len": 2048,
8
+ "rope_theta": 5000000.0,
9
+ "ffn_hidden": 11008,
10
+ "use_moe": false,
11
+ "moe_experts": 8,
12
+ "moe_top_k": 2,
13
+ "moe_shared": 1,
14
+ "moe_hidden": 1408,
15
+ "moe_every": 4,
16
+ "moe_bias_speed": 0.001,
17
+ "use_qk_norm": true,
18
+ "use_structural_bias": true,
19
+ "n_struct_rel": 4,
20
+ "use_fim": true,
21
+ "fim_rate": 0.5,
22
+ "fim_pre_id": 50269,
23
+ "fim_suf_id": 50270,
24
+ "fim_mid_id": 50271,
25
+ "grad_checkpointing": true,
26
+ "lr": 0.0002,
27
+ "min_lr": 2e-05,
28
+ "weight_decay": 0.1,
29
+ "beta1": 0.9,
30
+ "beta2": 0.95,
31
+ "grad_clip": 1.0,
32
+ "warmup_tokens": 500000000,
33
+ "target_total_tokens": 1000000000000,
34
+ "micro_batch": 4,
35
+ "grad_accum": 8,
36
+ "log_every": 10,
37
+ "ckpt_every_steps": 200,
38
+ "eval_every_steps": 200,
39
+ "seed": 42,
40
+ "max_session_hours": 11.5,
41
+ "final_step": 4240,
42
+ "final_epoch": 0,
43
+ "final_val_loss": 2.2819837987422944,
44
+ "total_tokens": 277872640
45
+ }