fitleech commited on
Commit
48c8a00
·
verified ·
1 Parent(s): 3ac2edd

Self-test hyperparams: bf16 dtype, use_cache=false for 16GB TDX

Browse files
Files changed (1) hide show
  1. config.json +73 -72
config.json CHANGED
@@ -1,73 +1,74 @@
1
  {
2
- "model_type": "sn38-nanochrono",
3
- "architectures": [
4
- "NanochronoForCausalLM"
5
- ],
6
- "attention_multiplier": 1.2,
7
- "aux_gate_dim": 12,
8
- "aux_layers": [
9
- 1,
10
- 3,
11
- 5,
12
- 7,
13
- 9,
14
- 11,
15
- 13,
16
- 15,
17
- 17,
18
- 19,
19
- 21,
20
- 23,
21
- 25,
22
- 27
23
- ],
24
- "bos_token_id": 32759,
25
- "dtype": "float32",
26
- "eos_token_id": 32763,
27
- "final_logit_softcapping": 15.0,
28
- "hidden_size": 1792,
29
- "intermediate_size": 7168,
30
- "layer_types": [
31
- "sliding_attention",
32
- "sliding_attention",
33
- "sliding_attention",
34
- "full_attention",
35
- "sliding_attention",
36
- "sliding_attention",
37
- "sliding_attention",
38
- "full_attention",
39
- "sliding_attention",
40
- "sliding_attention",
41
- "sliding_attention",
42
- "full_attention",
43
- "sliding_attention",
44
- "sliding_attention",
45
- "sliding_attention",
46
- "full_attention",
47
- "sliding_attention",
48
- "sliding_attention",
49
- "sliding_attention",
50
- "full_attention",
51
- "sliding_attention",
52
- "sliding_attention",
53
- "sliding_attention",
54
- "full_attention",
55
- "sliding_attention",
56
- "sliding_attention",
57
- "sliding_attention",
58
- "full_attention"
59
- ],
60
- "max_position_embeddings": 2048,
61
- "mix_gate_dim": 24,
62
- "num_attention_heads": 14,
63
- "num_hidden_layers": 28,
64
- "num_key_value_heads": 14,
65
- "pad_token_id": 32763,
66
- "rope_theta": 100000.0,
67
- "sliding_window": 512,
68
- "tap_layer": 14,
69
- "tie_word_embeddings": false,
70
- "transformers_version": "5.14.1",
71
- "use_cache": true,
72
- "vocab_size": 32768
73
- }
 
 
1
  {
2
+ "model_type": "sn38-nanochrono",
3
+ "architectures": [
4
+ "NanochronoForCausalLM"
5
+ ],
6
+ "attention_multiplier": 1.2,
7
+ "aux_gate_dim": 12,
8
+ "aux_layers": [
9
+ 1,
10
+ 3,
11
+ 5,
12
+ 7,
13
+ 9,
14
+ 11,
15
+ 13,
16
+ 15,
17
+ 17,
18
+ 19,
19
+ 21,
20
+ 23,
21
+ 25,
22
+ 27
23
+ ],
24
+ "bos_token_id": 32759,
25
+ "dtype": "bfloat16",
26
+ "eos_token_id": 32763,
27
+ "final_logit_softcapping": 15.0,
28
+ "hidden_size": 1792,
29
+ "intermediate_size": 7168,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "full_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "full_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "full_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "sliding_attention",
50
+ "full_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "full_attention"
59
+ ],
60
+ "max_position_embeddings": 2048,
61
+ "mix_gate_dim": 24,
62
+ "num_attention_heads": 14,
63
+ "num_hidden_layers": 28,
64
+ "num_key_value_heads": 14,
65
+ "pad_token_id": 32763,
66
+ "rope_theta": 100000.0,
67
+ "sliding_window": 512,
68
+ "tap_layer": 14,
69
+ "tie_word_embeddings": false,
70
+ "transformers_version": "5.14.1",
71
+ "use_cache": false,
72
+ "vocab_size": 32768,
73
+ "torch_dtype": "bfloat16"
74
+ }