Hannibal046 commited on
Commit
0bc24c3
·
verified ·
1 Parent(s): 060b67b

upload eagle3_gemma4_12b_ttt7 config

Browse files
Files changed (1) hide show
  1. config.json +70 -0
config.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma4Eagle3Model"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attention_k_eq_v": true,
8
+ "bos_token_id": 2,
9
+ "draft_num_hidden_layers": 1,
10
+ "dtype": "bfloat16",
11
+ "enable_moe_block": false,
12
+ "eos_token_id": 1,
13
+ "final_logit_softcapping": 30.0,
14
+ "global_head_dim": 512,
15
+ "head_dim": 256,
16
+ "hidden_activation": "gelu_pytorch_tanh",
17
+ "hidden_size": 3840,
18
+ "hidden_size_per_layer_input": 0,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 15360,
21
+ "layer_types": [
22
+ "full_attention"
23
+ ],
24
+ "max_position_embeddings": 262144,
25
+ "model_type": "gemma4_unified_text",
26
+ "moe_intermediate_size": null,
27
+ "num_attention_heads": 16,
28
+ "num_experts": null,
29
+ "num_global_key_value_heads": 1,
30
+ "num_hidden_layers": 1,
31
+ "num_key_value_heads": 8,
32
+ "num_kv_shared_layers": 0,
33
+ "num_target_layers": 48,
34
+ "pad_token_id": 0,
35
+ "rms_norm_eps": 1e-06,
36
+ "rope_parameters": {
37
+ "full_attention": {
38
+ "partial_rotary_factor": 0.25,
39
+ "rope_theta": 1000000.0,
40
+ "rope_type": "proportional"
41
+ },
42
+ "rope_theta": null,
43
+ "rope_type": "default",
44
+ "sliding_attention": {
45
+ "rope_theta": 10000.0,
46
+ "rope_type": "default"
47
+ }
48
+ },
49
+ "sliding_window": 1024,
50
+ "step_loss_decay": 0.8,
51
+ "target_layer_ids": [
52
+ 5,
53
+ 17,
54
+ 29,
55
+ 41,
56
+ 46
57
+ ],
58
+ "target_model_name_or_path": "google/gemma-4-12B-it",
59
+ "target_model_type": "gemma4_unified",
60
+ "target_text_model_type": "gemma4_unified_text",
61
+ "tie_word_embeddings": false,
62
+ "top_k_experts": null,
63
+ "transformers_version": "5.10.2",
64
+ "ttt_length": 7,
65
+ "use_bidirectional_attention": "vision",
66
+ "use_cache": true,
67
+ "use_double_wide_mlp": false,
68
+ "vocab_size": 262144,
69
+ "vocab_size_per_layer_input": 262144
70
+ }