danielhanchen commited on
Commit
1dca677
·
verified ·
1 Parent(s): 4af3c5c

Mirror worker 0

Browse files
config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DeepseekV4ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "eos_token_id": 1,
9
+ "hc_eps": 1e-06,
10
+ "hc_mult": 4,
11
+ "hc_sinkhorn_iters": 20,
12
+ "head_dim": 512,
13
+ "hidden_act": "silu",
14
+ "hidden_size": 4096,
15
+ "index_head_dim": 128,
16
+ "index_n_heads": 64,
17
+ "index_topk": 512,
18
+ "initializer_range": 0.02,
19
+ "max_position_embeddings": 1048576,
20
+ "model_type": "deepseek_v4",
21
+ "moe_intermediate_size": 2048,
22
+ "n_routed_experts": 256,
23
+ "n_shared_experts": 1,
24
+ "norm_topk_prob": true,
25
+ "num_attention_heads": 64,
26
+ "num_experts_per_tok": 6,
27
+ "num_hidden_layers": 43,
28
+ "num_hash_layers": 3,
29
+ "num_key_value_heads": 1,
30
+ "num_nextn_predict_layers": 1,
31
+ "o_groups": 8,
32
+ "o_lora_rank": 1024,
33
+ "q_lora_rank": 1024,
34
+ "qk_rope_head_dim": 64,
35
+ "quantization_config": {
36
+ "activation_scheme": "dynamic",
37
+ "fmt": "e4m3",
38
+ "quant_method": "fp8",
39
+ "scale_fmt": "ue8m0",
40
+ "weight_block_size": [
41
+ 128,
42
+ 128
43
+ ]
44
+ },
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_scaling": {
47
+ "beta_fast": 32,
48
+ "beta_slow": 1,
49
+ "factor": 16,
50
+ "original_max_position_embeddings": 65536,
51
+ "type": "yarn"
52
+ },
53
+ "rope_theta": 10000,
54
+ "routed_scaling_factor": 1.5,
55
+ "scoring_func": "sqrtsoftplus",
56
+ "sliding_window": 128,
57
+ "swiglu_limit": 10.0,
58
+ "tie_word_embeddings": false,
59
+ "topk_method": "noaux_tc",
60
+ "torch_dtype": "bfloat16",
61
+ "transformers_version": "4.57.1",
62
+ "use_cache": true,
63
+ "vocab_size": 129280,
64
+ "compress_rope_theta": 160000,
65
+ "compress_ratios": [0, 0, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 128, 4, 0]
66
+ }
model-00008-of-00046.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb62f0c8d724023c491bf4b443729458a0e9d678d64ff14bda1b9b24674719e8
3
+ size 6611517496
model-00016-of-00046.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52c3e8888b1dc7d6765f1252c0cea840cf136543da9579d86463a13af0b187f8
3
+ size 6611519072
model-00024-of-00046.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27af01cebf667dbb727534d9e2e82cb2978e8e6dba4ad905e90d32b74fc17a98
3
+ size 6611519072
model-00032-of-00046.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7495294a6d0d8ecc27fafd604244b2136b1e10cfe2a762e7b1110e0aa144b353
3
+ size 6611519072
model-00040-of-00046.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:673aa347cc9af732a6d6b13cc84dc586ef0396d8b8bbf6f5e3ce87fafce14c3e
3
+ size 6611519072
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff