RichardErkhov commited on
Commit
e333d7b
1 Parent(s): 514725d

uploaded model

Browse files
Files changed (1) hide show
  1. config.json +51 -0
config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen1.5-MoE-A2.7B",
3
+ "architectures": [
4
+ "Qwen2MoeForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "decoder_sparse_step": 1,
9
+ "eos_token_id": 151643,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 5632,
14
+ "max_position_embeddings": 8192,
15
+ "max_window_layers": 21,
16
+ "model_type": "qwen2_moe",
17
+ "moe_intermediate_size": 1408,
18
+ "norm_topk_prob": false,
19
+ "num_attention_heads": 16,
20
+ "num_experts": 60,
21
+ "num_experts_per_tok": 4,
22
+ "num_hidden_layers": 24,
23
+ "num_key_value_heads": 16,
24
+ "output_router_logits": false,
25
+ "quantization_config": {
26
+ "_load_in_4bit": true,
27
+ "_load_in_8bit": false,
28
+ "bnb_4bit_compute_dtype": "float32",
29
+ "bnb_4bit_quant_storage": "uint8",
30
+ "bnb_4bit_quant_type": "fp4",
31
+ "bnb_4bit_use_double_quant": false,
32
+ "llm_int8_enable_fp32_cpu_offload": false,
33
+ "llm_int8_has_fp16_weight": false,
34
+ "llm_int8_skip_modules": null,
35
+ "llm_int8_threshold": 6.0,
36
+ "load_in_4bit": true,
37
+ "load_in_8bit": false,
38
+ "quant_method": "bitsandbytes"
39
+ },
40
+ "rms_norm_eps": 1e-06,
41
+ "rope_theta": 1000000.0,
42
+ "router_aux_loss_coef": 0.001,
43
+ "shared_expert_intermediate_size": 5632,
44
+ "sliding_window": 32768,
45
+ "tie_word_embeddings": false,
46
+ "torch_dtype": "float16",
47
+ "transformers_version": "4.40.2",
48
+ "use_cache": true,
49
+ "use_sliding_window": false,
50
+ "vocab_size": 151936
51
+ }