gmonair commited on
Commit
bdb159f
·
verified ·
1 Parent(s): 71451b2

AutoGPTQ model for Qwen/Qwen2.5-Math-1.5B-Instruct: 8bits, gr128, desc_act=True

Browse files
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2.5-Math-1.5B-Instruct",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 8960,
13
+ "max_position_embeddings": 4096,
14
+ "max_window_layers": 21,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 2,
19
+ "quantization_config": {
20
+ "bits": 8,
21
+ "damp_percent": 0.01,
22
+ "desc_act": true,
23
+ "group_size": 128,
24
+ "is_marlin_format": false,
25
+ "model_file_base_name": null,
26
+ "model_name_or_path": null,
27
+ "quant_method": "gptq",
28
+ "static_groups": false,
29
+ "sym": true,
30
+ "true_sequential": true
31
+ },
32
+ "rms_norm_eps": 1e-06,
33
+ "rope_theta": 10000.0,
34
+ "sliding_window": 4096,
35
+ "tie_word_embeddings": true,
36
+ "torch_dtype": "float16",
37
+ "transformers_version": "4.38.0",
38
+ "use_cache": true,
39
+ "use_sliding_window": false,
40
+ "vocab_size": 151936
41
+ }
gptq_model-8bit-128g.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9be32c3bf9b94ee2535208b420e8820d383b98ce9e673eb269de78dcb592134
3
+ size 2278014312
quantize_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bits": 8,
3
+ "group_size": 128,
4
+ "damp_percent": 0.01,
5
+ "desc_act": true,
6
+ "static_groups": false,
7
+ "sym": true,
8
+ "true_sequential": true,
9
+ "model_name_or_path": null,
10
+ "model_file_base_name": null,
11
+ "is_marlin_format": false,
12
+ "quant_method": "gptq"
13
+ }