PerryCheng614 commited on
Commit
b6b7165
·
verified ·
1 Parent(s): 1ef5e0b

Upload folder using huggingface_hub

Browse files
Files changed (4) hide show
  1. config.json +62 -0
  2. model.safetensors +3 -0
  3. quant_log.csv +113 -0
  4. quantize_config.json +21 -0
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_attn_implementation_autoset": true,
3
+ "_name_or_path": "/home/azureuser/.cache/huggingface/hub/models--meta-llama--Llama-3.2-1B-Instruct/snapshots/9213176726f574b556790deb65791e0c5aa438b6",
4
+ "architectures": [
5
+ "LlamaForCausalLM"
6
+ ],
7
+ "attention_bias": false,
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 128000,
10
+ "eos_token_id": [
11
+ 128001,
12
+ 128008,
13
+ 128009
14
+ ],
15
+ "head_dim": 64,
16
+ "hidden_act": "silu",
17
+ "hidden_size": 2048,
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 8192,
20
+ "max_position_embeddings": 131072,
21
+ "mlp_bias": false,
22
+ "model_type": "llama",
23
+ "num_attention_heads": 32,
24
+ "num_hidden_layers": 16,
25
+ "num_key_value_heads": 8,
26
+ "pretraining_tp": 1,
27
+ "quantization_config": {
28
+ "bits": 4,
29
+ "checkpoint_format": "gptq",
30
+ "desc_act": false,
31
+ "dynamic": null,
32
+ "group_size": 32,
33
+ "lm_head": false,
34
+ "meta": {
35
+ "damp_auto_increment": 0.0025,
36
+ "damp_percent": 0.01,
37
+ "mse": 0.0,
38
+ "quantizer": [
39
+ "gptqmodel:1.5.1-dev"
40
+ ],
41
+ "static_groups": false,
42
+ "true_sequential": true,
43
+ "uri": "https://github.com/modelcloud/gptqmodel"
44
+ },
45
+ "quant_method": "gptq",
46
+ "sym": true
47
+ },
48
+ "rms_norm_eps": 1e-05,
49
+ "rope_scaling": {
50
+ "factor": 32.0,
51
+ "high_freq_factor": 4.0,
52
+ "low_freq_factor": 1.0,
53
+ "original_max_position_embeddings": 8192,
54
+ "rope_type": "llama3"
55
+ },
56
+ "rope_theta": 500000.0,
57
+ "tie_word_embeddings": true,
58
+ "torch_dtype": "bfloat16",
59
+ "transformers_version": "4.47.1",
60
+ "use_cache": true,
61
+ "vocab_size": 128256
62
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5435ed8cef43f02129c506f8b2e1dd08538eca737ad419c3ddfba23b81724499
3
+ size 1614733472
quant_log.csv ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ layer,module,loss,damp,time
2
+ 0,self_attn.k_proj,0.09601,0.01000,1.079
3
+ 0,self_attn.v_proj,0.00299,0.01000,0.841
4
+ 0,self_attn.q_proj,0.21113,0.01000,0.727
5
+ 0,self_attn.o_proj,0.00017,0.01000,0.727
6
+ 0,mlp.up_proj,0.14873,0.01000,0.763
7
+ 0,mlp.gate_proj,0.18908,0.01000,0.904
8
+ 0,mlp.down_proj,0.00190,0.01000,3.295
9
+ 1,self_attn.k_proj,0.16585,0.01000,0.711
10
+ 1,self_attn.v_proj,0.00992,0.01000,0.749
11
+ 1,self_attn.q_proj,0.32035,0.01000,0.690
12
+ 1,self_attn.o_proj,0.00039,0.01000,0.753
13
+ 1,mlp.up_proj,0.22062,0.01000,0.756
14
+ 1,mlp.gate_proj,0.30241,0.01000,0.732
15
+ 1,mlp.down_proj,0.16209,0.01000,3.030
16
+ 2,self_attn.k_proj,0.31228,0.01000,0.713
17
+ 2,self_attn.v_proj,0.02535,0.01000,0.677
18
+ 2,self_attn.q_proj,0.63971,0.01000,0.711
19
+ 2,self_attn.o_proj,0.00066,0.01000,0.730
20
+ 2,mlp.up_proj,0.28284,0.01000,0.746
21
+ 2,mlp.gate_proj,0.44825,0.01000,0.692
22
+ 2,mlp.down_proj,0.00335,0.01000,3.236
23
+ 3,self_attn.k_proj,0.19264,0.01000,0.758
24
+ 3,self_attn.v_proj,0.02762,0.01000,0.695
25
+ 3,self_attn.q_proj,0.42406,0.01000,0.738
26
+ 3,self_attn.o_proj,0.00111,0.01000,0.765
27
+ 3,mlp.up_proj,0.33513,0.01000,0.785
28
+ 3,mlp.gate_proj,0.66167,0.01000,0.750
29
+ 3,mlp.down_proj,0.00468,0.01000,3.254
30
+ 4,self_attn.k_proj,0.21151,0.01000,0.774
31
+ 4,self_attn.v_proj,0.02663,0.01000,0.746
32
+ 4,self_attn.q_proj,0.43425,0.01000,0.743
33
+ 4,self_attn.o_proj,0.00185,0.01000,0.801
34
+ 4,mlp.up_proj,0.33790,0.01000,0.835
35
+ 4,mlp.gate_proj,0.72013,0.01000,0.745
36
+ 4,mlp.down_proj,0.00553,0.01000,3.029
37
+ 5,self_attn.k_proj,0.30856,0.01000,0.716
38
+ 5,self_attn.v_proj,0.02561,0.01000,0.681
39
+ 5,self_attn.q_proj,0.57913,0.01000,0.702
40
+ 5,self_attn.o_proj,0.00256,0.01000,0.754
41
+ 5,mlp.up_proj,0.36168,0.01000,0.806
42
+ 5,mlp.gate_proj,0.66184,0.01000,0.734
43
+ 5,mlp.down_proj,0.00675,0.01000,3.046
44
+ 6,self_attn.k_proj,0.25328,0.01000,0.736
45
+ 6,self_attn.v_proj,0.02967,0.01000,0.681
46
+ 6,self_attn.q_proj,0.41055,0.01000,0.721
47
+ 6,self_attn.o_proj,0.00379,0.01000,0.724
48
+ 6,mlp.up_proj,0.37050,0.01000,0.751
49
+ 6,mlp.gate_proj,0.66534,0.01000,0.718
50
+ 6,mlp.down_proj,0.00709,0.01000,3.041
51
+ 7,self_attn.k_proj,0.25026,0.01000,0.748
52
+ 7,self_attn.v_proj,0.03261,0.01000,0.678
53
+ 7,self_attn.q_proj,0.47341,0.01000,0.708
54
+ 7,self_attn.o_proj,0.00349,0.01000,0.727
55
+ 7,mlp.up_proj,0.40085,0.01000,0.776
56
+ 7,mlp.gate_proj,0.65237,0.01000,0.749
57
+ 7,mlp.down_proj,0.00813,0.01000,3.014
58
+ 8,self_attn.k_proj,0.32256,0.01000,0.725
59
+ 8,self_attn.v_proj,0.03293,0.01000,0.690
60
+ 8,self_attn.q_proj,0.53772,0.01000,0.689
61
+ 8,self_attn.o_proj,0.00522,0.01000,0.755
62
+ 8,mlp.up_proj,0.47201,0.01000,0.770
63
+ 8,mlp.gate_proj,0.75260,0.01000,0.757
64
+ 8,mlp.down_proj,0.01079,0.01000,3.152
65
+ 9,self_attn.k_proj,0.24430,0.01000,0.727
66
+ 9,self_attn.v_proj,0.03725,0.01000,0.680
67
+ 9,self_attn.q_proj,0.60562,0.01000,0.724
68
+ 9,self_attn.o_proj,0.00523,0.01000,0.754
69
+ 9,mlp.up_proj,0.49043,0.01000,0.764
70
+ 9,mlp.gate_proj,0.80227,0.01000,0.715
71
+ 9,mlp.down_proj,0.01264,0.01000,3.103
72
+ 10,self_attn.k_proj,0.30944,0.01000,0.742
73
+ 10,self_attn.v_proj,0.04692,0.01000,0.668
74
+ 10,self_attn.q_proj,0.68625,0.01000,0.704
75
+ 10,self_attn.o_proj,0.00472,0.01000,0.736
76
+ 10,mlp.up_proj,0.57742,0.01000,0.771
77
+ 10,mlp.gate_proj,0.90810,0.01000,0.740
78
+ 10,mlp.down_proj,0.01584,0.01000,3.073
79
+ 11,self_attn.k_proj,0.36443,0.01000,0.712
80
+ 11,self_attn.v_proj,0.04592,0.01000,0.695
81
+ 11,self_attn.q_proj,0.66843,0.01000,0.697
82
+ 11,self_attn.o_proj,0.00372,0.01000,0.766
83
+ 11,mlp.up_proj,0.63095,0.01000,0.803
84
+ 11,mlp.gate_proj,0.97327,0.01000,0.726
85
+ 11,mlp.down_proj,0.01862,0.01000,3.065
86
+ 12,self_attn.k_proj,0.33120,0.01000,0.731
87
+ 12,self_attn.v_proj,0.04817,0.01000,0.694
88
+ 12,self_attn.q_proj,0.64928,0.01000,0.743
89
+ 12,self_attn.o_proj,0.00386,0.01000,0.726
90
+ 12,mlp.up_proj,0.66255,0.01000,0.791
91
+ 12,mlp.gate_proj,0.97375,0.01000,0.759
92
+ 12,mlp.down_proj,0.02400,0.01000,3.136
93
+ 13,self_attn.k_proj,0.36034,0.01000,0.763
94
+ 13,self_attn.v_proj,0.08636,0.01000,0.717
95
+ 13,self_attn.q_proj,0.78605,0.01000,0.696
96
+ 13,self_attn.o_proj,0.00579,0.01000,0.764
97
+ 13,mlp.up_proj,0.79367,0.01000,0.797
98
+ 13,mlp.gate_proj,1.07306,0.01000,0.757
99
+ 13,mlp.down_proj,0.03768,0.01000,3.024
100
+ 14,self_attn.k_proj,0.35830,0.01000,0.714
101
+ 14,self_attn.v_proj,0.16440,0.01000,0.697
102
+ 14,self_attn.q_proj,0.75050,0.01000,0.727
103
+ 14,self_attn.o_proj,0.01295,0.01000,0.740
104
+ 14,mlp.up_proj,0.92130,0.01000,0.780
105
+ 14,mlp.gate_proj,1.33264,0.01000,0.727
106
+ 14,mlp.down_proj,0.05907,0.01000,3.084
107
+ 15,self_attn.k_proj,0.35875,0.01000,0.726
108
+ 15,self_attn.v_proj,0.17096,0.01000,0.703
109
+ 15,self_attn.q_proj,0.67191,0.01000,0.680
110
+ 15,self_attn.o_proj,0.03647,0.01000,0.743
111
+ 15,mlp.up_proj,1.12717,0.01000,0.775
112
+ 15,mlp.gate_proj,1.49428,0.01000,0.742
113
+ 15,mlp.down_proj,0.22401,0.01000,2.976
quantize_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bits": 4,
3
+ "dynamic": null,
4
+ "group_size": 32,
5
+ "desc_act": false,
6
+ "sym": true,
7
+ "lm_head": false,
8
+ "quant_method": "gptq",
9
+ "checkpoint_format": "gptq",
10
+ "meta": {
11
+ "quantizer": [
12
+ "gptqmodel:1.5.1-dev"
13
+ ],
14
+ "uri": "https://github.com/modelcloud/gptqmodel",
15
+ "damp_percent": 0.01,
16
+ "damp_auto_increment": 0.0025,
17
+ "static_groups": false,
18
+ "true_sequential": true,
19
+ "mse": 0.0
20
+ }
21
+ }