Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- config.json +50 -0
- model.safetensors +3 -0
- quant_log.csv +281 -0
- quantize_config.json +21 -0
- special_tokens_map.json +24 -0
- tokenizer.json +3 -0
- tokenizer_config.json +0 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_attn_implementation_autoset": true,
|
3 |
+
"_name_or_path": "/fsx/ubuntu/.cache/huggingface/hub/models--cyberagent--Mistral-Nemo-Japanese-Instruct-2408/snapshots/8591f78522a5d651209fc9f354c848508db7a3eb",
|
4 |
+
"architectures": [
|
5 |
+
"MistralForCausalLM"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 1,
|
9 |
+
"eos_token_id": 131072,
|
10 |
+
"head_dim": 128,
|
11 |
+
"hidden_act": "silu",
|
12 |
+
"hidden_size": 5120,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 14336,
|
15 |
+
"max_position_embeddings": 1024000,
|
16 |
+
"model_type": "mistral",
|
17 |
+
"num_attention_heads": 32,
|
18 |
+
"num_hidden_layers": 40,
|
19 |
+
"num_key_value_heads": 8,
|
20 |
+
"pad_token_id": 10,
|
21 |
+
"quantization_config": {
|
22 |
+
"bits": 4,
|
23 |
+
"checkpoint_format": "gptq",
|
24 |
+
"desc_act": true,
|
25 |
+
"dynamic": null,
|
26 |
+
"group_size": 32,
|
27 |
+
"lm_head": false,
|
28 |
+
"meta": {
|
29 |
+
"damp_auto_increment": 0.0025,
|
30 |
+
"damp_percent": 0.01,
|
31 |
+
"mse": 0.0,
|
32 |
+
"quantizer": [
|
33 |
+
"gptqmodel:1.7.0"
|
34 |
+
],
|
35 |
+
"static_groups": false,
|
36 |
+
"true_sequential": true,
|
37 |
+
"uri": "https://github.com/modelcloud/gptqmodel"
|
38 |
+
},
|
39 |
+
"quant_method": "gptq",
|
40 |
+
"sym": true
|
41 |
+
},
|
42 |
+
"rms_norm_eps": 1e-05,
|
43 |
+
"rope_theta": 1000000.0,
|
44 |
+
"sliding_window": null,
|
45 |
+
"tie_word_embeddings": false,
|
46 |
+
"torch_dtype": "bfloat16",
|
47 |
+
"transformers_version": "4.48.0",
|
48 |
+
"use_cache": false,
|
49 |
+
"vocab_size": 131074
|
50 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2b1d20803f0cf850a9f43a30ddba3dc8940d8a8283ae7b5a4b66d1b57ad0e4b
|
3 |
+
size 8996969392
|
quant_log.csv
ADDED
@@ -0,0 +1,281 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
layer,module,loss,damp,time
|
2 |
+
0,self_attn.k_proj,0.35479,0.01000,2.349
|
3 |
+
0,self_attn.v_proj,0.00268,0.01000,1.606
|
4 |
+
0,self_attn.q_proj,0.82359,0.01000,1.709
|
5 |
+
0,self_attn.o_proj,0.00026,0.01000,1.405
|
6 |
+
0,mlp.up_proj,0.22933,0.01000,1.936
|
7 |
+
0,mlp.gate_proj,0.24757,0.01000,1.914
|
8 |
+
0,mlp.down_proj,1.37315,0.01000,4.905
|
9 |
+
1,self_attn.k_proj,0.54634,0.01000,1.653
|
10 |
+
1,self_attn.v_proj,0.06669,0.01000,1.636
|
11 |
+
1,self_attn.q_proj,1.12471,0.01000,1.694
|
12 |
+
1,self_attn.o_proj,0.00130,0.01000,1.373
|
13 |
+
1,mlp.up_proj,0.49362,0.01000,1.993
|
14 |
+
1,mlp.gate_proj,0.55734,0.01000,1.956
|
15 |
+
1,mlp.down_proj,0.00385,0.01000,5.071
|
16 |
+
2,self_attn.k_proj,0.42915,0.01000,1.694
|
17 |
+
2,self_attn.v_proj,0.07932,0.01000,1.647
|
18 |
+
2,self_attn.q_proj,0.94002,0.01000,1.693
|
19 |
+
2,self_attn.o_proj,0.00277,0.01000,1.394
|
20 |
+
2,mlp.up_proj,1.05308,0.01000,1.929
|
21 |
+
2,mlp.gate_proj,1.28850,0.01000,1.936
|
22 |
+
2,mlp.down_proj,0.06975,0.01000,4.864
|
23 |
+
3,self_attn.k_proj,0.92321,0.01000,1.623
|
24 |
+
3,self_attn.v_proj,0.14390,0.01000,1.649
|
25 |
+
3,self_attn.q_proj,2.24338,0.01000,1.682
|
26 |
+
3,self_attn.o_proj,0.00383,0.01000,1.375
|
27 |
+
3,mlp.up_proj,1.38623,0.01000,1.932
|
28 |
+
3,mlp.gate_proj,1.75863,0.01000,1.927
|
29 |
+
3,mlp.down_proj,0.01949,0.01000,4.850
|
30 |
+
4,self_attn.k_proj,0.70682,0.01000,1.651
|
31 |
+
4,self_attn.v_proj,0.17194,0.01000,1.583
|
32 |
+
4,self_attn.q_proj,1.66714,0.01000,1.630
|
33 |
+
4,self_attn.o_proj,0.00794,0.01000,1.341
|
34 |
+
4,mlp.up_proj,1.81410,0.01000,1.903
|
35 |
+
4,mlp.gate_proj,2.39854,0.01000,1.895
|
36 |
+
4,mlp.down_proj,0.03273,0.01000,4.847
|
37 |
+
5,self_attn.k_proj,0.92816,0.01000,1.608
|
38 |
+
5,self_attn.v_proj,0.18893,0.01000,1.555
|
39 |
+
5,self_attn.q_proj,2.33771,0.01000,1.629
|
40 |
+
5,self_attn.o_proj,0.00867,0.01000,1.318
|
41 |
+
5,mlp.up_proj,2.42875,0.01000,1.920
|
42 |
+
5,mlp.gate_proj,3.08451,0.01000,1.864
|
43 |
+
5,mlp.down_proj,0.04925,0.01000,4.845
|
44 |
+
6,self_attn.k_proj,1.07199,0.01000,1.631
|
45 |
+
6,self_attn.v_proj,0.18683,0.01000,1.573
|
46 |
+
6,self_attn.q_proj,2.46528,0.01000,1.628
|
47 |
+
6,self_attn.o_proj,0.01993,0.01000,1.387
|
48 |
+
6,mlp.up_proj,2.95284,0.01000,1.893
|
49 |
+
6,mlp.gate_proj,3.81204,0.01000,1.895
|
50 |
+
6,mlp.down_proj,0.07002,0.01000,4.853
|
51 |
+
7,self_attn.k_proj,1.08823,0.01000,1.636
|
52 |
+
7,self_attn.v_proj,0.28548,0.01000,1.569
|
53 |
+
7,self_attn.q_proj,2.76811,0.01000,1.654
|
54 |
+
7,self_attn.o_proj,0.03065,0.01000,1.345
|
55 |
+
7,mlp.up_proj,3.46371,0.01000,1.912
|
56 |
+
7,mlp.gate_proj,4.24435,0.01000,1.898
|
57 |
+
7,mlp.down_proj,0.08456,0.01000,4.945
|
58 |
+
8,self_attn.k_proj,1.16355,0.01000,1.617
|
59 |
+
8,self_attn.v_proj,0.28823,0.01000,1.582
|
60 |
+
8,self_attn.q_proj,2.91725,0.01000,1.675
|
61 |
+
8,self_attn.o_proj,0.03290,0.01000,1.433
|
62 |
+
8,mlp.up_proj,3.98309,0.01000,1.948
|
63 |
+
8,mlp.gate_proj,4.53147,0.01000,1.949
|
64 |
+
8,mlp.down_proj,0.10271,0.01000,5.016
|
65 |
+
9,self_attn.k_proj,1.29539,0.01000,1.722
|
66 |
+
9,self_attn.v_proj,0.42527,0.01000,1.761
|
67 |
+
9,self_attn.q_proj,3.25226,0.01000,1.695
|
68 |
+
9,self_attn.o_proj,0.05341,0.01000,1.411
|
69 |
+
9,mlp.up_proj,4.35932,0.01000,1.961
|
70 |
+
9,mlp.gate_proj,4.64491,0.01000,1.944
|
71 |
+
9,mlp.down_proj,0.12074,0.01000,4.965
|
72 |
+
10,self_attn.k_proj,1.25597,0.01000,1.706
|
73 |
+
10,self_attn.v_proj,0.39587,0.01000,1.630
|
74 |
+
10,self_attn.q_proj,3.27324,0.01000,1.729
|
75 |
+
10,self_attn.o_proj,0.06494,0.01000,1.398
|
76 |
+
10,mlp.up_proj,4.79890,0.01000,1.981
|
77 |
+
10,mlp.gate_proj,4.91185,0.01000,1.929
|
78 |
+
10,mlp.down_proj,0.13928,0.01000,5.106
|
79 |
+
11,self_attn.k_proj,1.56255,0.01000,1.750
|
80 |
+
11,self_attn.v_proj,0.44044,0.01000,1.676
|
81 |
+
11,self_attn.q_proj,3.97072,0.01000,1.734
|
82 |
+
11,self_attn.o_proj,0.07647,0.01000,1.422
|
83 |
+
11,mlp.up_proj,5.44356,0.01000,1.984
|
84 |
+
11,mlp.gate_proj,6.01119,0.01000,1.976
|
85 |
+
11,mlp.down_proj,0.15027,0.01000,5.132
|
86 |
+
12,self_attn.k_proj,1.50076,0.01000,1.713
|
87 |
+
12,self_attn.v_proj,0.55479,0.01000,1.642
|
88 |
+
12,self_attn.q_proj,3.89300,0.01000,1.739
|
89 |
+
12,self_attn.o_proj,0.07701,0.01000,1.434
|
90 |
+
12,mlp.up_proj,6.02853,0.01000,2.002
|
91 |
+
12,mlp.gate_proj,6.57857,0.01000,1.959
|
92 |
+
12,mlp.down_proj,0.16847,0.01000,5.181
|
93 |
+
13,self_attn.k_proj,1.90422,0.01000,1.729
|
94 |
+
13,self_attn.v_proj,0.58111,0.01000,1.654
|
95 |
+
13,self_attn.q_proj,4.91283,0.01000,1.749
|
96 |
+
13,self_attn.o_proj,0.09638,0.01000,1.414
|
97 |
+
13,mlp.up_proj,6.54751,0.01000,1.982
|
98 |
+
13,mlp.gate_proj,6.87143,0.01000,1.996
|
99 |
+
13,mlp.down_proj,0.21942,0.01000,5.136
|
100 |
+
14,self_attn.k_proj,2.09946,0.01000,1.731
|
101 |
+
14,self_attn.v_proj,0.65708,0.01000,1.623
|
102 |
+
14,self_attn.q_proj,5.52672,0.01000,1.734
|
103 |
+
14,self_attn.o_proj,0.08825,0.01000,1.415
|
104 |
+
14,mlp.up_proj,7.28541,0.01000,1.987
|
105 |
+
14,mlp.gate_proj,7.21690,0.01000,2.006
|
106 |
+
14,mlp.down_proj,0.22773,0.01000,5.153
|
107 |
+
15,self_attn.k_proj,1.91788,0.01000,1.722
|
108 |
+
15,self_attn.v_proj,0.63475,0.01000,1.677
|
109 |
+
15,self_attn.q_proj,4.70698,0.01000,1.727
|
110 |
+
15,self_attn.o_proj,0.10500,0.01000,1.423
|
111 |
+
15,mlp.up_proj,7.60786,0.01000,1.987
|
112 |
+
15,mlp.gate_proj,7.09275,0.01000,2.003
|
113 |
+
15,mlp.down_proj,0.26158,0.01000,5.205
|
114 |
+
16,self_attn.k_proj,1.99614,0.01000,1.711
|
115 |
+
16,self_attn.v_proj,0.73938,0.01000,1.662
|
116 |
+
16,self_attn.q_proj,4.97917,0.01000,1.740
|
117 |
+
16,self_attn.o_proj,0.13482,0.01000,1.417
|
118 |
+
16,mlp.up_proj,8.12699,0.01000,1.991
|
119 |
+
16,mlp.gate_proj,7.33698,0.01000,1.981
|
120 |
+
16,mlp.down_proj,0.31113,0.01000,5.161
|
121 |
+
17,self_attn.k_proj,1.86525,0.01000,1.723
|
122 |
+
17,self_attn.v_proj,0.90671,0.01000,1.678
|
123 |
+
17,self_attn.q_proj,5.01605,0.01000,1.680
|
124 |
+
17,self_attn.o_proj,0.17565,0.01000,1.379
|
125 |
+
17,mlp.up_proj,8.98972,0.01000,1.941
|
126 |
+
17,mlp.gate_proj,7.92179,0.01000,1.932
|
127 |
+
17,mlp.down_proj,0.38754,0.01000,4.989
|
128 |
+
18,self_attn.k_proj,2.20511,0.01000,1.657
|
129 |
+
18,self_attn.v_proj,0.90475,0.01000,1.591
|
130 |
+
18,self_attn.q_proj,5.73960,0.01000,1.678
|
131 |
+
18,self_attn.o_proj,0.23413,0.01000,1.366
|
132 |
+
18,mlp.up_proj,9.93882,0.01000,1.950
|
133 |
+
18,mlp.gate_proj,9.19216,0.01000,1.930
|
134 |
+
18,mlp.down_proj,0.46037,0.01000,4.896
|
135 |
+
19,self_attn.k_proj,2.24348,0.01000,1.629
|
136 |
+
19,self_attn.v_proj,1.30353,0.01000,1.628
|
137 |
+
19,self_attn.q_proj,6.65672,0.01000,1.692
|
138 |
+
19,self_attn.o_proj,0.17194,0.01000,1.385
|
139 |
+
19,mlp.up_proj,10.52798,0.01000,1.970
|
140 |
+
19,mlp.gate_proj,10.01086,0.01000,1.935
|
141 |
+
19,mlp.down_proj,0.50281,0.01000,5.060
|
142 |
+
20,self_attn.k_proj,2.35084,0.01000,1.742
|
143 |
+
20,self_attn.v_proj,1.16817,0.01000,1.616
|
144 |
+
20,self_attn.q_proj,6.67217,0.01000,1.681
|
145 |
+
20,self_attn.o_proj,0.15849,0.01000,1.382
|
146 |
+
20,mlp.up_proj,11.43906,0.01000,2.007
|
147 |
+
20,mlp.gate_proj,11.54784,0.01000,1.990
|
148 |
+
20,mlp.down_proj,0.56608,0.01000,5.002
|
149 |
+
21,self_attn.k_proj,2.25952,0.01000,1.673
|
150 |
+
21,self_attn.v_proj,1.03793,0.01000,1.616
|
151 |
+
21,self_attn.q_proj,6.24515,0.01000,1.695
|
152 |
+
21,self_attn.o_proj,0.13921,0.01000,1.371
|
153 |
+
21,mlp.up_proj,12.59370,0.01000,1.941
|
154 |
+
21,mlp.gate_proj,13.18203,0.01000,1.925
|
155 |
+
21,mlp.down_proj,0.65296,0.01000,4.990
|
156 |
+
22,self_attn.k_proj,2.22437,0.01000,1.709
|
157 |
+
22,self_attn.v_proj,1.47607,0.01000,1.619
|
158 |
+
22,self_attn.q_proj,6.59270,0.01000,1.687
|
159 |
+
22,self_attn.o_proj,0.17644,0.01000,1.374
|
160 |
+
22,mlp.up_proj,13.96070,0.01000,1.934
|
161 |
+
22,mlp.gate_proj,14.97387,0.01000,1.942
|
162 |
+
22,mlp.down_proj,0.82510,0.01000,5.001
|
163 |
+
23,self_attn.k_proj,2.38086,0.01000,1.676
|
164 |
+
23,self_attn.v_proj,1.46058,0.01000,1.642
|
165 |
+
23,self_attn.q_proj,6.76258,0.01000,1.692
|
166 |
+
23,self_attn.o_proj,0.17248,0.01000,1.387
|
167 |
+
23,mlp.up_proj,14.85585,0.01000,1.932
|
168 |
+
23,mlp.gate_proj,15.75251,0.01000,1.901
|
169 |
+
23,mlp.down_proj,0.95195,0.01000,4.966
|
170 |
+
24,self_attn.k_proj,2.63904,0.01000,1.661
|
171 |
+
24,self_attn.v_proj,1.58612,0.01000,1.607
|
172 |
+
24,self_attn.q_proj,7.52401,0.01000,1.681
|
173 |
+
24,self_attn.o_proj,0.23829,0.01000,1.377
|
174 |
+
24,mlp.up_proj,16.10369,0.01000,1.921
|
175 |
+
24,mlp.gate_proj,17.42771,0.01000,1.904
|
176 |
+
24,mlp.down_proj,0.99444,0.01000,4.953
|
177 |
+
25,self_attn.k_proj,2.90565,0.01000,1.666
|
178 |
+
25,self_attn.v_proj,1.69467,0.01000,1.588
|
179 |
+
25,self_attn.q_proj,8.14186,0.01000,1.673
|
180 |
+
25,self_attn.o_proj,0.16555,0.01000,1.373
|
181 |
+
25,mlp.up_proj,17.15473,0.01000,1.915
|
182 |
+
25,mlp.gate_proj,18.86563,0.01000,1.905
|
183 |
+
25,mlp.down_proj,1.09667,0.01000,4.972
|
184 |
+
26,self_attn.k_proj,2.89141,0.01000,1.663
|
185 |
+
26,self_attn.v_proj,1.96453,0.01000,1.535
|
186 |
+
26,self_attn.q_proj,8.08093,0.01000,1.618
|
187 |
+
26,self_attn.o_proj,0.18482,0.01000,1.328
|
188 |
+
26,mlp.up_proj,18.59406,0.01000,1.872
|
189 |
+
26,mlp.gate_proj,20.82703,0.01000,1.992
|
190 |
+
26,mlp.down_proj,1.17742,0.01000,4.828
|
191 |
+
27,self_attn.k_proj,3.05272,0.01000,1.609
|
192 |
+
27,self_attn.v_proj,2.29465,0.01000,1.547
|
193 |
+
27,self_attn.q_proj,8.93310,0.01000,1.618
|
194 |
+
27,self_attn.o_proj,0.14195,0.01000,1.318
|
195 |
+
27,mlp.up_proj,20.18908,0.01000,1.884
|
196 |
+
27,mlp.gate_proj,22.93357,0.01000,1.849
|
197 |
+
27,mlp.down_proj,1.29122,0.01000,4.823
|
198 |
+
28,self_attn.k_proj,3.17150,0.01000,1.604
|
199 |
+
28,self_attn.v_proj,2.32700,0.01000,1.546
|
200 |
+
28,self_attn.q_proj,8.92997,0.01000,1.621
|
201 |
+
28,self_attn.o_proj,0.27351,0.01000,1.319
|
202 |
+
28,mlp.up_proj,21.86856,0.01000,1.887
|
203 |
+
28,mlp.gate_proj,24.50107,0.01000,1.857
|
204 |
+
28,mlp.down_proj,1.48228,0.01000,4.848
|
205 |
+
29,self_attn.k_proj,3.09725,0.01000,1.607
|
206 |
+
29,self_attn.v_proj,3.17635,0.01000,1.547
|
207 |
+
29,self_attn.q_proj,9.09989,0.01000,1.616
|
208 |
+
29,self_attn.o_proj,0.37311,0.01000,1.341
|
209 |
+
29,mlp.up_proj,24.06158,0.01000,1.874
|
210 |
+
29,mlp.gate_proj,27.08915,0.01000,1.929
|
211 |
+
29,mlp.down_proj,1.66260,0.01000,5.058
|
212 |
+
30,self_attn.k_proj,3.52871,0.01000,1.690
|
213 |
+
30,self_attn.v_proj,2.85859,0.01000,1.629
|
214 |
+
30,self_attn.q_proj,9.94771,0.01000,1.708
|
215 |
+
30,self_attn.o_proj,0.23282,0.01000,1.413
|
216 |
+
30,mlp.up_proj,26.42622,0.01000,1.967
|
217 |
+
30,mlp.gate_proj,29.90111,0.01000,1.943
|
218 |
+
30,mlp.down_proj,1.96078,0.01000,5.088
|
219 |
+
31,self_attn.k_proj,3.17594,0.01000,1.723
|
220 |
+
31,self_attn.v_proj,2.83306,0.01000,1.621
|
221 |
+
31,self_attn.q_proj,9.12404,0.01000,1.697
|
222 |
+
31,self_attn.o_proj,0.36137,0.01000,1.397
|
223 |
+
31,mlp.up_proj,28.66065,0.01000,1.945
|
224 |
+
31,mlp.gate_proj,31.69213,0.01000,1.870
|
225 |
+
31,mlp.down_proj,2.30186,0.01000,4.915
|
226 |
+
32,self_attn.k_proj,3.19539,0.01000,1.673
|
227 |
+
32,self_attn.v_proj,3.98448,0.01000,1.636
|
228 |
+
32,self_attn.q_proj,9.97009,0.01000,1.714
|
229 |
+
32,self_attn.o_proj,0.34023,0.01000,1.404
|
230 |
+
32,mlp.up_proj,30.91885,0.01000,1.973
|
231 |
+
32,mlp.gate_proj,33.56028,0.01000,1.903
|
232 |
+
32,mlp.down_proj,2.74637,0.01000,4.990
|
233 |
+
33,self_attn.k_proj,3.47076,0.01000,1.670
|
234 |
+
33,self_attn.v_proj,3.22100,0.01000,1.623
|
235 |
+
33,self_attn.q_proj,10.57501,0.01000,1.701
|
236 |
+
33,self_attn.o_proj,0.47574,0.01000,1.392
|
237 |
+
33,mlp.up_proj,33.12611,0.01000,1.952
|
238 |
+
33,mlp.gate_proj,34.76408,0.01000,1.887
|
239 |
+
33,mlp.down_proj,3.54615,0.01000,5.018
|
240 |
+
34,self_attn.k_proj,3.05377,0.01000,1.673
|
241 |
+
34,self_attn.v_proj,6.53063,0.01000,1.612
|
242 |
+
34,self_attn.q_proj,10.03420,0.01000,1.736
|
243 |
+
34,self_attn.o_proj,0.97326,0.01000,1.377
|
244 |
+
34,mlp.up_proj,36.04812,0.01000,1.960
|
245 |
+
34,mlp.gate_proj,37.42268,0.01000,1.900
|
246 |
+
34,mlp.down_proj,4.00197,0.01000,5.060
|
247 |
+
35,self_attn.k_proj,3.36860,0.01000,1.674
|
248 |
+
35,self_attn.v_proj,5.95140,0.01000,1.608
|
249 |
+
35,self_attn.q_proj,10.86249,0.01000,1.679
|
250 |
+
35,self_attn.o_proj,0.65889,0.01000,1.392
|
251 |
+
35,mlp.up_proj,39.39637,0.01000,1.953
|
252 |
+
35,mlp.gate_proj,40.15826,0.01000,1.941
|
253 |
+
35,mlp.down_proj,4.86860,0.01000,5.032
|
254 |
+
36,self_attn.k_proj,3.07699,0.01000,1.678
|
255 |
+
36,self_attn.v_proj,6.95885,0.01000,1.641
|
256 |
+
36,self_attn.q_proj,10.35025,0.01000,1.687
|
257 |
+
36,self_attn.o_proj,0.92406,0.01000,1.377
|
258 |
+
36,mlp.up_proj,42.50772,0.01000,1.926
|
259 |
+
36,mlp.gate_proj,42.03286,0.01000,1.913
|
260 |
+
36,mlp.down_proj,6.17359,0.01000,4.961
|
261 |
+
37,self_attn.k_proj,2.74004,0.01000,1.705
|
262 |
+
37,self_attn.v_proj,7.82842,0.01000,1.647
|
263 |
+
37,self_attn.q_proj,10.62546,0.01000,1.710
|
264 |
+
37,self_attn.o_proj,2.19896,0.01000,1.450
|
265 |
+
37,mlp.up_proj,42.68120,0.01000,2.027
|
266 |
+
37,mlp.gate_proj,42.57842,0.01000,1.984
|
267 |
+
37,mlp.down_proj,7.93857,0.01000,5.068
|
268 |
+
38,self_attn.k_proj,3.25958,0.01000,1.681
|
269 |
+
38,self_attn.v_proj,11.66085,0.01000,1.620
|
270 |
+
38,self_attn.q_proj,11.31734,0.01000,1.676
|
271 |
+
38,self_attn.o_proj,2.05625,0.01000,1.383
|
272 |
+
38,mlp.up_proj,42.16687,0.01000,1.917
|
273 |
+
38,mlp.gate_proj,43.10248,0.01000,1.892
|
274 |
+
38,mlp.down_proj,9.73955,0.01000,4.959
|
275 |
+
39,self_attn.k_proj,2.71716,0.01000,1.660
|
276 |
+
39,self_attn.v_proj,6.01112,0.01000,1.611
|
277 |
+
39,self_attn.q_proj,9.59197,0.01000,1.676
|
278 |
+
39,self_attn.o_proj,1.59839,0.01000,1.366
|
279 |
+
39,mlp.up_proj,42.55232,0.01000,1.961
|
280 |
+
39,mlp.gate_proj,45.28327,0.01000,2.018
|
281 |
+
39,mlp.down_proj,15.42649,0.01000,4.955
|
quantize_config.json
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bits": 4,
|
3 |
+
"dynamic": null,
|
4 |
+
"group_size": 32,
|
5 |
+
"desc_act": true,
|
6 |
+
"sym": true,
|
7 |
+
"lm_head": false,
|
8 |
+
"quant_method": "gptq",
|
9 |
+
"checkpoint_format": "gptq",
|
10 |
+
"meta": {
|
11 |
+
"quantizer": [
|
12 |
+
"gptqmodel:1.7.0"
|
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 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "<|im_end|>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": "<pad>",
|
17 |
+
"unk_token": {
|
18 |
+
"content": "<unk>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": false,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
}
|
24 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea88f9940a84ab7e0100bc369506a28ec8d5d821691dc47d4dd63f1bbdf105ed
|
3 |
+
size 17078669
|
tokenizer_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|