Add files using upload-large-folder tool
Browse files- .gitattributes +27 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00001-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00002-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00003-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00004-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00005-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00006-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00007-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00008-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00009-of-00010.gguf +3 -0
- Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00010-of-00010.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00001-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00002-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00003-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00004-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00005-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00006-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00007-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00008-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00009-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00010-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00011-of-00012.gguf +3 -0
- Q6_K/DeepSeek-V3-0324-Q6_K-00012-of-00012.gguf +3 -0
- UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00001-of-00005.gguf +3 -0
- UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00002-of-00005.gguf +3 -0
- UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00003-of-00005.gguf +3 -0
- UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00004-of-00005.gguf +3 -0
- UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00005-of-00005.gguf +3 -0
- config.json +70 -0
.gitattributes
CHANGED
@@ -69,3 +69,30 @@ Q2_K/DeepSeek-V3-0324-Q2_K-00001-of-00005.gguf filter=lfs diff=lfs merge=lfs -te
|
|
69 |
Q2_K/DeepSeek-V3-0324-Q2_K-00002-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
70 |
Q3_K_M/DeepSeek-V3-0324-Q3_K_M-00001-of-00007.gguf filter=lfs diff=lfs merge=lfs -text
|
71 |
Q4_K_M/DeepSeek-V3-0324-Q4_K_M-00003-of-00009.gguf filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
69 |
Q2_K/DeepSeek-V3-0324-Q2_K-00002-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
70 |
Q3_K_M/DeepSeek-V3-0324-Q3_K_M-00001-of-00007.gguf filter=lfs diff=lfs merge=lfs -text
|
71 |
Q4_K_M/DeepSeek-V3-0324-Q4_K_M-00003-of-00009.gguf filter=lfs diff=lfs merge=lfs -text
|
72 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00012-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
73 |
+
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00005-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
74 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00010-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
75 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00011-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
76 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00010-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
77 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00004-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
78 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00009-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
79 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00001-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
80 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00008-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
81 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00005-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
82 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00006-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
83 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00004-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
84 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00005-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
85 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00003-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
86 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00007-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
87 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00008-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
88 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00007-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
89 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00003-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
90 |
+
Q6_K/DeepSeek-V3-0324-Q6_K-00002-of-00012.gguf filter=lfs diff=lfs merge=lfs -text
|
91 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00009-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
92 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00001-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
93 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00006-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
94 |
+
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00002-of-00010.gguf filter=lfs diff=lfs merge=lfs -text
|
95 |
+
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00001-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
96 |
+
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00002-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
97 |
+
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00004-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
98 |
+
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00003-of-00005.gguf filter=lfs diff=lfs merge=lfs -text
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00001-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3a1fd13f61dc17ab3e5a1b6f8d73390acc5cae36c6a8741e32040519ce68984b
|
3 |
+
size 49573632960
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00002-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:84e8614f0696443d04e8b31739922f960ca9837f4aa199d4b75b9af48de265f3
|
3 |
+
size 48506507776
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00003-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:db657e111a648e15328fb47dc0e50ade96bc162679670f9deb3fb12b44ddd64a
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00004-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f8ddf3018c75e7955617e794d2cd2fdec3d92410b559960144442c73cc2a32cc
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00005-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7a855e529690fbf8504bb47a88587243f58fc05a9a9944b70866828f7c2608cc
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00006-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:13899d58e2c13a6005928b9dc4ccfc48f543db955396882a6ebcbd2749e7b06b
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00007-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b76fc595be23703f6092cb051136df7c4c6cf317b30433d5540cfc49b944b1a
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00008-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:379b3dc2c215918442649d00a85430c4fbaa551de858c965c57a566815c911dc
|
3 |
+
size 48506507808
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00009-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e026951941323f399668efe99ab3647f0d8baaa39b1d4881b8bef8c8afabfe65
|
3 |
+
size 49999602080
|
Q5_K_M/DeepSeek-V3-0324-Q5_K_M-00010-of-00010.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:aca14897a0164d540f3cab79940c5fd9664363c307c43949c3911d6377f89160
|
3 |
+
size 36277768864
|
Q6_K/DeepSeek-V3-0324-Q6_K-00001-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:912f894b79597237d809a5e6034f09ffabfca78e321438e74bc2bf0b37f5bdb2
|
3 |
+
size 47094115200
|
Q6_K/DeepSeek-V3-0324-Q6_K-00002-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3c47d9a1933722505157935c299b89485e6de8a2f86ff61146777deae9d13213
|
3 |
+
size 47227297696
|
Q6_K/DeepSeek-V3-0324-Q6_K-00003-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fe8b8b5b37f4f6597563246b3192537f4a51bd0b0f25120fefb7fa1cea261dc0
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00004-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fce68bb41db1530b02c6bdfe2b19cef392922ecc2e04df1e2b31314b878c5ef4
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00005-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5a2c07235c80a9039d59f9f365c70a830c5144e89cb8d3ed9dc0fab39a0683ff
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00006-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:580b1c3dcd1ba43bda874b5e0373930ebaf737e83a2752092967075418e77d6f
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00007-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d03b6f8897a19f08bbbb2fcf16760c6d3616ba089c078fa2cb25f7956d481f29
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00008-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:617d9e94e91b67fafc3417692e68ca69049f0313143c6e17616ca0da8ae2aefe
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00009-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:42927844820ae7f91701617bf39024109eda9338bea4f40a99d47c4f62c0cb3e
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00010-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8fef82962924c0a8b6e3ea07df6b3def1912a747a4c0172a792c603b817ed3af
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00011-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c40dd87ff4f3e097041f32d187ba4377c1544a1e2df35f9d5bfe70ea589bd844
|
3 |
+
size 47227297760
|
Q6_K/DeepSeek-V3-0324-Q6_K-00012-of-00012.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:66631b3864ca7601d31fe0346161b62b91da5ce98dff1b42e051ab904e55df8c
|
3 |
+
size 31431234528
|
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00001-of-00005.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:94d972b3ec13ee8f3b37e4d087da1d49c1f0494125ef9f3b5c206eb9e6d223a1
|
3 |
+
size 49797297120
|
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00002-of-00005.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:449a92150fe18101352351d80cd954717f470611fa5b1206babc23d76f295361
|
3 |
+
size 49997709824
|
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00003-of-00005.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:deec2e79c58ca7bee52533c2fe1a91045244afe31572bf3ce4a4f342173f3726
|
3 |
+
size 49997709824
|
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00004-of-00005.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0dca617e27db5a0ceba9cb961cc64d5d61b28049fa11a6aea22e473275e1cec1
|
3 |
+
size 49997709824
|
UD-Q2_K_XL/DeepSeek-V3-0324-UD-Q2_K_XL-00005-of-00005.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3ad3af5a9111947e15c8f633cfad4a161a71be0c26b078a4692a89d8db5e52c9
|
3 |
+
size 26811168832
|
config.json
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"DeepseekV3ForCausalLM"
|
4 |
+
],
|
5 |
+
"attention_bias": false,
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"auto_map": {
|
8 |
+
"AutoConfig": "configuration_deepseek.DeepseekV3Config",
|
9 |
+
"AutoModel": "modeling_deepseek.DeepseekV3Model",
|
10 |
+
"AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
|
11 |
+
},
|
12 |
+
"aux_loss_alpha": 0.001,
|
13 |
+
"bos_token_id": 0,
|
14 |
+
"eos_token_id": 1,
|
15 |
+
"ep_size": 1,
|
16 |
+
"first_k_dense_replace": 3,
|
17 |
+
"hidden_act": "silu",
|
18 |
+
"hidden_size": 7168,
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 18432,
|
21 |
+
"kv_lora_rank": 512,
|
22 |
+
"max_position_embeddings": 163840,
|
23 |
+
"model_type": "deepseek_v3",
|
24 |
+
"moe_intermediate_size": 2048,
|
25 |
+
"moe_layer_freq": 1,
|
26 |
+
"n_group": 8,
|
27 |
+
"n_routed_experts": 256,
|
28 |
+
"n_shared_experts": 1,
|
29 |
+
"norm_topk_prob": true,
|
30 |
+
"num_attention_heads": 128,
|
31 |
+
"num_experts_per_tok": 8,
|
32 |
+
"num_hidden_layers": 61,
|
33 |
+
"num_key_value_heads": 128,
|
34 |
+
"num_nextn_predict_layers": 1,
|
35 |
+
"pretraining_tp": 1,
|
36 |
+
"q_lora_rank": 1536,
|
37 |
+
"qk_nope_head_dim": 128,
|
38 |
+
"qk_rope_head_dim": 64,
|
39 |
+
"quantization_config": {
|
40 |
+
"activation_scheme": "dynamic",
|
41 |
+
"fmt": "e4m3",
|
42 |
+
"quant_method": "fp8",
|
43 |
+
"weight_block_size": [
|
44 |
+
128,
|
45 |
+
128
|
46 |
+
]
|
47 |
+
},
|
48 |
+
"rms_norm_eps": 1e-06,
|
49 |
+
"rope_scaling": {
|
50 |
+
"beta_fast": 32,
|
51 |
+
"beta_slow": 1,
|
52 |
+
"factor": 40,
|
53 |
+
"mscale": 1.0,
|
54 |
+
"mscale_all_dim": 1.0,
|
55 |
+
"original_max_position_embeddings": 4096,
|
56 |
+
"type": "yarn"
|
57 |
+
},
|
58 |
+
"rope_theta": 10000,
|
59 |
+
"routed_scaling_factor": 2.5,
|
60 |
+
"scoring_func": "sigmoid",
|
61 |
+
"seq_aux": true,
|
62 |
+
"tie_word_embeddings": false,
|
63 |
+
"topk_group": 4,
|
64 |
+
"topk_method": "noaux_tc",
|
65 |
+
"torch_dtype": "bfloat16",
|
66 |
+
"transformers_version": "4.46.3",
|
67 |
+
"use_cache": true,
|
68 |
+
"v_head_dim": 128,
|
69 |
+
"vocab_size": 129280
|
70 |
+
}
|