alexxi19 commited on
Commit
3c6cea9
·
verified ·
1 Parent(s): c0c5514

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - Nitral-AI/Captain-Eris_Violet-V0.420-12B
4
+ - alexxi19/ft-v1-nemo-base
5
+ - anthracite-org/magnum-v2-12b
6
+ library_name: transformers
7
+ tags:
8
+ - mergekit
9
+ - merge
10
+
11
+ ---
12
+ # merged_llm
13
+
14
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
15
+
16
+ ## Merge Details
17
+ ### Merge Method
18
+
19
+ This model was merged using the linear [DELLA](https://arxiv.org/abs/2406.11617) merge method using [alexxi19/ft-v1-nemo-base](https://huggingface.co/alexxi19/ft-v1-nemo-base) as a base.
20
+
21
+ ### Models Merged
22
+
23
+ The following models were included in the merge:
24
+ * [Nitral-AI/Captain-Eris_Violet-V0.420-12B](https://huggingface.co/Nitral-AI/Captain-Eris_Violet-V0.420-12B)
25
+ * [anthracite-org/magnum-v2-12b](https://huggingface.co/anthracite-org/magnum-v2-12b)
26
+
27
+ ### Configuration
28
+
29
+ The following YAML configuration was used to produce this model:
30
+
31
+ ```yaml
32
+ # models:
33
+ # - model: anthracite-org/magnum-v2-12b # instruct model
34
+ # parameters:
35
+ # density: 0.6
36
+ # weight: 0.5
37
+ # # - model: /home/paperspace/projects/project/finetunellm/outputs/nemo-12b-creative/merged # creative writing model
38
+ # # parameters:
39
+ # # density: 0.3
40
+ # # weight: 0.3
41
+ # - model: alexxi19/ft-nemo-base-lora # sft model
42
+ # parameters:
43
+ # density: 0.7
44
+ # weight: 0.5
45
+ # merge_method: dare_ties
46
+ # base_model: alexxi19/ft-nemo-base-lora
47
+ # parameters:
48
+ # int8_mask: true
49
+ # rescale: true
50
+ # # normalize: true
51
+ # dtype: bfloat16
52
+ # chat_template: chatml
53
+
54
+
55
+ models:
56
+ - model: anthracite-org/magnum-v2-12b # instruct model
57
+ parameters:
58
+ density: 0.3
59
+ weight: 0.5
60
+ # - model: Nitral-AI/Captain_BMO-12B # instruct model
61
+ # parameters:
62
+ # density: 0.3
63
+ # weight: 0.5
64
+ - model: Nitral-AI/Captain-Eris_Violet-V0.420-12B # creative writing model
65
+ parameters:
66
+ density: 0.2
67
+ weight: 0.3
68
+ - model: alexxi19/ft-v1-nemo-base # sft model
69
+ parameters:
70
+ density: 0.5
71
+ weight: 0.5
72
+ base_model: alexxi19/ft-v1-nemo-base
73
+ dtype: bfloat16
74
+ chat_template: chatml
75
+ merge_method: della_linear
76
+ parameters:
77
+ epsilon: 0.05
78
+ int8_mask: true
79
+ rescale: true
80
+ lambda: 1.0
81
+
82
+
83
+ # models:
84
+ # - model: Nitral-AI/Captain_BMO-12B # another sft model
85
+ # - model: alexxi19/ft-v1-nemo-base # sft model
86
+ # merge_method: slerp
87
+ # base_model: alexxi19/ft-v1-nemo-base
88
+ # parameters:
89
+ # t:
90
+ # - filter: self_attn
91
+ # value: [0, 0.5, 0.3, 0.7, 1]
92
+ # - filter: mlp
93
+ # value: [1, 0.5, 0.7, 0.3, 0]
94
+ # - value: 0.3 # fallback for rest of tensors
95
+ # dtype: float16
96
+ # chat_template: chatml
97
+
98
+ ```
mergekit_config.yml CHANGED
@@ -26,10 +26,10 @@ models:
26
  parameters:
27
  density: 0.3
28
  weight: 0.5
29
- - model: Nitral-AI/Captain_BMO-12B # instruct model
30
- parameters:
31
- density: 0.3
32
- weight: 0.5
33
  - model: Nitral-AI/Captain-Eris_Violet-V0.420-12B # creative writing model
34
  parameters:
35
  density: 0.2
 
26
  parameters:
27
  density: 0.3
28
  weight: 0.5
29
+ # - model: Nitral-AI/Captain_BMO-12B # instruct model
30
+ # parameters:
31
+ # density: 0.3
32
+ # weight: 0.5
33
  - model: Nitral-AI/Captain-Eris_Violet-V0.420-12B # creative writing model
34
  parameters:
35
  density: 0.2
model-00001-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9e184f1641a32130ab4021f501ec2f0194cddd1ff86fec184ef03fac6cc93219
3
  size 1342177408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:108affc8d24fcd992171f56e8d2edd81f5dd09f9cae0873f3fcd9fc615d50591
3
  size 1342177408
model-00002-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3cbf44c3db9ff6a4c1624a7c16e5d0208b72a6311042b95cb4dfa5748e90169d
3
  size 1342177424
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f199bc51fc9a281f47da7601541f7db87e02df12c0f89efb82fcb13ec4d66fc
3
  size 1342177424
model-00003-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a2aa37ed3a7491de1e4d6d69f2b79643dee01d293507e1bd999614cd7dccb93
3
  size 996189888
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b806586cd71bfe471c9cf3cf7b8757a33c04b6f8ca01ead6eaa7bd912e1a234
3
  size 996189888
model-00004-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5150716ad2ecd7eeacc11f39dc81d1c5cf98f68427b51518bd7f6ab1a953e6ca
3
  size 933265104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5a409a3fd94e640d7a47685b81fbf28594cf9d20d71aa9408243f5e78d8b228
3
  size 933265104
model-00005-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e82bb44a066f39880ce90fe12417340a6c74e346f40e5b06e7047d0248eba7b9
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96f14b68f2e061603732d6eb8992384f0faadaec7e6da4335974bdbea844c9a9
3
  size 943761344
model-00006-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c836d76993f904cb3fcca1c118ad8a3225f6fea1c34028a99f4a9c27c2c78998
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:526c0d636acddb438b311b3ede3936ca92a48c3c4508064544adbb03e1b74505
3
  size 943761344
model-00007-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2fc86ea5b88bbc0879187ebe5b44f38dacd305049a94dae4315106d0c48ad872
3
  size 996179560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0233fd66040b8564cae8487155b98bc9dc7e39e4e755514f0887b80006b4e7e7
3
  size 996179560
model-00008-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:82505f6eb29cc3262ee97632f061a0c6d711dbdbea58ebdfa2df4dee95484869
3
  size 933265104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e1a34d52d49ef859b73672ee9f4e9f6c99986cc2b577c6ce4de2d9124cacd20
3
  size 933265104
model-00009-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ffae5d3e396566c5693ee2585ebf94fa0848c8ec2e797f124378d887fe298caf
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b008530bbd44acf16646bb14e9fdd0b39d322696014f73d6ae3bb96637ae3a6
3
  size 943761344
model-00010-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:30f6a6e101dbe5ac3dc0f738a2671a8de37750187459bd7b32d4a4b852f032ed
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7eb8348f50d3199371ae5933e514a183a1b71752e1663029431b3587cf0ad052
3
  size 943761344
model-00011-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ee9cee09a848532ec79f9f8b6052f26c1de2eee4ea5e905920e706b379c7f2d
3
  size 996179560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:068c8cbc978d80180f4f3f3be1d382d43c0f5323a713b8290f627d3000f2eee0
3
  size 996179560
model-00012-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed1d0d8f9cdd5e2b81811607fc80029c785e8c1dc2e4f3d16ee69d1550e14f7b
3
  size 933265104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:027b79e1995c11af05cdfd35d5cebeb737de7229078ac61edd7c48d42b415a0b
3
  size 933265104
model-00013-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:41c0d47aabcd52eb4aaa236598d11e31025f4ff7e65dfc695dcdd4747b7ff08e
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caacaa115632ed4a0f21b67c2e4d78b2c7738b51fbe951ccbf178a520642533e
3
  size 943761344
model-00014-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0bba56e4283455bdcddbfb0ce6a3469d8dc01b0f3be2a1542e3ad6f0ec82b6ce
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f02fbfae03aa00da0ccbf9b534720b4c4b7c20c460a76be76be9c0e723c31d9
3
  size 943761344
model-00015-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7769dabd42ac07f1a1662a3619ae8b6092fecb7a1d9ee08185b3fb9065d77141
3
  size 996179560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:226e20c12a80d2aaa03f4ccab94a8856d034861faaab44d1e476b6f1954aced4
3
  size 996179560
model-00016-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7ff69f83d84ab329dc9ed343237248eb7b5045ed12d2754ec5241dfc81102c3c
3
  size 933265096
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:12736ba86cb4b8d69af457812b2f6aa34d0d267522d34008492c4ffdb2769778
3
  size 933265096
model-00017-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9bb0027b9da8cf3fbc067c0645bd98070ce8810deb81e82b0632cb9af37bf5ed
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71ae2cec08d5d7e31ba0e451ec6e287560c04b85e58f67eb7353d25f91991745
3
  size 943761344
model-00018-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6785e81f1ffb0290fe3b028372c7b19e5f20e87c30f7115062820a049954cf22
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6caf29490f7b1232a36ac6de21dbfc591fa7752afe452610fd3b189522f304fe
3
  size 943761344
model-00019-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:602ebf799e962675498efaec27de1724ba0596ab84f8bdde66ea880158286b1d
3
  size 996179560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0a04541daa2feaa059fd013a3ac58c26edb5ccea65870ed3d3147c4cf21b25a
3
  size 996179560
model-00020-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:73e5e780dd6254d3ddadeffccf6017531d126dbfc66ccc75209cbf0ac51ef8ea
3
  size 933265104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bdb93b090c097e5ac5abf5f226d8a47d8f2f932a6899387a3e93ca3ba6274e3
3
  size 933265104
model-00021-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7ee0a3cb18cc80e8c6e455fb5e1f05717d63a59b64160871ec84194e94c924b
3
  size 943761344
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4616548382e2a76e4311cb904a3952ed395ccf7e1cad6fbbf054d6fb5f859cd8
3
  size 943761344
model-00022-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3dfae20da227a231d82e4a01c5172cceff8ba17bd73e2bc8e62787fade35d7b5
3
  size 943761336
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1beaa96a302191d9b25e33e37c659eb5a5c9657463765194b387a5eb7f2fe5bf
3
  size 943761336
model-00023-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cb3b1f15db9a66b2ecf1f8de85b0024026b6cc8e41caafe0f5228971ace2e0a9
3
  size 996179552
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb4883ac087bacbdf9d77d75640cab1c5bad271fbf58f8f1382595d3c90aa3a6
3
  size 996179552
model-00024-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7992887c819ecfd97b7f1a834572464870f4d3618e56b7b1e97769fc925ceda5
3
  size 933265088
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09f10360e7550aa1ffc9d98b22f6bcf32f957568224e85f81aead34e79812d6a
3
  size 933265088
model-00025-of-00025.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f2e57ade0795c8414e862dc8b42415c349c8a130a29cd07d9a4c5fa36e37256
3
  size 796960560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b64559765a3c0fc4a03d72f6149ee2018b638fd0cb79512a2e5bf291b0b44188
3
  size 796960560