grimjim commited on
Commit
6d35978
1 Parent(s): 2d2fb20

Initial release

Browse files
.gitattributes CHANGED
@@ -4,6 +4,7 @@
4
  *.bz2 filter=lfs diff=lfs merge=lfs -text
5
  *.ckpt filter=lfs diff=lfs merge=lfs -text
6
  *.ftz filter=lfs diff=lfs merge=lfs -text
 
7
  *.gz filter=lfs diff=lfs merge=lfs -text
8
  *.h5 filter=lfs diff=lfs merge=lfs -text
9
  *.joblib filter=lfs diff=lfs merge=lfs -text
 
4
  *.bz2 filter=lfs diff=lfs merge=lfs -text
5
  *.ckpt filter=lfs diff=lfs merge=lfs -text
6
  *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gguf filter=lfs diff=lfs merge=lfs -text
8
  *.gz filter=lfs diff=lfs merge=lfs -text
9
  *.h5 filter=lfs diff=lfs merge=lfs -text
10
  *.joblib filter=lfs diff=lfs merge=lfs -text
Magnolia-v1-12B.Q4_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76c310b5326107d86216dc48ab206f2e2fdde055907c518b15e74c634b097c5e
3
+ size 7477204064
Magnolia-v1-12B.Q5_K_M.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2095e101b7f3fa992a6cfc33403360e66368a13bbe7d99d8fb435146f56c94b4
3
+ size 8727630944
Magnolia-v1-12B.Q6_K.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d79850a5f3b5d5acf4196b6f967feaab9cab97712ace3430a1a6e92aa4f8582
3
+ size 10056209504
Magnolia-v1-12B.Q8_0.gguf ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:507d73c81da3f3f2316d6251ac49b3a374be7b7492f5b8d10f3011f05e583cda
3
+ size 13022368864
README.md ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - grimjim/magnum-consolidatum-v1-12b
4
+ - grimjim/mistralai-Mistral-Nemo-Instruct-2407
5
+ library_name: transformers
6
+ tags:
7
+ - mergekit
8
+ - merge
9
+ pipeline_tag: text-generation
10
+ license: apache-2.0
11
+ quanted_by: grimjim
12
+ ---
13
+ # Magnolia-v1-12B-GGUF
14
+
15
+ This repo contains a GGUF quants of a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
16
+ The base is a merge of two models trained for variety in text generation.
17
+ Instruct was added in at low weight in order to increase the steerability of the model;
18
+ safety has consequently been reinforced.
19
+
20
+ Tested at temperature 0.7 and minP 0.01, with ChatML prompting.
21
+
22
+ Mistral Nemo models tend to have repetition issues in general. For this model at least, various issues can be mitigated somewhat with additional sysprompting, e.g.:
23
+ ```
24
+ No passage shall exceed 10 lines of text, with turns limited to a maximum of 5 lines per speaker to ensure snappy and engaging dialog and action.
25
+ Ensure that all punctuation rules are adhered to without the introduction of spurious intervening spaces.
26
+ Avoid redundant phrasing and maintain forward narrative progression by utilizing varied sentence structure, alternative word choices, and active voice.
27
+ Employ descriptive details judiciously, ensuring they serve a purpose in advancing the story or revealing character or touching upon setting.
28
+ ```
29
+
30
+ ## Merge Details
31
+ ### Merge Method
32
+
33
+ This model was merged using the SLERP merge method.
34
+
35
+ ### Models Merged
36
+
37
+ The following models were included in the merge:
38
+ * [grimjim/magnum-consolidatum-v1-12b](https://huggingface.co/grimjim/magnum-consolidatum-v1-12b)
39
+ * [grimjim/mistralai-Mistral-Nemo-Instruct-2407](https://huggingface.co/grimjim/mistralai-Mistral-Nemo-Instruct-2407)
40
+
41
+ ### Configuration
42
+
43
+ The following YAML configuration was used to produce this model:
44
+
45
+ ```yaml
46
+ models:
47
+ - model: grimjim/mistralai-Mistral-Nemo-Instruct-2407
48
+ - model: grimjim/magnum-consolidatum-v1-12b
49
+ merge_method: slerp
50
+ base_model: grimjim/mistralai-Mistral-Nemo-Instruct-2407
51
+ parameters:
52
+ t:
53
+ - value: 0.1
54
+ dtype: bfloat16
55
+ ```