Hugging Face's logo

RichardErkhov
/
v000000_-_SyntheticMoist-11B-v2-gguf

GGUF
Model card Files Community
RichardErkhov commited on
Commit
e080b6c
•
1 Parent(s): 3b36f9f

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +120 -0
README.md ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ SyntheticMoist-11B-v2 - GGUF
11
+ - Model creator: https://huggingface.co/v000000/
12
+ - Original model: https://huggingface.co/v000000/SyntheticMoist-11B-v2/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [SyntheticMoist-11B-v2.Q2_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q2_K.gguf) | Q2_K | 3.73GB |
18
+ | [SyntheticMoist-11B-v2.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.IQ3_XS.gguf) | IQ3_XS | 4.14GB |
19
+ | [SyntheticMoist-11B-v2.IQ3_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.IQ3_S.gguf) | IQ3_S | 4.37GB |
20
+ | [SyntheticMoist-11B-v2.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q3_K_S.gguf) | Q3_K_S | 4.34GB |
21
+ | [SyntheticMoist-11B-v2.IQ3_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.IQ3_M.gguf) | IQ3_M | 4.51GB |
22
+ | [SyntheticMoist-11B-v2.Q3_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q3_K.gguf) | Q3_K | 4.84GB |
23
+ | [SyntheticMoist-11B-v2.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q3_K_M.gguf) | Q3_K_M | 4.84GB |
24
+ | [SyntheticMoist-11B-v2.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q3_K_L.gguf) | Q3_K_L | 5.26GB |
25
+ | [SyntheticMoist-11B-v2.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.IQ4_XS.gguf) | IQ4_XS | 5.43GB |
26
+ | [SyntheticMoist-11B-v2.Q4_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q4_0.gguf) | Q4_0 | 5.66GB |
27
+ | [SyntheticMoist-11B-v2.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.IQ4_NL.gguf) | IQ4_NL | 5.72GB |
28
+ | [SyntheticMoist-11B-v2.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q4_K_S.gguf) | Q4_K_S | 5.7GB |
29
+ | [SyntheticMoist-11B-v2.Q4_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q4_K.gguf) | Q4_K | 6.02GB |
30
+ | [SyntheticMoist-11B-v2.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q4_K_M.gguf) | Q4_K_M | 6.02GB |
31
+ | [SyntheticMoist-11B-v2.Q4_1.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q4_1.gguf) | Q4_1 | 6.27GB |
32
+ | [SyntheticMoist-11B-v2.Q5_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q5_0.gguf) | Q5_0 | 6.89GB |
33
+ | [SyntheticMoist-11B-v2.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q5_K_S.gguf) | Q5_K_S | 6.89GB |
34
+ | [SyntheticMoist-11B-v2.Q5_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q5_K.gguf) | Q5_K | 7.08GB |
35
+ | [SyntheticMoist-11B-v2.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q5_K_M.gguf) | Q5_K_M | 7.08GB |
36
+ | [SyntheticMoist-11B-v2.Q5_1.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q5_1.gguf) | Q5_1 | 7.51GB |
37
+ | [SyntheticMoist-11B-v2.Q6_K.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q6_K.gguf) | Q6_K | 8.2GB |
38
+ | [SyntheticMoist-11B-v2.Q8_0.gguf](https://huggingface.co/RichardErkhov/v000000_-_SyntheticMoist-11B-v2-gguf/blob/main/SyntheticMoist-11B-v2.Q8_0.gguf) | Q8_0 | 10.62GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ base_model:
46
+ - Sao10K/Fimbulvetr-11B-v2
47
+ - TheDrummer/Moistral-11B-v3
48
+ - Himitsui/MedMitsu-Instruct-11B
49
+ - Himitsui/Kaiju-11B
50
+ - migtissera/Synthia-v3.0-11B
51
+ - jeiku/Re-Host_Limarp_Mistral
52
+ library_name: transformers
53
+ tags:
54
+ - mergekit
55
+ - merge
56
+ - solar
57
+ - llama
58
+ - not-for-all-audiences
59
+ ---
60
+
61
+ ### SyntheticMoist-v2
62
+ RP Model, Solar.
63
+ Higher density+LimaRP led to better performance, Use Alpaca/Vicuna.
64
+
65
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/64f74b6e6389380c77562762/nipwgMMVVEWvJqN3TSWZP.png)
66
+
67
+ # Thanks mradermacher for the quants!
68
+ * [GGUF Q2-Q8](https://huggingface.co/mradermacher/SyntheticMoist-11B-v2-GGUF)
69
+ * [GGUF imatrix](https://huggingface.co/mradermacher/SyntheticMoist-11B-v2-i1-GGUF)
70
+
71
+ # Quants
72
+ * [GGUF Q6_K](https://huggingface.co/v000000/SyntheticMoist-11B-v2-Q6_K-GGUF)
73
+
74
+ # merge
75
+
76
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
77
+
78
+ ## Merge Details
79
+ ### Merge Method
80
+
81
+ This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [Sao10K/Fimbulvetr-11B-v2](https://huggingface.co/Sao10K/Fimbulvetr-11B-v2) as a base.
82
+
83
+ ### Models Merged
84
+
85
+ The following models were included in the merge:
86
+ * [TheDrummer/Moistral-11B-v3](https://huggingface.co/TheDrummer/Moistral-11B-v3)
87
+ * [Himitsui/MedMitsu-Instruct-11B](https://huggingface.co/Himitsui/MedMitsu-Instruct-11B)
88
+ * [Himitsui/Kaiju-11B](https://huggingface.co/Himitsui/Kaiju-11B)
89
+ * [migtissera/Synthia-v3.0-11B](https://huggingface.co/migtissera/Synthia-v3.0-11B) + [jeiku/Re-Host_Limarp_Mistral](https://huggingface.co/jeiku/Re-Host_Limarp_Mistral)
90
+
91
+ ### Configuration
92
+
93
+ The following YAML configuration was used to produce this model:
94
+
95
+ ```yaml
96
+ models:
97
+ - model: Himitsui/MedMitsu-Instruct-11B
98
+ parameters:
99
+ weight: 0.13
100
+ density: 0.60
101
+ - model: Himitsui/Kaiju-11B
102
+ parameters:
103
+ weight: 0.22
104
+ density: 0.73
105
+ - model: migtissera/Synthia-v3.0-11B+jeiku/Re-Host_Limarp_Mistral
106
+ parameters:
107
+ weight: 0.28
108
+ density: 0.80
109
+ - model: TheDrummer/Moistral-11B-v3
110
+ parameters:
111
+ weight: 0.37
112
+ density: 0.85
113
+ merge_method: dare_ties
114
+ base_model: Sao10K/Fimbulvetr-11B-v2
115
+ parameters:
116
+ int8_mask: true
117
+ dtype: bfloat16
118
+
119
+ ```
120
+