zerofata commited on
Commit
8abc196
·
verified ·
1 Parent(s): 3f06675

Upload folder using huggingface_hub

Browse files
Files changed (2) hide show
  1. README.md +19 -61
  2. measurement.json +0 -0
README.md CHANGED
@@ -1,79 +1,37 @@
1
  ---
2
  base_model:
3
- - zerofata/L3.3-GeneticLemonade-Final-70B
 
 
 
 
4
  library_name: transformers
5
  tags:
6
  - mergekit
7
  - merge
8
 
9
  ---
10
- # Genetic Lemonade Final
11
 
12
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65b19c6c638328850e12d38c/0-pxxp_xcaIecQFcaIzFI.png)
13
-
14
- Inspired to learn how to merge by the Nevoria series from [SteelSkull](https://huggingface.co/Steelskull).
15
-
16
- This model is the final result of the Genetic Lemonade series.
17
-
18
- Designed for RP, this model is mostly uncensored and focused around striking a balance between writing style, creativity and intelligence.
19
-
20
- When compared to the previous Genetic Lemonade Unleashed:
21
- - Higher instruct following
22
- - Increased coherence
23
- - *potentially* slightly more uncensored
24
-
25
- ## SillyTavern Settings
26
-
27
- [Llam@ception](https://huggingface.co/Konnect1221/The-Inception-Presets-Methception-LLamaception-Qwenception/tree/main/Llam%40ception) recommended for sane defaults if unsure, import them to SillyTavern and they're plug n play.
28
-
29
- ### Sampler Settings
30
- - Temp: 0.9-1.0
31
- - MinP: 0.03-0.05
32
- - Dry: 0.8, 1.75, 4
33
-
34
- Temperature last, neutralize other samplers. This model natively strikes a balance of creativity & intelligence.
35
-
36
- ### Instruct
37
-
38
- Llama-3-Instruct-Names but you will need to uncheck "System same as user".
39
-
40
- ## Quants
41
-
42
- ### GGUF
43
-
44
- - [Static quants by mradermacher](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Final-70B-GGUF)
45
- - [iMatrix quants by mradermacher](https://huggingface.co/mradermacher/L3.3-GeneticLemonade-Final-70B-i1-GGUF)
46
-
47
- ### EXL2
48
-
49
- - [4.5bpw](https://huggingface.co/zerofata/L3.3-GeneticLemonade-Final-70B-4.5bpw-h6-exl2)
50
- - [6bpw](https://huggingface.co/zerofata/L3.3-GeneticLemonade-Final-70B-6bpw-h8-exl2)
51
 
52
  ## Merge Details
53
  ### Merge Method
54
 
55
- This model was merged using the [SCE](https://arxiv.org/abs/2408.07990) merge method.
56
 
57
- The base aims to build a strong general purpose model using high performing models that are trained on various datasets from different languages / cultures. This is to reduce the chance of the same datasets appearing multiple times to build natural creativity into L3.3
58
- The second merge aims to impart specific RP / creative writing knowledge, again focusing on trying to find high performing models that use or likely use different datasets.
59
 
60
- ### Base_6_v2
61
- ```yaml
62
- models:
63
- - model: OpenBuddy/openbuddy-llama3.3-70b-v24.1-131k
64
- - model: nbeerbower/llama3.1-kartoffeldes-70B
65
- - model: tokyotech-llm/Llama-3.1-Swallow-70B-Instruct-v0.3
66
- - model: SicariusSicariiStuff/Negative_LLAMA_70B
67
- select_topk: .15
68
- merge_method: sce
69
- base_model: meta-llama/Llama-3.3-70B-Instruct
70
- out_dtype: bfloat16
71
- dype: float32
72
- tokenizer:
73
- source: base
74
- ```
75
 
76
- ### Genetic Lemonade Final
77
 
78
  ```yaml
79
  models:
@@ -89,4 +47,4 @@ out_dtype: bfloat16
89
  dype: float32
90
  tokenizer:
91
  source: union
92
- ```
 
1
  ---
2
  base_model:
3
+ - LatitudeGames/Wayfarer-Large-70B-Llama-3.3
4
+ - Sao10K/L3.3-70B-Euryale-v2.3
5
+ - crestf411/L3.1-nemotron-sunfall-v0.7.0
6
+ - SicariusSicariiStuff/Negative_LLAMA_70B
7
+ - EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1
8
  library_name: transformers
9
  tags:
10
  - mergekit
11
  - merge
12
 
13
  ---
14
+ # GeneticLemonadev2-c
15
 
16
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
17
 
18
  ## Merge Details
19
  ### Merge Method
20
 
21
+ This model was merged using the [SCE](https://arxiv.org/abs/2408.07990) merge method using ./Base_6_v2 as a base.
22
 
23
+ ### Models Merged
 
24
 
25
+ The following models were included in the merge:
26
+ * [LatitudeGames/Wayfarer-Large-70B-Llama-3.3](https://huggingface.co/LatitudeGames/Wayfarer-Large-70B-Llama-3.3)
27
+ * [Sao10K/L3.3-70B-Euryale-v2.3](https://huggingface.co/Sao10K/L3.3-70B-Euryale-v2.3)
28
+ * [crestf411/L3.1-nemotron-sunfall-v0.7.0](https://huggingface.co/crestf411/L3.1-nemotron-sunfall-v0.7.0)
29
+ * [SicariusSicariiStuff/Negative_LLAMA_70B](https://huggingface.co/SicariusSicariiStuff/Negative_LLAMA_70B)
30
+ * [EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1](https://huggingface.co/EVA-UNIT-01/EVA-LLaMA-3.33-70B-v0.1)
31
+
32
+ ### Configuration
 
 
 
 
 
 
 
33
 
34
+ The following YAML configuration was used to produce this model:
35
 
36
  ```yaml
37
  models:
 
47
  dype: float32
48
  tokenizer:
49
  source: union
50
+ ```
measurement.json ADDED
The diff for this file is too large to render. See raw diff