Hugging Face's logo

RichardErkhov
/
arcee-ai_-_Saul-Instruct-Clown-7b-gguf

GGUF
Inference Endpoints
Model card Files Community
RichardErkhov commited on
Commit
fb28a52
•
1 Parent(s): de7f606

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +93 -0
README.md ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ Saul-Instruct-Clown-7b - GGUF
11
+ - Model creator: https://huggingface.co/arcee-ai/
12
+ - Original model: https://huggingface.co/arcee-ai/Saul-Instruct-Clown-7b/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [Saul-Instruct-Clown-7b.Q2_K.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q2_K.gguf) | Q2_K | 2.53GB |
18
+ | [Saul-Instruct-Clown-7b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.IQ3_XS.gguf) | IQ3_XS | 2.81GB |
19
+ | [Saul-Instruct-Clown-7b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.IQ3_S.gguf) | IQ3_S | 2.96GB |
20
+ | [Saul-Instruct-Clown-7b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q3_K_S.gguf) | Q3_K_S | 2.95GB |
21
+ | [Saul-Instruct-Clown-7b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.IQ3_M.gguf) | IQ3_M | 3.06GB |
22
+ | [Saul-Instruct-Clown-7b.Q3_K.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q3_K.gguf) | Q3_K | 3.28GB |
23
+ | [Saul-Instruct-Clown-7b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q3_K_M.gguf) | Q3_K_M | 3.28GB |
24
+ | [Saul-Instruct-Clown-7b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q3_K_L.gguf) | Q3_K_L | 3.56GB |
25
+ | [Saul-Instruct-Clown-7b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.IQ4_XS.gguf) | IQ4_XS | 3.67GB |
26
+ | [Saul-Instruct-Clown-7b.Q4_0.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q4_0.gguf) | Q4_0 | 3.83GB |
27
+ | [Saul-Instruct-Clown-7b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.IQ4_NL.gguf) | IQ4_NL | 3.87GB |
28
+ | [Saul-Instruct-Clown-7b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q4_K_S.gguf) | Q4_K_S | 3.86GB |
29
+ | [Saul-Instruct-Clown-7b.Q4_K.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q4_K.gguf) | Q4_K | 4.07GB |
30
+ | [Saul-Instruct-Clown-7b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q4_K_M.gguf) | Q4_K_M | 4.07GB |
31
+ | [Saul-Instruct-Clown-7b.Q4_1.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q4_1.gguf) | Q4_1 | 4.24GB |
32
+ | [Saul-Instruct-Clown-7b.Q5_0.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q5_0.gguf) | Q5_0 | 4.65GB |
33
+ | [Saul-Instruct-Clown-7b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q5_K_S.gguf) | Q5_K_S | 4.65GB |
34
+ | [Saul-Instruct-Clown-7b.Q5_K.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q5_K.gguf) | Q5_K | 4.78GB |
35
+ | [Saul-Instruct-Clown-7b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q5_K_M.gguf) | Q5_K_M | 4.78GB |
36
+ | [Saul-Instruct-Clown-7b.Q5_1.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q5_1.gguf) | Q5_1 | 5.07GB |
37
+ | [Saul-Instruct-Clown-7b.Q6_K.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q6_K.gguf) | Q6_K | 5.53GB |
38
+ | [Saul-Instruct-Clown-7b.Q8_0.gguf](https://huggingface.co/RichardErkhov/arcee-ai_-_Saul-Instruct-Clown-7b-gguf/blob/main/Saul-Instruct-Clown-7b.Q8_0.gguf) | Q8_0 | 7.17GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ license: apache-2.0
46
+ tags:
47
+ - merge
48
+ - mergekit
49
+ - CorticalStack/pastiche-crown-clown-7b-dare-dpo
50
+ - Equall/Saul-Instruct-v1
51
+ ---
52
+
53
+ ![image/webp](https://cdn.prod.arcee.ai/images/saul-calme.jpeg)
54
+
55
+ # Saul-Instruct-Clown-7b
56
+
57
+ Saul-Instruct-Clown-7b is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
58
+ * [CorticalStack/pastiche-crown-clown-7b-dare-dpo](https://huggingface.co/CorticalStack/pastiche-crown-clown-7b-dare-dpo)
59
+ * [Equall/Saul-Instruct-v1](https://huggingface.co/Equall/Saul-Instruct-v1)
60
+
61
+ ## 🏆 Evaluation
62
+
63
+ ### OpenLLM
64
+
65
+ Saul-Instruct-Clown-7b OpenLLM benchmark suite
66
+
67
+ | Model | Average | arc | HellaSwag | mmlu | TruthfulQA | gsm8k |
68
+ |---|---:|---:|---:|---:|---:|---:|
69
+ | [arcee-ai/Saul-Instruct-Clown-7b](https://huggingface.co/arcee-ai/Saul-Instruct-Clown-7b/) | 72.79 | 68.26 | 86.28 | 63.12 | 64.68 | 83.43 | 70.96 |
70
+
71
+
72
+ ## 🧩 Configuration
73
+
74
+ ```yaml
75
+ slices:
76
+ - sources:
77
+ - model: CorticalStack/pastiche-crown-clown-7b-dare-dpo
78
+ layer_range: [0, 32]
79
+ - model: Equall/Saul-Instruct-v1
80
+ layer_range: [0, 32]
81
+ merge_method: slerp
82
+ base_model: CorticalStack/pastiche-crown-clown-7b-dare-dpo
83
+ parameters:
84
+ t:
85
+ - filter: self_attn
86
+ value: [0, 0.5, 0.3, 0.7, 1]
87
+ - filter: mlp
88
+ value: [1, 0.5, 0.7, 0.3, 0]
89
+ - value: 0.5
90
+ dtype: bfloat16
91
+
92
+ ```
93
+