afrideva commited on
Commit
7c59352
1 Parent(s): d2490bf

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +65 -0
README.md ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Aryanne/Astrea-RP-v1-3B
3
+ inference: false
4
+ language:
5
+ - en
6
+ library_name: transformers
7
+ license: other
8
+ model_creator: Aryanne
9
+ model_name: Astrea-RP-v1-3B
10
+ pipeline_tag: text-generation
11
+ quantized_by: afrideva
12
+ tags:
13
+ - gpt
14
+ - llm
15
+ - large language model
16
+ - gguf
17
+ - ggml
18
+ - quantized
19
+ - q2_k
20
+ - q3_k_m
21
+ - q4_k_m
22
+ - q5_k_m
23
+ - q6_k
24
+ - q8_0
25
+ ---
26
+ # Aryanne/Astrea-RP-v1-3B-GGUF
27
+
28
+ Quantized GGUF model files for [Astrea-RP-v1-3B](https://huggingface.co/Aryanne/Astrea-RP-v1-3B) from [Aryanne](https://huggingface.co/Aryanne)
29
+
30
+
31
+ | Name | Quant method | Size |
32
+ | ---- | ---- | ---- |
33
+ | [astrea-rp-v1-3b.fp16.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.fp16.gguf) | fp16 | 5.59 GB |
34
+ | [astrea-rp-v1-3b.q2_k.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q2_k.gguf) | q2_k | 1.20 GB |
35
+ | [astrea-rp-v1-3b.q3_k_m.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q3_k_m.gguf) | q3_k_m | 1.39 GB |
36
+ | [astrea-rp-v1-3b.q4_k_m.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q4_k_m.gguf) | q4_k_m | 1.71 GB |
37
+ | [astrea-rp-v1-3b.q5_k_m.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q5_k_m.gguf) | q5_k_m | 1.99 GB |
38
+ | [astrea-rp-v1-3b.q6_k.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q6_k.gguf) | q6_k | 2.30 GB |
39
+ | [astrea-rp-v1-3b.q8_0.gguf](https://huggingface.co/afrideva/Astrea-RP-v1-3B-GGUF/resolve/main/astrea-rp-v1-3b.q8_0.gguf) | q8_0 | 2.97 GB |
40
+
41
+
42
+
43
+ ## Original Model Card:
44
+ This model is a merge of [euclaise/Echo-3B](https://huggingface.co/euclaise/Echo-3B), [stabilityai/stablelm-zephyr-3b](https://huggingface.co/stabilityai/stablelm-zephyr-3b) and [Aryanne/Astridboros-3B](https://huggingface.co/Aryanne/Astridboros-3B) using task_arithmetic(see astrea-rp-v1-3b.yml or below).
45
+
46
+
47
+ ```yaml
48
+ merge_method: task_arithmetic
49
+ base_model: euclaise/Ferret-3B
50
+ models:
51
+ - model: euclaise/Ferret-3B
52
+ - model: stabilityai/stablelm-zephyr-3b
53
+ parameters:
54
+ weight: 0.33
55
+ - model: euclaise/Echo-3B
56
+ parameters:
57
+ weight: 0.66
58
+ - model: Aryanne/Astridboros-3B
59
+ parameters:
60
+ weight: 0.16
61
+ dtype: float16
62
+ ```
63
+ I recommend the use of Vicuna prompt format, but it's your choice to see what works for you.
64
+
65
+ I think zephyr license applies to this merge, for non commercial use.