Lewdiculous commited on
Commit
77558ed
1 Parent(s): bc7009c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +45 -0
README.md CHANGED
@@ -1,3 +1,48 @@
1
  ---
 
 
 
 
 
 
 
 
2
  license: cc-by-4.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - text-generation-inference
7
+ - instruct
8
+ - conversational
9
+ - roleplay
10
  license: cc-by-4.0
11
  ---
12
+
13
+ ## Description
14
+
15
+ This repo contains GGUF-IQ-Imatrix format model files for [Erosumika-7B-v3-0.2](https://huggingface.co/localfultonextractor/Erosumika-7B-v3-0.2).
16
+
17
+ Quants to be uploaded:
18
+
19
+ "Q4_K_M", "Q4_K_S", "IQ4_XS", "Q5_K_M", "Q5_K_S",
20
+ "Q6_K", "Q8_0", "IQ3_M", "IQ3_S", "IQ3_XXS"
21
+
22
+ <h1 style="text-align: center">Erosumika-7B-v3-0.2</h1>
23
+ <h2 style="text-align: center">~Mistral 0.2 Edition~</h1>
24
+
25
+ <div style="display: flex; justify-content: center;">
26
+ <img src="https://cdn-uploads.huggingface.co/production/uploads/6512681f4151fb1fa719e033/8YBKfcegQZliRlQNm0oir.gif" alt="Header GIF">
27
+ </div>
28
+
29
+ ## Model Details
30
+ The Mistral 0.2 version of Erosumika-7B-v3, a DARE TIES merge between Nitral's [Kunocchini-7b](https://huggingface.co/Nitral-AI/Kunocchini-7b), Endevor's [InfinityRP-v1-7B](https://huggingface.co/Endevor/InfinityRP-v1-7B) and my [FlatErosAlpha](https://huggingface.co/localfultonextractor/FlatErosAlpha), a flattened(in order to keep the vocab size 32000) version of tavtav's [eros-7B-ALPHA](https://huggingface.co/tavtav/eros-7B-ALPHA). Alpaca and ChatML work best. Slightly smarter and better prompt comprehension than Mistral 0.1 Erosumika-7B-v3. 32k context should work.
31
+
32
+ [GGUF quants](https://huggingface.co/localfultonextractor/Erosumika-7B-v3-0.2-GGUF)
33
+
34
+
35
+ ## Limitations and biases
36
+ The intended use-case for this model is fictional writing for entertainment purposes. Any other sort of usage is out of scope.
37
+ It may produce socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive. Outputs might often be factually wrong or misleading.
38
+
39
+
40
+ ```yaml
41
+ merge_method: task_arithmetic
42
+ base_model: alpindale/Mistral-7B-v0.2-hf
43
+ models:
44
+ - model: localfultonextractor/Erosumika-7B-v3
45
+ parameters:
46
+ weight: 1.0
47
+ dtype: float16
48
+ ```