inflatebot commited on
Commit
aae1948
·
verified ·
1 Parent(s): b77f50d

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +13 -1
README.md CHANGED
@@ -9,11 +9,23 @@ tags:
9
  - merge
10
 
11
  ---
 
 
12
  # merge
13
 
14
  This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
15
 
16
  ## Merge Details
 
 
 
 
 
 
 
 
 
 
17
  ### Merge Method
18
 
19
  This model was merged using the [task arithmetic](https://arxiv.org/abs/2212.04089) merge method using [NousResearch/Meta-Llama-3-8B](https://huggingface.co/NousResearch/Meta-Llama-3-8B) as a base.
@@ -35,7 +47,7 @@ models:
35
  weight: 0.5
36
  - model: Fizzarolli/L3-8b-Rosier-v1
37
  parameters:
38
- weight: 0.33
39
 
40
  merge_method: task_arithmetic
41
  base_model: NousResearch/Meta-Llama-3-8B
 
9
  - merge
10
 
11
  ---
12
+ ![By NovelAI](https://huggingface.co/inflatebot/helide-alpha-r2/resolve/main/img.png)
13
+ `"Helide" (say HE-lied) is an ion of helium -- famously a very unreactive element, which doesn't form ions in most conditions.`
14
  # merge
15
 
16
  This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
17
 
18
  ## Merge Details
19
+
20
+ An experimental merge of the legendary L3-8B-Stheno with Fizzarolli's Rosier. The aim is to improve Stheno's "ball-rolling" capabilities and reduce its awkwardness with more niche content. For a first go, I'm surprised at how well it's doing so far, but given that this is literally my first LLM project ever, probably temper your expectations.
21
+
22
+ Since R1: Changed to task-arithmetic. Snazzy new model card image.
23
+
24
+ Since R2: Fixed unnecessary conversion.
25
+
26
+ Since R3: Tweaked ratios, Rosier's influence cut in half.
27
+
28
+ Since R4: Scrubbin' it down. +0.13 to Rosier. Closing in on a good ratio.
29
  ### Merge Method
30
 
31
  This model was merged using the [task arithmetic](https://arxiv.org/abs/2212.04089) merge method using [NousResearch/Meta-Llama-3-8B](https://huggingface.co/NousResearch/Meta-Llama-3-8B) as a base.
 
47
  weight: 0.5
48
  - model: Fizzarolli/L3-8b-Rosier-v1
49
  parameters:
50
+ weight: 0.25
51
 
52
  merge_method: task_arithmetic
53
  base_model: NousResearch/Meta-Llama-3-8B