fuzzy-mittenz commited on
Commit
490f1a4
·
verified ·
1 Parent(s): a6096dc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -4
README.md CHANGED
@@ -8,11 +8,13 @@ tags:
8
  <div align="center">
9
  <h1>
10
  !-Open Source Week-!
 
 
11
  </h1>
12
  </div>
13
 
14
  > [!TIP]
15
- > This is an experimental Long context model, please use and transfer with caution.
16
 
17
 
18
 
@@ -20,13 +22,13 @@ tags:
20
 
21
  A transformative leap in Open Source tech and local intelligence
22
 
23
- Presenting in Ovation. With a reckless disregard for the rules this model brings accuracy in coding and mathematics to the masses. No GPU needed just electricity silicon and the urge to bend fate to your slimey little human hands. It will get you where you need to go and fast. This model beats o1/o3 and NVIDIA's Llama 3.1-nematron 70B and other models which look gigantically down at this little cabbi just doing it's thing. The sheer amount of ability this model provides can't be attributed to "Unhobbling gains" and reaches the frontier in IFEval and Mathematics beating all service AIs in short for non reasoning answerts. Be sure to open up the context and remove the governor on this puppy. It is uncensored and may be VERY steerable for the end user. System templates below for reasoning and multi-in-turn tool use.
24
  ![CAB driver.png](https://cdn-uploads.huggingface.co/production/uploads/6593502ca2607099284523db/uoYxBevaw4WKigvOacmHK.png)
25
 
26
 
27
- This model was converted to GGUF format from [`gz987/qwen2.5-7b-cabs-v0.3`](https://huggingface.co/gz987/qwen2.5-7b-cabs-v0.3) using a customized and curated Imatrix for smoothing using QAT and TTT* methods the math to increase code functionality and tool use don't be shy about commenting if you have questions or need help and you can reach us at [email protected]
28
 
29
- ## Custom system template for advanced reasoning and tool use
30
  ```
31
  {%- set system_message = 'You are a helpful assistant.' %}
32
  {%- if messages[0]['role'] == 'system' %}
 
8
  <div align="center">
9
  <h1>
10
  !-Open Source Week-!
11
+
12
+ UNCENSORED
13
  </h1>
14
  </div>
15
 
16
  > [!TIP]
17
+ > This is an experimental Long context model, please use S-AGI with caution.
18
 
19
 
20
 
 
22
 
23
  A transformative leap in Open Source tech and local intelligence
24
 
25
+ Presenting in Ovation. With a reckless disregard for the rules this model brings accuracy in coding and mathematics to the masses. No GPU needed just electricity, silicon and the urge to bend fate to your slimey little human hands. This VEGA base will get you where you need to go and fast. It might not be able to hang the drapes, walk your kids and teach your dog japanese all before coffee but this motherlover beats o1/o3 and NVIDIA's Llama 3.1-nematron 70B and other models which look gigantically down at this little cabbi just doing it's thing. The sheer amount of ability this model provides can't be attributed to "Unhobbling gains" and it reaches the frontier in IFEval and Mathematics beating all service AIs in short form non reasoning answers mathmatics and one shot coding. Be sure to open up the context and remove the governor on this puppy. It is uncensored and may be VERY steerable for the end user. System templates below for reasoning and multi-in-turn tool use.
26
  ![CAB driver.png](https://cdn-uploads.huggingface.co/production/uploads/6593502ca2607099284523db/uoYxBevaw4WKigvOacmHK.png)
27
 
28
 
29
+ This model was converted to GGUF format from [`gz987/qwen2.5-7b-cabs-v0.3`](https://huggingface.co/gz987/qwen2.5-7b-cabs-v0.3) using a customized and curated Impotance matrix for smoothing using QAT and TTT* methods to increase math/code functionality and tool use. Don't be shy about commenting if you have questions or need help and you can reach us at [email protected]
30
 
31
+ ## Custom alternate (Grok3 style) system template for advanced reasoning and tool use
32
  ```
33
  {%- set system_message = 'You are a helpful assistant.' %}
34
  {%- if messages[0]['role'] == 'system' %}