Nondzu commited on
Commit
315b039
1 Parent(s): 02903f8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -8
README.md CHANGED
@@ -7,14 +7,13 @@ tags:
7
 
8
  # Mistral-7B-codealpaca
9
 
10
- We are thrilled to introduce the Mistral-7B-codealpaca model. This variant is optimized and demonstrates potential in assisting developers as a coding companion. We welcome contributions from testers and enthusiasts to help evaluate its performance.
11
 
12
  ## Training Details
13
 
14
- The model was trained using 3xRTX 3090 for 118 hours.
15
  [![Built with Axolotl](https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png)](https://github.com/OpenAccess-AI-Collective/axolotl)
16
 
17
-
18
  ## Quantised Model Links:
19
 
20
  1.
@@ -44,11 +43,11 @@ Human eval plus: https://github.com/evalplus/evalplus
44
 
45
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/azE6LU0qQ9E9u60t5VrMk.png)
46
 
47
- Well, the results are better than I expected
48
  - Base: `{'pass@1': 0.47560975609756095}`
49
  - Base + Extra: `{'pass@1': 0.4329268292682927}`
50
 
51
- For reference, we've provided the performance of the original Mistral model alongside Mistral-7B-code-16k-qlora model.
52
 
53
  ** [Nondzu/Mistral-7B-code-16k-qlora](https://huggingface.co/Nondzu/Mistral-7B-code-16k-qlora)**:
54
 
@@ -62,7 +61,7 @@ For reference, we've provided the performance of the original Mistral model alon
62
 
63
  ## Model Configuration:
64
 
65
- The following are the configurations for the Mistral-7B-codealpaca-lora:
66
 
67
  ```yaml
68
  base_model: mistralai/Mistral-7B-Instruct-v0.1
@@ -90,10 +89,8 @@ lora_target_modules:
90
  lora_target_linear: true
91
  ```
92
 
93
-
94
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/5nPgL3ajROKf7dttf4BO0.png)
95
 
96
-
97
  ## Additional Projects:
98
 
99
  For other related projects, you can check out:
 
7
 
8
  # Mistral-7B-codealpaca
9
 
10
+ I am thrilled to introduce my Mistral-7B-codealpaca model. This variant is optimized and demonstrates potential in assisting developers as a coding companion. I welcome contributions from testers and enthusiasts to help evaluate its performance.
11
 
12
  ## Training Details
13
 
14
+ I trained the model using 3xRTX 3090 for 118 hours.
15
  [![Built with Axolotl](https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png)](https://github.com/OpenAccess-AI-Collective/axolotl)
16
 
 
17
  ## Quantised Model Links:
18
 
19
  1.
 
43
 
44
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/azE6LU0qQ9E9u60t5VrMk.png)
45
 
46
+ Well, the results are better than I expected:
47
  - Base: `{'pass@1': 0.47560975609756095}`
48
  - Base + Extra: `{'pass@1': 0.4329268292682927}`
49
 
50
+ For reference, I've provided the performance of the original Mistral model alongside my Mistral-7B-code-16k-qlora model.
51
 
52
  ** [Nondzu/Mistral-7B-code-16k-qlora](https://huggingface.co/Nondzu/Mistral-7B-code-16k-qlora)**:
53
 
 
61
 
62
  ## Model Configuration:
63
 
64
+ Here are the configurations for my Mistral-7B-codealpaca-lora:
65
 
66
  ```yaml
67
  base_model: mistralai/Mistral-7B-Instruct-v0.1
 
89
  lora_target_linear: true
90
  ```
91
 
 
92
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/63729f35acef705233c87909/5nPgL3ajROKf7dttf4BO0.png)
93
 
 
94
  ## Additional Projects:
95
 
96
  For other related projects, you can check out: