tjellm commited on
Commit
4bb10bc
1 Parent(s): cb8c261

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +39 -0
README.md CHANGED
@@ -1,3 +1,42 @@
1
  ---
2
  license: apache-2.0
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: apache-2.0
3
+ datasets:
4
+ - cognitivecomputations/dolphin
5
+ - cognitivecomputations/dolphin-coder
6
+ - ise-uiuc/Magicoder-OSS-Instruct-75K
7
+ - teknium/openhermes
8
+ - migtissera/Synthia-v1.3
9
+ language:
10
+ - en
11
  ---
12
+
13
+ # Model Description
14
+
15
+ This is fine-tuned model based on EmbeddedLLM/Mistral-7B-Merge-14-v0.3 for 9984 steps.
16
+
17
+ The dataset used are:
18
+
19
+ * dophin
20
+ * dolphin-coder
21
+ * Magicoder-OSS-Instruct-75K
22
+ * openhermes
23
+ * Synthia-v1.3
24
+
25
+ ## Chat Template
26
+
27
+ Prompt format: This model uses ChatML prompt format.
28
+ ```
29
+ <|im_start|>system
30
+ You are Dolphin, a helpful AI assistant.<|im_end|>
31
+ <|im_start|>user
32
+ {prompt}<|im_end|>
33
+ <|im_start|>assistant
34
+ ```
35
+
36
+ # Training
37
+
38
+ The model is scheduled to be fine-tuned for 3 epochs on 4 A100s using axolotl.
39
+
40
+ # Shout-Out to OSS
41
+
42
+ Thank you to the Open Source AI community for bringing together marvelous code frameworks and datasets.