teknium commited on
Commit
fbe9fc8
1 Parent(s): dc1633f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -1
README.md CHANGED
@@ -5,7 +5,7 @@ This is the GPTQ 4Bit Groupsize 128 Pre-Quantized Model, for the full model in f
5
 
6
  As a base model used https://huggingface.co/eachadea/vicuna-13b-1.1
7
 
8
- Finetuned on Teknium's GPTeacher dataset, unreleased Roleplay v2 dataset, GPT-4-LLM dataset, and Nous Research Instruct Dataset
9
 
10
  Approx 180k instructions, all from GPT-4, all cleaned of any OpenAI censorship/"As an AI Language Model" etc.
11
 
@@ -13,6 +13,25 @@ Base model still has OpenAI censorship. Soon, a new version will be released wit
13
 
14
  Trained on 8 A100-80GB GPUs for 5 epochs following Alpaca deepspeed training code.
15
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  Nous Research Instruct Dataset will be released soon.
17
 
18
  GPTeacher, Roleplay v2 by https://huggingface.co/teknium
 
5
 
6
  As a base model used https://huggingface.co/eachadea/vicuna-13b-1.1
7
 
8
+ Finetuned on Teknium's GPTeacher dataset, Teknium's unreleased Roleplay v2 dataset, WizardLM Uncensored, GPT-4-LLM Uncensored, and Nous Research Instruct Dataset
9
 
10
  Approx 180k instructions, all from GPT-4, all cleaned of any OpenAI censorship/"As an AI Language Model" etc.
11
 
 
13
 
14
  Trained on 8 A100-80GB GPUs for 5 epochs following Alpaca deepspeed training code.
15
 
16
+ Prompt format is Alpaca:
17
+ ```
18
+ ### Instruction:
19
+
20
+ ### Response:
21
+ ```
22
+
23
+ or
24
+
25
+ ```
26
+ ### Instruction:
27
+
28
+ ### Input:
29
+
30
+ ### Response:
31
+
32
+ ```
33
+
34
+
35
  Nous Research Instruct Dataset will be released soon.
36
 
37
  GPTeacher, Roleplay v2 by https://huggingface.co/teknium