Pankaj Mathur
commited on
Commit
•
a9e3300
1
Parent(s):
16843a5
Update README.md
Browse files
README.md
CHANGED
@@ -5,7 +5,7 @@ license: cc-by-sa-4.0
|
|
5 |
<div style="width: 800px; margin: auto;">
|
6 |
|
7 |
<h2>Model Description</h2>
|
8 |
-
<p>“Luna AI Llama2 Uncensored” is a Llama2 based Chat model <br />fine-tuned on over 40,000
|
9 |
This model was fine-tuned by Tap, the creator of Luna AI. <br />
|
10 |
The result is an enhanced Llama2 7b model that rivals ChatGPT in performance <br />across a variety of tasks.</p>
|
11 |
<p>This model stands out for its long responses, low hallucination rate, and absence of censorship mechanisms. <br />
|
@@ -20,16 +20,27 @@ license: cc-by-sa-4.0
|
|
20 |
<h2>Prompt Format</h2>
|
21 |
<p>The model follows the Vicuna 1.1/ OpenChat format:</p>
|
22 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
|
24 |
<h2>Future Plans</h2>
|
25 |
<p>The model is currently being uploaded in FP16 format, <br />and there are plans to convert the model to GGML and GPTQ 4bit quantizations.</p>
|
26 |
|
27 |
<h2>Benchmark Results</h2>
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
|
32 |
-
|
|
|
|
|
|
|
|
|
33 |
|
34 |
<h2>Ethical considerations</h2>
|
35 |
<p>The data used to train the model is collected from various sources, mostly from the Web. <br />
|
|
|
5 |
<div style="width: 800px; margin: auto;">
|
6 |
|
7 |
<h2>Model Description</h2>
|
8 |
+
<p>“Luna AI Llama2 Uncensored” is a Llama2 based Chat model <br />fine-tuned on over 40,000 long form chat discussions <br />
|
9 |
This model was fine-tuned by Tap, the creator of Luna AI. <br />
|
10 |
The result is an enhanced Llama2 7b model that rivals ChatGPT in performance <br />across a variety of tasks.</p>
|
11 |
<p>This model stands out for its long responses, low hallucination rate, and absence of censorship mechanisms. <br />
|
|
|
20 |
<h2>Prompt Format</h2>
|
21 |
<p>The model follows the Vicuna 1.1/ OpenChat format:</p>
|
22 |
|
23 |
+
```
|
24 |
+
USER: I have difficulties in making friends, and I really need someone to talk to. Would you be my friend?
|
25 |
+
|
26 |
+
ASSISTANT: Of course! Friends are always here for each other. What do you like to do?
|
27 |
+
|
28 |
+
```
|
29 |
+
|
30 |
|
31 |
<h2>Future Plans</h2>
|
32 |
<p>The model is currently being uploaded in FP16 format, <br />and there are plans to convert the model to GGML and GPTQ 4bit quantizations.</p>
|
33 |
|
34 |
<h2>Benchmark Results</h2>
|
35 |
+
|
36 |
+
||||||
|
37 |
+
|---:|---:|---:|---:|---:|
|
38 |
+
|Task|Version| Metric |Value |Stderr|
|
39 |
+
|arc_challenge|0|acc_norm|0.5512|0.0146|
|
40 |
+
|hellaswag|0||||
|
41 |
+
|mmlu|0||||
|
42 |
+
|truthfulqa_mc|1|mc2|0.4716|0.0155|
|
43 |
+
|Average|-|-|0.5114|0.0150|
|
44 |
|
45 |
<h2>Ethical considerations</h2>
|
46 |
<p>The data used to train the model is collected from various sources, mostly from the Web. <br />
|