Update README.md
Browse files
README.md
CHANGED
@@ -17,6 +17,26 @@ and HuggingFaceH4/ultrafeedback_binarized was created in addition to a series of
|
|
17 |
Thanks to the new type of training, Phoenix is not only able to compete with the Mistral model from LeoLM but also **beats the Llama-70b-chat model in 2 mt-bench categories**This model **wouldn't have been possible without the amazing work of Huggingface, LeoLM, openbnb, Argilla the Alma-Team and many others of the AI community**
|
18 |
|
19 |
## MT-Bench-DE Scores
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
20 |
|
21 |
## Model Details
|
22 |
|
|
|
17 |
Thanks to the new type of training, Phoenix is not only able to compete with the Mistral model from LeoLM but also **beats the Llama-70b-chat model in 2 mt-bench categories**This model **wouldn't have been possible without the amazing work of Huggingface, LeoLM, openbnb, Argilla the Alma-Team and many others of the AI community**
|
18 |
|
19 |
## MT-Bench-DE Scores
|
20 |
+
Phoenix beats the LeoLM-Mistral model in all categories except for coding and humanities.
|
21 |
+
Additionally it also Beats LeoLM/Llama-2-70b-chat in roleplay and reasoning which shows the power of DPO.
|
22 |
+
|
23 |
+
```
|
24 |
+
{
|
25 |
+
"first_turn": 6.39375,
|
26 |
+
"second_turn": 5.1625,
|
27 |
+
"categories": {
|
28 |
+
"writing": 7.45,
|
29 |
+
"roleplay": 7.9,
|
30 |
+
"reasoning": 4.3,
|
31 |
+
"math": 3.25,
|
32 |
+
"coding": 2.5,
|
33 |
+
"extraction": 5.9,
|
34 |
+
"stem": 7.125,
|
35 |
+
"humanities": 7.8
|
36 |
+
},
|
37 |
+
"average": 5.778124999999999
|
38 |
+
}
|
39 |
+
```
|
40 |
|
41 |
## Model Details
|
42 |
|