better readme
Browse files
README.md
CHANGED
@@ -8,6 +8,10 @@ license: apache-2.0
|
|
8 |
|
9 |
This is the model for SauerkrautLM-UNA-SOLAR-Instruct. I used [mergekit](https://github.com/cg123/mergekit) to merge models.
|
10 |
|
|
|
|
|
|
|
|
|
11 |
# Prompt Template(s)
|
12 |
|
13 |
```
|
@@ -42,3 +46,34 @@ tokenizer_source: union
|
|
42 |
|
43 |
dtype: bfloat16
|
44 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
This is the model for SauerkrautLM-UNA-SOLAR-Instruct. I used [mergekit](https://github.com/cg123/mergekit) to merge models.
|
10 |
|
11 |
+
🥳 This model is currently (24 December 2023) **1'st place** in [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard).
|
12 |
+
|
13 |
+
<h2><details><summary>Screenshot</summary><img src=https://cdn-uploads.huggingface.co/production/uploads/6468ce47e134d050a58aa89c/cVhjAJhuPoNgHo7CDCmA-.png></img></details></h2>
|
14 |
+
|
15 |
# Prompt Template(s)
|
16 |
|
17 |
```
|
|
|
46 |
|
47 |
dtype: bfloat16
|
48 |
```
|
49 |
+
|
50 |
+
# Quantizationed versions
|
51 |
+
|
52 |
+
Quantizationed versions of this model is available thanks to [TheBloke](https://hf.co/TheBloke).
|
53 |
+
|
54 |
+
##### GPTQ
|
55 |
+
|
56 |
+
- [TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-GPTQ](https://huggingface.co/TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-GPTQ)
|
57 |
+
|
58 |
+
##### GGUF
|
59 |
+
|
60 |
+
- [TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-GGUF](https://huggingface.co/TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-GGUF)
|
61 |
+
|
62 |
+
##### AWQ
|
63 |
+
|
64 |
+
- [TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-AWQ](https://huggingface.co/TheBloke/SauerkrautLM-UNA-SOLAR-Instruct-AWQ)
|
65 |
+
|
66 |
+
|
67 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
68 |
+
|
69 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_Weyaxi__SauerkrautLM-UNA-SOLAR-Instruct)
|
70 |
+
|
71 |
+
| Metric | Value |
|
72 |
+
|-----------------------|---------------------------|
|
73 |
+
| Avg. | 74.26 |
|
74 |
+
| ARC (25-shot) | 70.9 |
|
75 |
+
| HellaSwag (10-shot) | 88.3 |
|
76 |
+
| MMLU (5-shot) | 66.15 |
|
77 |
+
| TruthfulQA (0-shot) | 71.8 |
|
78 |
+
| Winogrande (5-shot) | 83.74 |
|
79 |
+
| GSM8K (5-shot) | 64.67 |
|