DavidGF commited on
Commit
911ad65
1 Parent(s): 0252f66

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -4
README.md CHANGED
@@ -89,7 +89,7 @@ Bitte erkläre mir, wie die Zusammenführung von Modellen durch bestehende Spitz
89
  ## Evaluation
90
 
91
  ### GPT4ALL:
92
- *Compared to Aleph Alpha Luminous Models, LeoLM and EM_German*
93
  ![GPT4ALL diagram](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All.png "SauerkrautLM-7b-HerO GPT4ALL Diagram")
94
 
95
  ![GPT4ALL table](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All-Tabelle.png "SauerkrautLM-7b-HerO GPT4ALL Table")
@@ -104,10 +104,10 @@ Bitte erkläre mir, wie die Zusammenführung von Modellen durch bestehende Spitz
104
  **performed with newest Language Model Evaluation Harness*
105
 
106
  ### MMLU:
107
- *Compared to Grok0,Grok1,GPT3.5,GPT4*
108
  ![MMLU](https://vago-solutions.de/wp-content/uploads/2023/11/MMLU-Benchmark.png "SauerkrautLM-7b-HerO MMLU")
109
  ### TruthfulQA:
110
- *Compared to GPT3.5,GPT4*
111
  ![TruthfulQA](https://vago-solutions.de/wp-content/uploads/2023/11/Truthfulqa-Benchmark.png "SauerkrautLM-7b-HerO TruthfulQA")
112
 
113
  ### MT-Bench (German):
@@ -170,6 +170,7 @@ SauerkrautLM-3b-v1 2.581250
170
  open_llama_3b_v2 1.456250
171
  Llama-2-7b 1.181250
172
  ```
 
173
  ### MT-Bench (English):
174
  ![MT-Bench English Diagram](https://vago-solutions.de/wp-content/uploads/2023/11/MT-Bench-Englisch.png "SauerkrautLM-7b-HerO MT-Bench English Diagram")
175
  ```
@@ -197,7 +198,7 @@ SauerkrautLM-7b-HerO <--- 7.409375
197
  Mistral-7B-OpenOrca 6.915625
198
  neural-chat-7b-v3-1 6.812500
199
  ```
200
-
201
 
202
  ### Additional German Benchmark results:
203
  ![GermanBenchmarks](https://vago-solutions.de/wp-content/uploads/2023/11/German-benchmarks.png "SauerkrautLM-7b-HerO German Benchmarks")
 
89
  ## Evaluation
90
 
91
  ### GPT4ALL:
92
+ *Compared to relevant German Closed and Open Source models*
93
  ![GPT4ALL diagram](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All.png "SauerkrautLM-7b-HerO GPT4ALL Diagram")
94
 
95
  ![GPT4ALL table](https://vago-solutions.de/wp-content/uploads/2023/11/GPT4All-Tabelle.png "SauerkrautLM-7b-HerO GPT4ALL Table")
 
104
  **performed with newest Language Model Evaluation Harness*
105
 
106
  ### MMLU:
107
+ *Compared to Big Boy LLMs (Grok0,Grok1,GPT3.5,GPT4)*
108
  ![MMLU](https://vago-solutions.de/wp-content/uploads/2023/11/MMLU-Benchmark.png "SauerkrautLM-7b-HerO MMLU")
109
  ### TruthfulQA:
110
+ *Compared to OpenAI Models (GPT3.5,GPT4)*
111
  ![TruthfulQA](https://vago-solutions.de/wp-content/uploads/2023/11/Truthfulqa-Benchmark.png "SauerkrautLM-7b-HerO TruthfulQA")
112
 
113
  ### MT-Bench (German):
 
170
  open_llama_3b_v2 1.456250
171
  Llama-2-7b 1.181250
172
  ```
173
+ **performed with the newest FastChat Version*
174
  ### MT-Bench (English):
175
  ![MT-Bench English Diagram](https://vago-solutions.de/wp-content/uploads/2023/11/MT-Bench-Englisch.png "SauerkrautLM-7b-HerO MT-Bench English Diagram")
176
  ```
 
198
  Mistral-7B-OpenOrca 6.915625
199
  neural-chat-7b-v3-1 6.812500
200
  ```
201
+ **performed with the newest FastChat Version*
202
 
203
  ### Additional German Benchmark results:
204
  ![GermanBenchmarks](https://vago-solutions.de/wp-content/uploads/2023/11/German-benchmarks.png "SauerkrautLM-7b-HerO German Benchmarks")