jphme commited on
Commit
3200f7c
1 Parent(s): 538382c

update readme

Browse files
Files changed (1) hide show
  1. README.md +5 -5
README.md CHANGED
@@ -22,7 +22,7 @@ tags:
22
 
23
  **Many thanks to the [LeoLM](https://huggingface.co/LeoLM) team for the publication of a base model that has received continued pretraining with German texts, greatly improving generation capabilities.**
24
 
25
- *If you get unsatisfying results with the LeoLM-based model version, please try setting `rope_scaling` to `2.0` manually, removing `repetition_penalty` and/or using a different model or version for your usecase.*
26
 
27
  # Table of Contents
28
 
@@ -42,14 +42,14 @@ tags:
42
 
43
  We offer versions based on 7b, 13b and 70b Llama-2, Mistral and LeoLM (Llama-2/Mistral with continued pretraining on German texts) models.
44
 
45
- Please find all Informations, Example Outputs, the special RAG prompt format, output examples and eval results for the EM German Model family in [our Github Repository](https://github.com/jphme/EM_German). ([Deutsche Version](https://github.com/jphme/EM_German/blob/main/README_DE.md))
46
 
47
 
48
  # Links & Demos
49
 
50
  ## Model Links
51
 
52
- Should you try only one model version, I strongly recommend the **LeoLM Mistral** model which offers by far the best combination of performance and computing requirements!
53
 
54
  | Base Model | HF | GPTQ | GGUF | AWQ |
55
  |-------|-------|-------|-------|-------|
@@ -59,7 +59,7 @@ Should you try only one model version, I strongly recommend the **LeoLM Mistral*
59
  | [Mistral 7b](https://huggingface.co/mistralai/Mistral-7B-v0.1) | [Link](https://huggingface.co/jphme/em_german_mistral_v01) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-GPTQ) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-GGUF) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-AWQ) |
60
  | [LeoLM 7b](https://huggingface.co/LeoLM/leo-hessianai-7b) | [Link](https://huggingface.co/jphme/em_german_7b_leo) | [Link](https://huggingface.co/jphme/em_german_7b_leo_gptq) | [Link](hhttps://huggingface.co/jphme/em_german_7b_leo_gguf) | tbc |
61
  | [LeoLM 13b](https://huggingface.co/LeoLM/leo-hessianai-13b) | soon | soon | [Link](https://huggingface.co/jphme/em_german_13b_leo_gguf) | tbc |
62
- | [LeoLM Mistral 7b](tbc) | [Link](https://huggingface.co/jphme/em_german_leo_mistral) | soon | [Link](https://huggingface.co/jphme/em_german_leo_mistral_gguf) | tbc |
63
 
64
  ### Notes about the different versions:
65
  See also the [comparison of example outputs](https://github.com/jphme/EM_German/blob/main/example_output_comparison.md) for a comparison of (7b) model capabilities.
@@ -133,7 +133,7 @@ The 70b model was trained with support of the [OVH Cloud Startup Program](https:
133
 
134
  # Contact
135
 
136
- I you are interested in customized LLMs for business applications, please get in contact with me via [my website](https://www.jph.me). I am also always happy about suggestions and feedback.
137
 
138
  *PS: We are also always interested in support for our startup [ellamind](https://ellamind.com), which will offer customized models for business applications in the future (we are currently still in stealth mode). If you use our models for business applications and have advanced needs for specialized capabilities, please get in touch.*
139
 
 
22
 
23
  **Many thanks to the [LeoLM](https://huggingface.co/LeoLM) team for the publication of a base model that has received continued pretraining with German texts, greatly improving generation capabilities.**
24
 
25
+ *If you get unsatisfying results with the LeoLM-based model version, please try setting `rope_scaling` to `2.0` manually, removing `repetition_penalty` and/or using a different model or version for your usecase (e.g. the Mistral-based version).*
26
 
27
  # Table of Contents
28
 
 
42
 
43
  We offer versions based on 7b, 13b and 70b Llama-2, Mistral and LeoLM (Llama-2/Mistral with continued pretraining on German texts) models.
44
 
45
+ Please find all Informations, Example Outputs, the special RAG prompt format, output examples and eval results for the EM German Model family in [our Github Repository](https://github.com/jphme/EM_German). ([Deutsche Version](https://github.com/jphme/EM_German/blob/main/README_DE.md)). You will also find instructions on how to run the models with a GUI (GPT4All/LM Studio).
46
 
47
 
48
  # Links & Demos
49
 
50
  ## Model Links
51
 
52
+ Should you only try one model version, I strongly recommend the **[LeoLM Mistral](https://huggingface.co/jphme/em_german_leo_mistral)** model which offers by far the best combination of performance and computing requirements!
53
 
54
  | Base Model | HF | GPTQ | GGUF | AWQ |
55
  |-------|-------|-------|-------|-------|
 
59
  | [Mistral 7b](https://huggingface.co/mistralai/Mistral-7B-v0.1) | [Link](https://huggingface.co/jphme/em_german_mistral_v01) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-GPTQ) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-GGUF) | [Link](https://huggingface.co/TheBloke/em_german_mistral_v01-AWQ) |
60
  | [LeoLM 7b](https://huggingface.co/LeoLM/leo-hessianai-7b) | [Link](https://huggingface.co/jphme/em_german_7b_leo) | [Link](https://huggingface.co/jphme/em_german_7b_leo_gptq) | [Link](hhttps://huggingface.co/jphme/em_german_7b_leo_gguf) | tbc |
61
  | [LeoLM 13b](https://huggingface.co/LeoLM/leo-hessianai-13b) | soon | soon | [Link](https://huggingface.co/jphme/em_german_13b_leo_gguf) | tbc |
62
+ | [LeoLM Mistral](https://huggingface.co/LeoLM/leo-mistral-hessianai-7b) | [Link](https://huggingface.co/jphme/em_german_leo_mistral) | [Link](https://huggingface.co/TheBloke/em_german_leo_mistral-GPTQ) | [Link](https://huggingface.co/TheBloke/em_german_leo_mistral-GGUF) | [Link](https://huggingface.co/TheBloke/em_german_leo_mistral-AWQ) |
63
 
64
  ### Notes about the different versions:
65
  See also the [comparison of example outputs](https://github.com/jphme/EM_German/blob/main/example_output_comparison.md) for a comparison of (7b) model capabilities.
 
133
 
134
  # Contact
135
 
136
+ For detailed feedback & feature requests, please open an issue or get in contact with me via [my website](https://www.jph.me).
137
 
138
  *PS: We are also always interested in support for our startup [ellamind](https://ellamind.com), which will offer customized models for business applications in the future (we are currently still in stealth mode). If you use our models for business applications and have advanced needs for specialized capabilities, please get in touch.*
139