Text Generation
Transformers
Safetensors
phi-msft
custom_code
Inference Endpoints
rhysjones commited on
Commit
b77cc20
1 Parent(s): 0adc64e

Add Ollama run prompt

Browse files
Files changed (1) hide show
  1. README.md +8 -0
README.md CHANGED
@@ -30,6 +30,14 @@ And then a DPO finetune using:
30
  - [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs)
31
  - [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned)
32
 
 
 
 
 
 
 
 
 
33
  # Prompt Format
34
 
35
  Phi-2 Orange uses ChatML as the prompt format, with or without the system instruction.
 
30
  - [Intel/orca_dpo_pairs](https://huggingface.co/datasets/Intel/orca_dpo_pairs)
31
  - [argilla/ultrafeedback-binarized-preferences-cleaned](https://huggingface.co/datasets/argilla/ultrafeedback-binarized-preferences-cleaned)
32
 
33
+
34
+ # Run within Ollama
35
+
36
+ If you're using [Ollama](https://ollama.ai), you can download and run using:
37
+ ```
38
+ ollama run rhysjones/phi-2-orange
39
+ ```
40
+
41
  # Prompt Format
42
 
43
  Phi-2 Orange uses ChatML as the prompt format, with or without the system instruction.