Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,11 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: llama3
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
#LLaMA-3-8B-RDF-Experiment
|
| 6 |
+
## Purpose
|
| 7 |
+
This model is an experimental model to see if LLaMA-3-8B can be used to construct knowledge graph triples. The model is a finetune of NousResearch/Hermes-2-Pro-Llama-3-8B. Finetuning was completed on Unsloth using qLoRA and then merged back to 16-bit.
|
| 8 |
+
|
| 9 |
+
## Prompt Template
|
| 10 |
+
It is recommended that you use the apply_chat_template feature.
|
| 11 |
+
This is the recommened system prompt: `"""You are an expert knowledge graph annotator and you respond in JSON. Here's the json schema you must adhere to where each element is a new triple if needed:\n<schema>\n[{"subject": str, "predicate": str, "object": str},...{"subject": str, "predicate": str, "object": str}]\n</schema>"""`
|