Spaces:
Paused
Paused
not starting
Browse files
README.md
CHANGED
@@ -10,7 +10,10 @@ pinned: false
|
|
10 |
preload_from_hub:
|
11 |
- "BAAI/bge-small-en-v1.5"
|
12 |
- "HuggingFaceH4/zephyr-7b-alpha"
|
13 |
-
- "microsoft/phi-2"
|
14 |
---
|
|
|
15 |
- "meta-llama/Meta-Llama-3-8B"
|
16 |
|
|
|
|
|
|
|
|
10 |
preload_from_hub:
|
11 |
- "BAAI/bge-small-en-v1.5"
|
12 |
- "HuggingFaceH4/zephyr-7b-alpha"
|
|
|
13 |
---
|
14 |
+
- "microsoft/phi-2"
|
15 |
- "meta-llama/Meta-Llama-3-8B"
|
16 |
|
17 |
+
Follow-up with langchain: https://medium.com/pythoneers/offline-rag-with-llamaindex-and-tiny-and-small-llms-ab2acac936b0
|
18 |
+
https://github.com/jeremy-k3/notebooks/blob/main/RAG_with_LlamaIndex_tiny_small_LLMS.ipynb
|
19 |
+
https://www.kaggle.com/code/iamleonie/advanced-rag-with-gemma-weaviate-and-llamaindex#Step-6:-Explore-a-Naive-RAG-Query-Engine
|
app.py
CHANGED
@@ -20,9 +20,10 @@ CHEAPMODE = torch.cuda.is_available()
|
|
20 |
# LLM = "HuggingFaceH4/zephyr-7b-alpha" if not CHEAPMODE else "microsoft/phi-2"
|
21 |
|
22 |
config = {
|
23 |
-
"LLM": "meta-llama/Meta-Llama-3-8B",
|
|
|
24 |
# "LLM": "microsoft/phi-2",
|
25 |
-
|
26 |
"embeddings": "BAAI/bge-small-en-v1.5",
|
27 |
"similarity_top_k": 2,
|
28 |
"context_window": 2048,
|
|
|
20 |
# LLM = "HuggingFaceH4/zephyr-7b-alpha" if not CHEAPMODE else "microsoft/phi-2"
|
21 |
|
22 |
config = {
|
23 |
+
# "LLM": "meta-llama/Meta-Llama-3-8B",
|
24 |
+
# "LLM": "google/gemma-2b",
|
25 |
# "LLM": "microsoft/phi-2",
|
26 |
+
"LLM": "HuggingFaceH4/zephyr-7b-alpha",
|
27 |
"embeddings": "BAAI/bge-small-en-v1.5",
|
28 |
"similarity_top_k": 2,
|
29 |
"context_window": 2048,
|