Push to HF space

README.md

@@ -12,6 +12,7 @@ datasets:
  - lilacai/lilac-open-assistant-conversations-2
  - lilacai/lilac-lmsys-chat-1m
  - lilacai/lilac-OpenOrca
+ - lilacai/lilac-OpenHermes-2.5
  - lilacai/lilac-SlimOrca
  - lilacai/lilac-UltraChat-200k
  - lilacai/lilac-roblox_luau_corpus

data/lilac.yml

@@ -110,6 +110,17 @@ datasets:
  - response
  tags:
  - datasets
+ - namespace: lilac
+ name: OpenHermes-2.5
+ source:
+ dataset_name: teknium/OpenHermes-2.5
+ source_name: huggingface
+ settings:
+ ui:
+ media_paths:
+ - - conversations
+ - '*'
+ - value
  - namespace: lilac
  name: SlimOrca
  source:
@@ -399,9 +410,6 @@ datasets:
  tags:
  - datasets
 use_garden: true
-signals:
- - signal_name: text_statistics
- - signal_name: lang_detection
 concept_model_cache_embeddings:
  - gte-small
  - gte-base
@@ -441,6 +449,13 @@ clusters:
  dataset_name: OpenOrca
  input_path: !!python/tuple
  - question
+ - dataset_namespace: lilac
+ dataset_name: OpenHermes-2.5
+ input_selector:
+ format: sharegpt
+ selector: human
+ output_path: !!python/tuple
+ - conversations__clusters
  - dataset_namespace: lilac
  dataset_name: SlimOrca
  input_selector: