feat: update yml chat_template to specify dataset field (#2001) [skip ci]

* feat: update yml chat_template to specify dataset field * feat: replace sharegpt references with chat_template
2024-10-29 21:26:03 +07:00
parent 107b67b852
commit 8c3a727f9d
7 changed files with 26 additions and 13 deletions
--- a/examples/jamba/qlora_fsdp_large.yaml
+++ b/examples/jamba/qlora_fsdp_large.yaml
@@ -4,11 +4,15 @@ tokenizer_type: AutoTokenizer
 load_in_4bit: true
 strict: false
 use_tensorboard: true
+chat_template: jamba
 datasets:
  - path: cgato/SlimOrcaDedupCleaned
    type: chat_template
-    chat_template: jamba
    drop_system_message: true
+    field_messages: conversations
+    message_field_role: from
+    message_field_content: value
+
 dataset_prepared_path: last_run_prepared
 val_set_size: 0.0
 output_dir: jamba-large-fsdp-qlora-ft