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

examples/deepseek-v2/qlora-fsdp-2_5.yaml

@@ -16,7 +16,10 @@ chat_template: deepseek_v2
 datasets:
   - path: mlabonne/FineTome-100k
     type: chat_template
-    split: train
+    split: train[:20%]
+    field_messages: conversations
+    message_field_role: from
+    message_field_content: value
 
 dataset_prepared_path: last_run_prepared
 val_set_size: 0.0

examples/gemma2/qlora.yml

@@ -11,8 +11,11 @@ chat_template: gemma
 datasets:
   - path: cgato/SlimOrcaDedupCleaned
     type: chat_template
-    chat_template: gemma
     drop_system_message: true
+    field_messages: conversations
+    message_field_role: from
+    message_field_content: value
+
 val_set_size: 0.0
 output_dir: ./outputs/out
 

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

examples/llama-3/fft-8b-liger-fsdp.yaml

@@ -14,6 +14,10 @@ datasets:
   - path: mlabonne/FineTome-100k
     type: chat_template
     split: train[:20%]
+    field_messages: conversations
+    message_field_role: from
+    message_field_content: value
+
 dataset_prepared_path: last_run_prepared
 val_set_size: 0.02
 output_dir: ./outputs/out

examples/phi/lora-3.5.yaml

@@ -10,7 +10,6 @@ chat_template: phi_3
 datasets:
   - path: fozziethebeat/alpaca_messages_2k_test
     type: chat_template
-    chat_template: phi_3
     field_messages: messages
     message_field_role: role
     message_field_content: content