Feat: Add support for tokenizer’s or custom jinja chat_template (#1970)

* Allow using tokenizer's default chat template with fallbacks

Summary of changes:

1. Adds `tokenizer_default` as option for `chat_template` in
   `chat_template` prompt strategy that allows using the chat template
   from tokenizer's config.json
2. Allows falling back to chat templates available in axolotl if
   tokenizer does not have a chat template
3. Adds a mistral chat template which supports system message - taken
   from https://github.com/chujiezheng/chat_templates/blob/main/chat_templates/mistral-instruct.jinja

---

Why?

Many popular models are not trained with chatml format. As a result for
the model to correctly learn chatml we have to turn on train_on_inputs
which requires more compute and time. If we can use the model's already
learned chat template we can just learn the output tokens

---

Todo:

- Write tests

* Add tests

* Fix lint and bug post merge from main

* Add option `chat_template_jinja` to provide a jinja template

* remove custom mistral template

* Address review comments and add docs

* Update docs/dataset-formats/conversation.qmd

Co-authored-by: NanoCode012 <kevinvong@rocketmail.com>

* fix: set default to tokenizer template

* Merge branch 'main' into cj_tokenizer_default_prompt_template

* chore: remove redundant function

* fix: re-arrange enum declaration position

* fix: refactor artifact left from main merge

* feat(doc): updated config with chat template options and clarified examples

* chore: clarify doc

* chore: added example for non-default template

* chore: refactor

* fix: test

* fix: config being dropped and unittest to catch that

* chore: lint

* chore: skip duplicate

* fix: rename var after merge

* feat: add test for levy's dpo case

* fix: remove default setting on edge case where chat template overriden in dataset section

* feat: handle sharegpt deprecation better in docs

* feat: add example using fallback

* feat: handles chat_template requiring specific user/assistant order

* fix: update test based on new defaults

* fix: imported name incorrectly updated on merge

* chore: lint

* fix: update dummy message to prevent potential overlap with real content

* fix(doc): formatting

* fix: update bradleyterry to use new chat_template

---------

Co-authored-by: Chirag Jain <jain.chirag925@gmail.com>

tests/prompt_strategies/test_chat_templates_advanced.py

@@ -12,7 +12,7 @@ from axolotl.prompt_strategies.chat_template import (
     ChatTemplateStrategy,
 )
 from axolotl.prompters import IGNORE_TOKEN_ID
-from axolotl.utils.chat_templates import chat_templates
+from axolotl.utils.chat_templates import get_chat_template
 
 logging.basicConfig(level=logging.DEBUG)
 LOG = logging.getLogger("axolotl")
@@ -35,7 +35,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_inputs=True")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=True,
@@ -80,7 +80,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_inputs=False")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -123,7 +123,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing roles_to_train with assistant only")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -151,7 +151,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing roles_to_train with all roles")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=True,
@@ -184,7 +184,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with empty roles_to_train")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -205,7 +205,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_eos='all'")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -232,7 +232,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_eos='turn'")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -282,7 +282,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_eos='last'")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -315,7 +315,7 @@ class TestChatTemplateConfigurations:
         LOG.info("Testing with train_on_eos='none'")
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
-                llama3_tokenizer, chat_template=chat_templates("llama3")
+                llama3_tokenizer, chat_template=get_chat_template("llama3")
             ),
             tokenizer=llama3_tokenizer,
             train_on_inputs=False,
@@ -343,7 +343,7 @@ class TestChatTemplateConfigurations:
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
                 llama3_tokenizer,
-                chat_template=chat_templates("llama3"),
+                chat_template=get_chat_template("llama3"),
                 drop_system_message=True,
             ),
             tokenizer=llama3_tokenizer,
@@ -371,7 +371,7 @@ class TestChatTemplateConfigurations:
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
                 llama3_tokenizer,
-                chat_template=chat_templates("llama3"),
+                chat_template=get_chat_template("llama3"),
                 roles=custom_roles,
             ),
             tokenizer=llama3_tokenizer,
@@ -424,7 +424,7 @@ class TestChatTemplateConfigurations:
         strategy = ChatTemplateStrategy(
             ChatTemplatePrompter(
                 llama3_tokenizer,
-                chat_template=chat_templates("llama3"),
+                chat_template=get_chat_template("llama3"),
                 message_field_training="train",
                 message_field_training_detail="train_detail",
             ),