diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index 29cd2556d..a43dbac41 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -25,6 +25,11 @@ jobs: python_version: "3.11" pytorch: 2.6.0 axolotl_extras: vllm + - cuda: 126 + cuda_version: 12.6.3 + python_version: "3.11" + pytorch: 2.7.0 + axolotl_extras: vllm - cuda: 126 cuda_version: 12.6.3 python_version: "3.11" @@ -93,6 +98,11 @@ jobs: pytorch: 2.6.0 axolotl_extras: is_latest: true + - cuda: 126 + cuda_version: 12.6.3 + python_version: "3.11" + pytorch: 2.7.0 + axolotl_extras: - cuda: 126 cuda_version: 12.6.3 python_version: "3.11" diff --git a/.nojekyll b/.nojekyll index ce3b02b59..fdc91bde4 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -2aa75668 \ No newline at end of file +f1b312ba \ No newline at end of file diff --git a/docs/api/prompt_strategies.chat_template.html b/docs/api/prompt_strategies.chat_template.html index cb2622b85..1c76b231a 100644 --- a/docs/api/prompt_strategies.chat_template.html +++ b/docs/api/prompt_strategies.chat_template.html @@ -695,23 +695,6 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true}); split_thinking=False, )

Mistral strategy for chat template.

-
-

Attributes

- - - - - - - - - - - - - -
NameDescription
supports_multiprocessingWhether this tokenizing strategy supports multiprocessing.
-

Methods

diff --git a/docs/api/prompt_tokenizers.html b/docs/api/prompt_tokenizers.html index a34b82249..a098be028 100644 --- a/docs/api/prompt_tokenizers.html +++ b/docs/api/prompt_tokenizers.html @@ -667,23 +667,6 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true}); sequence_len=2048,)

Abstract class for tokenizing strategies

-
-

Attributes

-
- - - - - - - - - - - - -
NameDescription
supports_multiprocessingWhether this tokenizing strategy supports multiprocessing.
-

ReflectionPromptTokenizingStrategy

diff --git a/docs/custom_integrations.html b/docs/custom_integrations.html index 44c037289..51843143c 100644 --- a/docs/custom_integrations.html +++ b/docs/custom_integrations.html @@ -559,7 +559,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true}); -
pip3 uninstall -y cut-cross-entropy && pip3 install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@622068a"
+
pip3 uninstall -y cut-cross-entropy && pip3 install "cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@865b899"

Usage

diff --git a/docs/fsdp_qlora.html b/docs/fsdp_qlora.html index a4f1546aa..674d2f738 100644 --- a/docs/fsdp_qlora.html +++ b/docs/fsdp_qlora.html @@ -481,7 +481,7 @@ See the example config file in addition to reading
  1. Set adapter: qlora in your axolotl config file.
    2. -
  2. Enable FSDP in your axolotl config, as described here.
    3. +
  3. Enable FSDP in your axolotl config, as described here.
  4. Use one of the supported model types: llama, mistral or mixtral.
diff --git a/search.json b/search.json index 4335f4077..35aecbb42 100644 --- a/search.json +++ b/search.json @@ -698,14 +698,14 @@ "href": "docs/api/prompt_strategies.chat_template.html", "title": "prompt_strategies.chat_template", "section": "", - "text": "prompt_strategies.chat_template\nHF Chat Templates prompt strategy\n\n\n\n\n\nName\nDescription\n\n\n\n\nChatTemplatePrompter\nPrompter for HF chat templates\n\n\nChatTemplateStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nMistralPrompter\nMistral prompter for chat template.\n\n\nMistralStrategy\nMistral strategy for chat template.\n\n\nStrategyLoader\nLoad chat template strategy based on configuration.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter(\n tokenizer,\n chat_template,\n processor=None,\n max_length=2048,\n message_property_mappings=None,\n message_field_training=None,\n message_field_training_detail=None,\n field_messages='messages',\n field_system='system',\n field_tools='tools',\n roles=None,\n chat_template_kwargs=None,\n drop_system_message=False,\n)\nPrompter for HF chat templates\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_prompt\nBuild a prompt from a conversation.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter.build_prompt(\n conversation,\n add_generation_prompt=False,\n images=None,\n tools=None,\n)\nBuild a prompt from a conversation.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconversation\nlist[dict]\nA list of messages.\nrequired\n\n\nadd_generation_prompt\n\nWhether to add a generation prompt.\nFalse\n\n\nimages\n\nA list of images. (optional)\nNone\n\n\ntools\n\nA list of tools. (optional)\nNone\n\n\n\n\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\nfind_turn\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\ntokenize_prompt\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_turn(\n turns,\n turn_idx,\n tools=None,\n)\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.tokenize_prompt(prompt)\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.MistralPrompter(*args, **kwargs)\nMistral prompter for chat template.\n\n\n\nprompt_strategies.chat_template.MistralStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nMistral strategy for chat template.\n\n\n\n\n\nName\nDescription\n\n\n\n\nsupports_multiprocessing\nWhether this tokenizing strategy supports multiprocessing.\n\n\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.MistralStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.StrategyLoader()\nLoad chat template strategy based on configuration." + "text": "prompt_strategies.chat_template\nHF Chat Templates prompt strategy\n\n\n\n\n\nName\nDescription\n\n\n\n\nChatTemplatePrompter\nPrompter for HF chat templates\n\n\nChatTemplateStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nMistralPrompter\nMistral prompter for chat template.\n\n\nMistralStrategy\nMistral strategy for chat template.\n\n\nStrategyLoader\nLoad chat template strategy based on configuration.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter(\n tokenizer,\n chat_template,\n processor=None,\n max_length=2048,\n message_property_mappings=None,\n message_field_training=None,\n message_field_training_detail=None,\n field_messages='messages',\n field_system='system',\n field_tools='tools',\n roles=None,\n chat_template_kwargs=None,\n drop_system_message=False,\n)\nPrompter for HF chat templates\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_prompt\nBuild a prompt from a conversation.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter.build_prompt(\n conversation,\n add_generation_prompt=False,\n images=None,\n tools=None,\n)\nBuild a prompt from a conversation.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconversation\nlist[dict]\nA list of messages.\nrequired\n\n\nadd_generation_prompt\n\nWhether to add a generation prompt.\nFalse\n\n\nimages\n\nA list of images. (optional)\nNone\n\n\ntools\n\nA list of tools. (optional)\nNone\n\n\n\n\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\nfind_turn\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\ntokenize_prompt\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_turn(\n turns,\n turn_idx,\n tools=None,\n)\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.tokenize_prompt(prompt)\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.MistralPrompter(*args, **kwargs)\nMistral prompter for chat template.\n\n\n\nprompt_strategies.chat_template.MistralStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nMistral strategy for chat template.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.MistralStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.StrategyLoader()\nLoad chat template strategy based on configuration." }, { "objectID": "docs/api/prompt_strategies.chat_template.html#classes", "href": "docs/api/prompt_strategies.chat_template.html#classes", "title": "prompt_strategies.chat_template", "section": "", - "text": "Name\nDescription\n\n\n\n\nChatTemplatePrompter\nPrompter for HF chat templates\n\n\nChatTemplateStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nMistralPrompter\nMistral prompter for chat template.\n\n\nMistralStrategy\nMistral strategy for chat template.\n\n\nStrategyLoader\nLoad chat template strategy based on configuration.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter(\n tokenizer,\n chat_template,\n processor=None,\n max_length=2048,\n message_property_mappings=None,\n message_field_training=None,\n message_field_training_detail=None,\n field_messages='messages',\n field_system='system',\n field_tools='tools',\n roles=None,\n chat_template_kwargs=None,\n drop_system_message=False,\n)\nPrompter for HF chat templates\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_prompt\nBuild a prompt from a conversation.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter.build_prompt(\n conversation,\n add_generation_prompt=False,\n images=None,\n tools=None,\n)\nBuild a prompt from a conversation.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconversation\nlist[dict]\nA list of messages.\nrequired\n\n\nadd_generation_prompt\n\nWhether to add a generation prompt.\nFalse\n\n\nimages\n\nA list of images. (optional)\nNone\n\n\ntools\n\nA list of tools. (optional)\nNone\n\n\n\n\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\nfind_turn\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\ntokenize_prompt\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_turn(\n turns,\n turn_idx,\n tools=None,\n)\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.tokenize_prompt(prompt)\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.MistralPrompter(*args, **kwargs)\nMistral prompter for chat template.\n\n\n\nprompt_strategies.chat_template.MistralStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nMistral strategy for chat template.\n\n\n\n\n\nName\nDescription\n\n\n\n\nsupports_multiprocessing\nWhether this tokenizing strategy supports multiprocessing.\n\n\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.MistralStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.StrategyLoader()\nLoad chat template strategy based on configuration." + "text": "Name\nDescription\n\n\n\n\nChatTemplatePrompter\nPrompter for HF chat templates\n\n\nChatTemplateStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nMistralPrompter\nMistral prompter for chat template.\n\n\nMistralStrategy\nMistral strategy for chat template.\n\n\nStrategyLoader\nLoad chat template strategy based on configuration.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter(\n tokenizer,\n chat_template,\n processor=None,\n max_length=2048,\n message_property_mappings=None,\n message_field_training=None,\n message_field_training_detail=None,\n field_messages='messages',\n field_system='system',\n field_tools='tools',\n roles=None,\n chat_template_kwargs=None,\n drop_system_message=False,\n)\nPrompter for HF chat templates\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_prompt\nBuild a prompt from a conversation.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplatePrompter.build_prompt(\n conversation,\n add_generation_prompt=False,\n images=None,\n tools=None,\n)\nBuild a prompt from a conversation.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconversation\nlist[dict]\nA list of messages.\nrequired\n\n\nadd_generation_prompt\n\nWhether to add a generation prompt.\nFalse\n\n\nimages\n\nA list of images. (optional)\nNone\n\n\ntools\n\nA list of tools. (optional)\nNone\n\n\n\n\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\nfind_turn\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\ntokenize_prompt\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.find_turn(\n turns,\n turn_idx,\n tools=None,\n)\nLocate the starting and ending indices of the specified turn in a conversation.\n\n\n\nprompt_strategies.chat_template.ChatTemplateStrategy.tokenize_prompt(prompt)\nPublic method that can handle either a single prompt or a batch of prompts.\n\n\n\n\n\nprompt_strategies.chat_template.MistralPrompter(*args, **kwargs)\nMistral prompter for chat template.\n\n\n\nprompt_strategies.chat_template.MistralStrategy(\n prompter,\n tokenizer,\n train_on_inputs,\n sequence_len,\n roles_to_train=None,\n train_on_eos=None,\n train_on_eot=None,\n eot_tokens=None,\n split_thinking=False,\n)\nMistral strategy for chat template.\n\n\n\n\n\nName\nDescription\n\n\n\n\nfind_first_eot_token\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.MistralStrategy.find_first_eot_token(\n input_ids,\n start_idx,\n)\nFind the first EOT token in the input_ids starting from start_idx.\n\n\n\n\n\nprompt_strategies.chat_template.StrategyLoader()\nLoad chat template strategy based on configuration." }, { "objectID": "docs/api/prompt_strategies.kto.user_defined.html", @@ -2959,14 +2959,14 @@ "href": "docs/api/prompt_tokenizers.html", "title": "prompt_tokenizers", "section": "", - "text": "prompt_tokenizers\nModule containing PromptTokenizingStrategy and Prompter classes\n\n\n\n\n\nName\nDescription\n\n\n\n\nAlpacaMultipleChoicePromptTokenizingStrategy\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\nAlpacaPromptTokenizingStrategy\nTokenizing strategy for Alpaca prompts.\n\n\nAlpacaReflectionPTStrategy\nTokenizing strategy for Alpaca Reflection prompts.\n\n\nDatasetWrappingStrategy\nAbstract class for wrapping datasets for Chat Messages\n\n\nGPTeacherPromptTokenizingStrategy\nTokenizing strategy for GPTeacher prompts.\n\n\nInstructionPromptTokenizingStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nInvalidDataException\nException raised when the data is invalid\n\n\nJeopardyPromptTokenizingStrategy\nTokenizing strategy for Jeopardy prompts.\n\n\nNomicGPT4AllPromptTokenizingStrategy\nTokenizing strategy for NomicGPT4All prompts.\n\n\nOpenAssistantPromptTokenizingStrategy\nTokenizing strategy for OpenAssistant prompts.\n\n\nPromptTokenizingStrategy\nAbstract class for tokenizing strategies\n\n\nReflectionPromptTokenizingStrategy\nTokenizing strategy for Reflection prompts.\n\n\nSummarizeTLDRPromptTokenizingStrategy\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\nprompt_tokenizers.AlpacaMultipleChoicePromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\n\nprompt_tokenizers.AlpacaPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca prompts.\n\n\n\nprompt_tokenizers.AlpacaReflectionPTStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Reflection prompts.\n\n\n\nprompt_tokenizers.DatasetWrappingStrategy()\nAbstract class for wrapping datasets for Chat Messages\n\n\n\nprompt_tokenizers.GPTeacherPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for GPTeacher prompts.\n\n\n\nprompt_tokenizers.InstructionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\nprompt_tokenizers.InvalidDataException()\nException raised when the data is invalid\n\n\n\nprompt_tokenizers.JeopardyPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Jeopardy prompts.\n\n\n\nprompt_tokenizers.NomicGPT4AllPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for NomicGPT4All prompts.\n\n\n\nprompt_tokenizers.OpenAssistantPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for OpenAssistant prompts.\n\n\n\nprompt_tokenizers.PromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nAbstract class for tokenizing strategies\n\n\n\n\n\nName\nDescription\n\n\n\n\nsupports_multiprocessing\nWhether this tokenizing strategy supports multiprocessing.\n\n\n\n\n\n\n\nprompt_tokenizers.ReflectionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Reflection prompts.\n\n\n\nprompt_tokenizers.SummarizeTLDRPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nparse_tokenized_to_result\nParses the tokenized prompt and append the tokenized input_ids, attention_mask and labels to the result\n\n\ntokenize_prompt_default\nReturns the default values for the tokenize prompt function\n\n\n\n\n\nprompt_tokenizers.parse_tokenized_to_result(\n result,\n current_len,\n res,\n labels,\n pad_token_id=None,\n)\nParses the tokenized prompt and append the tokenized input_ids, attention_mask and labels to the result\n\n\n\nprompt_tokenizers.tokenize_prompt_default()\nReturns the default values for the tokenize prompt function" + "text": "prompt_tokenizers\nModule containing PromptTokenizingStrategy and Prompter classes\n\n\n\n\n\nName\nDescription\n\n\n\n\nAlpacaMultipleChoicePromptTokenizingStrategy\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\nAlpacaPromptTokenizingStrategy\nTokenizing strategy for Alpaca prompts.\n\n\nAlpacaReflectionPTStrategy\nTokenizing strategy for Alpaca Reflection prompts.\n\n\nDatasetWrappingStrategy\nAbstract class for wrapping datasets for Chat Messages\n\n\nGPTeacherPromptTokenizingStrategy\nTokenizing strategy for GPTeacher prompts.\n\n\nInstructionPromptTokenizingStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nInvalidDataException\nException raised when the data is invalid\n\n\nJeopardyPromptTokenizingStrategy\nTokenizing strategy for Jeopardy prompts.\n\n\nNomicGPT4AllPromptTokenizingStrategy\nTokenizing strategy for NomicGPT4All prompts.\n\n\nOpenAssistantPromptTokenizingStrategy\nTokenizing strategy for OpenAssistant prompts.\n\n\nPromptTokenizingStrategy\nAbstract class for tokenizing strategies\n\n\nReflectionPromptTokenizingStrategy\nTokenizing strategy for Reflection prompts.\n\n\nSummarizeTLDRPromptTokenizingStrategy\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\nprompt_tokenizers.AlpacaMultipleChoicePromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\n\nprompt_tokenizers.AlpacaPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca prompts.\n\n\n\nprompt_tokenizers.AlpacaReflectionPTStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Reflection prompts.\n\n\n\nprompt_tokenizers.DatasetWrappingStrategy()\nAbstract class for wrapping datasets for Chat Messages\n\n\n\nprompt_tokenizers.GPTeacherPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for GPTeacher prompts.\n\n\n\nprompt_tokenizers.InstructionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\nprompt_tokenizers.InvalidDataException()\nException raised when the data is invalid\n\n\n\nprompt_tokenizers.JeopardyPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Jeopardy prompts.\n\n\n\nprompt_tokenizers.NomicGPT4AllPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for NomicGPT4All prompts.\n\n\n\nprompt_tokenizers.OpenAssistantPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for OpenAssistant prompts.\n\n\n\nprompt_tokenizers.PromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nAbstract class for tokenizing strategies\n\n\n\nprompt_tokenizers.ReflectionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Reflection prompts.\n\n\n\nprompt_tokenizers.SummarizeTLDRPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nparse_tokenized_to_result\nParses the tokenized prompt and append the tokenized input_ids, attention_mask and labels to the result\n\n\ntokenize_prompt_default\nReturns the default values for the tokenize prompt function\n\n\n\n\n\nprompt_tokenizers.parse_tokenized_to_result(\n result,\n current_len,\n res,\n labels,\n pad_token_id=None,\n)\nParses the tokenized prompt and append the tokenized input_ids, attention_mask and labels to the result\n\n\n\nprompt_tokenizers.tokenize_prompt_default()\nReturns the default values for the tokenize prompt function" }, { "objectID": "docs/api/prompt_tokenizers.html#classes", "href": "docs/api/prompt_tokenizers.html#classes", "title": "prompt_tokenizers", "section": "", - "text": "Name\nDescription\n\n\n\n\nAlpacaMultipleChoicePromptTokenizingStrategy\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\nAlpacaPromptTokenizingStrategy\nTokenizing strategy for Alpaca prompts.\n\n\nAlpacaReflectionPTStrategy\nTokenizing strategy for Alpaca Reflection prompts.\n\n\nDatasetWrappingStrategy\nAbstract class for wrapping datasets for Chat Messages\n\n\nGPTeacherPromptTokenizingStrategy\nTokenizing strategy for GPTeacher prompts.\n\n\nInstructionPromptTokenizingStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nInvalidDataException\nException raised when the data is invalid\n\n\nJeopardyPromptTokenizingStrategy\nTokenizing strategy for Jeopardy prompts.\n\n\nNomicGPT4AllPromptTokenizingStrategy\nTokenizing strategy for NomicGPT4All prompts.\n\n\nOpenAssistantPromptTokenizingStrategy\nTokenizing strategy for OpenAssistant prompts.\n\n\nPromptTokenizingStrategy\nAbstract class for tokenizing strategies\n\n\nReflectionPromptTokenizingStrategy\nTokenizing strategy for Reflection prompts.\n\n\nSummarizeTLDRPromptTokenizingStrategy\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\nprompt_tokenizers.AlpacaMultipleChoicePromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\n\nprompt_tokenizers.AlpacaPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca prompts.\n\n\n\nprompt_tokenizers.AlpacaReflectionPTStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Reflection prompts.\n\n\n\nprompt_tokenizers.DatasetWrappingStrategy()\nAbstract class for wrapping datasets for Chat Messages\n\n\n\nprompt_tokenizers.GPTeacherPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for GPTeacher prompts.\n\n\n\nprompt_tokenizers.InstructionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\nprompt_tokenizers.InvalidDataException()\nException raised when the data is invalid\n\n\n\nprompt_tokenizers.JeopardyPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Jeopardy prompts.\n\n\n\nprompt_tokenizers.NomicGPT4AllPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for NomicGPT4All prompts.\n\n\n\nprompt_tokenizers.OpenAssistantPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for OpenAssistant prompts.\n\n\n\nprompt_tokenizers.PromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nAbstract class for tokenizing strategies\n\n\n\n\n\nName\nDescription\n\n\n\n\nsupports_multiprocessing\nWhether this tokenizing strategy supports multiprocessing.\n\n\n\n\n\n\n\nprompt_tokenizers.ReflectionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Reflection prompts.\n\n\n\nprompt_tokenizers.SummarizeTLDRPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for SummarizeTLDR prompts." + "text": "Name\nDescription\n\n\n\n\nAlpacaMultipleChoicePromptTokenizingStrategy\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\nAlpacaPromptTokenizingStrategy\nTokenizing strategy for Alpaca prompts.\n\n\nAlpacaReflectionPTStrategy\nTokenizing strategy for Alpaca Reflection prompts.\n\n\nDatasetWrappingStrategy\nAbstract class for wrapping datasets for Chat Messages\n\n\nGPTeacherPromptTokenizingStrategy\nTokenizing strategy for GPTeacher prompts.\n\n\nInstructionPromptTokenizingStrategy\nTokenizing strategy for instruction-based prompts.\n\n\nInvalidDataException\nException raised when the data is invalid\n\n\nJeopardyPromptTokenizingStrategy\nTokenizing strategy for Jeopardy prompts.\n\n\nNomicGPT4AllPromptTokenizingStrategy\nTokenizing strategy for NomicGPT4All prompts.\n\n\nOpenAssistantPromptTokenizingStrategy\nTokenizing strategy for OpenAssistant prompts.\n\n\nPromptTokenizingStrategy\nAbstract class for tokenizing strategies\n\n\nReflectionPromptTokenizingStrategy\nTokenizing strategy for Reflection prompts.\n\n\nSummarizeTLDRPromptTokenizingStrategy\nTokenizing strategy for SummarizeTLDR prompts.\n\n\n\n\n\nprompt_tokenizers.AlpacaMultipleChoicePromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Multiple Choice prompts.\n\n\n\nprompt_tokenizers.AlpacaPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca prompts.\n\n\n\nprompt_tokenizers.AlpacaReflectionPTStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Alpaca Reflection prompts.\n\n\n\nprompt_tokenizers.DatasetWrappingStrategy()\nAbstract class for wrapping datasets for Chat Messages\n\n\n\nprompt_tokenizers.GPTeacherPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for GPTeacher prompts.\n\n\n\nprompt_tokenizers.InstructionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for instruction-based prompts.\n\n\n\nprompt_tokenizers.InvalidDataException()\nException raised when the data is invalid\n\n\n\nprompt_tokenizers.JeopardyPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Jeopardy prompts.\n\n\n\nprompt_tokenizers.NomicGPT4AllPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for NomicGPT4All prompts.\n\n\n\nprompt_tokenizers.OpenAssistantPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for OpenAssistant prompts.\n\n\n\nprompt_tokenizers.PromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nAbstract class for tokenizing strategies\n\n\n\nprompt_tokenizers.ReflectionPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for Reflection prompts.\n\n\n\nprompt_tokenizers.SummarizeTLDRPromptTokenizingStrategy(\n prompter,\n tokenizer,\n train_on_inputs=False,\n sequence_len=2048,\n)\nTokenizing strategy for SummarizeTLDR prompts." }, { "objectID": "docs/api/prompt_tokenizers.html#functions", @@ -3064,7 +3064,7 @@ "href": "docs/custom_integrations.html#cut-cross-entropy", "title": "Custom Integrations", "section": "Cut Cross Entropy", - "text": "Cut Cross Entropy\nCut Cross Entropy (CCE) reduces VRAM usage through optimization on the cross-entropy operation during loss calculation.\nSee https://github.com/apple/ml-cross-entropy\n\nRequirements\n\nPyTorch 2.4.0 or higher\n\n\n\nInstallation\nRun the following command to install cut_cross_entropy[transformers] if you don’t have it already.\n\nIf you are in dev environment\n\npython scripts/cutcrossentropy_install.py | sh\n\nIf you are installing from pip\n\npip3 uninstall -y cut-cross-entropy && pip3 install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@622068a\"\n\n\nUsage\nplugins:\n - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin\n\n\nSupported Models\n\ncohere\ncohere2\ngemma\ngemma2\ngemma3\ngemma3_text\nglm\nglm4\nllama\nllama4\nllama4_text\nmistral\nmistral3\nmllama\nphi\nphi3\nphi4_multimodal\nqwen2\nqwen2_vl\nqwen2_moe\nqwen2_5_vl\nqwen3\nqwen3_moe\n\n\n\nCitation\n@article{wijmans2024cut,\n author = {Erik Wijmans and\n Brody Huval and\n Alexander Hertzberg and\n Vladlen Koltun and\n Philipp Kr\\\"ahenb\\\"uhl},\n title = {Cut Your Losses in Large-Vocabulary Language Models},\n journal = {arXiv},\n year = {2024},\n url = {https://arxiv.org/abs/2411.09009},\n}\nPlease see reference here", + "text": "Cut Cross Entropy\nCut Cross Entropy (CCE) reduces VRAM usage through optimization on the cross-entropy operation during loss calculation.\nSee https://github.com/apple/ml-cross-entropy\n\nRequirements\n\nPyTorch 2.4.0 or higher\n\n\n\nInstallation\nRun the following command to install cut_cross_entropy[transformers] if you don’t have it already.\n\nIf you are in dev environment\n\npython scripts/cutcrossentropy_install.py | sh\n\nIf you are installing from pip\n\npip3 uninstall -y cut-cross-entropy && pip3 install \"cut-cross-entropy[transformers] @ git+https://github.com/axolotl-ai-cloud/ml-cross-entropy.git@865b899\"\n\n\nUsage\nplugins:\n - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin\n\n\nSupported Models\n\ncohere\ncohere2\ngemma\ngemma2\ngemma3\ngemma3_text\nglm\nglm4\nllama\nllama4\nllama4_text\nmistral\nmistral3\nmllama\nphi\nphi3\nphi4_multimodal\nqwen2\nqwen2_vl\nqwen2_moe\nqwen2_5_vl\nqwen3\nqwen3_moe\n\n\n\nCitation\n@article{wijmans2024cut,\n author = {Erik Wijmans and\n Brody Huval and\n Alexander Hertzberg and\n Vladlen Koltun and\n Philipp Kr\\\"ahenb\\\"uhl},\n title = {Cut Your Losses in Large-Vocabulary Language Models},\n journal = {arXiv},\n year = {2024},\n url = {https://arxiv.org/abs/2411.09009},\n}\nPlease see reference here", "crumbs": [ "Advanced Features", "Custom Integrations" diff --git a/sitemap.xml b/sitemap.xml index e3a39aca3..7176289fe 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -2,758 +2,758 @@ https://docs.axolotl.ai/docs/unsloth.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.221Z https://docs.axolotl.ai/docs/dataset-formats/conversation.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/dataset-formats/stepwise_supervised.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/dataset-formats/tokenized.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/mac.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/nccl.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/multi-node.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/docker.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/lr_groups.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/inference.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/cli.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/config-reference.html - 2025-07-07T21:08:49.663Z + 2025-07-08T20:32:09.686Z https://docs.axolotl.ai/docs/multi-gpu.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/debugging.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/multimodal.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/api/cli.sweeps.html - 2025-07-07T21:08:36.131Z + 2025-07-08T20:31:56.049Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.llama3.html - 2025-07-07T21:08:36.457Z + 2025-07-08T20:31:56.397Z https://docs.axolotl.ai/docs/api/utils.schedulers.html - 2025-07-07T21:08:36.840Z + 2025-07-08T20:31:56.787Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_xformers.html - 2025-07-07T21:08:36.655Z + 2025-07-08T20:31:56.601Z https://docs.axolotl.ai/docs/api/cli.cloud.modal_.html - 2025-07-07T21:08:36.180Z + 2025-07-08T20:31:56.098Z https://docs.axolotl.ai/docs/api/kernels.geglu.html - 2025-07-07T21:08:36.609Z + 2025-07-08T20:31:56.555Z https://docs.axolotl.ai/docs/api/core.trainers.utils.html - 2025-07-07T21:08:36.252Z + 2025-07-08T20:31:56.176Z https://docs.axolotl.ai/docs/api/core.datasets.chat.html - 2025-07-07T21:08:35.983Z + 2025-07-08T20:31:55.892Z https://docs.axolotl.ai/docs/api/utils.schemas.peft.html - 2025-07-07T21:08:36.954Z + 2025-07-08T20:31:56.905Z https://docs.axolotl.ai/docs/api/monkeypatch.btlm_attn_hijack_flash.html - 2025-07-07T21:08:36.718Z + 2025-07-08T20:31:56.662Z https://docs.axolotl.ai/docs/api/prompt_strategies.chat_template.html - 2025-07-07T21:08:36.360Z + 2025-07-08T20:31:56.287Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.user_defined.html - 2025-07-07T21:08:36.490Z + 2025-07-08T20:31:56.431Z https://docs.axolotl.ai/docs/api/cli.cloud.base.html - 2025-07-07T21:08:36.173Z + 2025-07-08T20:31:56.091Z https://docs.axolotl.ai/docs/api/kernels.swiglu.html - 2025-07-07T21:08:36.619Z + 2025-07-08T20:31:56.565Z https://docs.axolotl.ai/docs/api/prompt_strategies.stepwise_supervised.html - 2025-07-07T21:08:36.424Z + 2025-07-08T20:31:56.357Z https://docs.axolotl.ai/docs/api/prompt_strategies.bradley_terry.llama3.html - 2025-07-07T21:08:36.514Z + 2025-07-08T20:31:56.455Z https://docs.axolotl.ai/docs/api/prompt_strategies.completion.html - 2025-07-07T21:08:36.414Z + 2025-07-08T20:31:56.345Z https://docs.axolotl.ai/docs/api/kernels.utils.html - 2025-07-07T21:08:36.628Z + 2025-07-08T20:31:56.574Z https://docs.axolotl.ai/docs/api/common.datasets.html - 2025-07-07T21:08:37.165Z + 2025-07-08T20:31:57.120Z https://docs.axolotl.ai/docs/api/utils.schemas.datasets.html - 2025-07-07T21:08:36.945Z + 2025-07-08T20:31:56.895Z https://docs.axolotl.ai/docs/api/core.builders.rl.html - 2025-07-07T21:08:35.938Z + 2025-07-08T20:31:55.842Z https://docs.axolotl.ai/docs/api/evaluate.html - 2025-07-07T21:08:35.843Z + 2025-07-08T20:31:55.738Z https://docs.axolotl.ai/docs/api/kernels.quantize.html - 2025-07-07T21:08:36.626Z + 2025-07-08T20:31:56.572Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_flash.html - 2025-07-07T21:08:36.654Z + 2025-07-08T20:31:56.599Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.rng_state_loader.html - 2025-07-07T21:08:36.295Z + 2025-07-08T20:31:56.222Z https://docs.axolotl.ai/docs/api/integrations.base.html - 2025-07-07T21:08:37.125Z + 2025-07-08T20:31:57.079Z https://docs.axolotl.ai/docs/api/cli.merge_lora.html - 2025-07-07T21:08:36.105Z + 2025-07-08T20:31:56.023Z https://docs.axolotl.ai/docs/api/cli.merge_sharded_fsdp_weights.html - 2025-07-07T21:08:36.117Z + 2025-07-08T20:31:56.035Z https://docs.axolotl.ai/docs/api/monkeypatch.transformers_fa_utils.html - 2025-07-07T21:08:36.734Z + 2025-07-08T20:31:56.679Z https://docs.axolotl.ai/docs/api/prompt_strategies.llama2_chat.html - 2025-07-07T21:08:36.408Z + 2025-07-08T20:31:56.339Z https://docs.axolotl.ai/docs/api/utils.collators.mm_chat.html - 2025-07-07T21:08:37.195Z + 2025-07-08T20:31:57.149Z https://docs.axolotl.ai/docs/api/utils.data.sft.html - 2025-07-07T21:08:36.880Z + 2025-07-08T20:31:56.829Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_instruct.html - 2025-07-07T21:08:36.375Z + 2025-07-08T20:31:56.302Z https://docs.axolotl.ai/docs/api/integrations.liger.args.html - 2025-07-07T21:08:37.141Z + 2025-07-08T20:31:57.094Z https://docs.axolotl.ai/docs/api/monkeypatch.mistral_attn_hijack_flash.html - 2025-07-07T21:08:36.669Z + 2025-07-08T20:31:56.615Z https://docs.axolotl.ai/docs/api/cli.vllm_serve.html - 2025-07-07T21:08:36.170Z + 2025-07-08T20:31:56.088Z https://docs.axolotl.ai/docs/api/monkeypatch.utils.html - 2025-07-07T21:08:36.716Z + 2025-07-08T20:31:56.661Z https://docs.axolotl.ai/docs/api/loaders.patch_manager.html - 2025-07-07T21:08:36.285Z + 2025-07-08T20:31:56.211Z https://docs.axolotl.ai/docs/api/utils.schemas.integrations.html - 2025-07-07T21:08:36.975Z + 2025-07-08T20:31:56.926Z https://docs.axolotl.ai/docs/api/utils.callbacks.perplexity.html - 2025-07-07T21:08:37.242Z + 2025-07-08T20:31:57.197Z https://docs.axolotl.ai/docs/api/cli.utils.html - 2025-07-07T21:08:36.163Z + 2025-07-08T20:31:56.081Z https://docs.axolotl.ai/docs/api/utils.schemas.config.html - 2025-07-07T21:08:36.915Z + 2025-07-08T20:31:56.866Z https://docs.axolotl.ai/docs/api/prompt_strategies.input_output.html - 2025-07-07T21:08:36.420Z + 2025-07-08T20:31:56.351Z https://docs.axolotl.ai/docs/api/utils.distributed.html - 2025-07-07T21:08:36.861Z + 2025-07-08T20:31:56.808Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_disk.html - 2025-07-07T21:08:36.768Z + 2025-07-08T20:31:56.715Z https://docs.axolotl.ai/docs/api/monkeypatch.trainer_fsdp_optim.html - 2025-07-07T21:08:36.728Z + 2025-07-08T20:31:56.673Z https://docs.axolotl.ai/docs/api/core.builders.base.html - 2025-07-07T21:08:35.925Z + 2025-07-08T20:31:55.830Z https://docs.axolotl.ai/docs/api/core.trainers.trl.html - 2025-07-07T21:08:36.211Z + 2025-07-08T20:31:56.134Z https://docs.axolotl.ai/docs/api/cli.evaluate.html - 2025-07-07T21:08:36.038Z + 2025-07-08T20:31:55.954Z https://docs.axolotl.ai/docs/api/utils.optimizers.adopt.html - 2025-07-07T21:08:36.872Z + 2025-07-08T20:31:56.821Z https://docs.axolotl.ai/docs/api/utils.callbacks.qat.html - 2025-07-07T21:08:37.261Z + 2025-07-08T20:31:57.216Z https://docs.axolotl.ai/docs/api/core.trainers.dpo.trainer.html - 2025-07-07T21:08:36.228Z + 2025-07-08T20:31:56.150Z https://docs.axolotl.ai/docs/api/core.chat.format.shared.html - 2025-07-07T21:08:35.978Z + 2025-07-08T20:31:55.887Z https://docs.axolotl.ai/docs/api/monkeypatch.relora.html - 2025-07-07T21:08:36.678Z + 2025-07-08T20:31:56.623Z https://docs.axolotl.ai/docs/api/cli.config.html - 2025-07-07T21:08:36.082Z + 2025-07-08T20:31:55.998Z https://docs.axolotl.ai/docs/api/cli.preprocess.html - 2025-07-07T21:08:36.125Z + 2025-07-08T20:31:56.043Z https://docs.axolotl.ai/docs/api/core.trainers.base.html - 2025-07-07T21:08:36.195Z + 2025-07-08T20:31:56.113Z https://docs.axolotl.ai/docs/api/convert.html - 2025-07-07T21:08:35.867Z + 2025-07-08T20:31:55.763Z https://docs.axolotl.ai/docs/api/prompt_strategies.pygmalion.html - 2025-07-07T21:08:36.441Z + 2025-07-08T20:31:56.381Z https://docs.axolotl.ai/docs/api/utils.schemas.trl.html - 2025-07-07T21:08:36.957Z + 2025-07-08T20:31:56.908Z https://docs.axolotl.ai/docs/api/cli.args.html - 2025-07-07T21:08:36.058Z + 2025-07-08T20:31:55.974Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chat_template.html - 2025-07-07T21:08:36.447Z + 2025-07-08T20:31:56.387Z https://docs.axolotl.ai/docs/api/loaders.constants.html - 2025-07-07T21:08:36.287Z + 2025-07-08T20:31:56.213Z https://docs.axolotl.ai/docs/api/logging_config.html - 2025-07-07T21:08:35.919Z + 2025-07-08T20:31:55.823Z https://docs.axolotl.ai/docs/api/cli.inference.html - 2025-07-07T21:08:36.096Z + 2025-07-08T20:31:56.014Z https://docs.axolotl.ai/docs/api/utils.ctx_managers.sequence_parallel.html - 2025-07-07T21:08:36.326Z + 2025-07-08T20:31:56.252Z https://docs.axolotl.ai/docs/api/integrations.spectrum.args.html - 2025-07-07T21:08:37.147Z + 2025-07-08T20:31:57.101Z https://docs.axolotl.ai/docs/api/utils.schemas.training.html - 2025-07-07T21:08:36.927Z + 2025-07-08T20:31:56.878Z https://docs.axolotl.ai/docs/api/prompt_strategies.orcamini.html - 2025-07-07T21:08:36.435Z + 2025-07-08T20:31:56.375Z https://docs.axolotl.ai/docs/api/utils.freeze.html - 2025-07-07T21:08:36.798Z + 2025-07-08T20:31:56.745Z https://docs.axolotl.ai/docs/api/loaders.tokenizer.html - 2025-07-07T21:08:36.270Z + 2025-07-08T20:31:56.194Z https://docs.axolotl.ai/docs/api/utils.bench.html - 2025-07-07T21:08:36.790Z + 2025-07-08T20:31:56.737Z https://docs.axolotl.ai/docs/api/utils.quantization.html - 2025-07-07T21:08:36.901Z + 2025-07-08T20:31:56.850Z https://docs.axolotl.ai/docs/batch_vs_grad.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/input_output.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/sequence_parallelism.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.221Z https://docs.axolotl.ai/docs/reward_modelling.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/index.html - 2025-07-07T21:05:27.525Z + 2025-07-08T20:28:26.240Z https://docs.axolotl.ai/src/axolotl/integrations/LICENSE.html - 2025-07-07T21:05:27.529Z + 2025-07-08T20:28:26.244Z https://docs.axolotl.ai/FAQS.html - 2025-07-07T21:05:27.500Z + 2025-07-08T20:28:26.213Z https://docs.axolotl.ai/src/axolotl/integrations/cut_cross_entropy/ACKNOWLEDGEMENTS.html - 2025-07-07T21:05:27.529Z + 2025-07-08T20:28:26.244Z https://docs.axolotl.ai/TODO.html - 2025-07-07T21:05:27.500Z + 2025-07-08T20:28:26.213Z https://docs.axolotl.ai/examples/colab-notebooks/colab-axolotl-example.html - 2025-07-07T21:05:27.510Z + 2025-07-08T20:28:26.223Z https://docs.axolotl.ai/docs/torchao.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.221Z https://docs.axolotl.ai/docs/ray-integration.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/quantize.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/qat.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/api/utils.lora.html - 2025-07-07T21:08:36.781Z + 2025-07-08T20:31:56.728Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_w_system.html - 2025-07-07T21:08:36.387Z + 2025-07-08T20:31:56.318Z https://docs.axolotl.ai/docs/api/monkeypatch.stablelm_attn_hijack_flash.html - 2025-07-07T21:08:36.725Z + 2025-07-08T20:31:56.669Z https://docs.axolotl.ai/docs/api/utils.collators.core.html - 2025-07-07T21:08:37.168Z + 2025-07-08T20:31:57.122Z https://docs.axolotl.ai/docs/api/prompt_strategies.metharme.html - 2025-07-07T21:08:36.431Z + 2025-07-08T20:31:56.369Z https://docs.axolotl.ai/docs/api/utils.callbacks.profiler.html - 2025-07-07T21:08:37.245Z + 2025-07-08T20:31:57.201Z https://docs.axolotl.ai/docs/api/utils.data.pretraining.html - 2025-07-07T21:08:36.873Z + 2025-07-08T20:31:56.822Z https://docs.axolotl.ai/docs/api/utils.callbacks.lisa.html - 2025-07-07T21:08:37.247Z + 2025-07-08T20:31:57.202Z https://docs.axolotl.ai/docs/api/utils.trainer.html - 2025-07-07T21:08:36.816Z + 2025-07-08T20:31:56.763Z https://docs.axolotl.ai/docs/api/integrations.cut_cross_entropy.args.html - 2025-07-07T21:08:37.129Z + 2025-07-08T20:31:57.083Z https://docs.axolotl.ai/docs/api/utils.schemas.model.html - 2025-07-07T21:08:36.922Z + 2025-07-08T20:31:56.873Z https://docs.axolotl.ai/docs/api/monkeypatch.data.batch_dataset_fetcher.html - 2025-07-07T21:08:36.737Z + 2025-07-08T20:31:56.682Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.zephyr.html - 2025-07-07T21:08:36.469Z + 2025-07-08T20:31:56.410Z https://docs.axolotl.ai/docs/api/datasets.html - 2025-07-07T21:08:35.854Z + 2025-07-08T20:31:55.749Z https://docs.axolotl.ai/docs/api/utils.schemas.enums.html - 2025-07-07T21:08:36.985Z + 2025-07-08T20:31:56.936Z https://docs.axolotl.ai/docs/api/integrations.kd.trainer.html - 2025-07-07T21:08:37.137Z + 2025-07-08T20:31:57.091Z https://docs.axolotl.ai/docs/api/monkeypatch.lora_kernels.html - 2025-07-07T21:08:36.708Z + 2025-07-08T20:31:56.653Z https://docs.axolotl.ai/docs/api/utils.collators.batching.html - 2025-07-07T21:08:37.187Z + 2025-07-08T20:31:57.141Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.sampler.html - 2025-07-07T21:08:36.251Z + 2025-07-08T20:31:56.175Z https://docs.axolotl.ai/docs/api/prompt_strategies.base.html - 2025-07-07T21:08:36.327Z + 2025-07-08T20:31:56.254Z https://docs.axolotl.ai/docs/api/monkeypatch.multipack.html - 2025-07-07T21:08:36.671Z + 2025-07-08T20:31:56.617Z https://docs.axolotl.ai/docs/api/prompt_strategies.orpo.chat_template.html - 2025-07-07T21:08:36.510Z + 2025-07-08T20:31:56.451Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.passthrough.html - 2025-07-07T21:08:36.472Z + 2025-07-08T20:31:56.413Z https://docs.axolotl.ai/docs/api/core.chat.format.chatml.html - 2025-07-07T21:08:35.975Z + 2025-07-08T20:31:55.884Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.scheduler.html - 2025-07-07T21:08:36.302Z + 2025-07-08T20:31:56.229Z https://docs.axolotl.ai/docs/api/utils.model_shard_quant.html - 2025-07-07T21:08:36.787Z + 2025-07-08T20:31:56.733Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.chatml.html - 2025-07-07T21:08:36.488Z + 2025-07-08T20:31:56.429Z https://docs.axolotl.ai/docs/api/utils.tokenization.html - 2025-07-07T21:08:36.775Z + 2025-07-08T20:31:56.722Z https://docs.axolotl.ai/docs/api/loaders.model.html - 2025-07-07T21:08:36.262Z + 2025-07-08T20:31:56.186Z https://docs.axolotl.ai/docs/api/utils.callbacks.mlflow_.html - 2025-07-07T21:08:37.250Z + 2025-07-08T20:31:57.206Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.trainer.html - 2025-07-07T21:08:36.239Z + 2025-07-08T20:31:56.161Z https://docs.axolotl.ai/docs/api/cli.main.html - 2025-07-07T21:08:36.022Z + 2025-07-08T20:31:55.937Z https://docs.axolotl.ai/docs/api/utils.callbacks.comet_.html - 2025-07-07T21:08:37.254Z + 2025-07-08T20:31:57.210Z https://docs.axolotl.ai/docs/api/utils.chat_templates.html - 2025-07-07T21:08:36.776Z + 2025-07-08T20:31:56.723Z https://docs.axolotl.ai/docs/api/utils.schemas.utils.html - 2025-07-07T21:08:36.991Z + 2025-07-08T20:31:56.942Z https://docs.axolotl.ai/docs/api/common.architectures.html - 2025-07-07T21:08:37.149Z + 2025-07-08T20:31:57.102Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_expand_mask.html - 2025-07-07T21:08:36.679Z + 2025-07-08T20:31:56.625Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_chat.html - 2025-07-07T21:08:36.374Z + 2025-07-08T20:31:56.301Z https://docs.axolotl.ai/docs/api/utils.samplers.multipack.html - 2025-07-07T21:08:37.236Z + 2025-07-08T20:31:57.191Z https://docs.axolotl.ai/docs/api/integrations.grokfast.optimizer.html - 2025-07-07T21:08:37.130Z + 2025-07-08T20:31:57.084Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chatml.html - 2025-07-07T21:08:36.467Z + 2025-07-08T20:31:56.408Z https://docs.axolotl.ai/docs/api/monkeypatch.mixtral.html - 2025-07-07T21:08:36.739Z + 2025-07-08T20:31:56.683Z https://docs.axolotl.ai/docs/api/train.html - 2025-07-07T21:08:35.833Z + 2025-07-08T20:31:55.727Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_patch_multipack.html - 2025-07-07T21:08:36.719Z + 2025-07-08T20:31:56.664Z https://docs.axolotl.ai/docs/api/index.html - 2025-07-07T21:08:35.771Z + 2025-07-08T20:31:55.656Z https://docs.axolotl.ai/docs/api/loaders.adapter.html - 2025-07-07T21:08:36.277Z + 2025-07-08T20:31:56.201Z https://docs.axolotl.ai/docs/api/utils.schemas.multimodal.html - 2025-07-07T21:08:36.963Z + 2025-07-08T20:31:56.914Z https://docs.axolotl.ai/docs/api/kernels.lora.html - 2025-07-07T21:08:36.598Z + 2025-07-08T20:31:56.544Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.llama3.html - 2025-07-07T21:08:36.480Z + 2025-07-08T20:31:56.421Z https://docs.axolotl.ai/docs/api/cli.checks.html - 2025-07-07T21:08:36.065Z + 2025-07-08T20:31:55.980Z https://docs.axolotl.ai/docs/api/cli.quantize.html - 2025-07-07T21:08:36.185Z + 2025-07-08T20:31:56.103Z https://docs.axolotl.ai/docs/api/integrations.lm_eval.args.html - 2025-07-07T21:08:37.144Z + 2025-07-08T20:31:57.098Z https://docs.axolotl.ai/docs/api/core.chat.messages.html - 2025-07-07T21:08:35.974Z + 2025-07-08T20:31:55.883Z https://docs.axolotl.ai/docs/api/core.builders.causal.html - 2025-07-07T21:08:35.930Z + 2025-07-08T20:31:55.835Z https://docs.axolotl.ai/docs/api/core.trainers.relora.html - 2025-07-07T21:08:36.221Z + 2025-07-08T20:31:56.144Z https://docs.axolotl.ai/docs/api/models.mamba.modeling_mamba.html - 2025-07-07T21:08:37.166Z + 2025-07-08T20:31:57.121Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_cpu.html - 2025-07-07T21:08:36.742Z + 2025-07-08T20:31:56.687Z https://docs.axolotl.ai/docs/api/core.trainers.mamba.html - 2025-07-07T21:08:36.217Z + 2025-07-08T20:31:56.139Z https://docs.axolotl.ai/docs/api/core.datasets.transforms.chat_builder.html - 2025-07-07T21:08:35.991Z + 2025-07-08T20:31:55.900Z https://docs.axolotl.ai/docs/api/loaders.processor.html - 2025-07-07T21:08:36.272Z + 2025-07-08T20:31:56.196Z https://docs.axolotl.ai/docs/api/core.chat.format.llama3x.html - 2025-07-07T21:08:35.976Z + 2025-07-08T20:31:55.886Z https://docs.axolotl.ai/docs/api/prompt_strategies.messages.chat.html - 2025-07-07T21:08:36.445Z + 2025-07-08T20:31:56.386Z https://docs.axolotl.ai/docs/api/cli.train.html - 2025-07-07T21:08:36.030Z + 2025-07-08T20:31:55.945Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.optimizer.html - 2025-07-07T21:08:36.292Z + 2025-07-08T20:31:56.219Z https://docs.axolotl.ai/docs/api/utils.collators.mamba.html - 2025-07-07T21:08:37.190Z + 2025-07-08T20:31:57.145Z https://docs.axolotl.ai/docs/api/monkeypatch.unsloth_.html - 2025-07-07T21:08:36.736Z + 2025-07-08T20:31:56.680Z https://docs.axolotl.ai/docs/api/utils.dict.html - 2025-07-07T21:08:36.864Z + 2025-07-08T20:31:56.813Z https://docs.axolotl.ai/docs/api/prompt_strategies.user_defined.html - 2025-07-07T21:08:36.395Z + 2025-07-08T20:31:56.326Z https://docs.axolotl.ai/docs/api/core.training_args.html - 2025-07-07T21:08:35.951Z + 2025-07-08T20:31:55.855Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.user_defined.html - 2025-07-07T21:08:36.471Z + 2025-07-08T20:31:56.411Z https://docs.axolotl.ai/docs/api/prompt_tokenizers.html - 2025-07-07T21:08:35.910Z + 2025-07-08T20:31:55.809Z https://docs.axolotl.ai/docs/api/common.const.html - 2025-07-07T21:08:37.150Z + 2025-07-08T20:31:57.104Z https://docs.axolotl.ai/docs/fsdp_qlora.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/custom_integrations.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/getting-started.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/faq.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/lora_optims.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/rlhf.html - 2025-07-07T21:05:27.508Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/amd_hpc.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/installation.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/multipack.html - 2025-07-07T21:05:27.507Z + 2025-07-08T20:28:26.220Z https://docs.axolotl.ai/docs/dataset_preprocessing.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/dataset_loading.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.216Z https://docs.axolotl.ai/docs/dataset-formats/inst_tune.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/dataset-formats/template_free.html - 2025-07-07T21:05:27.503Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/dataset-formats/index.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z https://docs.axolotl.ai/docs/dataset-formats/pretraining.html - 2025-07-07T21:05:27.502Z + 2025-07-08T20:28:26.215Z