diff --git a/.nojekyll b/.nojekyll index 5a3685ab6..ec5b23aaa 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -3e16e425 \ No newline at end of file +55409b32 \ No newline at end of file diff --git a/docs/api/utils.distributed.html b/docs/api/utils.distributed.html index 428add302..5704040a9 100644 --- a/docs/api/utils.distributed.html +++ b/docs/api/utils.distributed.html @@ -459,7 +459,6 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
  • is_main_process
  • reduce_and_broadcast
  • zero_first
    • -
  • zero_only
    • @@ -521,10 +520,6 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin zero_first runs the wrapped context so that rank 0 runs first before other ranks - -zero_only -Context manager that only runs the enclosed block on the main rank. -
    @@ -579,9 +574,13 @@ The value is then broadcasted to all other ranks.

    is_main_process

    -
    utils.distributed.is_main_process()
    +
    utils.distributed.is_main_process(use_environ=False)

    Check if the current process is the main process. If not in distributed mode, always return True.

    +

    Args: +- use_environ (bool, optional): Use environment variable to determine main process.

    +

    Returns: +- bool: True if the current process is the main process, False otherwise.

    reduce_and_broadcast

    @@ -599,11 +598,6 @@ and then broadcast the reduced result to all ranks.

    zero_first

    utils.distributed.zero_first(is_main)

    runs the wrapped context so that rank 0 runs first before other ranks

    -
    -
    -

    zero_only

    -
    utils.distributed.zero_only()
    -

    Context manager that only runs the enclosed block on the main rank.

    diff --git a/search.json b/search.json index 7602408e7..eb5088008 100644 --- a/search.json +++ b/search.json @@ -554,14 +554,14 @@ "href": "docs/api/utils.distributed.html", "title": "utils.distributed", "section": "", - "text": "utils.distributed\nutility helpers for distributed checks\n\n\n\n\n\nName\nDescription\n\n\n\n\nbarrier\nActs as a barrier to wait for all processes. This ensures that all processes\n\n\ncleanup_distributed\nDestroy process group if torch distributed is initialized. Called in training early\n\n\ncompute_and_broadcast\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\n\n\ngather_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\ngather_scalar_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\nis_distributed\nCheck if distributed training is initialized.\n\n\nis_main_process\nCheck if the current process is the main process. If not in distributed mode,\n\n\nreduce_and_broadcast\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\n\n\nzero_first\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\nzero_only\nContext manager that only runs the enclosed block on the main rank.\n\n\n\n\n\nutils.distributed.barrier()\nActs as a barrier to wait for all processes. This ensures that all processes\nreach the barrier before proceeding further.\n\n\n\nutils.distributed.cleanup_distributed()\nDestroy process group if torch distributed is initialized. Called in training early\ntermination or when training successfully completes.\n\n\n\nutils.distributed.compute_and_broadcast(fn)\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\nThe value is then broadcasted to all other ranks.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that computes the value. Default is 0.\nReturns:\n- The computed value (int or float).\n\n\n\nutils.distributed.gather_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.gather_scalar_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.is_distributed()\nCheck if distributed training is initialized.\n\n\n\nutils.distributed.is_main_process()\nCheck if the current process is the main process. If not in distributed mode,\nalways return True.\n\n\n\nutils.distributed.reduce_and_broadcast(fn1, fn2)\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\nand then broadcast the reduced result to all ranks.\nArgs:\n- fn1 (callable): A function that computes the value on each rank.\n- fn2 (callable): A reduction function that takes a list of values and returns a single value.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- The reduced and broadcasted value.\n\n\n\nutils.distributed.zero_first(is_main)\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\n\nutils.distributed.zero_only()\nContext manager that only runs the enclosed block on the main rank." + "text": "utils.distributed\nutility helpers for distributed checks\n\n\n\n\n\nName\nDescription\n\n\n\n\nbarrier\nActs as a barrier to wait for all processes. This ensures that all processes\n\n\ncleanup_distributed\nDestroy process group if torch distributed is initialized. Called in training early\n\n\ncompute_and_broadcast\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\n\n\ngather_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\ngather_scalar_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\nis_distributed\nCheck if distributed training is initialized.\n\n\nis_main_process\nCheck if the current process is the main process. If not in distributed mode,\n\n\nreduce_and_broadcast\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\n\n\nzero_first\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\n\n\n\nutils.distributed.barrier()\nActs as a barrier to wait for all processes. This ensures that all processes\nreach the barrier before proceeding further.\n\n\n\nutils.distributed.cleanup_distributed()\nDestroy process group if torch distributed is initialized. Called in training early\ntermination or when training successfully completes.\n\n\n\nutils.distributed.compute_and_broadcast(fn)\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\nThe value is then broadcasted to all other ranks.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that computes the value. Default is 0.\nReturns:\n- The computed value (int or float).\n\n\n\nutils.distributed.gather_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.gather_scalar_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.is_distributed()\nCheck if distributed training is initialized.\n\n\n\nutils.distributed.is_main_process(use_environ=False)\nCheck if the current process is the main process. If not in distributed mode,\nalways return True.\nArgs:\n- use_environ (bool, optional): Use environment variable to determine main process.\nReturns:\n- bool: True if the current process is the main process, False otherwise.\n\n\n\nutils.distributed.reduce_and_broadcast(fn1, fn2)\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\nand then broadcast the reduced result to all ranks.\nArgs:\n- fn1 (callable): A function that computes the value on each rank.\n- fn2 (callable): A reduction function that takes a list of values and returns a single value.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- The reduced and broadcasted value.\n\n\n\nutils.distributed.zero_first(is_main)\nruns the wrapped context so that rank 0 runs first before other ranks" }, { "objectID": "docs/api/utils.distributed.html#functions", "href": "docs/api/utils.distributed.html#functions", "title": "utils.distributed", "section": "", - "text": "Name\nDescription\n\n\n\n\nbarrier\nActs as a barrier to wait for all processes. This ensures that all processes\n\n\ncleanup_distributed\nDestroy process group if torch distributed is initialized. Called in training early\n\n\ncompute_and_broadcast\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\n\n\ngather_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\ngather_scalar_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\nis_distributed\nCheck if distributed training is initialized.\n\n\nis_main_process\nCheck if the current process is the main process. If not in distributed mode,\n\n\nreduce_and_broadcast\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\n\n\nzero_first\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\nzero_only\nContext manager that only runs the enclosed block on the main rank.\n\n\n\n\n\nutils.distributed.barrier()\nActs as a barrier to wait for all processes. This ensures that all processes\nreach the barrier before proceeding further.\n\n\n\nutils.distributed.cleanup_distributed()\nDestroy process group if torch distributed is initialized. Called in training early\ntermination or when training successfully completes.\n\n\n\nutils.distributed.compute_and_broadcast(fn)\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\nThe value is then broadcasted to all other ranks.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that computes the value. Default is 0.\nReturns:\n- The computed value (int or float).\n\n\n\nutils.distributed.gather_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.gather_scalar_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.is_distributed()\nCheck if distributed training is initialized.\n\n\n\nutils.distributed.is_main_process()\nCheck if the current process is the main process. If not in distributed mode,\nalways return True.\n\n\n\nutils.distributed.reduce_and_broadcast(fn1, fn2)\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\nand then broadcast the reduced result to all ranks.\nArgs:\n- fn1 (callable): A function that computes the value on each rank.\n- fn2 (callable): A reduction function that takes a list of values and returns a single value.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- The reduced and broadcasted value.\n\n\n\nutils.distributed.zero_first(is_main)\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\n\nutils.distributed.zero_only()\nContext manager that only runs the enclosed block on the main rank." + "text": "Name\nDescription\n\n\n\n\nbarrier\nActs as a barrier to wait for all processes. This ensures that all processes\n\n\ncleanup_distributed\nDestroy process group if torch distributed is initialized. Called in training early\n\n\ncompute_and_broadcast\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\n\n\ngather_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\ngather_scalar_from_all_ranks\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\n\n\nis_distributed\nCheck if distributed training is initialized.\n\n\nis_main_process\nCheck if the current process is the main process. If not in distributed mode,\n\n\nreduce_and_broadcast\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\n\n\nzero_first\nruns the wrapped context so that rank 0 runs first before other ranks\n\n\n\n\n\nutils.distributed.barrier()\nActs as a barrier to wait for all processes. This ensures that all processes\nreach the barrier before proceeding further.\n\n\n\nutils.distributed.cleanup_distributed()\nDestroy process group if torch distributed is initialized. Called in training early\ntermination or when training successfully completes.\n\n\n\nutils.distributed.compute_and_broadcast(fn)\nCompute a value using the function ‘fn’ only on the specified rank (default is 0).\nThe value is then broadcasted to all other ranks.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that computes the value. Default is 0.\nReturns:\n- The computed value (int or float).\n\n\n\nutils.distributed.gather_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.gather_scalar_from_all_ranks(fn, world_size=1)\nRun a callable ‘fn’ on all ranks and gather the results on the specified rank.\nArgs:\n- fn (callable): A function that computes the value. This should not have any side effects.\n- rank (int, optional): The rank that gathers the values. Default is 0.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- A list of computed values from all ranks if on the gathering rank, otherwise None.\n\n\n\nutils.distributed.is_distributed()\nCheck if distributed training is initialized.\n\n\n\nutils.distributed.is_main_process(use_environ=False)\nCheck if the current process is the main process. If not in distributed mode,\nalways return True.\nArgs:\n- use_environ (bool, optional): Use environment variable to determine main process.\nReturns:\n- bool: True if the current process is the main process, False otherwise.\n\n\n\nutils.distributed.reduce_and_broadcast(fn1, fn2)\nRun a callable ‘fn1’ on all ranks, gather the results, reduce them using ‘fn2’,\nand then broadcast the reduced result to all ranks.\nArgs:\n- fn1 (callable): A function that computes the value on each rank.\n- fn2 (callable): A reduction function that takes a list of values and returns a single value.\n- world_size (int, optional): Total number of processes in the current distributed setup.\nReturns:\n- The reduced and broadcasted value.\n\n\n\nutils.distributed.zero_first(is_main)\nruns the wrapped context so that rank 0 runs first before other ranks" }, { "objectID": "docs/api/prompt_strategies.pygmalion.html", diff --git a/sitemap.xml b/sitemap.xml index 14aed0947..d548f409c 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -2,682 +2,682 @@ https://docs.axolotl.ai/examples/colab-notebooks/colab-axolotl-example.html - 2025-04-30T15:35:57.590Z + 2025-04-30T17:11:15.832Z https://docs.axolotl.ai/index.html - 2025-04-30T15:35:57.602Z + 2025-04-30T17:11:15.844Z https://docs.axolotl.ai/docs/rlhf.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/unsloth.html - 2025-04-30T15:35:57.590Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/dataset_preprocessing.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/input_output.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/dataset_loading.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/api/utils.collators.mamba.html - 2025-04-30T15:36:29.239Z + 2025-04-30T17:12:03.371Z https://docs.axolotl.ai/docs/api/utils.optimizers.adopt.html - 2025-04-30T15:36:29.013Z + 2025-04-30T17:12:03.145Z https://docs.axolotl.ai/docs/api/prompt_strategies.user_defined.html - 2025-04-30T15:36:28.522Z + 2025-04-30T17:12:02.657Z https://docs.axolotl.ai/docs/api/utils.chat_templates.html - 2025-04-30T15:36:28.914Z + 2025-04-30T17:12:03.045Z https://docs.axolotl.ai/docs/api/cli.merge_lora.html - 2025-04-30T15:36:28.346Z + 2025-04-30T17:12:02.488Z https://docs.axolotl.ai/docs/api/monkeypatch.multipack.html - 2025-04-30T15:36:28.797Z + 2025-04-30T17:12:02.928Z https://docs.axolotl.ai/docs/api/core.chat.format.shared.html - 2025-04-30T15:36:28.221Z + 2025-04-30T17:12:02.365Z https://docs.axolotl.ai/docs/api/utils.schemas.integrations.html - 2025-04-30T15:36:29.087Z + 2025-04-30T17:12:03.220Z https://docs.axolotl.ai/docs/api/utils.freeze.html - 2025-04-30T15:36:28.940Z + 2025-04-30T17:12:03.070Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_w_system.html - 2025-04-30T15:36:28.514Z + 2025-04-30T17:12:02.649Z https://docs.axolotl.ai/docs/api/monkeypatch.attention.mllama.html - 2025-04-30T15:36:28.865Z + 2025-04-30T17:12:02.996Z https://docs.axolotl.ai/docs/api/utils.schemas.model.html - 2025-04-30T15:36:29.034Z + 2025-04-30T17:12:03.169Z https://docs.axolotl.ai/docs/api/core.datasets.transforms.chat_builder.html - 2025-04-30T15:36:28.235Z + 2025-04-30T17:12:02.378Z https://docs.axolotl.ai/docs/api/monkeypatch.btlm_attn_hijack_flash.html - 2025-04-30T15:36:28.840Z + 2025-04-30T17:12:02.971Z https://docs.axolotl.ai/docs/api/models.mamba.modeling_mamba.html - 2025-04-30T15:36:29.215Z + 2025-04-30T17:12:03.347Z https://docs.axolotl.ai/docs/api/core.datasets.chat.html - 2025-04-30T15:36:28.227Z + 2025-04-30T17:12:02.370Z https://docs.axolotl.ai/docs/api/utils.model_shard_quant.html - 2025-04-30T15:36:28.928Z + 2025-04-30T17:12:03.059Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_patch_multipack.html - 2025-04-30T15:36:28.841Z + 2025-04-30T17:12:02.972Z https://docs.axolotl.ai/docs/api/utils.lora_embeddings.html - 2025-04-30T15:36:28.923Z + 2025-04-30T17:12:03.053Z https://docs.axolotl.ai/docs/api/utils.schemas.peft.html - 2025-04-30T15:36:29.066Z + 2025-04-30T17:12:03.200Z https://docs.axolotl.ai/docs/api/core.chat.format.llama3x.html - 2025-04-30T15:36:28.220Z + 2025-04-30T17:12:02.364Z https://docs.axolotl.ai/docs/api/monkeypatch.mistral_attn_hijack_flash.html - 2025-04-30T15:36:28.795Z + 2025-04-30T17:12:02.927Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chat_template.html - 2025-04-30T15:36:28.575Z + 2025-04-30T17:12:02.709Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.user_defined.html - 2025-04-30T15:36:28.598Z + 2025-04-30T17:12:02.732Z https://docs.axolotl.ai/docs/api/utils.distributed.html - 2025-04-30T15:36:29.003Z + 2025-04-30T17:12:03.131Z https://docs.axolotl.ai/docs/api/prompt_strategies.pygmalion.html - 2025-04-30T15:36:28.569Z + 2025-04-30T17:12:02.704Z https://docs.axolotl.ai/docs/api/utils.dict.html - 2025-04-30T15:36:29.006Z + 2025-04-30T17:12:03.134Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_flash.html - 2025-04-30T15:36:28.780Z + 2025-04-30T17:12:02.911Z https://docs.axolotl.ai/docs/api/prompt_strategies.base.html - 2025-04-30T15:36:28.468Z + 2025-04-30T17:12:02.604Z https://docs.axolotl.ai/docs/api/monkeypatch.data.batch_dataset_fetcher.html - 2025-04-30T15:36:28.867Z + 2025-04-30T17:12:02.997Z https://docs.axolotl.ai/docs/api/kernels.swiglu.html - 2025-04-30T15:36:28.745Z + 2025-04-30T17:12:02.877Z https://docs.axolotl.ai/docs/api/core.chat.messages.html - 2025-04-30T15:36:28.217Z + 2025-04-30T17:12:02.361Z https://docs.axolotl.ai/docs/api/index.html - 2025-04-30T15:36:27.950Z + 2025-04-30T17:12:02.097Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.zephyr.html - 2025-04-30T15:36:28.597Z + 2025-04-30T17:12:02.731Z https://docs.axolotl.ai/docs/api/convert.html - 2025-04-30T15:36:28.044Z + 2025-04-30T17:12:02.189Z https://docs.axolotl.ai/docs/api/utils.schemas.multimodal.html - 2025-04-30T15:36:29.075Z + 2025-04-30T17:12:03.208Z https://docs.axolotl.ai/docs/api/core.trainers.base.html - 2025-04-30T15:36:28.434Z + 2025-04-30T17:12:02.573Z https://docs.axolotl.ai/docs/api/evaluate.html - 2025-04-30T15:36:28.023Z + 2025-04-30T17:12:02.168Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_xformers.html - 2025-04-30T15:36:28.781Z + 2025-04-30T17:12:02.913Z https://docs.axolotl.ai/docs/api/kernels.quantize.html - 2025-04-30T15:36:28.752Z + 2025-04-30T17:12:02.884Z https://docs.axolotl.ai/docs/api/utils.callbacks.mlflow_.html - 2025-04-30T15:36:29.268Z + 2025-04-30T17:12:03.399Z https://docs.axolotl.ai/docs/api/utils.callbacks.profiler.html - 2025-04-30T15:36:29.262Z + 2025-04-30T17:12:03.394Z https://docs.axolotl.ai/docs/api/core.trainers.dpo.trainer.html - 2025-04-30T15:36:28.463Z + 2025-04-30T17:12:02.599Z https://docs.axolotl.ai/docs/api/cli.vllm_serve.html - 2025-04-30T15:36:28.409Z + 2025-04-30T17:12:02.549Z https://docs.axolotl.ai/docs/api/train.html - 2025-04-30T15:36:28.012Z + 2025-04-30T17:12:02.157Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chatml.html - 2025-04-30T15:36:28.595Z + 2025-04-30T17:12:02.729Z https://docs.axolotl.ai/docs/api/utils.schemas.trl.html - 2025-04-30T15:36:29.069Z + 2025-04-30T17:12:03.203Z https://docs.axolotl.ai/docs/api/kernels.geglu.html - 2025-04-30T15:36:28.735Z + 2025-04-30T17:12:02.867Z https://docs.axolotl.ai/docs/api/utils.bench.html - 2025-04-30T15:36:28.932Z + 2025-04-30T17:12:03.062Z https://docs.axolotl.ai/docs/api/monkeypatch.transformers_fa_utils.html - 2025-04-30T15:36:28.857Z + 2025-04-30T17:12:02.987Z https://docs.axolotl.ai/docs/api/integrations.liger.args.html - 2025-04-30T15:36:29.188Z + 2025-04-30T17:12:03.320Z https://docs.axolotl.ai/docs/api/core.trainer_builder.html - 2025-04-30T15:36:28.106Z + 2025-04-30T17:12:02.250Z https://docs.axolotl.ai/docs/api/utils.schemas.utils.html - 2025-04-30T15:36:29.099Z + 2025-04-30T17:12:03.233Z https://docs.axolotl.ai/docs/api/kernels.lora.html - 2025-04-30T15:36:28.724Z + 2025-04-30T17:12:02.856Z https://docs.axolotl.ai/docs/api/prompt_strategies.bradley_terry.llama3.html - 2025-04-30T15:36:28.641Z + 2025-04-30T17:12:02.775Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.trainer.html - 2025-04-30T15:36:28.467Z + 2025-04-30T17:12:02.603Z https://docs.axolotl.ai/docs/api/prompt_strategies.messages.chat.html - 2025-04-30T15:36:28.573Z + 2025-04-30T17:12:02.708Z https://docs.axolotl.ai/docs/api/utils.tokenization.html - 2025-04-30T15:36:28.904Z + 2025-04-30T17:12:03.035Z https://docs.axolotl.ai/docs/api/prompt_strategies.chat_template.html - 2025-04-30T15:36:28.487Z + 2025-04-30T17:12:02.622Z https://docs.axolotl.ai/docs/api/prompt_strategies.stepwise_supervised.html - 2025-04-30T15:36:28.552Z + 2025-04-30T17:12:02.686Z https://docs.axolotl.ai/docs/api/utils.samplers.multipack.html - 2025-04-30T15:36:29.252Z + 2025-04-30T17:12:03.384Z https://docs.axolotl.ai/docs/api/cli.args.html - 2025-04-30T15:36:28.299Z + 2025-04-30T17:12:02.442Z https://docs.axolotl.ai/docs/api/utils.callbacks.perplexity.html - 2025-04-30T15:36:29.259Z + 2025-04-30T17:12:03.391Z https://docs.axolotl.ai/docs/api/utils.gradient_checkpointing.unsloth.html - 2025-04-30T15:36:29.020Z + 2025-04-30T17:12:03.154Z https://docs.axolotl.ai/docs/mac.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/config.html - 2025-04-30T15:35:57.585Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/multimodal.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/lr_groups.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/dataset-formats/index.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/dataset-formats/stepwise_supervised.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/dataset-formats/pretraining.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/multi-node.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/sequence_parallelism.html - 2025-04-30T15:35:57.590Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/batch_vs_grad.html - 2025-04-30T15:35:57.585Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/amd_hpc.html - 2025-04-30T15:35:57.585Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/faq.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/custom_integrations.html - 2025-04-30T15:35:57.585Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/src/axolotl/integrations/LICENSE.html - 2025-04-30T15:35:57.606Z + 2025-04-30T17:11:15.848Z https://docs.axolotl.ai/TODO.html - 2025-04-30T15:35:57.584Z + 2025-04-30T17:11:15.826Z https://docs.axolotl.ai/src/axolotl/integrations/cut_cross_entropy/ACKNOWLEDGEMENTS.html - 2025-04-30T15:35:57.606Z + 2025-04-30T17:11:15.848Z https://docs.axolotl.ai/docs/getting-started.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/multipack.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/multi-gpu.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/installation.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/cli.html - 2025-04-30T15:35:57.585Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/dataset-formats/inst_tune.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/dataset-formats/tokenized.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/dataset-formats/conversation.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.827Z https://docs.axolotl.ai/docs/dataset-formats/template_free.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/reward_modelling.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/lora_optims.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/nccl.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/api/logging_config.html - 2025-04-30T15:36:28.091Z + 2025-04-30T17:12:02.235Z https://docs.axolotl.ai/docs/api/utils.trainer.html - 2025-04-30T15:36:28.957Z + 2025-04-30T17:12:03.087Z https://docs.axolotl.ai/docs/api/monkeypatch.unsloth_.html - 2025-04-30T15:36:28.858Z + 2025-04-30T17:12:02.989Z https://docs.axolotl.ai/docs/api/cli.evaluate.html - 2025-04-30T15:36:28.282Z + 2025-04-30T17:12:02.425Z https://docs.axolotl.ai/docs/api/kernels.utils.html - 2025-04-30T15:36:28.754Z + 2025-04-30T17:12:02.886Z https://docs.axolotl.ai/docs/api/datasets.html - 2025-04-30T15:36:28.030Z + 2025-04-30T17:12:02.175Z https://docs.axolotl.ai/docs/api/utils.models.html - 2025-04-30T15:36:28.897Z + 2025-04-30T17:12:03.028Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.llama3.html - 2025-04-30T15:36:28.608Z + 2025-04-30T17:12:02.742Z https://docs.axolotl.ai/docs/api/cli.preprocess.html - 2025-04-30T15:36:28.366Z + 2025-04-30T17:12:02.508Z https://docs.axolotl.ai/docs/api/cli.merge_sharded_fsdp_weights.html - 2025-04-30T15:36:28.358Z + 2025-04-30T17:12:02.500Z https://docs.axolotl.ai/docs/api/integrations.base.html - 2025-04-30T15:36:29.172Z + 2025-04-30T17:12:03.305Z https://docs.axolotl.ai/docs/api/prompt_strategies.orpo.chat_template.html - 2025-04-30T15:36:28.638Z + 2025-04-30T17:12:02.771Z https://docs.axolotl.ai/docs/api/utils.schemas.enums.html - 2025-04-30T15:36:29.093Z + 2025-04-30T17:12:03.227Z https://docs.axolotl.ai/docs/api/utils.callbacks.comet_.html - 2025-04-30T15:36:29.271Z + 2025-04-30T17:12:03.403Z https://docs.axolotl.ai/docs/api/prompt_strategies.input_output.html - 2025-04-30T15:36:28.547Z + 2025-04-30T17:12:02.682Z https://docs.axolotl.ai/docs/api/utils.schedulers.html - 2025-04-30T15:36:28.982Z + 2025-04-30T17:12:03.111Z https://docs.axolotl.ai/docs/api/cli.cloud.base.html - 2025-04-30T15:36:28.413Z + 2025-04-30T17:12:02.552Z https://docs.axolotl.ai/docs/api/cli.utils.html - 2025-04-30T15:36:28.405Z + 2025-04-30T17:12:02.545Z https://docs.axolotl.ai/docs/api/monkeypatch.lora_kernels.html - 2025-04-30T15:36:28.831Z + 2025-04-30T17:12:02.961Z https://docs.axolotl.ai/docs/api/prompt_strategies.completion.html - 2025-04-30T15:36:28.541Z + 2025-04-30T17:12:02.676Z https://docs.axolotl.ai/docs/api/utils.schemas.config.html - 2025-04-30T15:36:29.028Z + 2025-04-30T17:12:03.162Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_expand_mask.html - 2025-04-30T15:36:28.805Z + 2025-04-30T17:12:02.936Z https://docs.axolotl.ai/docs/api/cli.config.html - 2025-04-30T15:36:28.323Z + 2025-04-30T17:12:02.466Z https://docs.axolotl.ai/docs/api/utils.collators.core.html - 2025-04-30T15:36:29.216Z + 2025-04-30T17:12:03.349Z https://docs.axolotl.ai/docs/api/utils.data.sft.html - 2025-04-30T15:36:29.016Z + 2025-04-30T17:12:03.151Z https://docs.axolotl.ai/docs/api/integrations.spectrum.args.html - 2025-04-30T15:36:29.194Z + 2025-04-30T17:12:03.327Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.llama3.html - 2025-04-30T15:36:28.585Z + 2025-04-30T17:12:02.719Z https://docs.axolotl.ai/docs/api/cli.inference.html - 2025-04-30T15:36:28.338Z + 2025-04-30T17:12:02.480Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_instruct.html - 2025-04-30T15:36:28.502Z + 2025-04-30T17:12:02.637Z https://docs.axolotl.ai/docs/api/utils.collators.batching.html - 2025-04-30T15:36:29.236Z + 2025-04-30T17:12:03.368Z https://docs.axolotl.ai/docs/api/utils.schemas.datasets.html - 2025-04-30T15:36:29.057Z + 2025-04-30T17:12:03.191Z https://docs.axolotl.ai/docs/api/utils.data.pretraining.html - 2025-04-30T15:36:29.015Z + 2025-04-30T17:12:03.148Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.passthrough.html - 2025-04-30T15:36:28.600Z + 2025-04-30T17:12:02.734Z https://docs.axolotl.ai/docs/api/monkeypatch.utils.html - 2025-04-30T15:36:28.839Z + 2025-04-30T17:12:02.969Z https://docs.axolotl.ai/docs/api/utils.callbacks.lisa.html - 2025-04-30T15:36:29.264Z + 2025-04-30T17:12:03.396Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_chat.html - 2025-04-30T15:36:28.501Z + 2025-04-30T17:12:02.636Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.user_defined.html - 2025-04-30T15:36:28.617Z + 2025-04-30T17:12:02.751Z https://docs.axolotl.ai/docs/api/core.training_args.html - 2025-04-30T15:36:28.194Z + 2025-04-30T17:12:02.338Z https://docs.axolotl.ai/docs/api/utils.collators.mm_chat.html - 2025-04-30T15:36:29.244Z + 2025-04-30T17:12:03.376Z https://docs.axolotl.ai/docs/api/common.architectures.html - 2025-04-30T15:36:29.196Z + 2025-04-30T17:12:03.328Z https://docs.axolotl.ai/docs/api/integrations.kd.trainer.html - 2025-04-30T15:36:29.184Z + 2025-04-30T17:12:03.317Z https://docs.axolotl.ai/docs/api/prompt_strategies.llama2_chat.html - 2025-04-30T15:36:28.535Z + 2025-04-30T17:12:02.670Z https://docs.axolotl.ai/docs/api/core.trainers.trl.html - 2025-04-30T15:36:28.451Z + 2025-04-30T17:12:02.590Z https://docs.axolotl.ai/docs/api/prompt_strategies.orcamini.html - 2025-04-30T15:36:28.562Z + 2025-04-30T17:12:02.697Z https://docs.axolotl.ai/docs/api/integrations.cut_cross_entropy.args.html - 2025-04-30T15:36:29.176Z + 2025-04-30T17:12:03.308Z https://docs.axolotl.ai/docs/api/utils.lora.html - 2025-04-30T15:36:28.919Z + 2025-04-30T17:12:03.050Z https://docs.axolotl.ai/docs/api/cli.sweeps.html - 2025-04-30T15:36:28.372Z + 2025-04-30T17:12:02.514Z https://docs.axolotl.ai/docs/api/core.chat.format.chatml.html - 2025-04-30T15:36:28.219Z + 2025-04-30T17:12:02.362Z https://docs.axolotl.ai/docs/api/common.const.html - 2025-04-30T15:36:29.197Z + 2025-04-30T17:12:03.330Z https://docs.axolotl.ai/docs/api/prompt_strategies.metharme.html - 2025-04-30T15:36:28.559Z + 2025-04-30T17:12:02.693Z https://docs.axolotl.ai/docs/api/monkeypatch.stablelm_attn_hijack_flash.html - 2025-04-30T15:36:28.847Z + 2025-04-30T17:12:02.978Z https://docs.axolotl.ai/docs/api/integrations.grokfast.optimizer.html - 2025-04-30T15:36:29.177Z + 2025-04-30T17:12:03.309Z https://docs.axolotl.ai/docs/api/cli.checks.html - 2025-04-30T15:36:28.306Z + 2025-04-30T17:12:02.448Z https://docs.axolotl.ai/docs/api/common.datasets.html - 2025-04-30T15:36:29.214Z + 2025-04-30T17:12:03.346Z https://docs.axolotl.ai/docs/api/integrations.lm_eval.args.html - 2025-04-30T15:36:29.191Z + 2025-04-30T17:12:03.323Z https://docs.axolotl.ai/docs/api/monkeypatch.mixtral.html - 2025-04-30T15:36:28.868Z + 2025-04-30T17:12:02.999Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.chatml.html - 2025-04-30T15:36:28.616Z + 2025-04-30T17:12:02.750Z https://docs.axolotl.ai/docs/api/cli.train.html - 2025-04-30T15:36:28.274Z + 2025-04-30T17:12:02.417Z https://docs.axolotl.ai/docs/api/prompt_tokenizers.html - 2025-04-30T15:36:28.086Z + 2025-04-30T17:12:02.230Z https://docs.axolotl.ai/docs/api/utils.schemas.training.html - 2025-04-30T15:36:29.039Z + 2025-04-30T17:12:03.174Z https://docs.axolotl.ai/docs/api/monkeypatch.relora.html - 2025-04-30T15:36:28.804Z + 2025-04-30T17:12:02.935Z https://docs.axolotl.ai/docs/api/cli.cloud.modal_.html - 2025-04-30T15:36:28.420Z + 2025-04-30T17:12:02.559Z https://docs.axolotl.ai/docs/api/cli.main.html - 2025-04-30T15:36:28.266Z + 2025-04-30T17:12:02.409Z https://docs.axolotl.ai/docs/api/monkeypatch.trainer_fsdp_optim.html - 2025-04-30T15:36:28.850Z + 2025-04-30T17:12:02.981Z https://docs.axolotl.ai/docs/fsdp_qlora.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/debugging.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/ray-integration.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/docker.html - 2025-04-30T15:35:57.586Z + 2025-04-30T17:11:15.828Z https://docs.axolotl.ai/docs/inference.html - 2025-04-30T15:35:57.589Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/docs/torchao.html - 2025-04-30T15:35:57.590Z + 2025-04-30T17:11:15.831Z https://docs.axolotl.ai/FAQS.html - 2025-04-30T15:35:57.584Z + 2025-04-30T17:11:15.826Z