Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 5 additions & 5 deletions docs/source/api_ref_models.rst
Original file line number Diff line number Diff line change
Expand Up @@ -99,7 +99,7 @@ To download the Llama3.1-405B-Instruct model:
tune download meta-llama/Meta-Llama-3.1-405B-Instruct --ignore-patterns "original/consolidated*" --hf-token <HF_TOKEN>

To download the Llama3 weights of the above models, you can instead download from `Meta-Llama-3-8B-Instruct` and
`Meta-Llama-3-70B-Instruct`.
`Meta-Llama-3-70B-Instruct`, and remove the ignore patterns flag.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

unfortunately, 8b uses meta ckpt, and 70b uses hg ckpt. Ideally, we should update their configs to all use hf, so we can always ignore consolidated. To avoid adding too much info here, i suggested that the user can just remove the ignore patterns flag, downloading all models.


.. autosummary::
:toctree: generated/
Expand Down Expand Up @@ -217,7 +217,7 @@ To download the Qwen2.5 1.5B model, for example:

.. code-block:: bash

tune download Qwen/Qwen2.5-1.5B-Instruct --output-dir /tmp/Qwen2_5-1_5B-Instruct --ignore-patterns None
tune download Qwen/Qwen2.5-1.5B-Instruct --output-dir /tmp/Qwen2_5-1_5B-Instruct

.. autosummary::
:toctree: generated/
Expand Down Expand Up @@ -258,7 +258,7 @@ To download the Qwen2 1.5B model, for example:

.. code-block:: bash

tune download Qwen/Qwen2-1.5B-Instruct --output-dir /tmp/Qwen2-1.5B-Instruct --ignore-patterns None
tune download Qwen/Qwen2-1.5B-Instruct --output-dir /tmp/Qwen2-1.5B-Instruct

.. autosummary::
:toctree: generated/
Expand All @@ -283,7 +283,7 @@ To download the Phi-3 Mini 4k instruct model:

.. code-block:: bash

tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>

.. autosummary::
:toctree: generated/
Expand All @@ -307,7 +307,7 @@ To download the Mistral 7B v0.1 model:

.. code-block:: bash

tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --hf-token <HF_TOKEN>
tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>

.. autosummary::
:toctree: generated/
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/13B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 4 full_finetune_distributed --config llama2/13B_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/13B_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 4 lora_finetune_distributed --config llama2/13B_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/13B_qlora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-13b-hf --output-dir /tmp/Llama-2-13b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama2/13B_qlora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/70B_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-70b-hf --output-dir /tmp/Llama-2-70b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-70b-hf --output-dir /tmp/Llama-2-70b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# This config needs 8 GPUs to run
# # tune run --nproc_per_node 8 lora_finetune_distributed --config llama2/70B_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/70B_qlora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-70b-hf --output-dir /tmp/Llama-2-70b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-70b-hf --output-dir /tmp/Llama-2-70b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# This config needs 8 GPUs to run
# # tune run --nproc_per_node 8 lora_finetune_distributed --config llama2/70B_qlora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 4 full_finetune_distributed --config llama2/7B_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_full_low_memory.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
# you can install it with
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 2 lora_finetune_distributed --config llama2/7B_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_lora_dpo.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 2 lora_dpo_distributed --config llama2/7B_lora_dpo
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_lora_dpo_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_dpo_single_device --config llama2/7B_lora_dpo_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_lora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama2/7B_lora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_qat_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 4 qat_distributed --config llama2/7B_qat_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_qlora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 2 lora_finetune_distributed --config llama2/7B_qlora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/7B_qlora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --hf-token <HF_TOKEN>
# tune download meta-llama/Llama-2-7b-hf --output-dir /tmp/Llama-2-7b-hf --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama2/7B_qlora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama2/generation_v2.yaml
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
# Config for running the InferenceRecipe in generate_V2.py to generate output from an LLM
#
# This config assumes that you've run the following command before launching:
# tune download meta-llama/Llama-2-7b-chat-hf --output-dir /tmp/Llama-2-7b-chat-hf
# tune download meta-llama/Llama-2-7b-chat-hf --output-dir /tmp/Llama-2-7b-chat-hf --ignore-patterns "*.safetensors"
#
# To launch, run the following command:
# tune run dev/generate_v2 --config llama2/generation_v2
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/70B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-70B-Instruct --output-dir /tmp/Meta-Llama-3-70B-Instruct --ignore-patterns "original/consolidated*" --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-70B-Instruct --output-dir /tmp/Meta-Llama-3-70B-Instruct --ignore-patterns "original/consolidated*" --hf-token <HF_TOKEN>
#
# To launch on 8 devices, run the following command from root:
# tune run --nproc_per_node 8 full_finetune_distributed --config llama3/70B_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_dora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 2 lora_finetune_distributed --config llama3/8B_dora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nproc_per_node 4 full_finetune_distributed --config llama3/8B_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_full_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
# you can install it with
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nproc_per_node 2 lora_finetune_distributed --config llama3/8B_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_lora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama3/8B_lora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_qat_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on 4 devices, run the following command from root:
# tune run --nproc_per_node 4 qat_distributed --config llama3/8B_qat_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_qdora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama3/8B_qdora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/llama3/8B_qlora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --hf-token <HF_TOKEN>
# tune download meta-llama/Meta-Llama-3-8B-Instruct --output-dir /tmp/Meta-Llama-3-8B-Instruct --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config llama3/8B_qlora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/mistral/7B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download mistralai/Mistral-7B-v0.1 --hf-token <HF_TOKEN> --output-dir /tmp/Mistral-7B-v0.1
# tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# Run this config on 4 GPUs using the following:
# tune run --nnodes 1 --nproc_per_node 4 full_finetune_distributed --config mistral/7B_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/mistral/7B_full_low_memory.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download mistralai/Mistral-7B-v0.1 --hf-token <HF_TOKEN> --output-dir /tmp/Mistral-7B-v0.1
# tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
# you can install it with
Expand Down
4 changes: 2 additions & 2 deletions recipes/configs/mistral/7B_full_ppo_low_memory.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,8 +7,8 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download weqweasdas/RM-Mistral-7B --output-dir /tmp/RM-Mistral-7B/ --ignore-patterns None
# tune download mistralai/Mistral-7B-Instruct-v0.2 --output-dir /tmp/Mistral-7B-Instruct-v0.2/ --hf-token HF_TOKEN
# tune download weqweasdas/RM-Mistral-7B --output-dir /tmp/RM-Mistral-7B/
# tune download mistralai/Mistral-7B-Instruct-v0.2 --output-dir /tmp/Mistral-7B-Instruct-v0.2/ --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# You'll also need to ensure that {output_dir} exists beforehand, as checkpoints for policy and value models are saved in sub-folders.
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/mistral/7B_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download mistralai/Mistral-7B-v0.1 --hf-token <HF_TOKEN> --output-dir /tmp/Mistral-7B-v0.1
# tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# Run this config on 2 GPUs using the following:
# tune run --nnodes 1 --nproc_per_node 2 lora_finetune_distributed --config mistral/7B_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/mistral/7B_lora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download mistralai/Mistral-7B-v0.1 --hf-token <HF_TOKEN> --output-dir /tmp/Mistral-7B-v0.1
# tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config mistral/7B_lora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/mistral/7B_qlora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download mistralai/Mistral-7B-v0.1 --hf-token <HF_TOKEN> --output-dir /tmp/Mistral-7B-v0.1
# tune download mistralai/Mistral-7B-v0.1 --output-dir /tmp/Mistral-7B-v0.1 --ignore-patterns "*.safetensors" --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config mistral/7B_qlora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/phi3/mini_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>
#
# Run this config on 4 GPUs using the following:
# tune run --nproc_per_node 4 full_finetune_distributed --config phi3/mini_full
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/phi3/mini_full_low_memory.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>
#
# The default config uses an optimizer from bitsandbytes. If you do not have it installed,
# you can install it with
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/phi3/mini_lora.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>
#
# To launch on 2 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 2 lora_finetune_distributed --config phi3/mini_lora
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/phi3/mini_lora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config phi3/mini_lora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/phi3/mini_qlora_single_device.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --ignore-patterns None --hf-token <HF_TOKEN>
# tune download microsoft/Phi-3-mini-4k-instruct --output-dir /tmp/Phi-3-mini-4k-instruct --hf-token <HF_TOKEN>
#
# To launch on a single device, run the following command from root:
# tune run lora_finetune_single_device --config phi3/mini_qlora_single_device
Expand Down
2 changes: 1 addition & 1 deletion recipes/configs/qwen2/0.5B_full.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
#
# This config assumes that you've run the following command before launching
# this run:
# tune download Qwen/Qwen2-0.5B-Instruct --output-dir /tmp/Qwen2-0.5B-Instruct --ignore-patterns None
# tune download Qwen/Qwen2-0.5B-Instruct --output-dir /tmp/Qwen2-0.5B-Instruct
#
# To launch on 4 devices, run the following command from root:
# tune run --nnodes 1 --nproc_per_node 4 full_finetune_distributed --config qwen2/0.5B_full
Expand Down
Loading
Loading