Skip to content
Merged
Show file tree
Hide file tree
Changes from 3 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
3 changes: 2 additions & 1 deletion scripts/train/build_image_and_launch.sh
Original file line number Diff line number Diff line change
Expand Up @@ -69,6 +69,7 @@ echo "Installing dependencies with uv..."
uv sync

# Run the provided script with the image name and all remaining arguments
export BEAKER_IMAGE="$beaker_user/$image_name"
script="$1"
shift
bash "$script" "$beaker_user/$image_name" "$@"
bash "$script" "$BEAKER_IMAGE" "$@"
Comment thread
finbarrtimbers marked this conversation as resolved.
Outdated
3 changes: 2 additions & 1 deletion scripts/train/build_image_and_launch_dirty.sh
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,7 @@ echo "Installing dependencies with uv..."
uv sync

# Run the provided script with the image name and all remaining arguments
export BEAKER_IMAGE="$beaker_user/$image_name"
script="$1"
shift
bash "$script" "$beaker_user/$image_name" "$@"
bash "$script" "$BEAKER_IMAGE" "$@"
9 changes: 5 additions & 4 deletions scripts/train/qwen/qwen3_4b_dapo_math.sh
Original file line number Diff line number Diff line change
Expand Up @@ -6,14 +6,15 @@ RUN_NAME="${RUN_NAME:-${EXP_NAME}_$(date +%Y%m%d_%H%M%S)}"
NUM_GPUS="${NUM_GPUS:-8}"
BEAKER_IMAGE="${BEAKER_IMAGE:-nathanl/open_instruct_auto}"

CLUSTER="${CLUSTER:-ai2/jupiter ai2/ceres}"
PRIORITY="${PRIORITY:-high}"
CLUSTER="${CLUSTER:-ai2/jupiter}"
PRIORITY="${PRIORITY:-urgent}"
WORKSPACE="${WORKSPACE:-ai2/olmo-instruct}"

uv run mason.py \
--task_name ${EXP_NAME} \
--description "${RUN_NAME}" \
--cluster ${CLUSTER} \
--workspace ai2/oe-adapt-code \
--workspace ${WORKSPACE} \
--priority ${PRIORITY} \
--pure_docker_mode \
--no_auto_dataset_cache \
Expand Down Expand Up @@ -43,7 +44,7 @@ uv run open_instruct/grpo_fast.py \
--per_device_train_batch_size 1 \
--dataset_mixer_list hamishivi/DAPO-Math-17k-Processed_filtered 1.0 \
--dataset_mixer_list_splits "train" \
--dataset_mixer_eval_list mnoukhov/aime_2025_openinstruct 1.0 mnoukhov/brumo_2025_openinstruct 1.0 \
--dataset_mixer_eval_list allenai/aime_2025_openinstruct 1.0 allenai/brumo_2025_openinstruct 1.0 \
--dataset_mixer_eval_list_splits "train" \
--max_prompt_token_length 2048 \
--response_length 8192 \
Expand Down
Loading