diff --git a/run_repro.sh b/run_repro.sh
deleted file mode 100755
index 221bcbaa16..0000000000
--- a/run_repro.sh
+++ /dev/null
@@ -1,57 +0,0 @@
-#!/bin/bash
-python mason.py \
-	   --cluster ai2/jupiter-cirrascale-2 --image nathanl/open_instruct_auto \
-	   --pure_docker_mode \
-	   --workspace ai2/tulu-thinker \
-	   --priority high \
-	   --preemptible \
-	   --num_nodes 6 \
-	   --max_retries 0 \
-	   --env VLLM_ALLOW_LONG_MAX_MODEL_LEN=1 \
-	   --budget ai2/oe-adapt \
-	   --gpus 8 -- source configs/beaker_configs/ray_node_setup.sh \&\& source configs/beaker_configs/code_api_setup.sh \&\& python open_instruct/grpo_fast.py \
-	   --exp_name repro_1106rl_int_all_qwen_integration_mix_11607 \
-	   --beta 0.0 \
-	   --add_bos False \
-	   --num_samples_per_prompt_rollout 16 \
-	   --num_unique_prompts_rollout 128 \
-	   --num_mini_batches 1 \
-	   --gather_whole_model False \
-	   --num_epochs 1 \
-	   --learning_rate 5e-7 \
-	   --per_device_train_batch_size 1 \
-	   --output_dir /output \
-	   --kl_estimator kl3 \
-	   --dataset_mixer_list "saurabh5/rlvr_acecoder 56878 hamishivi/rlvr_orz_math_57k_collected 56878 hamishivi/tulu_3_rewritten_400k_string_f1_only_v2 56878 allenai/IF_multi_constraints_upto5 56878" \
-	   --dataset_mixer_list_splits train \
-	   --dataset_mixer_eval_list \
-       hamishivi/tulu_3_rewritten_100k 32 \
-	   --dataset_mixer_eval_list_splits train \
-	   --max_token_length 10240 \
-	   --max_prompt_token_length 2048 \
-	   --response_length 8192 \
-	   --pack_length 16384 \
-	   --model_name_or_path Qwen/Qwen2.5-7B \
-	   --model_revision main \
-	   --tokenizer_name_or_path hamishivi/Qwen-2.5-7b-tokenizer \
-	   --chat_template_name tulu_thinker_r1_style \
-	   --stop_strings "</answer>" \
-	   --non_stop_penalty False \
-	   --temperature 1.0 \
-	   --ground_truths_key ground_truth \
-	   --sft_messages_key messages \
-	   --total_episodes 10000000 \
-	   --deepspeed_stage 2 \
-	   --num_learners_per_node 8 8 \
-	   --vllm_num_engines 16 \
-	   --vllm_tensor_parallel_size 1 \
-	   --lr_scheduler_type constant \
-	   --apply_verifiable_reward true \
-	   --seed 1 \
-	   --num_evals 5 \
-	   --save_freq 100 \
-	   --try_launch_beaker_eval_jobs_on_weka True \
-	   --gradient_checkpointing \
-	   --with_tracking \
-	   --oe_eval_max_length 32768 \
-	   --oe_eval_tasks minerva_math::hamish_zs_reasoning,bbh:cot::hamish_zs_reasoning,gsm8k::hamish_zs_reasoning,minerva_math_500::hamish_zs_reasoning,zebralogic::hamish_zs_reasoning,aime::hamish_zs_reasoning,agi_eval_english:0shot_cot::hamish_zs_reasoning,gpqa:0shot_cot::hamish_zs_reasoning,codex_humanevalplus:0-shot-chat::tulu-thinker,ifeval::tulu,popqa::tulu,mmlu:mc::tulu,bbh:cot-v1::tulu