### 0
pkill -9 sglang; sleep 3; ray stop
  --force; pkill -9 ray; pkill -9 miles; sleep 3; pkill -9 ray; pkill -9 miles; pkill -9 redis; true; 

### 1
export PYTHONUNBUFFERED=1 && ray start
  --head
  --node-ip-address 127.0.0.1
  --num-gpus 8
  --disable-usage-stats

### 2
nvidia-smi topo -m 2>/dev/null | grep -o 'NV[0-9][0-9]*' | wc -l

### 3
export no_proxy=127.0.0.1 && export PYTHONUNBUFFERED=1 && ray job submit
  --address="http://127.0.0.1:8265"
  --runtime-env-json='{"env_vars": {"PYTHONUNBUFFERED": "1", "CUDA_DEVICE_MAX_CONNECTIONS": "1", "NCCL_NVLS_ENABLE": "0", "no_proxy": "127.0.0.1,127.0.0.1", "MASTER_ADDR": "127.0.0.1", "SGLANG_ENABLE_SPEC_V2": "1", "PYTHONPATH": "<REPO_ROOT>:/root/Megatron-LM:/frozen/pythonpath"}}'
  -- python3 <REPO_ROOT>/train.py
  --spec miles_plugins.models.qwen3_5 get_qwen3_5_spec
  --disable-bias-linear
  --qk-layernorm
  --group-query-attention
  --num-attention-heads 16
  --num-query-groups 2
  --kv-channels 256
  --num-layers 40
  --hidden-size 2048
  --ffn-hidden-size 512
  --normalization RMSNorm
  --apply-layernorm-1p
  --position-embedding-type rope
  --norm-epsilon 1e-6
  --rotary-percent 0.25
  --swiglu
  --untie-embeddings-and-output-weights
  --vocab-size 248320
  --rotary-base 10000000
  --moe-ffn-hidden-size 512
  --moe-shared-expert-intermediate-size 512
  --moe-router-score-function softmax
  --moe-token-dispatcher-type alltoall
  --moe-router-topk 8
  --moe-layer-freq '[1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]'
  --num-experts 256
  --moe-grouped-gemm
  --moe-token-drop-policy probs
  --moe-router-dtype fp32
  --moe-permute-fusion
  --moe-aux-loss-coeff 0
  --attention-output-gate
  --moe-shared-expert-gate
  --mtp-num-layers 1
  --hf-checkpoint /root/models/Qwen3.6-35B-A3B
  --ref-load /root/models/Qwen3.6-35B-A3B_torch_dist 
  --prompt-data /root/datasets/dapo-math-17k/dapo-math-17k.jsonl
  --input-key prompt
  --label-key label
  --apply-chat-template
  --rollout-shuffle
  --rm-type deepscaler
  --num-rollout 10
  --rollout-batch-size 8
  --n-samples-per-prompt 2
  --rollout-max-response-len 1024
  --rollout-temperature 1
  --global-batch-size 16
  --balance-data 
  --optimizer adam
  --lr 1e-6
  --lr-decay-style constant
  --weight-decay 0.1
  --adam-beta1 0.9
  --adam-beta2 0.98
  --optimizer-cpu-offload
  --overlap-cpu-optimizer-d2h-h2d
  --use-precision-aware-optimizer 
  --advantage-estimator grpo
  --use-kl-loss
  --kl-loss-coef 0.00
  --kl-loss-type low_var_kl
  --entropy-coef 0.00
  --eps-clip 0.2
  --eps-clip-high 0.28 
  --tensor-model-parallel-size 1
  --sequence-parallel
  --pipeline-model-parallel-size 1
  --context-parallel-size 1
  --expert-model-parallel-size 8
  --expert-tensor-parallel-size 1
  --recompute-granularity full
  --recompute-method uniform
  --recompute-num-layers 1
  --use-dynamic-batch-size
  --max-tokens-per-gpu 8192  
  --rollout-num-gpus-per-engine 8
  --sglang-mem-fraction-static 0.7
  --sglang-ep-size 8
  --sglang-cuda-graph-bs-decode 1 2 4 8 16 24 32 40 48 56 64 72 80 88 96 104 112 120 128
  --sglang-speculative-algorithm EAGLE
  --sglang-speculative-num-steps 2
  --sglang-speculative-eagle-topk 1
  --sglang-speculative-num-draft-tokens 3
  --sglang-max-running-requests 256
  --sglang-mamba-radix-cache-strategy extra_buffer 
  --enable-mtp-training
  --mtp-num-layers 1
  --mtp-loss-scaling-factor 0.2 
  --attention-dropout 0.0
  --hidden-dropout 0.0
  --accumulate-allreduce-grads-in-fp32
  --attention-softmax-in-fp32
  --attention-backend flash
  --moe-token-dispatcher-type flex
  --actor-num-nodes 1
  --actor-num-gpus-per-node 8
  --num-gpus-per-node 8
  --colocate   
  --deploy-component all
