chain_name: "Simple DPO Chain"
description: "A simple 3-job training chain"

# Global settings
settings:
  base_dir: "/home/tony/Work/tony/slurm/bluejay"
  log_dir: "/home/tony/slurm/logs"
  
  # SLURM defaults - no partition specified
  slurm_defaults:
    time: "48:00:00"
    ntasks_per_node: 8
    cpus_per_task: 4
    gpus_per_node: 8

# Job definitions
jobs:
  # - name: "round_1"
  #   script: "run_ipo_bluejay_og_r1.sh"
  #   nodes: 6

  # - name: "round_2"
  #   script: "run_ipo_bluejay_r2_task.sh"
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_1"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r2"
    
  # - name: "round_3"
  #   script: "run_ipo_bluejay_r3_exp.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_2"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r3"

  # - name: "round_4"
  #   script: "run_ipo_bluejay_r4_mix.sh"
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_3"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r4"
    
  # - name: "round_5"
  #   script: "run_ipo_bluejay_r5_mix_2.sh"  
  #   nodes: 10
  #   model_cache_loss_name: "6b_bjay_27_4"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r5"

  # - name: "round_6"
  #   script: "run_ipo_bluejay_r6_auk.sh"
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_1_sft"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r2_sft"
  #   sft_loss_scale: 0.001
    
  # - name: "round_7"
  #   script: "run_ipo_bluejay_r7_auk_2.sh"
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_2_sft"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r3_sft"
  #   sft_loss_scale: 0.001
    
  # - name: "round_8"
  #   script: "run_ipo_bluejay_r8_auk_3.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_3_sft"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r4_sft"
  #   sft_loss_scale: 0.001
    
  # - name: "round_9"
  #   script: "run_ipo_bluejay_r9_auk_4.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_4_sft"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_5_sft"
  #   sft_loss_scale: 0.001
  
  # - name: "round_2"
  #   script: "run_ipo_bluejay_test_1.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_1"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r2_v18_sft01_beta5"
  #   sft_loss_scale: 0.01
    
  # - name: "round_3"
  #   script: "run_ipo_bluejay_test_2.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_2_v18_sft01_beta5"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r3_v19_sft01_beta5"
  #   sft_loss_scale: 0.01
    
  # - name: "round_4"
  #   script: "run_ipo_bluejay_test_3.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_3_v19_sft01_beta5"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r4_v20_sft01_beta5"
  #   sft_loss_scale: 0.01
    
  # - name: "round_5"  
  #   script: "run_ipo_bluejay_test_4.sh"
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_4_v20_sft006"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r5_v21_sft006"
  #   sft_loss_scale: 0.006

  # - name: "round_6"  
  #   script: "run_ipo_bluejay_test_5.sh"
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_5_v21_sft006"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r6_v22_sft006"
  #   sft_loss_scale: 0.006

  # - name: "round_7"
  #   script: "run_ipo_bluejay_test_6.sh"
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_27_6_v22_sft006"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_r7_v23_sft006"
  #   sft_loss_scale: 0.006

  # - name: "round_10"
  #   script: "run_ipo_bluejay_r10_bluejay_1.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_20_r13_v25"
  #   wandb_run_name: "dpo_6b_base0519_bjay_20_r14_v24"
    
  # - name: "round_11"
  #   script: "run_ipo_bluejay_r11_bluejay_2.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_27_d43_r8_v30"
  #   wandb_run_name: "dpo_6b_base0519_bjay_27_d43_r9_v28"

  - name: "round_5"
    script: "run_ipo_bluejay_r12_bluejay_3.sh"  
    nodes: 16
    model_cache_loss_name: "6b_bjay_27_d49_r6"
    wandb_run_name: "dpo_6b_base0519_bjay_27_d49_r7"
    
  - name: "round_6"
    script: "run_ipo_bluejay_r13_bluejay_4.sh"  
    nodes: 16
    model_cache_loss_name: "6b_bjay_27_d49_r7"
    wandb_run_name: "dpo_6b_base0519_bjay_27_d49_r8"
