chain_name: "Bluejay DPO"
description: "Bluejay DPO chain, reproducable"

# 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: 4
  #   model_cache_loss_name: "6b_bjay_30_0"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_auk_v19_r1"

  # - name: "round_2"
  #   script: "run_ipo_bluejay_test_1.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_30_1"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_sft006_bjay_v19_r2"
  #   sft_loss_scale: 0.006
    
  # - name: "round_3"
  #   script: "run_ipo_bluejay_test_2.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_30_2"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_sft006_bjay_v20_r3"
  #   sft_loss_scale: 0.006
    
  # - name: "round_4"
  #   script: "run_ipo_bluejay_test_3.sh"  
  #   nodes: 8
  #   model_cache_loss_name: "6b_bjay_30_3"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_sft006_bjay_v21_r4"
  #   sft_loss_scale: 0.006

  # - name: "round_5"
  #   script: "run_ipo_bluejay_r10_bluejay_1.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_30_4"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_bjay_v29_r5"
    
  # - name: "round_6"
  #   script: "run_ipo_bluejay_r11_bluejay_2.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_30_5"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_bjay_v30_r6"

  # - name: "round_7"
  #   script: "run_ipo_bluejay_r12_bluejay_3.sh"  
  #   nodes: 16
  #   model_cache_loss_name: "6b_bjay_30_6"
  #   wandb_run_name: "dpo_6b_base0519_bjay_30_bjay_v31_r7"