seed: 7
device: auto
output_dir: outputs/overfit

data:
  source_manifest: ../../alpamayo/alpamayo-qwen05b-validation/data/processed/overfit_manifest.jsonl
  max_samples: 1
  augmentation_factor: 1
  image_size: 64
  history_frames: 6
  action_horizon: 16
  policy_variants:
    - expert
    - fast
    - slow
    - understeer
    - stop

model:
  patch_size: 16
  d_model: 96
  n_heads: 4
  n_layers: 3
  memory_layers: 2
  mlp_ratio: 3
  dropout: 0.0
  max_text_tokens: 96
  action_dim: 2
  action_horizon: 16
  goal_heatmap_size: 32

memory:
  enabled: true
  history_frames: 6
  history_dropout: 0.3
  ambiguous_pair_weight: 0.25

goal:
  enabled: true
  target: future_observation_proxy

high_level:
  enabled: true
  predict_goal: true
  predict_metadata: true
  metadata_fields:
    - speed
    - quality
    - mistake
    - control

steering:
  subgoal_probability: 0.25
  subtask_dropout_with_goal: 0.3
  metadata_full_dropout: 0.15
  metadata_component_dropout: 0.05
  cfg_beta: 1.7

rtc:
  enabled: true
  max_delay_steps: 12
  include_executed_prefix: true
  masked_loss: true

flow:
  timestep_beta_alpha: 1.5
  timestep_beta_beta: 1.0
  integration_steps: 5

train:
  goal_steps: 800
  high_level_steps: 200
  policy_steps: 900
  batch_size: 5
  learning_rate: 0.001
  weight_decay: 0.0
  grad_clip_norm: 1.0
  goal_loss_weight: 1.0
  high_level_loss_weight: 1.0
  flow_loss_weight: 2.0
  rtc_loss_weight: 1.0
  log_every: 20

evaluation:
  fixed_noise_seed: 1
  goal_flow_steps: 10
  cfg_betas: [0.0, 1.3, 1.7, 2.2]
  ablations:
    - full
    - no_metadata
    - no_goal
    - no_history
  compare_rtc_delays: [0, 4, 8, 12]
  save_trajectory_plot: true
  save_goal_plot: true
  save_ablation_plot: true