experiment: name: controller_v2s5 seed: 1 mode: prisme model: from_checkpoint checkpoint: experiments/runs/p4v2_prisme_seed05_20261030001839/checkpoint labels: train: experiments/runs/labels_v2s5_seed01_20261030004544/labels_train.pt calib: experiments/runs/labels_v2s5_seed01_20261030004544/labels_calib.pt train: steps: 2000 lr: 0.001 weight_decay: 0.01 alpha: 1.0 beta: 0.5 gamma: 0.1 huber_delta: 0.1 log_every: 100 eval_every: 250 seq_len: 320 lambda_grid: - 0.0 - 0.01 - 0.03 - 0.1 - 0.3 kappa_grid: - 0.0 - 0.5 sweep_batches: 10 max_budget: 8 selection_rule: max_loss_reduction_per_ops model_overrides: ctrl_input: h ctrl_use_entropy: true model: name: tiny_p4v2 vocab_size: 128 d_model: 256 n_layers: 6 n_heads: 8 n_kv_heads: 4 ffn_width: 768 max_seq_len: 320 rope_theta: 10000.0 norm_eps: 1.0e-05 tie_embeddings: true init_std: 0.02 use_workspace: true w: 64 ffn_w: 192 n_slots: 16 n_editor_blocks: 2 n_editor_heads: 2 r_max: 8 eta_max: 0.25 clip_c: 1.0 clip_eps: 1.0e-06 readout_zero_init: false readout_init_std: 0.2 readout_scale: 39.13882827758789 controller_hidden: 0 ctrl_input: h ctrl_use_entropy: true lambda_cost: 0.0 kappa_risk: 0.0 dense_writes: false