experiment: name: labels_103m seed: 1 mode: prisme model: from_checkpoint checkpoint: experiments/runs/editor_pilot_103m_lr3e4_seed01_20261030021200/checkpoint use_policy: false data: kind: tokens collect: batch_size: 8 seq_len: 512 max_prefix: 3 audit_every: 4 partitions: train: bin: data/processed/fineweb_edu_300m/train.bin n_batches: 120 stream_seed: 777 calib: bin: data/processed/fineweb_edu_300m/val.bin n_batches: 30 stream_seed: 778 model: name: prisme_103m vocab_size: 32768 d_model: 768 n_layers: 12 n_heads: 12 n_kv_heads: 3 ffn_width: 2048 max_seq_len: 4096 rope_theta: 10000.0 norm_eps: 1.0e-05 tie_embeddings: true init_std: 0.02 use_workspace: true w: 256 ffn_w: 768 n_slots: 16 n_editor_blocks: 2 n_editor_heads: 4 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: 8.851566314697266 controller_hidden: 0 ctrl_input: h ctrl_use_entropy: true lambda_cost: 0.0 kappa_risk: 0.0 dense_writes: false