# Locked evaluation of the 103M pilot: held-out token cross-entropy at forced budgets and under the controller.
# Splits: val (development) now; test (LOCKED) only after protocol/preregistration.md is frozen and the manifest unsealed.
out_dir: results/raw
data: {kind: tokens, dir: data/processed/fineweb_edu_300m}
splits: [val]
seq_len: 512
batch_size: 16
n_batches: 40          # 40 x 16 x 512 = 327,680 held-out tokens
budgets: [0, 1, 2, 4, 8]
systems:
  backbone_103m:
    mode: backbone_only
    model: from_checkpoint
    checkpoint: PRETRAIN103M_CKPT
    controller: false
  enlarged_102m:
    mode: backbone_only
    model: from_checkpoint
    checkpoint: ENLARGED_CKPT
    controller: false
  prisme_103m_editor:
    mode: prisme
    model: from_checkpoint
    checkpoint: EDITOR103M_CKPT
    controller: false
  gr_103m:
    mode: global_recurrence
    baseline_kwargs: {ffn_w: 2560}
    model: from_checkpoint
    checkpoint: GR103M_CKPT
    controller: false