visual_tokenizer.json

{
  "dataset": {
    "frame_count": 64,
    "fps": 8,
    "physics_hz": 120,
    "radius": 16,
    "velocity_min": 38.0,
    "velocity_max": 72.0,
    "angular_velocity_min": 0.55,
    "angular_velocity_max": 1.8,
    "empty_probability": 0.005,
    "duration_seconds": 8.0,
    "circle_count_distribution": "uniform_1_through_6_conditional_on_nonempty",
    "colors": [
      [
        255,
        0,
        0
      ],
      [
        0,
        0,
        255
      ],
      [
        0,
        255,
        0
      ],
      [
        255,
        255,
        0
      ],
      [
        0,
        255,
        255
      ],
      [
        255,
        0,
        255
      ]
    ],
    "colors_unique_within_video": true,
    "renderer": "cached_quarter_pixel_antialiased_circle_stamps",
    "online": true,
    "deterministic_key": "data_seed_and_global_document_index",
    "empty_sampling": "one_pseudorandom_slot_per_200_document_block"
  },
  "fsq_checkpoint": "/home/nathanbreslow/tinyWorldModel/runs/fsq_8x8x6_5p98m_circles_r16_physics_mix_e11/checkpoints/best.pt",
  "fsq_checkpoint_sha256": "8562c5b69d151904404f964275a1a31df8a4ef1399f125a7785b7d7a2fdd7188",
  "fsq_checkpoint_epoch": 3,
  "fsq_checkpoint_global_step": 1053,
  "fsq_levels": [
    5,
    5,
    5,
    5,
    5,
    5
  ],
  "fsq_latent_shape": [
    8,
    8,
    6
  ],
  "codebook_size": 15625,
  "vocab_size": 15629,
  "tokens_per_frame": 64,
  "tokens_per_document": 4162,
  "loss_tokens_per_document": 4161,
  "sequence_format": "BOS (FRAME_START 64xFSQ_TOKEN) repeated 64 times EOS"
}