addition_big_sharp/position_coupling_s0/config.json

{
  "name": "addition_big_sharp position coupling seed 0",
  "task": {
    "op": "add",
    "max_digits": 30,
    "min_digits": 1,
    "carry_heavy": 0.0,
    "reverse": true,
    "zero_pad": true,
    "positions": "coupled",
    "blanks": 0,
    "blanks_fixed": true,
    "offset_max": 200,
    "test_digits": [
      10,
      20,
      30,
      40,
      50,
      60,
      80,
      100,
      150,
      200
    ],
    "test_examples": 256,
    "kind": "arithmetic"
  },
  "model": {
    "arch": "transformer",
    "d_model": 384,
    "n_layers": 6,
    "n_heads": 6,
    "d_mlp": 1536,
    "max_positions": 1024,
    "pos_embed": "absolute",
    "attn_scale": 2.0,
    "dropout": 0.0
  },
  "train": {
    "steps": 60000,
    "batch_size": 384,
    "lr": 0.0002,
    "weight_decay": 0.1,
    "betas": [
      0.9,
      0.98
    ],
    "warmup_steps": 2000,
    "cosine": true,
    "amp": true,
    "eval_every": 5000,
    "checkpoint_every": 20000,
    "seed": 0,
    "wandb": false
  }
}