parameters.json

{
    "force": true,
    "dump_path": "data/distilgpt2-e7",
    "student_type": "gpt2",
    "student_config": "distilgpt2-ja.json",
    "student_pretrained_weights": "data/distilgpt2-e6/checkpoint.pth",
    "teacher_type": "gpt2",
    "teacher_name": "rinna/japanese-gpt2-medium",
    "temperature": 2.0,
    "alpha_ce": 5.0,
    "alpha_mlm": 0.0,
    "alpha_clm": 0.5,
    "alpha_mse": 0.0,
    "alpha_cos": 1.0,
    "mlm": false,
    "mlm_mask_prop": 0.15,
    "word_mask": 0.8,
    "word_keep": 0.1,
    "word_rand": 0.1,
    "mlm_smoothing": 0.7,
    "restrict_ce_to_mask": false,
    "freeze_pos_embs": true,
    "freeze_token_type_embds": false,
    "n_epoch": 2,
    "batch_size": 16,
    "group_by_size": false,
    "gradient_accumulation_steps": 50,
    "warmup_prop": 0.0,
    "weight_decay": 0.0,
    "learning_rate": 4.9e-05,
    "adam_epsilon": 1e-06,
    "max_grad_norm": 5.0,
    "initializer_range": 0.02,
    "fp16": false,
    "fp16_opt_level": "O1",
    "n_gpu": 4,
    "local_rank": 0,
    "seed": 56,
    "log_interval": 500,
    "checkpoint_interval": 4000,
    "world_size": 4,
    "n_gpu_per_node": 4,
    "global_rank": 0,
    "n_nodes": 1,
    "node_id": 0,
    "multi_gpu": true,
    "is_master": true,
    "multi_node": false
}