training_args.json

{
  "activate_conditionality": "PKV",
  "adam_beta1": 0.9,
  "adam_beta2": 0.999,
  "attention_dropout": 0.1,
  "codecarbon": false,
  "cond_dropout": 0.01,
  "cond_lr": 0.0005,
  "cond_wd": 0.01,
  "condition_columns": "['norm_Density (g/cm^3)']",
  "config": "data/config.jsonc",
  "context_length": 1024,
  "data_seed": 1,
  "dataset_HF": "c-bone/mp-20-1k_final",
  "deepspeed_config": null,
  "do_sample": "True",
  "early_stopping_patience": 2,
  "early_stopping_threshold": 1e-05,
  "embedding_dropout": 0.1,
  "eval_batch_size": 32,
  "eval_steps": 100,
  "eval_strategy": "steps",
  "fp16": true,
  "gen_max_length": 1024,
  "grad_clip": 1.0,
  "gradient_accumulation_steps": 1,
  "greater_is_better": false,
  "input_parquet": null,
  "learning_rate": 5e-06,
  "load_best_model_at_end": true,
  "logging_steps": 50,
  "lr_scheduler_kwargs": {
    "min_lr_rate": 0.01
  },
  "lr_scheduler_type": "cosine_with_min_lr",
  "max_return_attempts": 1,
  "max_samples": null,
  "max_steps": 500,
  "metric_for_best_model": "eval_loss",
  "model_ckpt_dir": "model_ckpts/cif-gpt2-small/checkpoint-400",
  "muon_lr": 0.02,
  "muon_momentum": 0.95,
  "n_embd": 512,
  "n_head": 8,
  "n_heads_sharing_slider": null,
  "n_hidden_cond": 256,
  "n_layer": 8,
  "n_prefix_tokens": 2,
  "num_return_sequences": 1,
  "optimizer": "adamw",
  "output_dir": "data/MP-20-Density/",
  "output_parquet": null,
  "pretrained_model_dir": "/home/cyprien/CrystaLLM-pi-paper-v2/model_ckpts/mpdb-small-base-lematerial/checkpoint-1250000",
  "pretrained_tokenizer_dir": "HF-cif-tokenizer",
  "remove_CIFs_above_context": false,
  "remove_CIFs_with_unk": false,
  "report_to": "none",
  "residual_dropout": 0.1,
  "save_strategy": "steps",
  "save_total_limit": 2,
  "scoring_mode": "None",
  "seed": 2,
  "share_layers": true,
  "target_valid_cifs": 1,
  "temperature": 1.0,
  "top_k": 15,
  "top_p": 0.95,
  "torch_compile": true,
  "tracker_project": "CrystaLLM_pi",
  "train_batch_size": 32,
  "wandb_project_folder": "CrystaLLM_pi",
  "warmup_ratio": 0.02,
  "warmup_steps": null,
  "weight_decay": 0.01
}