| {"name": "dream", "dump_dir": "/scratch/dyvm6xra/dyvm6xrauseryuhao/dream-training/checkpoint/", "seed": 777, "grad_acc_steps": 1, "gc_collect_freq": 1000, "probe_freq": null, "steps": 2500, "eval_steps": 10, "data": {"root_dir": "/scratch/dyvm6xra/dyvm6xrauseryuhao/dataset/Pretraining_Dataset", "sources": {"data1": 1.0}, "val_sources": {}, "batch_size": 1, "seq_len": 4096, "val_seq_len": 2048, "val_batch_size": 10, "n_views": 2, "seed": 42, "add_bos": true, "add_eos": true, "load_async": true, "prefetch_size": 1024, "tokenizer": {"name": "hf", "path": "/scratch/dyvm6xra/dyvm6xrauseryuhao/dream-training/Qwen2.5-7B-dcp"}}, "optim": {"lr": 1e-05, "weight_decay": 0.033, "epsilon": 1e-08, "beta1": 0.9, "beta2": 0.95, "clip": 1.0, "scheduler": "wsd", "warmup": 5000, "lr_min_ratio": 0.0, "cycle_length": 1.0, "cosine_theta": 1.0, "annealing_step": 1000, "decay_fraction": 0.2, "exp_factor": 0.5}, "distributed": {"dp_shard": 1, "dp_replicate": 8, "tp_size": 1, "selective_activation_checkpointing": false, "compile": true, "fsdp_type": "full_shard", "model_dtype": "bf16", "float8_recipe": null, "float8_filter": "layers\\.[0-9]+\\.", "matmul_allow_tf32": false, "detect_anomaly": false, "compile_cache_size_limit": 8, "spawn_method": "forkserver"}, "env": {"MKL_SERVICE_FORCE_INTEL": "GNU", "OMP_NUM_THREADS": "1", "MKL_NUM_THREADS": "1", "ENABLE_INTRA_NODE_COMM": "1", "TORCH_NCCL_AVOID_RECORD_STREAMS": "1", "NCCL_IB_TIMEOUT": "22", "NCCL_DEBUG": "INFO", "TORCH_NCCL_ASYNC_ERROR_HANDLING": "1"}, "checkpoint": {"dump": {"every": 500, "keep": 20}, "eval": {"every": 1000, "keep": 0}, "path": "/scratch/dyvm6xra/dyvm6xrauseryuhao/dream-training/checkpoint/checkpoints", "init_ckpt_path": "/scratch/dyvm6xra/dyvm6xrauseryuhao/dream-training/Qwen2.5-7B-dcp", "continue_training_from_init": false}, "reset_lr": false, "log_start_step": 0, "profiling": {"run": false, "trace_folder": "profiling", "mem_warmup": 100, "mem_steps": 2, "profile_warmup": 102, "profile_steps": 2}, "logging": {"freq": 1, "acc_freq": null, "wandb": {"job_type": null, "dir": null, "project": "diffllm-2", "entity": null, "tags": null, "group": null, "name": "dream", "notes": null, "config_exclude_keys": null, "config_include_keys": null, "anonymous": null, "mode": null, "allow_val_change": null, "resume": null, "force": null, "tensorboard": null, "sync_tensorboard": null, "monitor_gym": null, "save_code": null, "id": null, "fork_from": null, "resume_from": null}}, "async_eval_gpus": null, "eval": null, "weighting": "cart", "cart_p": 0.1} |