| { | |
| "name": "olmo3-lingua", | |
| "dump_dir": "logs/debug", | |
| "seed": 777, | |
| "grad_acc_steps": 1, | |
| "gc_collect_freq": 1000, | |
| "probe_freq": null, | |
| "steps": 4882812500, | |
| "data": { | |
| "root_dir": "data", | |
| "sources": { | |
| "dclm-baseline_shuffled": 1.0 | |
| }, | |
| "batch_size": 32, | |
| "seq_len": 4096, | |
| "n_views": 2, | |
| "seed": 42, | |
| "add_bos": true, | |
| "add_eos": true, | |
| "load_async": true, | |
| "prefetch_size": 1024, | |
| "tokenizer": { | |
| "name": "huggingface", | |
| "path": "/data/rsadhukh/checkpoints/olmo2-1b-base-token4T" | |
| } | |
| }, | |
| "optim": { | |
| "lr": 0.0003, | |
| "weight_decay": 0.1, | |
| "epsilon": 1e-08, | |
| "beta1": 0.9, | |
| "beta2": 0.95, | |
| "clip": 1.0, | |
| "scheduler": "cosine", | |
| "warmup": 8192000, | |
| "lr_min_ratio": 0.1, | |
| "cycle_length": 1.0, | |
| "cosine_theta": 1.0, | |
| "annealing_step": 0, | |
| "decay_fraction": 0.1, | |
| "exp_factor": 0.5 | |
| }, | |
| "distributed": { | |
| "dp_shard": 1, | |
| "dp_replicate": 1, | |
| "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" | |
| }, | |
| "checkpoint": { | |
| "dump": { | |
| "every": 10000, | |
| "keep": 1 | |
| }, | |
| "eval": { | |
| "every": 10000, | |
| "keep": 1 | |
| }, | |
| "path": "logs/debug/checkpoints", | |
| "init_ckpt_path": "", | |
| "continue_training_from_init": false | |
| }, | |
| "logging": { | |
| "freq": 10, | |
| "acc_freq": null, | |
| "wandb": null | |
| }, | |
| "model": { | |
| "dim": 4096, | |
| "n_layers": 32, | |
| "head_dim": 128, | |
| "n_heads": 32, | |
| "n_kv_heads": 32, | |
| "ffn_dim_multiplier": null, | |
| "multiple_of": 256, | |
| "norm_eps": 1e-06, | |
| "rope_theta": 500000, | |
| "rope_scaling": null, | |
| "init_base_std": 0.02, | |
| "init_std_factor": "disabled", | |
| "max_seqlen": 4096, | |
| "seed": 42, | |
| "vocab_size": 100352, | |
| "weight_tying": false, | |
| "sliding_window": null | |
| } | |
| } |