{ "step": 60, "stage": "sft", "val_loss": 0.5093614831566811, "version": 5, "params": 1523202304, "dtype": "bf16", "config": { "vocab_size": 8192, "block_size": 1024, "n_layer": 34, "n_head": 32, "n_embd": 2048, "dropout": 0.0, "n_kv_head": 8, "qk_norm": true, "arch_version": 5 }, "files": [ "model.pt", "model-00001-of-00103.pt", "model-00002-of-00103.pt", "model-00003-of-00103.pt", "model-00004-of-00103.pt", "model-00005-of-00103.pt", "model-00006-of-00103.pt", "model-00007-of-00103.pt", "model-00008-of-00103.pt", "model-00009-of-00103.pt", "model-00010-of-00103.pt", "model-00011-of-00103.pt", "model-00012-of-00103.pt", "model-00013-of-00103.pt", "model-00014-of-00103.pt", "model-00015-of-00103.pt", "model-00016-of-00103.pt", "model-00017-of-00103.pt", "model-00018-of-00103.pt", "model-00019-of-00103.pt", "model-00020-of-00103.pt", "model-00021-of-00103.pt", "model-00022-of-00103.pt", "model-00023-of-00103.pt", "model-00024-of-00103.pt", "model-00025-of-00103.pt", "model-00026-of-00103.pt", "model-00027-of-00103.pt", "model-00028-of-00103.pt", "model-00029-of-00103.pt", "model-00030-of-00103.pt", "model-00031-of-00103.pt", "model-00032-of-00103.pt", "model-00033-of-00103.pt", "model-00034-of-00103.pt", "model-00035-of-00103.pt", "model-00036-of-00103.pt", "model-00037-of-00103.pt", "model-00038-of-00103.pt", "model-00039-of-00103.pt", "model-00040-of-00103.pt", "model-00041-of-00103.pt", "model-00042-of-00103.pt", "model-00043-of-00103.pt", "model-00044-of-00103.pt", "model-00045-of-00103.pt", "model-00046-of-00103.pt", "model-00047-of-00103.pt", "model-00048-of-00103.pt", "model-00049-of-00103.pt", "model-00050-of-00103.pt", "model-00051-of-00103.pt", "model-00052-of-00103.pt", "model-00053-of-00103.pt", "model-00054-of-00103.pt", "model-00055-of-00103.pt", "model-00056-of-00103.pt", "model-00057-of-00103.pt", "model-00058-of-00103.pt", "model-00059-of-00103.pt", "model-00060-of-00103.pt", "model-00061-of-00103.pt", "model-00062-of-00103.pt", "model-00063-of-00103.pt", "model-00064-of-00103.pt", "model-00065-of-00103.pt", "model-00066-of-00103.pt", "model-00067-of-00103.pt", "model-00068-of-00103.pt", "model-00069-of-00103.pt", "model-00070-of-00103.pt", "model-00071-of-00103.pt", "model-00072-of-00103.pt", "model-00073-of-00103.pt", "model-00074-of-00103.pt", "model-00075-of-00103.pt", "model-00076-of-00103.pt", "model-00077-of-00103.pt", "model-00078-of-00103.pt", "model-00079-of-00103.pt", "model-00080-of-00103.pt", "model-00081-of-00103.pt", "model-00082-of-00103.pt", "model-00083-of-00103.pt", "model-00084-of-00103.pt", "model-00085-of-00103.pt", "model-00086-of-00103.pt", "model-00087-of-00103.pt", "model-00088-of-00103.pt", "model-00089-of-00103.pt", "model-00090-of-00103.pt", "model-00091-of-00103.pt", "model-00092-of-00103.pt", "model-00093-of-00103.pt", "model-00094-of-00103.pt", "model-00095-of-00103.pt", "model-00096-of-00103.pt", "model-00097-of-00103.pt", "model-00098-of-00103.pt", "model-00099-of-00103.pt", "model-00100-of-00103.pt", "model-00101-of-00103.pt", "model-00102-of-00103.pt", "model-00103-of-00103.pt" ] }