{ "model": "KUTE-Mini", "dataset": "OpenLLM-Ro/fineweb2-ro-llm", "from_scratch": true, "vocab_size": 16000, "d_model": 256, "n_heads": 8, "n_layers": 6, "d_ff": 1024, "max_seq_len": 128, "parameters": 8867840, "global_step": 5000, "loss": 3.936351478099823 }