{ "implementation": "megablocks_results", "config": { "warmup": 10, "iters": 50, "device": "cuda", "dtype": "torch.float32", "tokens": 100 }, "stats": { "avg_ms": 4.133109538815916, "min_ms": 0.822786008939147, "max_ms": 8.589233038946986, "std_ms": 3.7810751831568976, "p50_ms": 0.8644890040159225, "p95_ms": 8.578622492495924, "p99_ms": 8.588659723172896, "num_iters": 50, "tokens_per_s": 24194.858389513854, "throughput_variance": 52511.72634241489 }, "output_sum": -0.5972487926483154 }