{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_perplexity": 13.74486255645752,
    "base_token_generation_latency_sync": 36.7140287399292,
    "base_token_generation_latency_async": 37.803952395915985,
    "base_token_generation_throughput_sync": 0.027237544729391865,
    "base_token_generation_throughput_async": 0.02645226058712399,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 122.47920608520508,
    "base_inference_latency_async": 43.92876625061035,
    "base_inference_throughput_sync": 0.008164651224995125,
    "base_inference_throughput_async": 0.02276412668398366,
    "base_inference_CO2_emissions": null,
    "base_inference_energy_consumption": null,
    "smashed_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "smashed_current_gpu_total_memory": 40339.3125,
    "smashed_perplexity": 205742.3125,
    "smashed_token_generation_latency_sync": 169.95355529785155,
    "smashed_token_generation_latency_async": 170.05142513662577,
    "smashed_token_generation_throughput_sync": 0.005883960463477526,
    "smashed_token_generation_throughput_async": 0.005880574062796369,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 268.46238861083987,
    "smashed_inference_latency_async": 212.37447261810303,
    "smashed_inference_throughput_sync": 0.003724916570900325,
    "smashed_inference_throughput_async": 0.0047086638411492346,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}