{
    "base_current_gpu_type": "NVIDIA A100-PCIE-40GB",
    "base_current_gpu_total_memory": 40339.3125,
    "base_perplexity": 6.637787342071533,
    "base_token_generation_latency_sync": 37.664760208129884,
    "base_token_generation_latency_async": 37.534007616341114,
    "base_token_generation_throughput_sync": 0.026550016367398815,
    "base_token_generation_throughput_async": 0.026642505383960964,
    "base_token_generation_CO2_emissions": null,
    "base_token_generation_energy_consumption": null,
    "base_inference_latency_sync": 118.52748794555664,
    "base_inference_latency_async": 38.34338188171387,
    "base_inference_throughput_sync": 0.00843686150219712,
    "base_inference_throughput_async": 0.026080119982241433,
    "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": 7.146455764770508,
    "smashed_token_generation_latency_sync": 44.163258743286136,
    "smashed_token_generation_latency_async": 44.04097441583872,
    "smashed_token_generation_throughput_sync": 0.022643256599628164,
    "smashed_token_generation_throughput_async": 0.022706127947077485,
    "smashed_token_generation_CO2_emissions": null,
    "smashed_token_generation_energy_consumption": null,
    "smashed_inference_latency_sync": 161.99608306884767,
    "smashed_inference_latency_async": 69.91302967071533,
    "smashed_inference_throughput_sync": 0.006172988760320854,
    "smashed_inference_throughput_async": 0.014303485411945647,
    "smashed_inference_CO2_emissions": null,
    "smashed_inference_energy_consumption": null
}