Upload folder using huggingface_hub
#2
by
begumcig
- opened
- base_results.json +19 -0
- plots.png +0 -0
- smashed_results.json +19 -0
base_results.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"current_gpu_type": "Tesla T4",
|
3 |
+
"current_gpu_total_memory": 15095.0625,
|
4 |
+
"perplexity": 3.4586403369903564,
|
5 |
+
"memory_inference_first": 810.0,
|
6 |
+
"memory_inference": 808.0,
|
7 |
+
"token_generation_latency_sync": 37.52931938171387,
|
8 |
+
"token_generation_latency_async": 37.49031759798527,
|
9 |
+
"token_generation_throughput_sync": 0.02664583361688273,
|
10 |
+
"token_generation_throughput_async": 0.026673553708537804,
|
11 |
+
"token_generation_CO2_emissions": 1.9795767676085165e-05,
|
12 |
+
"token_generation_energy_consumption": 0.0018361529394885828,
|
13 |
+
"inference_latency_sync": 122.5987979888916,
|
14 |
+
"inference_latency_async": 48.247456550598145,
|
15 |
+
"inference_throughput_sync": 0.008156686822415727,
|
16 |
+
"inference_throughput_async": 0.020726481176293273,
|
17 |
+
"inference_CO2_emissions": 1.9329542916591693e-05,
|
18 |
+
"inference_energy_consumption": 6.80833538907916e-05
|
19 |
+
}
|
plots.png
ADDED
![]() |
smashed_results.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"current_gpu_type": "Tesla T4",
|
3 |
+
"current_gpu_total_memory": 15095.0625,
|
4 |
+
"perplexity": 3.472892999649048,
|
5 |
+
"memory_inference_first": 402.0,
|
6 |
+
"memory_inference": 402.0,
|
7 |
+
"token_generation_latency_sync": 58.756526947021484,
|
8 |
+
"token_generation_latency_async": 58.4649283438921,
|
9 |
+
"token_generation_throughput_sync": 0.01701938579354191,
|
10 |
+
"token_generation_throughput_async": 0.017104271369631656,
|
11 |
+
"token_generation_CO2_emissions": 2.0461894299703792e-05,
|
12 |
+
"token_generation_energy_consumption": 0.0028872648917871455,
|
13 |
+
"inference_latency_sync": 131.08970794677734,
|
14 |
+
"inference_latency_async": 64.51253890991211,
|
15 |
+
"inference_throughput_sync": 0.007628363932323366,
|
16 |
+
"inference_throughput_async": 0.015500862574893231,
|
17 |
+
"inference_CO2_emissions": 1.9946509275385296e-05,
|
18 |
+
"inference_energy_consumption": 6.95395721998155e-05
|
19 |
+
}
|