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": 808.0,
|
6 |
+
"memory_inference": 808.0,
|
7 |
+
"token_generation_latency_sync": 36.94038696289063,
|
8 |
+
"token_generation_latency_async": 37.75534201413393,
|
9 |
+
"token_generation_throughput_sync": 0.027070642248674184,
|
10 |
+
"token_generation_throughput_async": 0.026486318138123188,
|
11 |
+
"token_generation_CO2_emissions": 2.9548104595618817e-05,
|
12 |
+
"token_generation_energy_consumption": 0.0024448525932904515,
|
13 |
+
"inference_latency_sync": 259.3957290649414,
|
14 |
+
"inference_latency_async": 82.87992477416992,
|
15 |
+
"inference_throughput_sync": 0.0038551135888195117,
|
16 |
+
"inference_throughput_async": 0.012065648016993093,
|
17 |
+
"inference_CO2_emissions": 3.8677334298918255e-05,
|
18 |
+
"inference_energy_consumption": 0.00013983032998305092
|
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": 343688032.0,
|
5 |
+
"memory_inference_first": 344.0,
|
6 |
+
"memory_inference": 344.0,
|
7 |
+
"token_generation_latency_sync": 94.73909301757813,
|
8 |
+
"token_generation_latency_async": 94.9408158659935,
|
9 |
+
"token_generation_throughput_sync": 0.010555304765419883,
|
10 |
+
"token_generation_throughput_async": 0.010532877676251214,
|
11 |
+
"token_generation_CO2_emissions": 2.0721120007850446e-05,
|
12 |
+
"token_generation_energy_consumption": 0.004636361917186246,
|
13 |
+
"inference_latency_sync": 147.19097137451172,
|
14 |
+
"inference_latency_async": 94.34385299682617,
|
15 |
+
"inference_throughput_sync": 0.006793894969655487,
|
16 |
+
"inference_throughput_async": 0.010599524698589966,
|
17 |
+
"inference_CO2_emissions": 2.0695969686373567e-05,
|
18 |
+
"inference_energy_consumption": 7.278868671603012e-05
|
19 |
+
}
|