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": 790.0,
|
6 |
+
"memory_inference": 790.0,
|
7 |
+
"token_generation_latency_sync": 37.40723571777344,
|
8 |
+
"token_generation_latency_async": 37.517561949789524,
|
9 |
+
"token_generation_throughput_sync": 0.026732795963452235,
|
10 |
+
"token_generation_throughput_async": 0.026654184014897324,
|
11 |
+
"token_generation_CO2_emissions": 1.942063140736845e-05,
|
12 |
+
"token_generation_energy_consumption": 0.0018457401360101422,
|
13 |
+
"inference_latency_sync": 122.10714988708496,
|
14 |
+
"inference_latency_async": 47.79155254364014,
|
15 |
+
"inference_throughput_sync": 0.008189528630589782,
|
16 |
+
"inference_throughput_async": 0.020924199921876677,
|
17 |
+
"inference_CO2_emissions": 1.9370730362422426e-05,
|
18 |
+
"inference_energy_consumption": 6.823837478184117e-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": 5.245799541473389,
|
5 |
+
"memory_inference_first": 362.0,
|
6 |
+
"memory_inference": 362.0,
|
7 |
+
"token_generation_latency_sync": 95.34894561767578,
|
8 |
+
"token_generation_latency_async": 95.61750814318657,
|
9 |
+
"token_generation_throughput_sync": 0.010487792953786162,
|
10 |
+
"token_generation_throughput_async": 0.010458335710887872,
|
11 |
+
"token_generation_CO2_emissions": 2.0800614727756692e-05,
|
12 |
+
"token_generation_energy_consumption": 0.004631853474529401,
|
13 |
+
"inference_latency_sync": 146.7707092285156,
|
14 |
+
"inference_latency_async": 94.17455196380615,
|
15 |
+
"inference_throughput_sync": 0.006813348557463488,
|
16 |
+
"inference_throughput_async": 0.010618579851426608,
|
17 |
+
"inference_CO2_emissions": 2.1171634369010966e-05,
|
18 |
+
"inference_energy_consumption": 7.173729223505599e-05
|
19 |
+
}
|