begumcig commited on
Commit
97ab671
·
verified ·
1 Parent(s): 7eed328

Upload folder using huggingface_hub (#2)

Browse files

- e6c68e93a28adc079b58231ba80da60b9ff60b0fb940223c82d1c4daac04a3db (6b074cb063a69c1935adb35d139ff826f6b62181)

Files changed (3) hide show
  1. base_results.json +19 -0
  2. plots.png +0 -0
  3. 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": 2.2062017917633057,
5
+ "memory_inference_first": 7298.0,
6
+ "memory_inference": 7298.0,
7
+ "token_generation_latency_sync": 35.75132808685303,
8
+ "token_generation_latency_async": 35.67290659993887,
9
+ "token_generation_throughput_sync": 0.027970988869857785,
10
+ "token_generation_throughput_async": 0.02803247885614439,
11
+ "token_generation_CO2_emissions": 0.00018359090889950798,
12
+ "token_generation_energy_consumption": 0.002221255088823526,
13
+ "inference_latency_sync": 1058.6026702880858,
14
+ "inference_latency_async": 136.83900833129883,
15
+ "inference_throughput_sync": 0.000944641486430279,
16
+ "inference_throughput_async": 0.007307857694926547,
17
+ "inference_CO2_emissions": 0.00018326470280637688,
18
+ "inference_energy_consumption": 0.000631884688384694
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": 2.7251362800598145,
5
+ "memory_inference_first": 1690.0,
6
+ "memory_inference": 1690.0,
7
+ "token_generation_latency_sync": 155.3000717163086,
8
+ "token_generation_latency_async": 155.34533187747002,
9
+ "token_generation_throughput_sync": 0.006439147058648696,
10
+ "token_generation_throughput_async": 0.0064372710007711,
11
+ "token_generation_CO2_emissions": 0.00020285744923935242,
12
+ "token_generation_energy_consumption": 0.009233308010788546,
13
+ "inference_latency_sync": 1166.3823852539062,
14
+ "inference_latency_async": 429.13405895233154,
15
+ "inference_throughput_sync": 0.0008573517678615431,
16
+ "inference_throughput_async": 0.0023302741396041943,
17
+ "inference_CO2_emissions": 0.000202412501568058,
18
+ "inference_energy_consumption": 0.0006962767357209158
19
+ }