Hugging Face's logo

Datasets:
meg
/
HUGS_energy

Modalities:
Text
Formats:
parquet
Size:
< 1K
Libraries:
Datasets
pandas
Dataset card Data Studio Files
xet
Community
2
Dataset Viewer
Auto-converted to Parquet
task
stringclasses
1 value
org
stringclasses
7 values
model
stringclasses
9 values
hardware
stringclasses
2 values
date
stringclasses
9 values
prefill
dict
decode
dict
preprocess
dict
text_generation
NousResearch
Hermes-3-Llama-3.1-8B
a100-large
2024-10-30-23-07-10
{ "efficency": { "unit": "tokens/kWh", "value": 7896383.83303204 }, "energy": { "cpu": 0.010771365165906027, "gpu": 0.03499995191660332, "ram": 0.000042438637758223256, "total": 0.04581375572026757, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 383715.9018998692 }, "energy": { "cpu": 0.006909195393007252, "gpu": 0.016518416020279855, "ram": 0.000027241396927033754, "total": 0.023454852810214137, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 19314595.743536726 }, "energy": { "cpu": 0.00003136402598271767, "gpu": 0.000020315571802598242, "ram": 9.47190721638247e-8, "total": 0.000051774316857479744, "unit": "kWh" } }
text_generation
meta-llama
Llama-3.1-8B-Instruct
a100-large
2024-10-30-19-38-09
{ "efficency": { "unit": "tokens/kWh", "value": 7875937.087230494 }, "energy": { "cpu": 0.010796036717298752, "gpu": 0.035094443769978056, "ram": 0.00004221247271874213, "total": 0.04593269295999555, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 373534.5783406723 }, "energy": { "cpu": 0.006907515534659228, "gpu": 0.01715961075545351, "ram": 0.000027028633857338117, "total": 0.024094154923970088, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 18903508.14097379 }, "energy": { "cpu": 0.000031577332566181814, "gpu": 0.000021227516981525696, "ram": 9.538423654320006e-8, "total": 0.000052900233784250705, "unit": "kWh" } }
text_generation
EleutherAI
pythia-1.4b
a10g-large
2024-10-25-14-19-11
{ "efficency": { "unit": "tokens/kWh", "value": 54946841.77704233 }, "energy": { "cpu": 0.0006950257205501962, "gpu": 0.004783902188229749, "ram": 0.000007229079517594969, "total": 0.00548615698829754, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 1185499.0738062232 }, "energy": { "cpu": 0.001532609044249083, "gpu": 0.006043184112321255, "ram": 0.00001594622041976216, "total": 0.0075917393769901025, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 33069647.189926323 }, "energy": { "cpu": 0.000011265942924405358, "gpu": 0.00001888834844376852, "ram": 8.491845032760329e-8, "total": 0.00003023920981850148, "unit": "kWh" } }
text_generation
microsoft
phi-2
a10g-large
2024-10-25-00-12-06
{ "efficency": { "unit": "tokens/kWh", "value": 25532119.335130304 }, "energy": { "cpu": 0.001485621188480941, "gpu": 0.010244187250898752, "ram": 0.000014692518964651205, "total": 0.011744500958344343, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 653974.0928690574 }, "energy": { "cpu": 0.002664958041045172, "gpu": 0.011070685912097256, "ram": 0.000026369096206358968, "total": 0.013762013049348782, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 2594586258.0208464 }, "energy": { "cpu": 3.829965067173665e-7, "gpu": 0, "ram": 2.4213983180210174e-9, "total": 3.8541790503538754e-7, "unit": "kWh" } }
text_generation
allenai
OLMo-1B-hf
a10g-large
2024-10-24-18-23-56
{ "efficency": { "unit": "tokens/kWh", "value": 62297846.58810162 }, "energy": { "cpu": 0.0006078373521613038, "gpu": 0.004224683213077207, "ram": 0.0000062823229751878135, "total": 0.0048388028882136985, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 1390874.1413723528 }, "energy": { "cpu": 0.0012634316170816028, "gpu": 0.005194258099847594, "ram": 0.000013061109683948865, "total": 0.006470750826613145, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 35562851.946256354 }, "energy": { "cpu": 0.00001108087887082042, "gpu": 0.000016956680231938748, "ram": 8.167052233473494e-8, "total": 0.000028119229625093902, "unit": "kWh" } }
text_generation
openai-community
gpt2-large
a10g-large
2024-10-25-15-05-20
{ "efficency": { "unit": "tokens/kWh", "value": 77907728.2032169 }, "energy": { "cpu": 0.00042952262899040045, "gpu": 0.0027924934839930414, "ram": 0.000004223514289139063, "total": 0.0032262396272725808, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 1343266.704857417 }, "energy": { "cpu": 0.0016164460666061049, "gpu": 0.005067730581958951, "ram": 0.00001590753237599822, "total": 0.006700084180941058, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 34837547.66517286 }, "energy": { "cpu": 0.000010432217698123875, "gpu": 0.000018202236783837478, "ram": 7.020663859453463e-8, "total": 0.00002870466112055589, "unit": "kWh" } }
text_generation
HuggingFaceTB
SmolLM-135M
a10g-large
2024-10-23-19-09-15
{ "efficency": { "unit": "tokens/kWh", "value": 218923397.1364946 }, "energy": { "cpu": 0.00028216661832921095, "gpu": 0.0011204146741087939, "ram": 0.000002878352463094162, "total": 0.001405459644901099, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 1478710.1205686298 }, "energy": { "cpu": 0.0020209151121617142, "gpu": 0.004044847485875458, "ram": 0.000020623012510417966, "total": 0.00608638561054759, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 35187752.05342098 }, "energy": { "cpu": 0.000010718486329682895, "gpu": 0.000017620569652265772, "ram": 7.992339706452299e-8, "total": 0.00002841897937901319, "unit": "kWh" } }
text_generation
HuggingFaceTB
SmolLM-1.7B
a10g-large
2024-10-24-15-16-02
{ "efficency": { "unit": "tokens/kWh", "value": 41960978.223952904 }, "energy": { "cpu": 0.0009210179793032473, "gpu": 0.006402295760721532, "ram": 0.000009403775706541059, "total": 0.007332717515731321, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 986792.3688060087 }, "energy": { "cpu": 0.0017958659208215804, "gpu": 0.007306251844996492, "ram": 0.000018341901791392123, "total": 0.009120459667609458, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 37376068.58504866 }, "energy": { "cpu": 0.000010438337355784218, "gpu": 0.000016238901880072376, "ram": 7.78486547950319e-8, "total": 0.000026755087890651626, "unit": "kWh" } }
text_generation
HuggingFaceTB
SmolLM-360M
a10g-large
2024-10-24-14-22-22
{ "efficency": { "unit": "tokens/kWh", "value": 124481811.63546507 }, "energy": { "cpu": 0.00038417323045180534, "gpu": 0.0020836598613710013, "ram": 0.000003917567631278797, "total": 0.002471750659454085, "unit": "kWh" } }
{ "efficiency": { "unit": "tokens/kWh", "value": 1315130.308155782 }, "energy": { "cpu": 0.0021575579513180274, "gpu": 0.004663859981084995, "ram": 0.00002201039115281833, "total": 0.0068434283235558405, "unit": "kWh" } }
{ "efficiency": { "unit": "samples/kWh", "value": 36616831.97480331 }, "energy": { "cpu": 0.000010503983233461947, "gpu": 0.000016727513381886716, "ram": 7.834823109019483e-8, "total": 0.00002730984484643886, "unit": "kWh" } }

Analysis of energy usage for HUGS models

Based on the energy_star branch of optimum-benchmark, and using codecarbon.

Fields

  • task: Task the model was benchmarked on.
  • org: Organization hosting the model.
  • model: The specific model. Model names at HF are usually constructed with {org}/{model}.
  • date: The date that the benchmark was run.
  • prefill: The esimated energy and efficiency for prefilling.
  • decode: The estimated energy and efficiency for decoding.
  • preprocess: The estimated energy and efficiency for preprocessing.

Code to Reproduce

As I'm devving, I'm hopping between https://huggingface.co/spaces/AIEnergyScore/benchmark-hugs-models and https://huggingface.co/spaces/meg/CalculateCarbon

From there, python code/make_pretty_dataset.py (included in this repository) takes the raw results and uploads them to the dataset here.

Downloads last month
24
Size of downloaded dataset files:
30.8 kB
Size of the auto-converted Parquet files:
30.8 kB
Number of rows:
18