drbh HF Staff commited on
Commit
b19bcf6
·
verified ·
1 Parent(s): 93ca832

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ artifacts/visualization/moe_performance_comparison.png filter=lfs diff=lfs merge=lfs -text
artifacts/binned_run/binned_results.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "implementation": "binned_results",
3
+ "config": {
4
+ "warmup": 10,
5
+ "iters": 50,
6
+ "device": "cuda",
7
+ "dtype": "torch.float32",
8
+ "tokens": 100
9
+ },
10
+ "stats": {
11
+ "avg_ms": 105.61800455325283,
12
+ "min_ms": 103.41683897422627,
13
+ "max_ms": 107.80877695651725,
14
+ "std_ms": 1.4579030267805855,
15
+ "p50_ms": 105.04751797998324,
16
+ "p95_ms": 107.72936256835237,
17
+ "p99_ms": 107.78988109494094,
18
+ "num_iters": 50,
19
+ "tokens_per_s": 946.8082683722715,
20
+ "throughput_variance": 13.040579048845922
21
+ },
22
+ "output_sum": -0.5972483158111572
23
+ }
artifacts/gptoss_run/gptoss_results.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "implementation": "gptoss_results",
3
+ "config": {
4
+ "warmup": 10,
5
+ "iters": 50,
6
+ "device": "cuda",
7
+ "dtype": "torch.float32",
8
+ "tokens": 100
9
+ },
10
+ "stats": {
11
+ "avg_ms": 47.135205678641796,
12
+ "min_ms": 46.581514994613826,
13
+ "max_ms": 47.89501999039203,
14
+ "std_ms": 0.5025885809879256,
15
+ "p50_ms": 46.78934201365337,
16
+ "p95_ms": 47.8006230579922,
17
+ "p99_ms": 47.856090962304734,
18
+ "num_iters": 50,
19
+ "tokens_per_s": 2121.5564578582635,
20
+ "throughput_variance": 22.54050463266689
21
+ },
22
+ "output_sum": -0.5972497463226318
23
+ }
artifacts/gptoss_training_run/gptoss_training_results.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "implementation": "gptoss_training_results",
3
+ "config": {
4
+ "warmup": 10,
5
+ "iters": 50,
6
+ "device": "cuda",
7
+ "dtype": "torch.float32",
8
+ "tokens": 100
9
+ },
10
+ "stats": {
11
+ "avg_ms": 48.6301564052701,
12
+ "min_ms": 47.534799959976226,
13
+ "max_ms": 49.413569970056415,
14
+ "std_ms": 0.5585868637338623,
15
+ "p50_ms": 48.39514396735467,
16
+ "p95_ms": 49.34615618258249,
17
+ "p99_ms": 49.389877034700476,
18
+ "num_iters": 50,
19
+ "tokens_per_s": 2056.3372070331834,
20
+ "throughput_variance": 23.60225801703488
21
+ },
22
+ "output_sum": -0.5972497463226318
23
+ }
artifacts/megablocks_run/megablocks_results.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "implementation": "megablocks_results",
3
+ "config": {
4
+ "warmup": 10,
5
+ "iters": 50,
6
+ "device": "cuda",
7
+ "dtype": "torch.float32",
8
+ "tokens": 100
9
+ },
10
+ "stats": {
11
+ "avg_ms": 4.133109538815916,
12
+ "min_ms": 0.822786008939147,
13
+ "max_ms": 8.589233038946986,
14
+ "std_ms": 3.7810751831568976,
15
+ "p50_ms": 0.8644890040159225,
16
+ "p95_ms": 8.578622492495924,
17
+ "p99_ms": 8.588659723172896,
18
+ "num_iters": 50,
19
+ "tokens_per_s": 24194.858389513854,
20
+ "throughput_variance": 52511.72634241489
21
+ },
22
+ "output_sum": -0.5972487926483154
23
+ }
artifacts/visualization/moe_performance_comparison.png ADDED

Git LFS Details

  • SHA256: 3ec2f6a6635799f826e6d05687476446e6b1e694b24c78ac1b197d9dfa4f32a9
  • Pointer size: 131 Bytes
  • Size of remote file: 311 kB
artifacts/yamoe_run/yamoe_results.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "implementation": "yamoe_results",
3
+ "config": {
4
+ "warmup": 10,
5
+ "iters": 50,
6
+ "device": "cuda",
7
+ "dtype": "torch.float32",
8
+ "tokens": 100
9
+ },
10
+ "stats": {
11
+ "avg_ms": 8.6306691612117,
12
+ "min_ms": 8.526118996087462,
13
+ "max_ms": 8.66098003461957,
14
+ "std_ms": 0.021547447475258306,
15
+ "p50_ms": 8.635657490231097,
16
+ "p95_ms": 8.65293950191699,
17
+ "p99_ms": 8.657966026803479,
18
+ "num_iters": 50,
19
+ "tokens_per_s": 11586.58710374672,
20
+ "throughput_variance": 29.13130136142299
21
+ },
22
+ "output_sum": -0.5972495079040527
23
+ }
compare.html ADDED
The diff for this file is too large to render. See raw diff