HabibAhmed commited on
Commit
c686070
·
verified ·
1 Parent(s): 562909c

Upload folder using huggingface_hub

Browse files
.ipynb_checkpoints/Llama3.2-1B-INT8-lora-BF16-Evaluation_Results-checkpoint.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eval_loss": 0.7597980586406545,
3
+ "perplexity": 2.1378444576870472,
4
+ "performance_metrics": {
5
+ "accuracy:": 1.0,
6
+ "precision:": 1.0,
7
+ "recall:": 1.0,
8
+ "f1:": 1.0,
9
+ "bleu:": 0.9716335135223347,
10
+ "rouge:": {
11
+ "rouge1": 0.9847093827042951,
12
+ "rouge2": 0.9846102568018172,
13
+ "rougeL": 0.9847093827042951
14
+ },
15
+ "semantic_similarity_avg:": 0.9987004399299622
16
+ },
17
+ "mauve": 0.8832777311669464,
18
+ "inference_performance": {
19
+ "min_latency_ms": 70.59454917907715,
20
+ "max_latency_ms": 2859.173536300659,
21
+ "lower_quartile_ms": 75.28102397918701,
22
+ "median_latency_ms": 76.46441459655762,
23
+ "upper_quartile_ms": 516.7384743690491,
24
+ "avg_latency_ms": 626.5141687023942,
25
+ "min_memory_mb": 1818372.0,
26
+ "max_memory_mb": 1819316.0,
27
+ "lower_quartile_mb": 1818658.5,
28
+ "median_memory_mb": 1818923.0,
29
+ "upper_quartile_mb": 1819046.0,
30
+ "avg_memory_mb": 1818866.592790387
31
+ }
32
+ }
Llama3.2-1B-INT8-lora-BF16-Evaluation_Results.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eval_loss:": 0.7597980586406545,
3
+ "perplexity:": 2.1378444576870472,
4
+ "performance_metrics:": {
5
+ "accuracy:": 1.0,
6
+ "precision:": 1.0,
7
+ "recall:": 1.0,
8
+ "f1:": 1.0,
9
+ "bleu:": 0.9716128952473347,
10
+ "rouge:": {
11
+ "rouge1": 0.9847181247843387,
12
+ "rouge2": 0.9846230727757008,
13
+ "rougeL": 0.9847181247843387
14
+ },
15
+ "semantic_similarity_avg:": 0.9988512396812439
16
+ },
17
+ "mauve:": 0.8832777311669464,
18
+ "inference_performance:": {
19
+ "min_latency_ms": 73.21643829345703,
20
+ "max_latency_ms": 3008.4731578826904,
21
+ "lower_quartile_ms": 77.86405086517334,
22
+ "median_latency_ms": 79.05101776123047,
23
+ "upper_quartile_ms": 568.7307715415955,
24
+ "avg_latency_ms": 646.2314533136874,
25
+ "min_memory_gb": 0.07419586181640625,
26
+ "max_memory_gb": 0.07562446594238281,
27
+ "lower_quartile_gb": 0.0746617317199707,
28
+ "median_memory_gb": 0.07486915588378906,
29
+ "upper_quartile_gb": 0.07507467269897461,
30
+ "avg_memory_gb": 0.07487648606141196,
31
+ "model_load_memory_gb": 4.578568935394287,
32
+ "avg_inference_memory_gb": 0.07487648606141196
33
+ }
34
+ }
Llama3.2-1B-INT8-lora-BF16-Inference_Curve.png ADDED
Llama3.2-1B-INT8-lora-BF16-Latency_Histogram.png ADDED
Llama3.2-1B-INT8-lora-BF16-Memory_Histogram.png ADDED
Llama3.2-1B-INT8-lora-BF16-Memory_Usage_Curve.png ADDED