Hugging Face's logo

lapp0
/
distily_bench_gpt2_optim

TensorBoard
Safetensors
distily
gpt2
Generated from Trainer
8-bit precision
bitsandbytes
Model card Files Metrics Community
lapp0 commited on
Commit
d5a3ccf
·
verified ·
1 Parent(s): 7ab1146

End of training

Browse files
Files changed (2) hide show
  1. README.md +22 -22
  2. runs/Aug10_08-03-02_93d6cbb3ad53/events.out.tfevents.1723281203.93d6cbb3ad53 +3 -0
README.md CHANGED
@@ -16,13 +16,13 @@ This student model is distilled from the teacher model [gpt2](https://huggingfac
16
  The [Distily](https://github.com/lapp0/distily) library was used for this distillation.
17
 
18
  It achieves the following results on the evaluation set:
19
- - eval_enwikippl: 26003.4414
20
- - eval_frwikippl: 43473.625
21
- - eval_zhwikippl: 54798.5430
22
- - eval_loss: 21585.9199
23
- - eval_runtime: 21.7886
24
- - eval_samples_per_second: 45.896
25
- - eval_steps_per_second: 11.474
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
  should probably proofread and complete it, then remove this comment.
@@ -59,26 +59,26 @@ The following hyperparameters were used during training:
59
  - num_epochs: 1.0
60
 
61
  ### Resource Usage
62
- Peak GPU Memory: 4.5037 GB
63
 
64
  ### Eval-Phase Metrics
65
  | step | epoch | enwikippl | frwikippl | loss | runtime | samples_per_second | steps_per_second | zhwikippl |
66
  | --- | --- | --- | --- | --- | --- | --- | --- | --- |
67
  | **teacher eval** | | 30.2385 | 57.2728 | | | | | 18.1772 |
68
- | 0 | 0 | 55339.3672 | 57682.5742 | 331776.0 | 21.609 | 46.277 | 11.569 | 57080.2930 |
69
- | 500 | 0.0808 | 53840.9336 | 57103.8711 | 31504.6406 | 21.8206 | 45.828 | 11.457 | 60063.5586 |
70
- | 1000 | 0.1616 | 46110.3789 | 54346.3320 | 25851.3926 | 21.7004 | 46.082 | 11.521 | 58033.3359 |
71
- | 1500 | 0.2424 | 39930.7539 | 50785.9883 | 24363.0078 | 21.7826 | 45.908 | 11.477 | 56878.6953 |
72
- | 2000 | 0.3232 | 35821.5273 | 48514.4766 | 23500.8008 | 21.6304 | 46.231 | 11.558 | 56064.2539 |
73
- | 2500 | 0.4040 | 33513.9102 | 47385.3516 | 23009.5352 | 22.046 | 45.36 | 11.34 | 55873.6484 |
74
- | 3000 | 0.4848 | 31516.0898 | 46269.4453 | 22568.4473 | 21.8604 | 45.745 | 11.436 | 55709.7695 |
75
- | 3500 | 0.5657 | 30457.4590 | 45776.25 | 22369.2793 | 21.741 | 45.996 | 11.499 | 55598.2578 |
76
- | 4000 | 0.6465 | 29546.6035 | 45307.4453 | 22169.5996 | 21.7185 | 46.044 | 11.511 | 55524.0742 |
77
- | 4500 | 0.7273 | 28461.1484 | 44691.9258 | 21980.1602 | 21.6611 | 46.166 | 11.541 | 55228.2812 |
78
- | 5000 | 0.8081 | 27586.4121 | 44246.7188 | 21925.6328 | 21.7331 | 46.013 | 11.503 | 55025.875 |
79
- | 5500 | 0.8889 | 26811.3066 | 43867.7734 | 21713.1523 | 21.755 | 45.966 | 11.492 | 54930.3984 |
80
- | 6000 | 0.9697 | 26139.0703 | 43621.0156 | 21624.0645 | 21.6556 | 46.177 | 11.544 | 54864.4336 |
81
- | 6187 | 0.9999 | 26003.4414 | 43473.625 | 21585.9199 | 21.7886 | 45.896 | 11.474 | 54798.5430 |
82
 
83
  ### Framework versions
84
  - Distily 0.2.0
 
16
  The [Distily](https://github.com/lapp0/distily) library was used for this distillation.
17
 
18
  It achieves the following results on the evaluation set:
19
+ - eval_enwikippl: 854.1211
20
+ - eval_frwikippl: 4984.0498
21
+ - eval_zhwikippl: 8071.4624
22
+ - eval_loss: 7592.3198
23
+ - eval_runtime: 22.093
24
+ - eval_samples_per_second: 45.263
25
+ - eval_steps_per_second: 11.316
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
28
  should probably proofread and complete it, then remove this comment.
 
59
  - num_epochs: 1.0
60
 
61
  ### Resource Usage
62
+ Peak GPU Memory: 4.6175 GB
63
 
64
  ### Eval-Phase Metrics
65
  | step | epoch | enwikippl | frwikippl | loss | runtime | samples_per_second | steps_per_second | zhwikippl |
66
  | --- | --- | --- | --- | --- | --- | --- | --- | --- |
67
  | **teacher eval** | | 30.2385 | 57.2728 | | | | | 18.1772 |
68
+ | 0 | 0 | 55339.3672 | 57682.5742 | 331776.0 | 21.5321 | 46.442 | 11.611 | 57080.2930 |
69
+ | 500 | 0.0808 | 2252.7444 | 10031.8955 | 12590.8477 | 21.6698 | 46.147 | 11.537 | 54057.1758 |
70
+ | 1000 | 0.1616 | 1772.2999 | 6312.1328 | 10713.8564 | 21.7782 | 45.917 | 11.479 | 20088.1348 |
71
+ | 1500 | 0.2424 | 1526.4294 | 5867.9150 | 9839.7441 | 21.9107 | 45.64 | 11.41 | 12735.1455 |
72
+ | 2000 | 0.3232 | 1386.3311 | 5905.2700 | 9335.0400 | 21.6766 | 46.133 | 11.533 | 12883.1006 |
73
+ | 2500 | 0.4040 | 1285.9930 | 5870.4004 | 9057.4082 | 21.7814 | 45.911 | 11.478 | 12968.5391 |
74
+ | 3000 | 0.4848 | 1184.0109 | 5485.2373 | 8730.4961 | 21.825 | 45.819 | 11.455 | 11484.6025 |
75
+ | 3500 | 0.5657 | 1126.2782 | 5563.9180 | 8546.9443 | 21.8317 | 45.805 | 11.451 | 12904.6191 |
76
+ | 4000 | 0.6465 | 1054.3176 | 5538.6753 | 8247.4883 | 21.7217 | 46.037 | 11.509 | 12877.9414 |
77
+ | 4500 | 0.7273 | 994.4172 | 5374.4102 | 8100.9922 | 21.713 | 46.055 | 11.514 | 11938.875 |
78
+ | 5000 | 0.8081 | 946.4249 | 5192.8228 | 7962.6880 | 21.8256 | 45.818 | 11.454 | 9305.625 |
79
+ | 5500 | 0.8889 | 910.4888 | 5210.4282 | 7757.3120 | 22.1477 | 45.152 | 11.288 | 9479.9629 |
80
+ | 6000 | 0.9697 | 871.6422 | 5126.9775 | 7617.0239 | 22.8323 | 43.798 | 10.949 | 8336.0049 |
81
+ | 6187 | 0.9999 | 854.1211 | 4984.0498 | 7592.3198 | 22.093 | 45.263 | 11.316 | 8071.4624 |
82
 
83
  ### Framework versions
84
  - Distily 0.2.0
runs/Aug10_08-03-02_93d6cbb3ad53/events.out.tfevents.1723281203.93d6cbb3ad53 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fd785d490400e047e911f4728357f95327a79505e21b50bde6624a3a5945764
3
+ size 249