End of training
Browse files
README.md
CHANGED
@@ -16,14 +16,14 @@ 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:
|
20 |
-
- eval_frwikippl:
|
21 |
-
- eval_zhwikippl:
|
22 |
-
- eval_tinystoriesppl:
|
23 |
-
- eval_loss: 0.
|
24 |
-
- eval_runtime: 16.
|
25 |
-
- eval_samples_per_second:
|
26 |
-
- eval_steps_per_second: 7.
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
should probably proofread and complete it, then remove this comment.
|
@@ -46,7 +46,7 @@ More information needed
|
|
46 |
### Training hyperparameters
|
47 |
|
48 |
The following hyperparameters were used during training:
|
49 |
-
- distillation_objective: DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl, layer_mapper=None, projector=None), hs_loss_component=LossComponent(label=hs, weight=0, loss_fn=
|
50 |
- train_embeddings: True
|
51 |
- learning_rate: 0.0001
|
52 |
- train_batch_size: 4
|
@@ -58,38 +58,38 @@ The following hyperparameters were used during training:
|
|
58 |
- num_epochs: 1.0
|
59 |
|
60 |
### Resource Usage
|
61 |
-
Peak GPU Memory: 7.
|
62 |
|
63 |
### Eval-Phase Metrics
|
64 |
| step | epoch | enwikippl | frwikippl | loss | runtime | samples_per_second | steps_per_second | tinystoriesppl | zhwikippl |
|
65 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
66 |
| **teacher eval** | | 43.75 | 61.75 | | | | | 11.8125 | 19.125 |
|
67 |
-
| 0 | 0 |
|
68 |
-
| 1000 | 0.0404 |
|
69 |
-
| 2000 | 0.0808 |
|
70 |
-
| 3000 | 0.1212 |
|
71 |
-
| 4000 | 0.1616 |
|
72 |
-
| 5000 | 0.2020 |
|
73 |
-
| 6000 | 0.2424 |
|
74 |
-
| 7000 | 0.2828 |
|
75 |
-
| 8000 | 0.3232 |
|
76 |
-
| 9000 | 0.3636 |
|
77 |
-
| 10000 | 0.4040 |
|
78 |
-
| 11000 | 0.4444 |
|
79 |
-
| 12000 | 0.4848 |
|
80 |
-
| 13000 | 0.5253 |
|
81 |
-
| 14000 | 0.5657 |
|
82 |
-
| 15000 | 0.6061 |
|
83 |
-
| 16000 | 0.6465 | 70.0 |
|
84 |
-
| 17000 | 0.6869 |
|
85 |
-
| 18000 | 0.7273 |
|
86 |
-
| 19000 | 0.7677 | 68.
|
87 |
-
| 20000 | 0.8081 | 65.
|
88 |
-
| 21000 | 0.8485 |
|
89 |
-
| 22000 | 0.8889 |
|
90 |
-
| 23000 | 0.9293 |
|
91 |
-
| 24000 | 0.9697 |
|
92 |
-
| 24750 | 1.0 |
|
93 |
|
94 |
### Framework versions
|
95 |
- 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: 84.0
|
20 |
+
- eval_frwikippl: 342.0
|
21 |
+
- eval_zhwikippl: 217.0
|
22 |
+
- eval_tinystoriesppl: 69.5
|
23 |
+
- eval_loss: 0.6877
|
24 |
+
- eval_runtime: 16.9969
|
25 |
+
- eval_samples_per_second: 58.834
|
26 |
+
- eval_steps_per_second: 7.354
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
should probably proofread and complete it, then remove this comment.
|
|
|
46 |
### Training hyperparameters
|
47 |
|
48 |
The following hyperparameters were used during training:
|
49 |
+
- distillation_objective: DistillationObjective(logits_loss_component=LossComponent(label=logits, weight=1, loss_fn=kl, layer_mapper=None, projector=None), hs_loss_component=LossComponent(label=hs, weight=1.0, loss_fn=mse, layer_mapper=last, projector=None), attn_loss_component=LossComponent(label=attn, weight=0, loss_fn=None, layer_mapper=None, projector=None))
|
50 |
- train_embeddings: True
|
51 |
- learning_rate: 0.0001
|
52 |
- train_batch_size: 4
|
|
|
58 |
- num_epochs: 1.0
|
59 |
|
60 |
### Resource Usage
|
61 |
+
Peak GPU Memory: 7.7252 GB
|
62 |
|
63 |
### Eval-Phase Metrics
|
64 |
| step | epoch | enwikippl | frwikippl | loss | runtime | samples_per_second | steps_per_second | tinystoriesppl | zhwikippl |
|
65 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
|
66 |
| **teacher eval** | | 43.75 | 61.75 | | | | | 11.8125 | 19.125 |
|
67 |
+
| 0 | 0 | 2473901162496.0 | 170424302305280.0 | 20.7680 | 17.0409 | 58.682 | 7.335 | 4060086272.0 | 71468255805440.0 |
|
68 |
+
| 1000 | 0.0404 | 334.0 | 1464.0 | 1.5419 | 17.0178 | 58.762 | 7.345 | 243.0 | 596.0 |
|
69 |
+
| 2000 | 0.0808 | 232.0 | 756.0 | 1.3235 | 16.9755 | 58.909 | 7.364 | 189.0 | 250.0 |
|
70 |
+
| 3000 | 0.1212 | 180.0 | 628.0 | 1.1620 | 16.9923 | 58.85 | 7.356 | 149.0 | 171.0 |
|
71 |
+
| 4000 | 0.1616 | 150.0 | 576.0 | 1.0434 | 16.9803 | 58.892 | 7.361 | 121.5 | 172.0 |
|
72 |
+
| 5000 | 0.2020 | 130.0 | 504.0 | 0.9520 | 17.0128 | 58.779 | 7.347 | 100.5 | 144.0 |
|
73 |
+
| 6000 | 0.2424 | 113.5 | 420.0 | 0.8702 | 17.0074 | 58.798 | 7.35 | 91.0 | 137.0 |
|
74 |
+
| 7000 | 0.2828 | 106.0 | 408.0 | 0.8100 | 16.9821 | 58.885 | 7.361 | 80.5 | 160.0 |
|
75 |
+
| 8000 | 0.3232 | 96.5 | 396.0 | 0.7421 | 16.9749 | 58.911 | 7.364 | 70.5 | 127.0 |
|
76 |
+
| 9000 | 0.3636 | 84.0 | 342.0 | 0.6877 | 16.9969 | 58.834 | 7.354 | 69.5 | 217.0 |
|
77 |
+
| 10000 | 0.4040 | 78.0 | 300.0 | 0.6467 | 16.9846 | 58.877 | 7.36 | 65.0 | 139.0 |
|
78 |
+
| 11000 | 0.4444 | 77.0 | 278.0 | 0.5957 | 16.9903 | 58.857 | 7.357 | 60.0 | 127.5 |
|
79 |
+
| 12000 | 0.4848 | 75.0 | 272.0 | 0.5789 | 16.9858 | 58.873 | 7.359 | 56.5 | 140.0 |
|
80 |
+
| 13000 | 0.5253 | 71.5 | 266.0 | 0.5525 | 16.9418 | 59.026 | 7.378 | 56.5 | 116.0 |
|
81 |
+
| 14000 | 0.5657 | 71.0 | 252.0 | 0.5416 | 17.088 | 58.521 | 7.315 | 53.75 | 132.0 |
|
82 |
+
| 15000 | 0.6061 | 68.0 | 221.0 | 0.5283 | 16.9524 | 58.989 | 7.374 | 50.25 | 112.5 |
|
83 |
+
| 16000 | 0.6465 | 70.0 | 244.0 | 0.5200 | 17.0495 | 58.653 | 7.332 | 52.5 | 109.5 |
|
84 |
+
| 17000 | 0.6869 | 67.0 | 225.0 | 0.5097 | 17.0223 | 58.747 | 7.343 | 51.5 | 109.0 |
|
85 |
+
| 18000 | 0.7273 | 71.0 | 239.0 | 0.5016 | 17.0519 | 58.644 | 7.331 | 49.5 | 150.0 |
|
86 |
+
| 19000 | 0.7677 | 68.0 | 212.0 | 0.4887 | 17.0831 | 58.537 | 7.317 | 51.25 | 98.0 |
|
87 |
+
| 20000 | 0.8081 | 65.0 | 211.0 | 0.4865 | 17.0098 | 58.789 | 7.349 | 49.0 | 101.5 |
|
88 |
+
| 21000 | 0.8485 | 64.5 | 217.0 | 0.4791 | 17.0253 | 58.736 | 7.342 | 47.5 | 142.0 |
|
89 |
+
| 22000 | 0.8889 | 66.5 | 230.0 | 0.4798 | 16.9954 | 58.839 | 7.355 | 48.5 | 147.0 |
|
90 |
+
| 23000 | 0.9293 | 62.5 | 212.0 | 0.4675 | 16.9835 | 58.881 | 7.36 | 45.5 | 134.0 |
|
91 |
+
| 24000 | 0.9697 | 63.5 | 220.0 | 0.4712 | 16.9973 | 58.833 | 7.354 | 47.0 | 138.0 |
|
92 |
+
| 24750 | 1.0 | 63.75 | 247.0 | 0.4679 | 17.0597 | 58.618 | 7.327 | 45.75 | 205.0 |
|
93 |
|
94 |
### Framework versions
|
95 |
- Distily 0.2.0
|
logs/copy_teacher_modules=_(_lm_head___True)_, hs_layer_mapper=last, hs_loss_fn=mse, hs_weight=1.0, learning_rate=0.0001, per_device_train_batch_size=4/events.out.tfevents.1724156440.02dbb11e2dcc
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e1097da824e2c3d8de78fa43ea9a9d6095ba8a8f0527d4011fd4e466792212c4
|
3 |
+
size 312
|