Downloads · 30 days
8
29% of all-time downloads
damgomz/fp_bs4_lr2e4_x4
fp_bs4_lr2e4_x4 is a fill-mask model from damgomz. Use it when you need the model to fill a missing word. It is set up for transformers.
Downloads · 30 days
8
29% of all-time downloads
All-time downloads
28
Public
Parameters
32.5M
1.7 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors130 MB · 98%
From the Hugging Face model README
| Metric | Value |
|---|---|
| Duration (in seconds) | 194303.9660050869 |
| Emissions (Co2eq in kg) | 0.1461757517129581 |
| CPU power (W) | 42.5 |
| GPU power (W) | [No GPU] |
| RAM power (W) | 15.0 |
| CPU energy (kWh) | 2.293861236172571 |
| GPU energy (kWh) | [No GPU] |
| RAM energy (kWh) | 0.8095920177489497 |
| Consumed energy (kWh) | 3.1034532539215345 |
| Country name | Switzerland |
| Cloud provider | nan |
| Cloud region | nan |
| CPU count | 6 |
| CPU model | Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz |
| GPU count | nan |
| GPU model | nan |
| Metric | Value |
|---|---|
| CPU energy (kWh) | 0.37403513455979226 |
| Emissions (Co2eq in kg) | 0.07610238668532569 |
5 juillet 2024 !
| Config | Value |
|---|---|
| checkpoint | albert-base-v2 |
| model_name | fp_bs4_lr2e4_x4 |
| sequence_length | 400 |
| num_epoch | 6 |
| learning_rate | 0.0002 |
| batch_size | 4 |
| weight_decay | 0.0 |
| warm_up_prop | 0.0 |
| drop_out_prob | 0.1 |
| packing_length | 100 |
| train_test_split | 0.2 |
| num_steps | 165228 |
| Epoch | Train Loss | Test Loss |
|---|---|---|
| 0.0 | 14.615252 | 8.929798 |
| 0.5 | 7.183832 | 7.124656 |
| 1.0 | 7.048965 | 7.024166 |
| 1.5 | 7.011350 | 7.016570 |
| 2.0 | 7.003679 | 7.006157 |
| 2.5 | 6.997983 | 7.014537 |
| 3.0 | 6.991793 | 7.001341 |
| 3.5 | 6.991715 | 6.998451 |
| 4.0 | 6.989365 | 6.990131 |
| 4.5 | 6.984105 | 6.998179 |
| 5.0 | 6.982990 | 6.989589 |
| 5.5 | 6.979184 | 6.982454 |
| 6.0 | 6.976470 | 6.992018 |