Aller au contenu
ModelVérifiéExterne
Télécharger

Child Mlm 13 Age 7 English20000

This model is a fine-tuned version of []( on an unknown dataset. It achieves the following results on the evaluation set: Loss: 2.5704

TypeOtherVersionRevision 5bac12eÉditeur@HaerimCClasse

Source du modèle

Type
Other
Taille du fichier
80,0 MB
Version
vRevision 5bac12e
Mis à jour
5 oct. 2026
Éditeur
@Haerim
Téléchargements
1
Source
Hugging Face

Description de la source

child_mlm_13_age_7_english20000

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 2.5704

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 13
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 40000
  • training_steps: 100000
  • mixed_precision_training: Native AMP

Training results

Training LossEpochStepValidation Loss
8.28413.358520007.7270
6.02916.717040006.1305
5.809110.075660005.9916
5.69813.434180005.8868
5.616616.7926100005.8108
5.55820.1511120005.7603
5.525423.5097140005.7123
5.498826.8682160005.6968
5.480830.2267180005.6992
5.464333.5852200005.6614
5.453436.9437220005.6807
5.440840.3023240005.6420
5.435343.6608260005.6259
5.432247.0193280005.6326
5.423450.3778300005.6014
5.415353.7364320005.6256
5.415557.0949340005.6147
5.404960.4534360005.6082
5.400563.8119380005.5832
5.385567.1704400005.5802
5.365870.5290420005.5462
4.081973.8875440003.9602
3.406477.2460460003.5232
3.14180.6045480003.3000
2.979883.9631500003.1618
2.858687.3216520003.0844
2.774690.6801540003.0142
2.706794.0386560002.9721
2.647397.3971580002.9117
2.5961100.7557600002.8787
2.5554104.1142620002.8349
2.5181107.4727640002.7934
2.4887110.8312660002.7957
2.4563114.1898680002.7555
2.4331117.5483700002.7204
2.4109120.9068720002.7106
2.391124.2653740002.6970
2.3751127.6238760002.6673
2.3531130.9824780002.6641
2.3408134.3409800002.6521
2.3283137.6994820002.6417
2.316141.0579840002.6322
2.3059144.4165860002.6191
2.2941147.7750880002.6216
2.2863151.1335900002.6146
2.2793154.4920920002.6057
2.2756157.8505940002.6026
2.266161.2091960002.5884
2.2653164.5676980002.5904
2.26167.92611000002.5704

Framework versions

  • Transformers 4.45.2
  • Pytorch 2.11.0+cu128
  • Datasets 3.0.1
  • Tokenizers 0.20.1

Sources

1 source
Hugging FacePrincipalVersion 5bac12e17fdc516cc054451730355aef4ae9b1d0
huggingface.co/Haerim/child_mlm_13_age_7_english20000/tree/5bac12e17fdc516cc054451730355aef4ae9b1d0

Vérifié 2 sept.

Artefacts du modèle

1 artefact

model.safetensors

safetensors · 80,0 MB · SHA-256 91e6dad20f22…4b52 · Hugging Face

Télécharger

Extraits de sources

3 extraits

--- library_name: transformers tags: - generated_from_trainer model-index: - name: child_mlm_13_age_7_english20000 results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # child_mlm_13_age_7_english20000 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset. It achieves the following results on the evaluation set: - Loss: 2.5704 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 0.0001 - train_batch_size: 16 - eval_batch_size: 16 - seed: 13 - gradient_accumulation_steps: 2 - total_train_batch_size: 32 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - lr_scheduler_warmup_steps: 40000 - training_steps: 100000 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:--------:|:------:|:---------------:| | 8.2841 | 3.3585 | 2000 | 7.7270 | | 6.0291 | 6.7170 | 4000 | 6.1305 | | 5.8091 | 10.0756 | 6000 | 5.9916 | | 5.698 | 13.4341 | 8000 | 5.8868 | | 5.6166 | 16.7926 | 10000 | 5.8108 | | 5.558 | 20.1511 | 12000 | 5.7603 | | 5.5254 | 23.5097 | 14000 | 5.7123 | | 5.4988 | 26.8682 | 16000 | 5.6968 | | 5.4808 | 30.2267 | 18000 | 5.6992 | | 5.4643 | 33.5852 | 20000 | 5.6614 | | 5.4534 | 36.9437 | 22000 | 5.6807 | | 5.4408 | 40.3023 | 24000 | 5.6420 | | 5.4353 | 43.6608 | 26000 | 5.6259 | | 5.4322 | 47.0193 | 28000 | 5.6326 | | 5.4234 | 50.3778 | 30000 | 5.6014 | | 5.4153 | 53.7364 | 32000 | 5.6256 | | 5.4155 | 57.0949 | 34000 | 5.6147 | | 5.4049 | 60.4534 | 36000 | 5.6082 | | 5.4005 | 63.8119 | 38000 | 5.5832 | | 5.3855 | 67.1704 | 40000 | 5.5802 | | 5.3658 | 70.5290 | 42000 | 5.5462 | | 4.0819 | 73.8875 | 44000 | 3.9602 | | 3.4064 | 77.2460 | 46000 | 3.5232 | | 3.141 | 80.6045 | 48000 | 3.3000 | | 2.9798 | 83.9631 | 50000 | 3.1618 | | 2.8586 | 87.3216 | 52000 | 3.0844 | | 2.7746 | 90.6801 | 54000 | 3.0142 | | 2.7067 | 94.0386 | 56000 | 2.9721 | | 2.6473 | 97.3971 | 58000 | 2.9117 | | 2.5961 | 100.7557 | 60000 | 2.8787 | | 2.5554 | 104.11...

Source context: 1 downloads · 0 likes · Pipeline fill-mask · Library transformers · Repo Haerim/child_mlm_13_age_7_english20000

Source context: 2 downloads · 0 likes · Pipeline fill-mask · Library transformers · Repo Haerim/child_mlm_13_age_7_english20000