Fill-Mask
Transformers
Safetensors
English
theo_bert_base
masked-language-modeling
bible
theology
christianity
trust-remote-code
custom_code
Eval Results (legacy)
Instructions to use toranb/theo-bert-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use toranb/theo-bert-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="toranb/theo-bert-base", trust_remote_code=True)# Load model directly from transformers import AutoModelForMaskedLM model = AutoModelForMaskedLM.from_pretrained("toranb/theo-bert-base", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "release_stage": "mlmcontinued (stage 2, epoch 25)", | |
| "reported_loss": 0.8958267427277438, | |
| "parameter_count": 273051864, | |
| "fp16_export": true, | |
| "tokenizer": "google-bert/bert-base-uncased", | |
| "pretraining_stages": [ | |
| { | |
| "stage": 1, | |
| "name": "encoder", | |
| "objective": "token-level masked language modeling at 20% mask rate 80/10/10 split", | |
| "epochs": 24, | |
| "seq_len": 256, | |
| "final_train_loss": 1.0678829201169648, | |
| "final_train_accuracy": 76.41802635495705 | |
| }, | |
| { | |
| "stage": 2, | |
| "name": "mlmcontinued", | |
| "objective": "whole-word-masking continued pretraining at 18% mask rate", | |
| "epochs": 25, | |
| "seq_len": 256, | |
| "final_train_loss": 0.8958267427277438, | |
| "final_train_accuracy": 79.66191907459469 | |
| } | |
| ], | |
| "mlm_eval_overall_pass_rate": 0.947, | |
| "mlm_eval_passed_cases": 517, | |
| "mlm_eval_total_cases": 546 | |
| } | |