898 MB

Ctrl+K

3 contributors

History: 21 commits

diyclassics

Claude Opus 4.6 (1M context)

fix: apply do_lower_case to root-level HF-facing files

f25becd 10 days ago

data
refactor: extract shared case study utils and move data to tracked paths 14 days ago
src
fix: add do_lower_case=True to tokenizer (v1.1.1) 10 days ago
tests
fix: add do_lower_case=True to tokenizer (v1.1.1) 10 days ago
.gitattributes

1.52 kB
chore: add HF model repo files (config, tokenizer, encoder, README) 14 days ago
.gitignore

635 Bytes
chore: gitignore cluster/, track test vocab fixture 12 days ago
.python-version

5 Bytes
Initial: HF-compatible Latin BERT tokenizer (Bamman & Burns 2020) 17 days ago
README.md

4.22 kB
fix: add do_lower_case=True to tokenizer (v1.1.1) 10 days ago
config.json

562 Bytes
fix: revert tie_word_embeddings — safetensors needs tying for decoder weights 12 days ago
latin.subword.encoder

287 kB
chore: add HF model repo files (config, tokenizer, encoder, README) 14 days ago
model.safetensors

448 MB
xet

Adding `safetensors` variant of this model (#1) 12 days ago
pyproject.toml

680 Bytes
fix: add do_lower_case=True to tokenizer (v1.1.1) 10 days ago
pytorch_model.bin
Detected Pickle imports (3)
- "torch._utils._rebuild_tensor_v2",
- "collections.OrderedDict",
- "torch.FloatStorage"
What is a pickle import?
448 MB
xet

chore: re-upload model weights (pytorch_model.bin) 14 days ago
tokenization_latin_bert.py

11.3 kB
fix: apply do_lower_case to root-level HF-facing files 10 days ago
tokenization_latin_bert_fast.py

9.96 kB
fix: add decode/unescape to fast tokenizer, silence tied-weights warning 12 days ago
tokenizer_config.json

389 Bytes
fix: apply do_lower_case to root-level HF-facing files 10 days ago
uv.lock

644 kB
fix: add do_lower_case=True to tokenizer (v1.1.1) 10 days ago

Detected Pickle imports (3)