Publish portable current-study preparation, guarded training, and model loader

Add fixed current-study preparation inputs and shared tokenizer

Publish audited October 2026 code and protein data release

Upload from_scratch/tokenizers/mixed_bpe/tokenizer.json with huggingface_hub

Upload from_scratch/tokenizers/bytelevel/vocab.json with huggingface_hub

Upload from_scratch/tokenizers/bytelevel/merges.txt with huggingface_hub

Upload from_scratch/corpora/mixed_bpe_50M/shuffled.bin with huggingface_hub

Upload from_scratch/corpora/mixed_bpe_50M/randomaa.bin with huggingface_hub

Upload from_scratch/corpora/mixed_bpe_50M/protein.bin with huggingface_hub

Upload from_scratch/corpora/mixed_bpe_50M/manifest.json with huggingface_hub

Upload from_scratch/corpora/mixed_bpe_50M/english.bin with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/shuffled.bin with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/randomaa.bin with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/protein.bin with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/manifest.json with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/english.bin with huggingface_hub

Upload from_scratch/corpora/bytelevel_50M/dna.bin with huggingface_hub

Upload cpt_corpus/sprot.fasta with huggingface_hub

Upload cpt_corpus/cpt_text.txt with huggingface_hub

Upload cpt_corpus/cpt_randomaa.txt with huggingface_hub

Upload cpt_corpus/cpt_shuffled.txt with huggingface_hub

Upload cpt_corpus/cpt_protein.txt with huggingface_hub

Upload cpt_corpus/manifest.json with huggingface_hub

Upload folder using huggingface_hub

Upload README.md with huggingface_hub

initial commit