Add models/<name>/tokenizer.json for the five tokbench models
tokbench's make models downloads models/<name>/tokenizer.json for its five benchmark models from this dataset. None of those paths exist yet, so a fresh clone fails at make models. This PR adds them.
Each file is the unmodified tokenizer.json from the official model repo (full vocabularies — all five pass tokbench's truncated-vocab check, and the full benchmark matrix was run locally against these exact files before opening this PR):
| path | source | revision | sha256 |
|---|---|---|---|
models/gpt2/tokenizer.json |
openai-community/gpt2 | 607a30d783df |
8414cab924d8b9b3… |
models/llama-3/tokenizer.json |
meta-llama/Meta-Llama-3-8B | 8cde5ca83804 |
e134af98b985517b… |
models/deepseek-v4/tokenizer.json |
deepseek-ai/DeepSeek-V4-Pro | b5968e9190ef |
8f9f37ca37fdc4f5… |
models/bert-base-uncased/tokenizer.json |
google-bert/bert-base-uncased | 86b5e0934494 |
ce64fce797c24f68… |
models/t5-base/tokenizer.json |
google-t5/t5-base | a9723ea7f1b3 |
d2acde0d8d71dd30… |
Notes:
deepseek-v4: DeepSeek-V4-Pro and DeepSeek-V4-Flash serve a byte-identical tokenizer.json (same LFS object), so one file covers the family.llama-3: taken from the gated meta-llama/Meta-Llama-3-8B repo rather than reusing the root-levelllama-3-tokenizer.jsonalready in this dataset, because that file is a Llama-3.1 config: same 128000-entry vocab and identical merges, but ids 128004/128008/128010 hold<|finetune_right_pad_id|>/<|eom_id|>/<|python_tag|>where Llama-3 has<|reserved_special_token_*|>. tokbench names this modelllama-3and renders its chat fixtures from the Llama-3 (not 3.1) template, so the Llama-3 config is the consistent choice. With the current fixture set the two are interchangeable anyway — no fixture contains any of the differing special tokens.
Closing: this was the wrong fix, and the files were never missing.
All five configs are already in this dataset as flat root-level files (gpt2.json, deepseek-v4.json, bert-base-uncased.json, t5-base.json, llama-3-tokenizer.json). The first four are byte-identical to the official model repos' tokenizer.json. llama-3-tokenizer.json carries the Llama-3.1 special-token names (<|eom_id|>, <|python_tag|>, <|finetune_right_pad_id|> at ids 128008/128010/128004), but its model, pre_tokenizer, normalizer, post_processor and decoder are identical to Llama-3's, and it derives a byte-identical ranks.tiktoken and pattern.txt.
What actually broke was tokbench's Makefile: it fetched models/<name>/tokenizer.json, a layout this dataset does not use. That is fixed in tokbench instead, so no data is duplicated here.