Upload 5 files
Browse files- crumb_metadata.json +38 -0
- merges.txt +0 -0
- tokenizer.json +0 -0
- tokenizer_config.json +69 -0
- vocab.json +0 -0
crumb_metadata.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"name": "CRUMB Python BPE \u2014 TOK-1 v7.4.0",
|
| 3 |
+
"script_version": "v7.4.0",
|
| 4 |
+
"vocab_size": 32768,
|
| 5 |
+
"domain": "python_source_code",
|
| 6 |
+
"corpus": "stack-dedup-python-parquet-27GB",
|
| 7 |
+
"min_frequency": 2,
|
| 8 |
+
"trainer": "ByteLevelBPETokenizer via train_from_iterator (Python str)",
|
| 9 |
+
"protected_keywords": 38,
|
| 10 |
+
"protected_dunders": 21,
|
| 11 |
+
"protected_builtins": 73,
|
| 12 |
+
"protected_exceptions": 69,
|
| 13 |
+
"protected_stdlib": 125,
|
| 14 |
+
"protected_ds_tokens": 119,
|
| 15 |
+
"total_never_split": 474,
|
| 16 |
+
"special_tokens": 20,
|
| 17 |
+
"reserved_future_tokens": 4,
|
| 18 |
+
"target_model_sizes": [
|
| 19 |
+
"124M",
|
| 20 |
+
"250M",
|
| 21 |
+
"370M",
|
| 22 |
+
"500M"
|
| 23 |
+
],
|
| 24 |
+
"fim_tokens": [
|
| 25 |
+
"<fim_prefix>",
|
| 26 |
+
"<fim_suffix>",
|
| 27 |
+
"<fim_middle>"
|
| 28 |
+
],
|
| 29 |
+
"notebook_tokens": [
|
| 30 |
+
"<cell>",
|
| 31 |
+
"</cell>",
|
| 32 |
+
"<cell_output>",
|
| 33 |
+
"</cell_output>"
|
| 34 |
+
],
|
| 35 |
+
"comments_in_training": true,
|
| 36 |
+
"docstrings_in_training": true,
|
| 37 |
+
"training_date": "2026-05-11"
|
| 38 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"extra_special_tokens": [
|
| 6 |
+
"<func>",
|
| 7 |
+
"</func>",
|
| 8 |
+
"<class>",
|
| 9 |
+
"</class>",
|
| 10 |
+
"<module>",
|
| 11 |
+
"</module>",
|
| 12 |
+
"<cell>",
|
| 13 |
+
"</cell>",
|
| 14 |
+
"<cell_output>",
|
| 15 |
+
"</cell_output>",
|
| 16 |
+
"<traceback>",
|
| 17 |
+
"</traceback>",
|
| 18 |
+
"<fim_prefix>",
|
| 19 |
+
"<fim_suffix>",
|
| 20 |
+
"<fim_middle>",
|
| 21 |
+
"<comment>",
|
| 22 |
+
"<docstring>",
|
| 23 |
+
"</docstring>",
|
| 24 |
+
"<type_hint>"
|
| 25 |
+
],
|
| 26 |
+
"mask_token": "<mask>",
|
| 27 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 28 |
+
"pad_token": "<pad>",
|
| 29 |
+
"tokenizer_class": "TokenizersBackend",
|
| 30 |
+
"unk_token": "<unk>",
|
| 31 |
+
"crumb_metadata": {
|
| 32 |
+
"name": "CRUMB Python BPE \u2014 TOK-1 v7.4.0",
|
| 33 |
+
"script_version": "v7.4.0",
|
| 34 |
+
"vocab_size": 32768,
|
| 35 |
+
"domain": "python_source_code",
|
| 36 |
+
"corpus": "stack-dedup-python-parquet-27GB",
|
| 37 |
+
"min_frequency": 2,
|
| 38 |
+
"trainer": "ByteLevelBPETokenizer via train_from_iterator (Python str)",
|
| 39 |
+
"protected_keywords": 38,
|
| 40 |
+
"protected_dunders": 21,
|
| 41 |
+
"protected_builtins": 73,
|
| 42 |
+
"protected_exceptions": 69,
|
| 43 |
+
"protected_stdlib": 125,
|
| 44 |
+
"protected_ds_tokens": 119,
|
| 45 |
+
"total_never_split": 474,
|
| 46 |
+
"special_tokens": 20,
|
| 47 |
+
"reserved_future_tokens": 4,
|
| 48 |
+
"target_model_sizes": [
|
| 49 |
+
"124M",
|
| 50 |
+
"250M",
|
| 51 |
+
"370M",
|
| 52 |
+
"500M"
|
| 53 |
+
],
|
| 54 |
+
"fim_tokens": [
|
| 55 |
+
"<fim_prefix>",
|
| 56 |
+
"<fim_suffix>",
|
| 57 |
+
"<fim_middle>"
|
| 58 |
+
],
|
| 59 |
+
"notebook_tokens": [
|
| 60 |
+
"<cell>",
|
| 61 |
+
"</cell>",
|
| 62 |
+
"<cell_output>",
|
| 63 |
+
"</cell_output>"
|
| 64 |
+
],
|
| 65 |
+
"comments_in_training": true,
|
| 66 |
+
"docstrings_in_training": true,
|
| 67 |
+
"training_date": "2026-05-11"
|
| 68 |
+
}
|
| 69 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|