climbmix-code-v1: 32K BPE fitted on the ClimbMix plus code recipe
Browse files- config.json +16 -0
- evaluation.json +81 -0
- freeze-manifest.json +12 -0
- sample.json +30 -0
- tokenizer.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format": "huggingface-tokenizers-json",
|
| 3 |
+
"vocab_size": 32768,
|
| 4 |
+
"special_tokens": {
|
| 5 |
+
"<|pad|>": 0,
|
| 6 |
+
"<|bos|>": 1,
|
| 7 |
+
"<|eos|>": 2,
|
| 8 |
+
"<|unk|>": 3
|
| 9 |
+
},
|
| 10 |
+
"normalization": null,
|
| 11 |
+
"pretokenization_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?\\p{L}+|\\p{N}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
|
| 12 |
+
"add_prefix_space": false,
|
| 13 |
+
"automatic_bos_eos": false,
|
| 14 |
+
"padding": null,
|
| 15 |
+
"truncation": null
|
| 16 |
+
}
|
evaluation.json
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_sha256": "71e377e1008755b83876ebab70f509df46974387d270ca218ceb7a7263ade543",
|
| 3 |
+
"tokenizers": {
|
| 4 |
+
"climbmix-code-v1": {
|
| 5 |
+
"tokenizer_sha256": "48aa6fb6d01d04067570637ca8a5ecced608f5df5585f9ce44a12c05c5281ae5",
|
| 6 |
+
"sources": {
|
| 7 |
+
"climbmix": {
|
| 8 |
+
"bytes": 14743727,
|
| 9 |
+
"tokens": 3217386,
|
| 10 |
+
"roundtrip_failures": 0,
|
| 11 |
+
"bytes_per_token": 4.582517298204194
|
| 12 |
+
},
|
| 13 |
+
"code_curated": {
|
| 14 |
+
"bytes": 3682462,
|
| 15 |
+
"tokens": 1023424,
|
| 16 |
+
"roundtrip_failures": 0,
|
| 17 |
+
"bytes_per_token": 3.598178272153086
|
| 18 |
+
},
|
| 19 |
+
"code_mit": {
|
| 20 |
+
"bytes": 1733957,
|
| 21 |
+
"tokens": 503413,
|
| 22 |
+
"roundtrip_failures": 0,
|
| 23 |
+
"bytes_per_token": 3.4444025084771352
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"digits": [
|
| 27 |
+
"1",
|
| 28 |
+
"2",
|
| 29 |
+
"3",
|
| 30 |
+
"4",
|
| 31 |
+
"5",
|
| 32 |
+
"\u0120+",
|
| 33 |
+
"\u0120",
|
| 34 |
+
"6",
|
| 35 |
+
"7",
|
| 36 |
+
"8",
|
| 37 |
+
"\u0120=",
|
| 38 |
+
"\u0120",
|
| 39 |
+
"1",
|
| 40 |
+
"3",
|
| 41 |
+
"0",
|
| 42 |
+
"2",
|
| 43 |
+
"3"
|
| 44 |
+
]
|
| 45 |
+
},
|
| 46 |
+
"structured-code-v1": {
|
| 47 |
+
"tokenizer_sha256": "add5b2c7542e4efded329735825ecdcb94df1432617c3d3e4b7508aa459432f9",
|
| 48 |
+
"sources": {
|
| 49 |
+
"climbmix": {
|
| 50 |
+
"bytes": 14743727,
|
| 51 |
+
"tokens": 4022037,
|
| 52 |
+
"roundtrip_failures": 0,
|
| 53 |
+
"bytes_per_token": 3.6657362923314727
|
| 54 |
+
},
|
| 55 |
+
"code_curated": {
|
| 56 |
+
"bytes": 3682462,
|
| 57 |
+
"tokens": 1046393,
|
| 58 |
+
"roundtrip_failures": 0,
|
| 59 |
+
"bytes_per_token": 3.519195942633408
|
| 60 |
+
},
|
| 61 |
+
"code_mit": {
|
| 62 |
+
"bytes": 1733957,
|
| 63 |
+
"tokens": 514537,
|
| 64 |
+
"roundtrip_failures": 0,
|
| 65 |
+
"bytes_per_token": 3.3699364671539658
|
| 66 |
+
}
|
| 67 |
+
},
|
| 68 |
+
"digits": [
|
| 69 |
+
"123",
|
| 70 |
+
"45",
|
| 71 |
+
"\u0120+",
|
| 72 |
+
"\u0120",
|
| 73 |
+
"678",
|
| 74 |
+
"\u0120=",
|
| 75 |
+
"\u0120",
|
| 76 |
+
"130",
|
| 77 |
+
"23"
|
| 78 |
+
]
|
| 79 |
+
}
|
| 80 |
+
}
|
| 81 |
+
}
|
freeze-manifest.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"id": "climbmix-code-bpe-v1",
|
| 3 |
+
"status": "frozen",
|
| 4 |
+
"vocab_size": 32768,
|
| 5 |
+
"files": {
|
| 6 |
+
"config.json": "ef1f7e15eb8451f989ed2b380a09ce84aa3960ea03d2e472aa1f25a12cdb9a18",
|
| 7 |
+
"sample.json": "cf6e2241aba3ea6f3c779e54087cb01ffa3394cf5e5562807c244561d3de5cf9",
|
| 8 |
+
"tokenizer.json": "48aa6fb6d01d04067570637ca8a5ecced608f5df5585f9ce44a12c05c5281ae5",
|
| 9 |
+
"evaluation.json": "57a207587ae46a9a4293629a52f5653b6de980353f320b639c52e926c1c6ddc8"
|
| 10 |
+
},
|
| 11 |
+
"contract": "Immutable. Record tokenizer SHA256 in shards and checkpoints. Prior tokenizer data must be retokenized in a new version."
|
| 12 |
+
}
|
sample.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"recipe_sha256": "320b4e5219aba8a78dd675847041653c55eb96f74ef57abdc54d472f2c90ea5f",
|
| 3 |
+
"train_bytes_target": 268435456,
|
| 4 |
+
"eval_bytes_target": 16777216,
|
| 5 |
+
"files_per_source": 64,
|
| 6 |
+
"sources": {
|
| 7 |
+
"climbmix": {
|
| 8 |
+
"train": 228529615,
|
| 9 |
+
"validation": 14743727,
|
| 10 |
+
"documents": 81269,
|
| 11 |
+
"files": 64
|
| 12 |
+
},
|
| 13 |
+
"code_curated": {
|
| 14 |
+
"train": 31454416,
|
| 15 |
+
"validation": 3682462,
|
| 16 |
+
"documents": 5130,
|
| 17 |
+
"files": 64
|
| 18 |
+
},
|
| 19 |
+
"code_mit": {
|
| 20 |
+
"train": 13964697,
|
| 21 |
+
"validation": 1733957,
|
| 22 |
+
"documents": 1586,
|
| 23 |
+
"files": 54
|
| 24 |
+
}
|
| 25 |
+
},
|
| 26 |
+
"train_sha256": "f4b611bc901131b6dd4a8824b30024fb7e38c3f56be6a83a1fed482549f4f1df",
|
| 27 |
+
"eval_sha256": "71e377e1008755b83876ebab70f509df46974387d270ca218ceb7a7263ade543",
|
| 28 |
+
"recipe_id": "climbmix-code-v1",
|
| 29 |
+
"fit_seconds": 31.667878392000034
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|