QT_V.2_96K / training_report.json
JamesQuartz's picture
Upload folder using huggingface_hub
bbf1368 verified
{
"vocab_size": 96000,
"avg_token_length_chars": 6.1,
"max_token_length_chars": 15,
"single_byte_tokens": 256,
"special_tokens": [
{
"token": "<pad>",
"id": 0
},
{
"token": "<|bos|>",
"id": 1
},
{
"token": "<|endoftext|>",
"id": 2
},
{
"token": "<",
"id": 30
},
{
"token": "[",
"id": 61
},
{
"token": "[User",
"id": 1157
},
{
"token": "[System",
"id": 1217
},
{
"token": "[]",
"id": 3325
},
{
"token": "</",
"id": 4477
},
{
"token": "['",
"id": 4641
},
{
"token": "[\"",
"id": 5119
},
{
"token": "[i",
"id": 5453
},
{
"token": "<T",
"id": 7073
},
{
"token": "[Oracle",
"id": 9667
},
{
"token": "[:",
"id": 11907
},
{
"token": "<String",
"id": 13683
},
{
"token": "<string",
"id": 14770
},
{
"token": "[string",
"id": 15770
},
{
"token": "[str",
"id": 16887
},
{
"token": "<int",
"id": 18424
}
],
"compression_ratio_chars_per_token": 3.174,
"fertility_tokens_per_word": 4.711,
"training_files": 1085,
"training_time_seconds": 647.3,
"config": {
"preset": "llama3",
"vocab_size": 96000,
"min_frequency": 2,
"bpe_dropout": null,
"single_digit": true,
"max_token_length": 16,
"special_tokens": [
"<pad>",
"<|bos|>",
"<|endoftext|>"
],
"sample_ratio": 0.15
},
"per_language_evaluation": {
"ar": {
"chars": 135,
"bytes": 250,
"tokens": 40,
"words": 20,
"chars_per_token": 3.375,
"bytes_per_token": 6.25,
"tokens_per_word": 2.0,
"byte_fragmentation_ratio": 0.1
},
"arithmetic": {
"chars": 88,
"bytes": 88,
"tokens": 78,
"words": 18,
"chars_per_token": 1.128,
"bytes_per_token": 1.128,
"tokens_per_word": 4.333,
"byte_fragmentation_ratio": 0.987
},
"bn": {
"chars": 129,
"bytes": 347,
"tokens": 68,
"words": 21,
"chars_per_token": 1.897,
"bytes_per_token": 5.103,
"tokens_per_word": 3.238,
"byte_fragmentation_ratio": 0.0
},
"de": {
"chars": 177,
"bytes": 180,
"tokens": 30,
"words": 23,
"chars_per_token": 5.9,
"bytes_per_token": 6.0,
"tokens_per_word": 1.304,
"byte_fragmentation_ratio": 0.1
},
"el": {
"chars": 158,
"bytes": 289,
"tokens": 56,
"words": 26,
"chars_per_token": 2.821,
"bytes_per_token": 5.161,
"tokens_per_word": 2.154,
"byte_fragmentation_ratio": 0.179
},
"en": {
"chars": 224,
"bytes": 224,
"tokens": 41,
"words": 30,
"chars_per_token": 5.463,
"bytes_per_token": 5.463,
"tokens_per_word": 1.367,
"byte_fragmentation_ratio": 0.171
},
"es": {
"chars": 162,
"bytes": 168,
"tokens": 37,
"words": 26,
"chars_per_token": 4.378,
"bytes_per_token": 4.541,
"tokens_per_word": 1.423,
"byte_fragmentation_ratio": 0.243
},
"fa": {
"chars": 153,
"bytes": 281,
"tokens": 45,
"words": 28,
"chars_per_token": 3.4,
"bytes_per_token": 6.244,
"tokens_per_word": 1.607,
"byte_fragmentation_ratio": 0.089
},
"fr": {
"chars": 175,
"bytes": 183,
"tokens": 40,
"words": 26,
"chars_per_token": 4.375,
"bytes_per_token": 4.575,
"tokens_per_word": 1.538,
"byte_fragmentation_ratio": 0.35
},
"he": {
"chars": 136,
"bytes": 249,
"tokens": 48,
"words": 22,
"chars_per_token": 2.833,
"bytes_per_token": 5.188,
"tokens_per_word": 2.182,
"byte_fragmentation_ratio": 0.229
},
"hi": {
"chars": 137,
"bytes": 353,
"tokens": 67,
"words": 30,
"chars_per_token": 2.045,
"bytes_per_token": 5.269,
"tokens_per_word": 2.233,
"byte_fragmentation_ratio": 0.0
},
"hy": {
"chars": 140,
"bytes": 257,
"tokens": 47,
"words": 20,
"chars_per_token": 2.979,
"bytes_per_token": 5.468,
"tokens_per_word": 2.35,
"byte_fragmentation_ratio": 0.234
},
"ja": {
"chars": 70,
"bytes": 210,
"tokens": 44,
"words": 1,
"chars_per_token": 1.591,
"bytes_per_token": 4.773,
"tokens_per_word": 44.0,
"byte_fragmentation_ratio": 0.0
},
"javascript": {
"chars": 212,
"bytes": 212,
"tokens": 53,
"words": 31,
"chars_per_token": 4.0,
"bytes_per_token": 4.0,
"tokens_per_word": 1.71,
"byte_fragmentation_ratio": 0.245
},
"ka": {
"chars": 126,
"bytes": 346,
"tokens": 47,
"words": 14,
"chars_per_token": 2.681,
"bytes_per_token": 7.362,
"tokens_per_word": 3.357,
"byte_fragmentation_ratio": 0.085
},
"ko": {
"chars": 73,
"bytes": 185,
"tokens": 37,
"words": 16,
"chars_per_token": 1.973,
"bytes_per_token": 5.0,
"tokens_per_word": 2.312,
"byte_fragmentation_ratio": 0.054
},
"math": {
"chars": 152,
"bytes": 172,
"tokens": 85,
"words": 28,
"chars_per_token": 1.788,
"bytes_per_token": 2.024,
"tokens_per_word": 3.036,
"byte_fragmentation_ratio": 0.776
},
"python": {
"chars": 202,
"bytes": 202,
"tokens": 61,
"words": 32,
"chars_per_token": 3.311,
"bytes_per_token": 3.311,
"tokens_per_word": 1.906,
"byte_fragmentation_ratio": 0.492
},
"ru": {
"chars": 190,
"bytes": 357,
"tokens": 46,
"words": 23,
"chars_per_token": 4.13,
"bytes_per_token": 7.761,
"tokens_per_word": 2.0,
"byte_fragmentation_ratio": 0.043
},
"th": {
"chars": 123,
"bytes": 367,
"tokens": 53,
"words": 2,
"chars_per_token": 2.321,
"bytes_per_token": 6.925,
"tokens_per_word": 26.5,
"byte_fragmentation_ratio": 0.0
},
"uk": {
"chars": 157,
"bytes": 291,
"tokens": 42,
"words": 22,
"chars_per_token": 3.738,
"bytes_per_token": 6.929,
"tokens_per_word": 1.909,
"byte_fragmentation_ratio": 0.095
},
"vi": {
"chars": 143,
"bytes": 190,
"tokens": 42,
"words": 34,
"chars_per_token": 3.405,
"bytes_per_token": 4.524,
"tokens_per_word": 1.235,
"byte_fragmentation_ratio": 0.167
},
"zh": {
"chars": 62,
"bytes": 186,
"tokens": 35,
"words": 1,
"chars_per_token": 1.771,
"bytes_per_token": 5.314,
"tokens_per_word": 35.0,
"byte_fragmentation_ratio": 0.0
}
}
}