Add deep benchmark results (30 test cases, 4 tokenizer comparison)
Browse files- deep_benchmark_results.json +1124 -0
deep_benchmark_results.json
ADDED
|
@@ -0,0 +1,1124 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"per_sample": {
|
| 3 |
+
"Sentinel-SUT (61K)": {
|
| 4 |
+
"English": {
|
| 5 |
+
"tokens": 55,
|
| 6 |
+
"bytes": 251,
|
| 7 |
+
"words": 27,
|
| 8 |
+
"fertility": 2.037037037037037,
|
| 9 |
+
"compression": 4.5636363636363635,
|
| 10 |
+
"roundtrip": true
|
| 11 |
+
},
|
| 12 |
+
"French": {
|
| 13 |
+
"tokens": 58,
|
| 14 |
+
"bytes": 221,
|
| 15 |
+
"words": 29,
|
| 16 |
+
"fertility": 2.0,
|
| 17 |
+
"compression": 3.810344827586207,
|
| 18 |
+
"roundtrip": true
|
| 19 |
+
},
|
| 20 |
+
"German": {
|
| 21 |
+
"tokens": 56,
|
| 22 |
+
"bytes": 204,
|
| 23 |
+
"words": 17,
|
| 24 |
+
"fertility": 3.2941176470588234,
|
| 25 |
+
"compression": 3.642857142857143,
|
| 26 |
+
"roundtrip": true
|
| 27 |
+
},
|
| 28 |
+
"Spanish": {
|
| 29 |
+
"tokens": 49,
|
| 30 |
+
"bytes": 209,
|
| 31 |
+
"words": 24,
|
| 32 |
+
"fertility": 2.0416666666666665,
|
| 33 |
+
"compression": 4.26530612244898,
|
| 34 |
+
"roundtrip": true
|
| 35 |
+
},
|
| 36 |
+
"Portuguese": {
|
| 37 |
+
"tokens": 37,
|
| 38 |
+
"bytes": 145,
|
| 39 |
+
"words": 19,
|
| 40 |
+
"fertility": 1.9473684210526316,
|
| 41 |
+
"compression": 3.918918918918919,
|
| 42 |
+
"roundtrip": true
|
| 43 |
+
},
|
| 44 |
+
"Italian": {
|
| 45 |
+
"tokens": 41,
|
| 46 |
+
"bytes": 146,
|
| 47 |
+
"words": 17,
|
| 48 |
+
"fertility": 2.411764705882353,
|
| 49 |
+
"compression": 3.5609756097560976,
|
| 50 |
+
"roundtrip": true
|
| 51 |
+
},
|
| 52 |
+
"Dutch": {
|
| 53 |
+
"tokens": 41,
|
| 54 |
+
"bytes": 123,
|
| 55 |
+
"words": 12,
|
| 56 |
+
"fertility": 3.4166666666666665,
|
| 57 |
+
"compression": 3.0,
|
| 58 |
+
"roundtrip": true
|
| 59 |
+
},
|
| 60 |
+
"Polish": {
|
| 61 |
+
"tokens": 41,
|
| 62 |
+
"bytes": 122,
|
| 63 |
+
"words": 13,
|
| 64 |
+
"fertility": 3.1538461538461537,
|
| 65 |
+
"compression": 2.975609756097561,
|
| 66 |
+
"roundtrip": true
|
| 67 |
+
},
|
| 68 |
+
"Swedish": {
|
| 69 |
+
"tokens": 34,
|
| 70 |
+
"bytes": 109,
|
| 71 |
+
"words": 7,
|
| 72 |
+
"fertility": 4.857142857142857,
|
| 73 |
+
"compression": 3.2058823529411766,
|
| 74 |
+
"roundtrip": true
|
| 75 |
+
},
|
| 76 |
+
"Turkish": {
|
| 77 |
+
"tokens": 42,
|
| 78 |
+
"bytes": 134,
|
| 79 |
+
"words": 13,
|
| 80 |
+
"fertility": 3.230769230769231,
|
| 81 |
+
"compression": 3.1904761904761907,
|
| 82 |
+
"roundtrip": true
|
| 83 |
+
},
|
| 84 |
+
"Ukrainian": {
|
| 85 |
+
"tokens": 44,
|
| 86 |
+
"bytes": 225,
|
| 87 |
+
"words": 13,
|
| 88 |
+
"fertility": 3.3846153846153846,
|
| 89 |
+
"compression": 5.113636363636363,
|
| 90 |
+
"roundtrip": true
|
| 91 |
+
},
|
| 92 |
+
"Chinese": {
|
| 93 |
+
"tokens": 48,
|
| 94 |
+
"bytes": 173,
|
| 95 |
+
"words": 1,
|
| 96 |
+
"fertility": 48.0,
|
| 97 |
+
"compression": 3.6041666666666665,
|
| 98 |
+
"roundtrip": true
|
| 99 |
+
},
|
| 100 |
+
"Japanese": {
|
| 101 |
+
"tokens": 74,
|
| 102 |
+
"bytes": 296,
|
| 103 |
+
"words": 1,
|
| 104 |
+
"fertility": 74.0,
|
| 105 |
+
"compression": 4.0,
|
| 106 |
+
"roundtrip": true
|
| 107 |
+
},
|
| 108 |
+
"Korean": {
|
| 109 |
+
"tokens": 62,
|
| 110 |
+
"bytes": 285,
|
| 111 |
+
"words": 24,
|
| 112 |
+
"fertility": 2.5833333333333335,
|
| 113 |
+
"compression": 4.596774193548387,
|
| 114 |
+
"roundtrip": true
|
| 115 |
+
},
|
| 116 |
+
"Vietnamese": {
|
| 117 |
+
"tokens": 31,
|
| 118 |
+
"bytes": 144,
|
| 119 |
+
"words": 25,
|
| 120 |
+
"fertility": 1.24,
|
| 121 |
+
"compression": 4.645161290322581,
|
| 122 |
+
"roundtrip": true
|
| 123 |
+
},
|
| 124 |
+
"Thai": {
|
| 125 |
+
"tokens": 71,
|
| 126 |
+
"bytes": 339,
|
| 127 |
+
"words": 1,
|
| 128 |
+
"fertility": 71.0,
|
| 129 |
+
"compression": 4.774647887323944,
|
| 130 |
+
"roundtrip": false
|
| 131 |
+
},
|
| 132 |
+
"Hindi": {
|
| 133 |
+
"tokens": 91,
|
| 134 |
+
"bytes": 325,
|
| 135 |
+
"words": 16,
|
| 136 |
+
"fertility": 5.6875,
|
| 137 |
+
"compression": 3.5714285714285716,
|
| 138 |
+
"roundtrip": true
|
| 139 |
+
},
|
| 140 |
+
"Arabic": {
|
| 141 |
+
"tokens": 55,
|
| 142 |
+
"bytes": 265,
|
| 143 |
+
"words": 22,
|
| 144 |
+
"fertility": 2.5,
|
| 145 |
+
"compression": 4.818181818181818,
|
| 146 |
+
"roundtrip": true
|
| 147 |
+
},
|
| 148 |
+
"Russian": {
|
| 149 |
+
"tokens": 72,
|
| 150 |
+
"bytes": 390,
|
| 151 |
+
"words": 19,
|
| 152 |
+
"fertility": 3.789473684210526,
|
| 153 |
+
"compression": 5.416666666666667,
|
| 154 |
+
"roundtrip": true
|
| 155 |
+
},
|
| 156 |
+
"Python": {
|
| 157 |
+
"tokens": 115,
|
| 158 |
+
"bytes": 291,
|
| 159 |
+
"words": 39,
|
| 160 |
+
"fertility": 2.948717948717949,
|
| 161 |
+
"compression": 2.5304347826086957,
|
| 162 |
+
"roundtrip": true
|
| 163 |
+
},
|
| 164 |
+
"JavaScript": {
|
| 165 |
+
"tokens": 121,
|
| 166 |
+
"bytes": 314,
|
| 167 |
+
"words": 40,
|
| 168 |
+
"fertility": 3.025,
|
| 169 |
+
"compression": 2.5950413223140494,
|
| 170 |
+
"roundtrip": true
|
| 171 |
+
},
|
| 172 |
+
"Rust": {
|
| 173 |
+
"tokens": 136,
|
| 174 |
+
"bytes": 289,
|
| 175 |
+
"words": 44,
|
| 176 |
+
"fertility": 3.090909090909091,
|
| 177 |
+
"compression": 2.125,
|
| 178 |
+
"roundtrip": true
|
| 179 |
+
},
|
| 180 |
+
"LaTeX_Complex": {
|
| 181 |
+
"tokens": 143,
|
| 182 |
+
"bytes": 248,
|
| 183 |
+
"words": 26,
|
| 184 |
+
"fertility": 5.5,
|
| 185 |
+
"compression": 1.7342657342657342,
|
| 186 |
+
"roundtrip": true
|
| 187 |
+
},
|
| 188 |
+
"Unicode_Math": {
|
| 189 |
+
"tokens": 58,
|
| 190 |
+
"bytes": 115,
|
| 191 |
+
"words": 17,
|
| 192 |
+
"fertility": 3.411764705882353,
|
| 193 |
+
"compression": 1.9827586206896552,
|
| 194 |
+
"roundtrip": false
|
| 195 |
+
},
|
| 196 |
+
"Mixed_Notation": {
|
| 197 |
+
"tokens": 69,
|
| 198 |
+
"bytes": 138,
|
| 199 |
+
"words": 20,
|
| 200 |
+
"fertility": 3.45,
|
| 201 |
+
"compression": 2.0,
|
| 202 |
+
"roundtrip": false
|
| 203 |
+
},
|
| 204 |
+
"Emoji_Heavy": {
|
| 205 |
+
"tokens": 68,
|
| 206 |
+
"bytes": 155,
|
| 207 |
+
"words": 17,
|
| 208 |
+
"fertility": 4.0,
|
| 209 |
+
"compression": 2.2794117647058822,
|
| 210 |
+
"roundtrip": true
|
| 211 |
+
},
|
| 212 |
+
"Numbers_Heavy": {
|
| 213 |
+
"tokens": 67,
|
| 214 |
+
"bytes": 130,
|
| 215 |
+
"words": 15,
|
| 216 |
+
"fertility": 4.466666666666667,
|
| 217 |
+
"compression": 1.9402985074626866,
|
| 218 |
+
"roundtrip": false
|
| 219 |
+
},
|
| 220 |
+
"URL_Path": {
|
| 221 |
+
"tokens": 46,
|
| 222 |
+
"bytes": 113,
|
| 223 |
+
"words": 1,
|
| 224 |
+
"fertility": 46.0,
|
| 225 |
+
"compression": 2.4565217391304346,
|
| 226 |
+
"roundtrip": true
|
| 227 |
+
},
|
| 228 |
+
"Mixed_Script": {
|
| 229 |
+
"tokens": 44,
|
| 230 |
+
"bytes": 122,
|
| 231 |
+
"words": 17,
|
| 232 |
+
"fertility": 2.588235294117647,
|
| 233 |
+
"compression": 2.772727272727273,
|
| 234 |
+
"roundtrip": true
|
| 235 |
+
},
|
| 236 |
+
"Repetition": {
|
| 237 |
+
"tokens": 17,
|
| 238 |
+
"bytes": 109,
|
| 239 |
+
"words": 14,
|
| 240 |
+
"fertility": 1.2142857142857142,
|
| 241 |
+
"compression": 6.411764705882353,
|
| 242 |
+
"roundtrip": true
|
| 243 |
+
},
|
| 244 |
+
"Whitespace": {
|
| 245 |
+
"tokens": 22,
|
| 246 |
+
"bytes": 54,
|
| 247 |
+
"words": 6,
|
| 248 |
+
"fertility": 3.6666666666666665,
|
| 249 |
+
"compression": 2.4545454545454546,
|
| 250 |
+
"roundtrip": true
|
| 251 |
+
},
|
| 252 |
+
"Empty_Adjacent": {
|
| 253 |
+
"tokens": 14,
|
| 254 |
+
"bytes": 39,
|
| 255 |
+
"words": 5,
|
| 256 |
+
"fertility": 2.8,
|
| 257 |
+
"compression": 2.7857142857142856,
|
| 258 |
+
"roundtrip": true
|
| 259 |
+
}
|
| 260 |
+
},
|
| 261 |
+
"GPT-2 (50K)": {
|
| 262 |
+
"English": {
|
| 263 |
+
"tokens": 34,
|
| 264 |
+
"bytes": 251,
|
| 265 |
+
"words": 27,
|
| 266 |
+
"fertility": 1.2592592592592593,
|
| 267 |
+
"compression": 7.382352941176471,
|
| 268 |
+
"roundtrip": true
|
| 269 |
+
},
|
| 270 |
+
"French": {
|
| 271 |
+
"tokens": 60,
|
| 272 |
+
"bytes": 221,
|
| 273 |
+
"words": 29,
|
| 274 |
+
"fertility": 2.0689655172413794,
|
| 275 |
+
"compression": 3.683333333333333,
|
| 276 |
+
"roundtrip": true
|
| 277 |
+
},
|
| 278 |
+
"German": {
|
| 279 |
+
"tokens": 66,
|
| 280 |
+
"bytes": 204,
|
| 281 |
+
"words": 17,
|
| 282 |
+
"fertility": 3.8823529411764706,
|
| 283 |
+
"compression": 3.090909090909091,
|
| 284 |
+
"roundtrip": true
|
| 285 |
+
},
|
| 286 |
+
"Spanish": {
|
| 287 |
+
"tokens": 66,
|
| 288 |
+
"bytes": 209,
|
| 289 |
+
"words": 24,
|
| 290 |
+
"fertility": 2.75,
|
| 291 |
+
"compression": 3.1666666666666665,
|
| 292 |
+
"roundtrip": true
|
| 293 |
+
},
|
| 294 |
+
"Portuguese": {
|
| 295 |
+
"tokens": 52,
|
| 296 |
+
"bytes": 145,
|
| 297 |
+
"words": 19,
|
| 298 |
+
"fertility": 2.736842105263158,
|
| 299 |
+
"compression": 2.7884615384615383,
|
| 300 |
+
"roundtrip": true
|
| 301 |
+
},
|
| 302 |
+
"Italian": {
|
| 303 |
+
"tokens": 51,
|
| 304 |
+
"bytes": 146,
|
| 305 |
+
"words": 17,
|
| 306 |
+
"fertility": 3.0,
|
| 307 |
+
"compression": 2.8627450980392157,
|
| 308 |
+
"roundtrip": true
|
| 309 |
+
},
|
| 310 |
+
"Dutch": {
|
| 311 |
+
"tokens": 36,
|
| 312 |
+
"bytes": 123,
|
| 313 |
+
"words": 12,
|
| 314 |
+
"fertility": 3.0,
|
| 315 |
+
"compression": 3.4166666666666665,
|
| 316 |
+
"roundtrip": true
|
| 317 |
+
},
|
| 318 |
+
"Polish": {
|
| 319 |
+
"tokens": 53,
|
| 320 |
+
"bytes": 122,
|
| 321 |
+
"words": 13,
|
| 322 |
+
"fertility": 4.076923076923077,
|
| 323 |
+
"compression": 2.30188679245283,
|
| 324 |
+
"roundtrip": true
|
| 325 |
+
},
|
| 326 |
+
"Swedish": {
|
| 327 |
+
"tokens": 36,
|
| 328 |
+
"bytes": 109,
|
| 329 |
+
"words": 7,
|
| 330 |
+
"fertility": 5.142857142857143,
|
| 331 |
+
"compression": 3.0277777777777777,
|
| 332 |
+
"roundtrip": true
|
| 333 |
+
},
|
| 334 |
+
"Turkish": {
|
| 335 |
+
"tokens": 55,
|
| 336 |
+
"bytes": 134,
|
| 337 |
+
"words": 13,
|
| 338 |
+
"fertility": 4.230769230769231,
|
| 339 |
+
"compression": 2.4363636363636365,
|
| 340 |
+
"roundtrip": true
|
| 341 |
+
},
|
| 342 |
+
"Ukrainian": {
|
| 343 |
+
"tokens": 142,
|
| 344 |
+
"bytes": 225,
|
| 345 |
+
"words": 13,
|
| 346 |
+
"fertility": 10.923076923076923,
|
| 347 |
+
"compression": 1.5845070422535212,
|
| 348 |
+
"roundtrip": true
|
| 349 |
+
},
|
| 350 |
+
"Chinese": {
|
| 351 |
+
"tokens": 117,
|
| 352 |
+
"bytes": 173,
|
| 353 |
+
"words": 1,
|
| 354 |
+
"fertility": 117.0,
|
| 355 |
+
"compression": 1.4786324786324787,
|
| 356 |
+
"roundtrip": true
|
| 357 |
+
},
|
| 358 |
+
"Japanese": {
|
| 359 |
+
"tokens": 150,
|
| 360 |
+
"bytes": 296,
|
| 361 |
+
"words": 1,
|
| 362 |
+
"fertility": 150.0,
|
| 363 |
+
"compression": 1.9733333333333334,
|
| 364 |
+
"roundtrip": true
|
| 365 |
+
},
|
| 366 |
+
"Korean": {
|
| 367 |
+
"tokens": 242,
|
| 368 |
+
"bytes": 285,
|
| 369 |
+
"words": 24,
|
| 370 |
+
"fertility": 10.083333333333334,
|
| 371 |
+
"compression": 1.177685950413223,
|
| 372 |
+
"roundtrip": true
|
| 373 |
+
},
|
| 374 |
+
"Vietnamese": {
|
| 375 |
+
"tokens": 95,
|
| 376 |
+
"bytes": 144,
|
| 377 |
+
"words": 25,
|
| 378 |
+
"fertility": 3.8,
|
| 379 |
+
"compression": 1.5157894736842106,
|
| 380 |
+
"roundtrip": true
|
| 381 |
+
},
|
| 382 |
+
"Thai": {
|
| 383 |
+
"tokens": 226,
|
| 384 |
+
"bytes": 339,
|
| 385 |
+
"words": 1,
|
| 386 |
+
"fertility": 226.0,
|
| 387 |
+
"compression": 1.5,
|
| 388 |
+
"roundtrip": true
|
| 389 |
+
},
|
| 390 |
+
"Hindi": {
|
| 391 |
+
"tokens": 203,
|
| 392 |
+
"bytes": 325,
|
| 393 |
+
"words": 16,
|
| 394 |
+
"fertility": 12.6875,
|
| 395 |
+
"compression": 1.6009852216748768,
|
| 396 |
+
"roundtrip": true
|
| 397 |
+
},
|
| 398 |
+
"Arabic": {
|
| 399 |
+
"tokens": 142,
|
| 400 |
+
"bytes": 265,
|
| 401 |
+
"words": 22,
|
| 402 |
+
"fertility": 6.454545454545454,
|
| 403 |
+
"compression": 1.8661971830985915,
|
| 404 |
+
"roundtrip": true
|
| 405 |
+
},
|
| 406 |
+
"Russian": {
|
| 407 |
+
"tokens": 228,
|
| 408 |
+
"bytes": 390,
|
| 409 |
+
"words": 19,
|
| 410 |
+
"fertility": 12.0,
|
| 411 |
+
"compression": 1.7105263157894737,
|
| 412 |
+
"roundtrip": true
|
| 413 |
+
},
|
| 414 |
+
"Python": {
|
| 415 |
+
"tokens": 135,
|
| 416 |
+
"bytes": 291,
|
| 417 |
+
"words": 39,
|
| 418 |
+
"fertility": 3.4615384615384617,
|
| 419 |
+
"compression": 2.1555555555555554,
|
| 420 |
+
"roundtrip": true
|
| 421 |
+
},
|
| 422 |
+
"JavaScript": {
|
| 423 |
+
"tokens": 118,
|
| 424 |
+
"bytes": 314,
|
| 425 |
+
"words": 40,
|
| 426 |
+
"fertility": 2.95,
|
| 427 |
+
"compression": 2.6610169491525424,
|
| 428 |
+
"roundtrip": true
|
| 429 |
+
},
|
| 430 |
+
"Rust": {
|
| 431 |
+
"tokens": 144,
|
| 432 |
+
"bytes": 289,
|
| 433 |
+
"words": 44,
|
| 434 |
+
"fertility": 3.272727272727273,
|
| 435 |
+
"compression": 2.0069444444444446,
|
| 436 |
+
"roundtrip": true
|
| 437 |
+
},
|
| 438 |
+
"LaTeX_Complex": {
|
| 439 |
+
"tokens": 130,
|
| 440 |
+
"bytes": 248,
|
| 441 |
+
"words": 26,
|
| 442 |
+
"fertility": 5.0,
|
| 443 |
+
"compression": 1.9076923076923078,
|
| 444 |
+
"roundtrip": true
|
| 445 |
+
},
|
| 446 |
+
"Unicode_Math": {
|
| 447 |
+
"tokens": 74,
|
| 448 |
+
"bytes": 115,
|
| 449 |
+
"words": 17,
|
| 450 |
+
"fertility": 4.352941176470588,
|
| 451 |
+
"compression": 1.554054054054054,
|
| 452 |
+
"roundtrip": true
|
| 453 |
+
},
|
| 454 |
+
"Mixed_Notation": {
|
| 455 |
+
"tokens": 78,
|
| 456 |
+
"bytes": 138,
|
| 457 |
+
"words": 20,
|
| 458 |
+
"fertility": 3.9,
|
| 459 |
+
"compression": 1.7692307692307692,
|
| 460 |
+
"roundtrip": true
|
| 461 |
+
},
|
| 462 |
+
"Emoji_Heavy": {
|
| 463 |
+
"tokens": 70,
|
| 464 |
+
"bytes": 155,
|
| 465 |
+
"words": 17,
|
| 466 |
+
"fertility": 4.117647058823529,
|
| 467 |
+
"compression": 2.2142857142857144,
|
| 468 |
+
"roundtrip": true
|
| 469 |
+
},
|
| 470 |
+
"Numbers_Heavy": {
|
| 471 |
+
"tokens": 67,
|
| 472 |
+
"bytes": 130,
|
| 473 |
+
"words": 15,
|
| 474 |
+
"fertility": 4.466666666666667,
|
| 475 |
+
"compression": 1.9402985074626866,
|
| 476 |
+
"roundtrip": true
|
| 477 |
+
},
|
| 478 |
+
"URL_Path": {
|
| 479 |
+
"tokens": 39,
|
| 480 |
+
"bytes": 113,
|
| 481 |
+
"words": 1,
|
| 482 |
+
"fertility": 39.0,
|
| 483 |
+
"compression": 2.8974358974358974,
|
| 484 |
+
"roundtrip": true
|
| 485 |
+
},
|
| 486 |
+
"Mixed_Script": {
|
| 487 |
+
"tokens": 47,
|
| 488 |
+
"bytes": 122,
|
| 489 |
+
"words": 17,
|
| 490 |
+
"fertility": 2.764705882352941,
|
| 491 |
+
"compression": 2.595744680851064,
|
| 492 |
+
"roundtrip": true
|
| 493 |
+
},
|
| 494 |
+
"Repetition": {
|
| 495 |
+
"tokens": 14,
|
| 496 |
+
"bytes": 109,
|
| 497 |
+
"words": 14,
|
| 498 |
+
"fertility": 1.0,
|
| 499 |
+
"compression": 7.785714285714286,
|
| 500 |
+
"roundtrip": true
|
| 501 |
+
},
|
| 502 |
+
"Whitespace": {
|
| 503 |
+
"tokens": 24,
|
| 504 |
+
"bytes": 54,
|
| 505 |
+
"words": 6,
|
| 506 |
+
"fertility": 4.0,
|
| 507 |
+
"compression": 2.25,
|
| 508 |
+
"roundtrip": true
|
| 509 |
+
},
|
| 510 |
+
"Empty_Adjacent": {
|
| 511 |
+
"tokens": 20,
|
| 512 |
+
"bytes": 39,
|
| 513 |
+
"words": 5,
|
| 514 |
+
"fertility": 4.0,
|
| 515 |
+
"compression": 1.95,
|
| 516 |
+
"roundtrip": true
|
| 517 |
+
}
|
| 518 |
+
},
|
| 519 |
+
"Gemma (256K)": {
|
| 520 |
+
"English": {
|
| 521 |
+
"tokens": 32,
|
| 522 |
+
"bytes": 251,
|
| 523 |
+
"words": 27,
|
| 524 |
+
"fertility": 1.1851851851851851,
|
| 525 |
+
"compression": 7.84375,
|
| 526 |
+
"roundtrip": true
|
| 527 |
+
},
|
| 528 |
+
"French": {
|
| 529 |
+
"tokens": 41,
|
| 530 |
+
"bytes": 221,
|
| 531 |
+
"words": 29,
|
| 532 |
+
"fertility": 1.4137931034482758,
|
| 533 |
+
"compression": 5.390243902439025,
|
| 534 |
+
"roundtrip": true
|
| 535 |
+
},
|
| 536 |
+
"German": {
|
| 537 |
+
"tokens": 39,
|
| 538 |
+
"bytes": 204,
|
| 539 |
+
"words": 17,
|
| 540 |
+
"fertility": 2.2941176470588234,
|
| 541 |
+
"compression": 5.230769230769231,
|
| 542 |
+
"roundtrip": true
|
| 543 |
+
},
|
| 544 |
+
"Spanish": {
|
| 545 |
+
"tokens": 33,
|
| 546 |
+
"bytes": 209,
|
| 547 |
+
"words": 24,
|
| 548 |
+
"fertility": 1.375,
|
| 549 |
+
"compression": 6.333333333333333,
|
| 550 |
+
"roundtrip": true
|
| 551 |
+
},
|
| 552 |
+
"Portuguese": {
|
| 553 |
+
"tokens": 28,
|
| 554 |
+
"bytes": 145,
|
| 555 |
+
"words": 19,
|
| 556 |
+
"fertility": 1.4736842105263157,
|
| 557 |
+
"compression": 5.178571428571429,
|
| 558 |
+
"roundtrip": true
|
| 559 |
+
},
|
| 560 |
+
"Italian": {
|
| 561 |
+
"tokens": 26,
|
| 562 |
+
"bytes": 146,
|
| 563 |
+
"words": 17,
|
| 564 |
+
"fertility": 1.5294117647058822,
|
| 565 |
+
"compression": 5.615384615384615,
|
| 566 |
+
"roundtrip": true
|
| 567 |
+
},
|
| 568 |
+
"Dutch": {
|
| 569 |
+
"tokens": 27,
|
| 570 |
+
"bytes": 123,
|
| 571 |
+
"words": 12,
|
| 572 |
+
"fertility": 2.25,
|
| 573 |
+
"compression": 4.555555555555555,
|
| 574 |
+
"roundtrip": true
|
| 575 |
+
},
|
| 576 |
+
"Polish": {
|
| 577 |
+
"tokens": 33,
|
| 578 |
+
"bytes": 122,
|
| 579 |
+
"words": 13,
|
| 580 |
+
"fertility": 2.5384615384615383,
|
| 581 |
+
"compression": 3.696969696969697,
|
| 582 |
+
"roundtrip": true
|
| 583 |
+
},
|
| 584 |
+
"Swedish": {
|
| 585 |
+
"tokens": 25,
|
| 586 |
+
"bytes": 109,
|
| 587 |
+
"words": 7,
|
| 588 |
+
"fertility": 3.5714285714285716,
|
| 589 |
+
"compression": 4.36,
|
| 590 |
+
"roundtrip": true
|
| 591 |
+
},
|
| 592 |
+
"Turkish": {
|
| 593 |
+
"tokens": 30,
|
| 594 |
+
"bytes": 134,
|
| 595 |
+
"words": 13,
|
| 596 |
+
"fertility": 2.3076923076923075,
|
| 597 |
+
"compression": 4.466666666666667,
|
| 598 |
+
"roundtrip": true
|
| 599 |
+
},
|
| 600 |
+
"Ukrainian": {
|
| 601 |
+
"tokens": 35,
|
| 602 |
+
"bytes": 225,
|
| 603 |
+
"words": 13,
|
| 604 |
+
"fertility": 2.6923076923076925,
|
| 605 |
+
"compression": 6.428571428571429,
|
| 606 |
+
"roundtrip": true
|
| 607 |
+
},
|
| 608 |
+
"Chinese": {
|
| 609 |
+
"tokens": 33,
|
| 610 |
+
"bytes": 173,
|
| 611 |
+
"words": 1,
|
| 612 |
+
"fertility": 33.0,
|
| 613 |
+
"compression": 5.242424242424242,
|
| 614 |
+
"roundtrip": true
|
| 615 |
+
},
|
| 616 |
+
"Japanese": {
|
| 617 |
+
"tokens": 55,
|
| 618 |
+
"bytes": 296,
|
| 619 |
+
"words": 1,
|
| 620 |
+
"fertility": 55.0,
|
| 621 |
+
"compression": 5.381818181818182,
|
| 622 |
+
"roundtrip": true
|
| 623 |
+
},
|
| 624 |
+
"Korean": {
|
| 625 |
+
"tokens": 74,
|
| 626 |
+
"bytes": 285,
|
| 627 |
+
"words": 24,
|
| 628 |
+
"fertility": 3.0833333333333335,
|
| 629 |
+
"compression": 3.8513513513513513,
|
| 630 |
+
"roundtrip": true
|
| 631 |
+
},
|
| 632 |
+
"Vietnamese": {
|
| 633 |
+
"tokens": 28,
|
| 634 |
+
"bytes": 144,
|
| 635 |
+
"words": 25,
|
| 636 |
+
"fertility": 1.12,
|
| 637 |
+
"compression": 5.142857142857143,
|
| 638 |
+
"roundtrip": true
|
| 639 |
+
},
|
| 640 |
+
"Thai": {
|
| 641 |
+
"tokens": 51,
|
| 642 |
+
"bytes": 339,
|
| 643 |
+
"words": 1,
|
| 644 |
+
"fertility": 51.0,
|
| 645 |
+
"compression": 6.647058823529412,
|
| 646 |
+
"roundtrip": true
|
| 647 |
+
},
|
| 648 |
+
"Hindi": {
|
| 649 |
+
"tokens": 54,
|
| 650 |
+
"bytes": 325,
|
| 651 |
+
"words": 16,
|
| 652 |
+
"fertility": 3.375,
|
| 653 |
+
"compression": 6.018518518518518,
|
| 654 |
+
"roundtrip": true
|
| 655 |
+
},
|
| 656 |
+
"Arabic": {
|
| 657 |
+
"tokens": 51,
|
| 658 |
+
"bytes": 265,
|
| 659 |
+
"words": 22,
|
| 660 |
+
"fertility": 2.3181818181818183,
|
| 661 |
+
"compression": 5.196078431372549,
|
| 662 |
+
"roundtrip": true
|
| 663 |
+
},
|
| 664 |
+
"Russian": {
|
| 665 |
+
"tokens": 43,
|
| 666 |
+
"bytes": 390,
|
| 667 |
+
"words": 19,
|
| 668 |
+
"fertility": 2.263157894736842,
|
| 669 |
+
"compression": 9.069767441860465,
|
| 670 |
+
"roundtrip": true
|
| 671 |
+
},
|
| 672 |
+
"Python": {
|
| 673 |
+
"tokens": 115,
|
| 674 |
+
"bytes": 291,
|
| 675 |
+
"words": 39,
|
| 676 |
+
"fertility": 2.948717948717949,
|
| 677 |
+
"compression": 2.5304347826086957,
|
| 678 |
+
"roundtrip": true
|
| 679 |
+
},
|
| 680 |
+
"JavaScript": {
|
| 681 |
+
"tokens": 105,
|
| 682 |
+
"bytes": 314,
|
| 683 |
+
"words": 40,
|
| 684 |
+
"fertility": 2.625,
|
| 685 |
+
"compression": 2.9904761904761905,
|
| 686 |
+
"roundtrip": true
|
| 687 |
+
},
|
| 688 |
+
"Rust": {
|
| 689 |
+
"tokens": 142,
|
| 690 |
+
"bytes": 289,
|
| 691 |
+
"words": 44,
|
| 692 |
+
"fertility": 3.227272727272727,
|
| 693 |
+
"compression": 2.035211267605634,
|
| 694 |
+
"roundtrip": true
|
| 695 |
+
},
|
| 696 |
+
"LaTeX_Complex": {
|
| 697 |
+
"tokens": 110,
|
| 698 |
+
"bytes": 248,
|
| 699 |
+
"words": 26,
|
| 700 |
+
"fertility": 4.230769230769231,
|
| 701 |
+
"compression": 2.2545454545454544,
|
| 702 |
+
"roundtrip": true
|
| 703 |
+
},
|
| 704 |
+
"Unicode_Math": {
|
| 705 |
+
"tokens": 61,
|
| 706 |
+
"bytes": 115,
|
| 707 |
+
"words": 17,
|
| 708 |
+
"fertility": 3.588235294117647,
|
| 709 |
+
"compression": 1.8852459016393444,
|
| 710 |
+
"roundtrip": true
|
| 711 |
+
},
|
| 712 |
+
"Mixed_Notation": {
|
| 713 |
+
"tokens": 61,
|
| 714 |
+
"bytes": 138,
|
| 715 |
+
"words": 20,
|
| 716 |
+
"fertility": 3.05,
|
| 717 |
+
"compression": 2.262295081967213,
|
| 718 |
+
"roundtrip": true
|
| 719 |
+
},
|
| 720 |
+
"Emoji_Heavy": {
|
| 721 |
+
"tokens": 42,
|
| 722 |
+
"bytes": 155,
|
| 723 |
+
"words": 17,
|
| 724 |
+
"fertility": 2.4705882352941178,
|
| 725 |
+
"compression": 3.6904761904761907,
|
| 726 |
+
"roundtrip": true
|
| 727 |
+
},
|
| 728 |
+
"Numbers_Heavy": {
|
| 729 |
+
"tokens": 115,
|
| 730 |
+
"bytes": 130,
|
| 731 |
+
"words": 15,
|
| 732 |
+
"fertility": 7.666666666666667,
|
| 733 |
+
"compression": 1.1304347826086956,
|
| 734 |
+
"roundtrip": true
|
| 735 |
+
},
|
| 736 |
+
"URL_Path": {
|
| 737 |
+
"tokens": 34,
|
| 738 |
+
"bytes": 113,
|
| 739 |
+
"words": 1,
|
| 740 |
+
"fertility": 34.0,
|
| 741 |
+
"compression": 3.323529411764706,
|
| 742 |
+
"roundtrip": true
|
| 743 |
+
},
|
| 744 |
+
"Mixed_Script": {
|
| 745 |
+
"tokens": 33,
|
| 746 |
+
"bytes": 122,
|
| 747 |
+
"words": 17,
|
| 748 |
+
"fertility": 1.9411764705882353,
|
| 749 |
+
"compression": 3.696969696969697,
|
| 750 |
+
"roundtrip": true
|
| 751 |
+
},
|
| 752 |
+
"Repetition": {
|
| 753 |
+
"tokens": 14,
|
| 754 |
+
"bytes": 109,
|
| 755 |
+
"words": 14,
|
| 756 |
+
"fertility": 1.0,
|
| 757 |
+
"compression": 7.785714285714286,
|
| 758 |
+
"roundtrip": true
|
| 759 |
+
},
|
| 760 |
+
"Whitespace": {
|
| 761 |
+
"tokens": 16,
|
| 762 |
+
"bytes": 54,
|
| 763 |
+
"words": 6,
|
| 764 |
+
"fertility": 2.6666666666666665,
|
| 765 |
+
"compression": 3.375,
|
| 766 |
+
"roundtrip": true
|
| 767 |
+
},
|
| 768 |
+
"Empty_Adjacent": {
|
| 769 |
+
"tokens": 14,
|
| 770 |
+
"bytes": 39,
|
| 771 |
+
"words": 5,
|
| 772 |
+
"fertility": 2.8,
|
| 773 |
+
"compression": 2.7857142857142856,
|
| 774 |
+
"roundtrip": true
|
| 775 |
+
}
|
| 776 |
+
},
|
| 777 |
+
"Qwen2 (152K)": {
|
| 778 |
+
"English": {
|
| 779 |
+
"tokens": 31,
|
| 780 |
+
"bytes": 251,
|
| 781 |
+
"words": 27,
|
| 782 |
+
"fertility": 1.1481481481481481,
|
| 783 |
+
"compression": 8.096774193548388,
|
| 784 |
+
"roundtrip": true
|
| 785 |
+
},
|
| 786 |
+
"French": {
|
| 787 |
+
"tokens": 50,
|
| 788 |
+
"bytes": 221,
|
| 789 |
+
"words": 29,
|
| 790 |
+
"fertility": 1.7241379310344827,
|
| 791 |
+
"compression": 4.42,
|
| 792 |
+
"roundtrip": true
|
| 793 |
+
},
|
| 794 |
+
"German": {
|
| 795 |
+
"tokens": 50,
|
| 796 |
+
"bytes": 204,
|
| 797 |
+
"words": 17,
|
| 798 |
+
"fertility": 2.9411764705882355,
|
| 799 |
+
"compression": 4.08,
|
| 800 |
+
"roundtrip": true
|
| 801 |
+
},
|
| 802 |
+
"Spanish": {
|
| 803 |
+
"tokens": 46,
|
| 804 |
+
"bytes": 209,
|
| 805 |
+
"words": 24,
|
| 806 |
+
"fertility": 1.9166666666666667,
|
| 807 |
+
"compression": 4.543478260869565,
|
| 808 |
+
"roundtrip": true
|
| 809 |
+
},
|
| 810 |
+
"Portuguese": {
|
| 811 |
+
"tokens": 34,
|
| 812 |
+
"bytes": 145,
|
| 813 |
+
"words": 19,
|
| 814 |
+
"fertility": 1.7894736842105263,
|
| 815 |
+
"compression": 4.264705882352941,
|
| 816 |
+
"roundtrip": true
|
| 817 |
+
},
|
| 818 |
+
"Italian": {
|
| 819 |
+
"tokens": 39,
|
| 820 |
+
"bytes": 146,
|
| 821 |
+
"words": 17,
|
| 822 |
+
"fertility": 2.2941176470588234,
|
| 823 |
+
"compression": 3.7435897435897436,
|
| 824 |
+
"roundtrip": true
|
| 825 |
+
},
|
| 826 |
+
"Dutch": {
|
| 827 |
+
"tokens": 30,
|
| 828 |
+
"bytes": 123,
|
| 829 |
+
"words": 12,
|
| 830 |
+
"fertility": 2.5,
|
| 831 |
+
"compression": 4.1,
|
| 832 |
+
"roundtrip": true
|
| 833 |
+
},
|
| 834 |
+
"Polish": {
|
| 835 |
+
"tokens": 39,
|
| 836 |
+
"bytes": 122,
|
| 837 |
+
"words": 13,
|
| 838 |
+
"fertility": 3.0,
|
| 839 |
+
"compression": 3.128205128205128,
|
| 840 |
+
"roundtrip": true
|
| 841 |
+
},
|
| 842 |
+
"Swedish": {
|
| 843 |
+
"tokens": 30,
|
| 844 |
+
"bytes": 109,
|
| 845 |
+
"words": 7,
|
| 846 |
+
"fertility": 4.285714285714286,
|
| 847 |
+
"compression": 3.6333333333333333,
|
| 848 |
+
"roundtrip": true
|
| 849 |
+
},
|
| 850 |
+
"Turkish": {
|
| 851 |
+
"tokens": 40,
|
| 852 |
+
"bytes": 134,
|
| 853 |
+
"words": 13,
|
| 854 |
+
"fertility": 3.076923076923077,
|
| 855 |
+
"compression": 3.35,
|
| 856 |
+
"roundtrip": true
|
| 857 |
+
},
|
| 858 |
+
"Ukrainian": {
|
| 859 |
+
"tokens": 57,
|
| 860 |
+
"bytes": 225,
|
| 861 |
+
"words": 13,
|
| 862 |
+
"fertility": 4.384615384615385,
|
| 863 |
+
"compression": 3.9473684210526314,
|
| 864 |
+
"roundtrip": true
|
| 865 |
+
},
|
| 866 |
+
"Chinese": {
|
| 867 |
+
"tokens": 33,
|
| 868 |
+
"bytes": 173,
|
| 869 |
+
"words": 1,
|
| 870 |
+
"fertility": 33.0,
|
| 871 |
+
"compression": 5.242424242424242,
|
| 872 |
+
"roundtrip": true
|
| 873 |
+
},
|
| 874 |
+
"Japanese": {
|
| 875 |
+
"tokens": 77,
|
| 876 |
+
"bytes": 296,
|
| 877 |
+
"words": 1,
|
| 878 |
+
"fertility": 77.0,
|
| 879 |
+
"compression": 3.844155844155844,
|
| 880 |
+
"roundtrip": true
|
| 881 |
+
},
|
| 882 |
+
"Korean": {
|
| 883 |
+
"tokens": 70,
|
| 884 |
+
"bytes": 285,
|
| 885 |
+
"words": 24,
|
| 886 |
+
"fertility": 2.9166666666666665,
|
| 887 |
+
"compression": 4.071428571428571,
|
| 888 |
+
"roundtrip": true
|
| 889 |
+
},
|
| 890 |
+
"Vietnamese": {
|
| 891 |
+
"tokens": 32,
|
| 892 |
+
"bytes": 144,
|
| 893 |
+
"words": 25,
|
| 894 |
+
"fertility": 1.28,
|
| 895 |
+
"compression": 4.5,
|
| 896 |
+
"roundtrip": true
|
| 897 |
+
},
|
| 898 |
+
"Thai": {
|
| 899 |
+
"tokens": 68,
|
| 900 |
+
"bytes": 339,
|
| 901 |
+
"words": 1,
|
| 902 |
+
"fertility": 68.0,
|
| 903 |
+
"compression": 4.985294117647059,
|
| 904 |
+
"roundtrip": true
|
| 905 |
+
},
|
| 906 |
+
"Hindi": {
|
| 907 |
+
"tokens": 115,
|
| 908 |
+
"bytes": 325,
|
| 909 |
+
"words": 16,
|
| 910 |
+
"fertility": 7.1875,
|
| 911 |
+
"compression": 2.8260869565217392,
|
| 912 |
+
"roundtrip": true
|
| 913 |
+
},
|
| 914 |
+
"Arabic": {
|
| 915 |
+
"tokens": 54,
|
| 916 |
+
"bytes": 265,
|
| 917 |
+
"words": 22,
|
| 918 |
+
"fertility": 2.4545454545454546,
|
| 919 |
+
"compression": 4.907407407407407,
|
| 920 |
+
"roundtrip": true
|
| 921 |
+
},
|
| 922 |
+
"Russian": {
|
| 923 |
+
"tokens": 65,
|
| 924 |
+
"bytes": 390,
|
| 925 |
+
"words": 19,
|
| 926 |
+
"fertility": 3.4210526315789473,
|
| 927 |
+
"compression": 6.0,
|
| 928 |
+
"roundtrip": true
|
| 929 |
+
},
|
| 930 |
+
"Python": {
|
| 931 |
+
"tokens": 97,
|
| 932 |
+
"bytes": 291,
|
| 933 |
+
"words": 39,
|
| 934 |
+
"fertility": 2.4871794871794872,
|
| 935 |
+
"compression": 3.0,
|
| 936 |
+
"roundtrip": true
|
| 937 |
+
},
|
| 938 |
+
"JavaScript": {
|
| 939 |
+
"tokens": 91,
|
| 940 |
+
"bytes": 314,
|
| 941 |
+
"words": 40,
|
| 942 |
+
"fertility": 2.275,
|
| 943 |
+
"compression": 3.4505494505494507,
|
| 944 |
+
"roundtrip": true
|
| 945 |
+
},
|
| 946 |
+
"Rust": {
|
| 947 |
+
"tokens": 132,
|
| 948 |
+
"bytes": 289,
|
| 949 |
+
"words": 44,
|
| 950 |
+
"fertility": 3.0,
|
| 951 |
+
"compression": 2.1893939393939394,
|
| 952 |
+
"roundtrip": true
|
| 953 |
+
},
|
| 954 |
+
"LaTeX_Complex": {
|
| 955 |
+
"tokens": 114,
|
| 956 |
+
"bytes": 248,
|
| 957 |
+
"words": 26,
|
| 958 |
+
"fertility": 4.384615384615385,
|
| 959 |
+
"compression": 2.175438596491228,
|
| 960 |
+
"roundtrip": true
|
| 961 |
+
},
|
| 962 |
+
"Unicode_Math": {
|
| 963 |
+
"tokens": 64,
|
| 964 |
+
"bytes": 115,
|
| 965 |
+
"words": 17,
|
| 966 |
+
"fertility": 3.764705882352941,
|
| 967 |
+
"compression": 1.796875,
|
| 968 |
+
"roundtrip": true
|
| 969 |
+
},
|
| 970 |
+
"Mixed_Notation": {
|
| 971 |
+
"tokens": 66,
|
| 972 |
+
"bytes": 138,
|
| 973 |
+
"words": 20,
|
| 974 |
+
"fertility": 3.3,
|
| 975 |
+
"compression": 2.090909090909091,
|
| 976 |
+
"roundtrip": true
|
| 977 |
+
},
|
| 978 |
+
"Emoji_Heavy": {
|
| 979 |
+
"tokens": 46,
|
| 980 |
+
"bytes": 155,
|
| 981 |
+
"words": 17,
|
| 982 |
+
"fertility": 2.7058823529411766,
|
| 983 |
+
"compression": 3.369565217391304,
|
| 984 |
+
"roundtrip": true
|
| 985 |
+
},
|
| 986 |
+
"Numbers_Heavy": {
|
| 987 |
+
"tokens": 114,
|
| 988 |
+
"bytes": 130,
|
| 989 |
+
"words": 15,
|
| 990 |
+
"fertility": 7.6,
|
| 991 |
+
"compression": 1.1403508771929824,
|
| 992 |
+
"roundtrip": true
|
| 993 |
+
},
|
| 994 |
+
"URL_Path": {
|
| 995 |
+
"tokens": 30,
|
| 996 |
+
"bytes": 113,
|
| 997 |
+
"words": 1,
|
| 998 |
+
"fertility": 30.0,
|
| 999 |
+
"compression": 3.7666666666666666,
|
| 1000 |
+
"roundtrip": true
|
| 1001 |
+
},
|
| 1002 |
+
"Mixed_Script": {
|
| 1003 |
+
"tokens": 37,
|
| 1004 |
+
"bytes": 122,
|
| 1005 |
+
"words": 17,
|
| 1006 |
+
"fertility": 2.176470588235294,
|
| 1007 |
+
"compression": 3.2972972972972974,
|
| 1008 |
+
"roundtrip": true
|
| 1009 |
+
},
|
| 1010 |
+
"Repetition": {
|
| 1011 |
+
"tokens": 14,
|
| 1012 |
+
"bytes": 109,
|
| 1013 |
+
"words": 14,
|
| 1014 |
+
"fertility": 1.0,
|
| 1015 |
+
"compression": 7.785714285714286,
|
| 1016 |
+
"roundtrip": true
|
| 1017 |
+
},
|
| 1018 |
+
"Whitespace": {
|
| 1019 |
+
"tokens": 15,
|
| 1020 |
+
"bytes": 54,
|
| 1021 |
+
"words": 6,
|
| 1022 |
+
"fertility": 2.5,
|
| 1023 |
+
"compression": 3.6,
|
| 1024 |
+
"roundtrip": true
|
| 1025 |
+
},
|
| 1026 |
+
"Empty_Adjacent": {
|
| 1027 |
+
"tokens": 14,
|
| 1028 |
+
"bytes": 39,
|
| 1029 |
+
"words": 5,
|
| 1030 |
+
"fertility": 2.8,
|
| 1031 |
+
"compression": 2.7857142857142856,
|
| 1032 |
+
"roundtrip": true
|
| 1033 |
+
}
|
| 1034 |
+
}
|
| 1035 |
+
},
|
| 1036 |
+
"overall": {
|
| 1037 |
+
"Sentinel-SUT (61K)": {
|
| 1038 |
+
"avg_fertility": 10.210548371110242,
|
| 1039 |
+
"std_fertility": 19.290933783132925,
|
| 1040 |
+
"median_fertility": 3.262443438914027,
|
| 1041 |
+
"avg_compression": 3.4607235916418793,
|
| 1042 |
+
"median_compression": 3.3834289813486373,
|
| 1043 |
+
"fairness": 0.04928309414874054,
|
| 1044 |
+
"wins": 2.0,
|
| 1045 |
+
"total_tests": 32.0
|
| 1046 |
+
},
|
| 1047 |
+
"GPT-2 (50K)": {
|
| 1048 |
+
"avg_fertility": 20.730707859469526,
|
| 1049 |
+
"std_fertility": 48.701493147558395,
|
| 1050 |
+
"median_fertility": 4.038461538461538,
|
| 1051 |
+
"avg_compression": 2.5703998033314455,
|
| 1052 |
+
"median_compression": 2.184920634920635,
|
| 1053 |
+
"fairness": 0.020120119873081224,
|
| 1054 |
+
"wins": 0.0,
|
| 1055 |
+
"total_tests": 32.0
|
| 1056 |
+
},
|
| 1057 |
+
"Gemma (256K)": {
|
| 1058 |
+
"avg_fertility": 7.687682759598745,
|
| 1059 |
+
"std_fertility": 13.936255807387061,
|
| 1060 |
+
"median_fertility": 2.645833333333333,
|
| 1061 |
+
"avg_compression": 4.543616791377602,
|
| 1062 |
+
"median_compression": 4.511111111111111,
|
| 1063 |
+
"fairness": 0.0669511832748223,
|
| 1064 |
+
"wins": 20.0,
|
| 1065 |
+
"total_tests": 32.0
|
| 1066 |
+
},
|
| 1067 |
+
"Qwen2 (152K)": {
|
| 1068 |
+
"avg_fertility": 9.134830991971093,
|
| 1069 |
+
"std_fertility": 17.843274072731045,
|
| 1070 |
+
"median_fertility": 2.928921568627451,
|
| 1071 |
+
"avg_compression": 3.8791477128080354,
|
| 1072 |
+
"median_compression": 3.755128205128205,
|
| 1073 |
+
"fairness": 0.05306933371240114,
|
| 1074 |
+
"wins": 6.0,
|
| 1075 |
+
"total_tests": 32.0
|
| 1076 |
+
}
|
| 1077 |
+
},
|
| 1078 |
+
"categories": {
|
| 1079 |
+
"European": [
|
| 1080 |
+
"English",
|
| 1081 |
+
"French",
|
| 1082 |
+
"German",
|
| 1083 |
+
"Spanish",
|
| 1084 |
+
"Portuguese",
|
| 1085 |
+
"Italian",
|
| 1086 |
+
"Dutch",
|
| 1087 |
+
"Polish",
|
| 1088 |
+
"Swedish",
|
| 1089 |
+
"Turkish",
|
| 1090 |
+
"Ukrainian"
|
| 1091 |
+
],
|
| 1092 |
+
"Asian": [
|
| 1093 |
+
"Chinese",
|
| 1094 |
+
"Japanese",
|
| 1095 |
+
"Korean",
|
| 1096 |
+
"Vietnamese",
|
| 1097 |
+
"Thai",
|
| 1098 |
+
"Hindi"
|
| 1099 |
+
],
|
| 1100 |
+
"Semitic/RTL": [
|
| 1101 |
+
"Arabic",
|
| 1102 |
+
"Russian"
|
| 1103 |
+
],
|
| 1104 |
+
"Code": [
|
| 1105 |
+
"Python",
|
| 1106 |
+
"JavaScript",
|
| 1107 |
+
"Rust"
|
| 1108 |
+
],
|
| 1109 |
+
"Mathematics": [
|
| 1110 |
+
"LaTeX_Complex",
|
| 1111 |
+
"Unicode_Math",
|
| 1112 |
+
"Mixed_Notation"
|
| 1113 |
+
],
|
| 1114 |
+
"Edge Cases": [
|
| 1115 |
+
"Emoji_Heavy",
|
| 1116 |
+
"Numbers_Heavy",
|
| 1117 |
+
"URL_Path",
|
| 1118 |
+
"Mixed_Script",
|
| 1119 |
+
"Repetition",
|
| 1120 |
+
"Whitespace",
|
| 1121 |
+
"Empty_Adjacent"
|
| 1122 |
+
]
|
| 1123 |
+
}
|
| 1124 |
+
}
|