Upload safe/flexitok--bpe_pol_Latn_8000_overlap.json with huggingface_hub
Browse files
safe/flexitok--bpe_pol_Latn_8000_overlap.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"1": {"ratio_to_total_tokens": 0.66375, "expected_training_ratio_in_superset": 0.031607142857142854, "num_tokens": 5310}, "2": {"ratio_to_total_tokens": 0.071, "expected_training_ratio_in_superset": 0.0067619047619047615, "num_tokens": 568}, "3": {"ratio_to_total_tokens": 0.034625, "expected_training_ratio_in_superset": 0.004946428571428571, "num_tokens": 277}, "4": {"ratio_to_total_tokens": 0.025125, "expected_training_ratio_in_superset": 0.0047857142857142855, "num_tokens": 201}, "5": {"ratio_to_total_tokens": 0.01775, "expected_training_ratio_in_superset": 0.004226190476190476, "num_tokens": 142}, "6": {"ratio_to_total_tokens": 0.013625, "expected_training_ratio_in_superset": 0.003892857142857143, "num_tokens": 109}, "7": {"ratio_to_total_tokens": 0.01175, "expected_training_ratio_in_superset": 0.003916666666666666, "num_tokens": 94}, "8": {"ratio_to_total_tokens": 0.012125, "expected_training_ratio_in_superset": 0.004619047619047619, "num_tokens": 97}, "9": {"ratio_to_total_tokens": 0.01175, "expected_training_ratio_in_superset": 0.005035714285714286, "num_tokens": 94}, "10": {"ratio_to_total_tokens": 0.009625, "expected_training_ratio_in_superset": 0.004583333333333333, "num_tokens": 77}, "11": {"ratio_to_total_tokens": 0.009625, "expected_training_ratio_in_superset": 0.005041666666666667, "num_tokens": 77}, "12": {"ratio_to_total_tokens": 0.010625, "expected_training_ratio_in_superset": 0.006071428571428571, "num_tokens": 85}, "13": {"ratio_to_total_tokens": 0.008375, "expected_training_ratio_in_superset": 0.005184523809523809, "num_tokens": 67}, "14": {"ratio_to_total_tokens": 0.0115, "expected_training_ratio_in_superset": 0.007666666666666666, "num_tokens": 92}, "15": {"ratio_to_total_tokens": 0.011625, "expected_training_ratio_in_superset": 0.00830357142857143, "num_tokens": 93}, "16": {"ratio_to_total_tokens": 0.006875, "expected_training_ratio_in_superset": 0.005238095238095238, "num_tokens": 55}, "17": {"ratio_to_total_tokens": 0.004625, "expected_training_ratio_in_superset": 0.003744047619047619, "num_tokens": 37}, "18": {"ratio_to_total_tokens": 0.0055, "expected_training_ratio_in_superset": 0.004714285714285714, "num_tokens": 44}, "19": {"ratio_to_total_tokens": 0.004375, "expected_training_ratio_in_superset": 0.003958333333333334, "num_tokens": 35}, "20": {"ratio_to_total_tokens": 0.005125, "expected_training_ratio_in_superset": 0.004880952380952381, "num_tokens": 41}, "21": {"ratio_to_total_tokens": 0.050625, "expected_training_ratio_in_superset": 0.050625, "num_tokens": 405}, "total_training_compared_to_full_model": 0.17980357142857142}
|