Hugging Face's logo

AdaptiveChunking
/
specfull_A_s2

PyTorch
Safetensors
dynamic-chunking
tokenization
hnet
byte-level
beetle-hnet
legacy-trainer
Model card Files
xet
 Community
suchirsalhan commited on
Commit
67c24fa
·
verified ·
1 Parent(s): 8a548fc

run records (split from bundled legacy repo)

Browse files
Files changed (2) hide show
  1. final.json +114 -0
  2. log.jsonl +0 -0
final.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "event": "final",
3
+ "run": "specfull_A_s2",
4
+ "condition": "A",
5
+ "step": 76300,
6
+ "params": 82470528,
7
+ "per_lang": {
8
+ "en": {
9
+ "cpc": 0.17491237143021318,
10
+ "cps": 23.78125,
11
+ "bpb": 1.4459496566972647,
12
+ "bytes_per_chunk": 5.552104457775824
13
+ },
14
+ "de": {
15
+ "cpc": 0.1814152827156285,
16
+ "cps": 29.1015625,
17
+ "bpb": 1.5502144021312767,
18
+ "bytes_per_chunk": 5.632720592185133
19
+ },
20
+ "vi": {
21
+ "cpc": 0.2257119609438568,
22
+ "cps": 32.5078125,
23
+ "bpb": 1.1977475303956786,
24
+ "bytes_per_chunk": 5.7664125999494065
25
+ },
26
+ "tr": {
27
+ "cpc": 0.18820085068773132,
28
+ "cps": 26.6171875,
29
+ "bpb": 1.4290408491613016,
30
+ "bytes_per_chunk": 5.68389327956115
31
+ },
32
+ "fi": {
33
+ "cpc": 0.18063898782627186,
34
+ "cps": 26.546875,
35
+ "bpb": 1.5827007424985364,
36
+ "bytes_per_chunk": 5.751704542337924
37
+ },
38
+ "ru": {
39
+ "cpc": 0.3015303430079156,
40
+ "cps": 44.640625,
41
+ "bpb": 0.8939889669747939,
42
+ "bytes_per_chunk": 5.815535811342939
43
+ },
44
+ "el": {
45
+ "cpc": 0.32782731569458656,
46
+ "cps": 52.5625,
47
+ "bpb": 0.8818085429524961,
48
+ "bytes_per_chunk": 5.717301694619528
49
+ },
50
+ "hi": {
51
+ "cpc": 0.4349479940564636,
52
+ "cps": 57.171875,
53
+ "bpb": 0.7264628568609015,
54
+ "bytes_per_chunk": 5.7822799887506
55
+ },
56
+ "ta": {
57
+ "cpc": 0.4586787911206205,
58
+ "cps": 66.9921875,
59
+ "bpb": 0.6170430255202892,
60
+ "bytes_per_chunk": 5.838786555894603
61
+ },
62
+ "th": {
63
+ "cpc": 0.49835250233136463,
64
+ "cps": 62.625,
65
+ "bpb": 0.7054821638980503,
66
+ "bytes_per_chunk": 5.787290408751228
67
+ },
68
+ "zh": {
69
+ "cpc": 0.4615645106671182,
70
+ "cps": 21.296875,
71
+ "bpb": 2.1749229147532927,
72
+ "bytes_per_chunk": 5.809832505000471
73
+ },
74
+ "my": {
75
+ "cpc": 0.5495566814439519,
76
+ "cps": 81.3515625,
77
+ "bpb": 0.6715674592452543,
78
+ "bytes_per_chunk": 5.296721170700168
79
+ }
80
+ },
81
+ "gini_cpc": 0.2290363327631324,
82
+ "gini_cps": 0.24393082930457433,
83
+ "bpic": 5.708902937895128,
84
+ "seed": 2,
85
+ "mean_boundary_entropy": 0.039606429636478424,
86
+ "dechunk": "ema",
87
+ "router_noise": 0.0,
88
+ "transitions": [
89
+ {
90
+ "step": 0,
91
+ "phase": 1,
92
+ "objective": "global",
93
+ "router_trainable": true,
94
+ "encoder_trainable": true,
95
+ "frozen_router_params": 0
96
+ },
97
+ {
98
+ "step": 30520,
99
+ "phase": 2,
100
+ "objective": "global",
101
+ "router_trainable": true,
102
+ "encoder_trainable": true,
103
+ "frozen_router_params": 0
104
+ },
105
+ {
106
+ "step": 53410,
107
+ "phase": 3,
108
+ "objective": "global",
109
+ "router_trainable": true,
110
+ "encoder_trainable": true,
111
+ "frozen_router_params": 0
112
+ }
113
+ ]
114
+ }
log.jsonl ADDED
The diff for this file is too large to render. See raw diff