andrew-healey commited on
Commit
681d536
·
verified ·
1 Parent(s): 32014ad

Upload folder using huggingface_hub

Browse files
lr3e-3_total_batch_size20480_seq_len512/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep_2/lr3e-3_total_batch_size20480_seq_len512", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 2, "protect_bos_token": true, "prevent_from_masking_myself": true, "max_steps": null, "group": "proxy_model_sweep_2", "use_wandb": true, "kill_self_after_run": false, "random_seed": 1337, "memory_penalty_epsilon": 0.1, "selection_head_linear_combo": "none", "selection_head_linear_combo_scale": 1.0, "protection_kind": "none", "leaky_relu_alpha": null, "leaky_relu_bias": null, "use_compile": true, "use_mini_model": false, "upload_to_hf": true, "seq_len": 512, "batch_size": null, "total_batch_size": 20480, "protection_head_scaling_factor": 1.0, "protection_head_bias": 0.0, "n_sliced_masks": null, "n_latent_masks": null, "mask_layernorm": false, "residual_attention_masks": false, "compute_base_shapes": false, "base_shapes_savefile": null, "mup": true, "disable_selection": true, "mup_enable_coord_check_logging": false, "max_lr": 0.003, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr3e-3_total_batch_size20480_seq_len512/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8513131d2bf732f8f6d97ff82d32c505b2a2f3f9f2247bac315bb09ce0e1f97
3
+ size 964
lr3e-3_total_batch_size20480_seq_len512/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.7800
3
+ 0 val perplexity 130608.6406
4
+ 0 train 11.774407 (lr=1.0490e-05) (hash(x)=27234506)
5
+ 1 train 11.734957 (lr=2.0979e-05) (hash(x)=20782690)
6
+ 2 train 11.728184 (lr=3.1469e-05) (hash(x)=26532095)
7
+ 3 train 11.657902 (lr=4.1958e-05) (hash(x)=23890908)
8
+ 4 train 11.624018 (lr=5.2448e-05) (hash(x)=24011372)
9
+ 5 train 11.491178 (lr=6.2937e-05) (hash(x)=24611628)
10
+ 6 train 11.382491 (lr=7.3427e-05) (hash(x)=19245352)
11
+ 7 train 11.221613 (lr=8.3916e-05) (hash(x)=25767553)
12
+ 8 train 11.101578 (lr=9.4406e-05) (hash(x)=24809873)
13
+ 9 train 10.976390 (lr=1.0490e-04) (hash(x)=22512166)
14
+ 10 train 10.794262 (lr=1.1538e-04) (hash(x)=28092148)
15
+ 11 train 10.702009 (lr=1.2587e-04) (hash(x)=24732833)
16
+ 12 train 10.615644 (lr=1.3636e-04) (hash(x)=22694718)
17
+ 13 train 10.514603 (lr=1.4685e-04) (hash(x)=25129410)
18
+ 14 train 10.476822 (lr=1.5734e-04) (hash(x)=27858570)
19
+ 15 train 10.356726 (lr=1.6783e-04) (hash(x)=26582391)
20
+ 16 train 10.326472 (lr=1.7832e-04) (hash(x)=27279806)
21
+ 17 train 10.284825 (lr=1.8881e-04) (hash(x)=26472336)
22
+ 18 train 10.251107 (lr=1.9930e-04) (hash(x)=28301589)
23
+ 19 train 10.207313 (lr=2.0979e-04) (hash(x)=27654289)
24
+ 20 train 10.162497 (lr=2.2028e-04) (hash(x)=24995715)
25
+ 21 train 10.107504 (lr=2.3077e-04) (hash(x)=26559876)
26
+ 22 train 10.024425 (lr=2.4126e-04) (hash(x)=28364895)
27
+ 23 train 10.025440 (lr=2.5175e-04) (hash(x)=21770329)
28
+ 24 train 10.051819 (lr=2.6224e-04) (hash(x)=24628606)
29
+ 25 train 9.938678 (lr=2.7273e-04) (hash(x)=22723459)
30
+ 26 train 9.795247 (lr=2.8322e-04) (hash(x)=22897919)
31
+ 27 train 9.950988 (lr=2.9371e-04) (hash(x)=30338342)
32
+ 28 train 9.787160 (lr=3.0420e-04) (hash(x)=27063280)
33
+ 29 train 9.717479 (lr=3.1469e-04) (hash(x)=25678059)
34
+ 30 train 9.552927 (lr=3.2517e-04) (hash(x)=21853028)
35
+ 31 train 9.525123 (lr=3.3566e-04) (hash(x)=26109335)
36
+ 32 train 9.512399 (lr=3.4615e-04) (hash(x)=25334848)
37
+ 33 train 9.387586 (lr=3.5664e-04) (hash(x)=25044885)
38
+ 34 train 9.374243 (lr=3.6713e-04) (hash(x)=23678349)
39
+ 35 train 9.381580 (lr=3.7762e-04) (hash(x)=28870690)
40
+ 36 train 9.301183 (lr=3.8811e-04) (hash(x)=32982615)
41
+ 37 train 9.346842 (lr=3.9860e-04) (hash(x)=35279941)
42
+ 38 train 9.190553 (lr=4.0909e-04) (hash(x)=26400041)
43
+ 39 train 8.973139 (lr=4.1958e-04) (hash(x)=25015923)
44
+ 40 train 8.849054 (lr=4.3007e-04) (hash(x)=24853586)
45
+ 41 train 8.825850 (lr=4.4056e-04) (hash(x)=26740855)
46
+ 42 train 8.594728 (lr=4.5105e-04) (hash(x)=24584116)
47
+ 43 train 8.624332 (lr=4.6154e-04) (hash(x)=22262151)
48
+ 44 train 8.542684 (lr=4.7203e-04) (hash(x)=26627860)
49
+ 45 train 8.401657 (lr=4.8252e-04) (hash(x)=26621834)
50
+ 46 train 8.439075 (lr=4.9301e-04) (hash(x)=26930630)
51
+ 47 train 8.273438 (lr=5.0350e-04) (hash(x)=23234741)
52
+ 48 train 8.313539 (lr=5.1399e-04) (hash(x)=28102443)
53
+ 49 train 8.539777 (lr=5.2448e-04) (hash(x)=25470563)
54
+ 50 val loss 8.2499
55
+ 50 val perplexity 3827.0601
56
+ 50 train 8.317603 (lr=5.3497e-04) (hash(x)=28282861)
57
+ 51 train 8.245094 (lr=5.4545e-04) (hash(x)=21799102)
58
+ 52 train 8.115021 (lr=5.5594e-04) (hash(x)=23471769)
59
+ 53 train 8.076307 (lr=5.6643e-04) (hash(x)=23275479)
60
+ 54 train 8.150254 (lr=5.7692e-04) (hash(x)=24473581)
61
+ 55 train 8.054639 (lr=5.8741e-04) (hash(x)=22091862)
62
+ 56 train 8.046779 (lr=5.9790e-04) (hash(x)=26018202)
63
+ 57 train 7.948277 (lr=6.0839e-04) (hash(x)=17430373)
64
+ 58 train 7.703851 (lr=6.1888e-04) (hash(x)=24078083)
65
+ 59 train 7.911138 (lr=6.2937e-04) (hash(x)=22954861)
66
+ 60 train 8.077868 (lr=6.3986e-04) (hash(x)=28695016)
67
+ 61 train 8.054693 (lr=6.5035e-04) (hash(x)=26671799)
68
+ 62 train 8.055908 (lr=6.6084e-04) (hash(x)=23166092)
69
+ 63 train 7.887484 (lr=6.7133e-04) (hash(x)=26924723)
70
+ 64 train 7.959543 (lr=6.8182e-04) (hash(x)=25138659)
71
+ 65 train 8.051960 (lr=6.9231e-04) (hash(x)=23557495)
72
+ 66 train 7.868886 (lr=7.0280e-04) (hash(x)=25288123)
73
+ 67 train 7.752593 (lr=7.1329e-04) (hash(x)=27804380)
74
+ 68 train 7.990045 (lr=7.2378e-04) (hash(x)=26993263)
75
+ 69 train 8.049271 (lr=7.3427e-04) (hash(x)=26654908)
76
+ 70 train 7.857317 (lr=7.4476e-04) (hash(x)=27347722)
77
+ 71 train 7.890500 (lr=7.5524e-04) (hash(x)=23627518)
78
+ 72 train 7.745518 (lr=7.6573e-04) (hash(x)=21768671)
79
+ 73 train 7.844417 (lr=7.7622e-04) (hash(x)=21889990)
80
+ 74 train 7.733242 (lr=7.8671e-04) (hash(x)=23665838)
81
+ 75 train 8.116943 (lr=7.9720e-04) (hash(x)=22528001)
82
+ 76 train 7.883326 (lr=8.0769e-04) (hash(x)=22487328)
83
+ 77 train 8.071822 (lr=8.1818e-04) (hash(x)=29248942)
84
+ 78 train 7.824840 (lr=8.2867e-04) (hash(x)=23543273)
85
+ 79 train 7.795044 (lr=8.3916e-04) (hash(x)=25533417)
86
+ 80 train 7.932984 (lr=8.4965e-04) (hash(x)=25552036)
87
+ 81 train 7.722288 (lr=8.6014e-04) (hash(x)=20612533)
88
+ 82 train 7.469649 (lr=8.7063e-04) (hash(x)=20216795)
89
+ 83 train 7.836080 (lr=8.8112e-04) (hash(x)=24677740)
90
+ 84 train 7.544708 (lr=8.9161e-04) (hash(x)=16722715)
91
+ 85 train 7.815243 (lr=9.0210e-04) (hash(x)=25347203)
92
+ 86 train 7.822248 (lr=9.1259e-04) (hash(x)=24272489)
93
+ 87 train 7.946124 (lr=9.2308e-04) (hash(x)=28922363)
94
+ 88 train 7.877947 (lr=9.3357e-04) (hash(x)=25510798)
95
+ 89 train 7.850759 (lr=9.4406e-04) (hash(x)=25631397)
96
+ 90 train 7.664144 (lr=9.5455e-04) (hash(x)=21333676)
97
+ 91 train 7.969671 (lr=9.6503e-04) (hash(x)=26237983)
98
+ 92 train 7.808350 (lr=9.7552e-04) (hash(x)=25976502)
99
+ 93 train 7.820788 (lr=9.8601e-04) (hash(x)=24742645)
100
+ 94 train 7.935299 (lr=9.9650e-04) (hash(x)=27086333)
101
+ 95 train 7.816861 (lr=1.0070e-03) (hash(x)=26157660)
102
+ 96 train 7.846170 (lr=1.0175e-03) (hash(x)=23616370)
103
+ 97 train 7.578517 (lr=1.0280e-03) (hash(x)=22582169)
104
+ 98 train 7.672332 (lr=1.0385e-03) (hash(x)=26233189)
105
+ 99 train 7.871157 (lr=1.0490e-03) (hash(x)=26431960)
106
+ 100 val loss 7.8077
107
+ 100 val perplexity 2459.3525
108
+ 100 train 7.924748 (lr=1.0594e-03) (hash(x)=30707498)
109
+ 101 train 7.828882 (lr=1.0699e-03) (hash(x)=25716176)
110
+ 102 train 7.823374 (lr=1.0804e-03) (hash(x)=26256420)
111
+ 103 train 7.855412 (lr=1.0909e-03) (hash(x)=28019788)
112
+ 104 train 7.791597 (lr=1.1014e-03) (hash(x)=23510110)
113
+ 105 train 7.778003 (lr=1.1119e-03) (hash(x)=20382963)
114
+ 106 train 7.665036 (lr=1.1224e-03) (hash(x)=25339466)
115
+ 107 train 7.689520 (lr=1.1329e-03) (hash(x)=24489647)
116
+ 108 train 7.670721 (lr=1.1434e-03) (hash(x)=22927699)
117
+ 109 train 7.808705 (lr=1.1538e-03) (hash(x)=25128502)
118
+ 110 train 7.796720 (lr=1.1643e-03) (hash(x)=25808969)
119
+ 111 train 7.781571 (lr=1.1748e-03) (hash(x)=25231390)
120
+ 112 train 7.626089 (lr=1.1853e-03) (hash(x)=21154388)
121
+ 113 train 7.566475 (lr=1.1958e-03) (hash(x)=22708977)
122
+ 114 train 8.024735 (lr=1.2063e-03) (hash(x)=27258353)
123
+ 115 train 7.779997 (lr=1.2168e-03) (hash(x)=24364920)
124
+ 116 train 7.889498 (lr=1.2273e-03) (hash(x)=27243972)
125
+ 117 train 7.936576 (lr=1.2378e-03) (hash(x)=24932925)
126
+ 118 train 7.773919 (lr=1.2483e-03) (hash(x)=26371091)
127
+ 119 train 7.727289 (lr=1.2587e-03) (hash(x)=26913684)
128
+ 120 train 7.923633 (lr=1.2692e-03) (hash(x)=29239611)
129
+ 121 train 7.577542 (lr=1.2797e-03) (hash(x)=18017792)
130
+ 122 train 7.756227 (lr=1.2902e-03) (hash(x)=27574089)
131
+ 123 train 7.837070 (lr=1.3007e-03) (hash(x)=28613882)
132
+ 124 train 7.810490 (lr=1.3112e-03) (hash(x)=26981914)
133
+ 125 train 7.775507 (lr=1.3217e-03) (hash(x)=22652243)
134
+ 126 train 7.572736 (lr=1.3322e-03) (hash(x)=15446025)
135
+ 127 train 7.672573 (lr=1.3427e-03) (hash(x)=27194521)
136
+ 128 train 7.670260 (lr=1.3531e-03) (hash(x)=25264518)
137
+ 129 train 7.890049 (lr=1.3636e-03) (hash(x)=24936836)
138
+ 130 train 7.921948 (lr=1.3741e-03) (hash(x)=28197282)
139
+ 131 train 7.859503 (lr=1.3846e-03) (hash(x)=24765578)
140
+ 132 train 7.915826 (lr=1.3951e-03) (hash(x)=27448185)
141
+ 133 train 7.881388 (lr=1.4056e-03) (hash(x)=23107416)
142
+ 134 train 7.784338 (lr=1.4161e-03) (hash(x)=26755138)
143
+ 135 train 7.807302 (lr=1.4266e-03) (hash(x)=29314255)
144
+ 136 train 7.764961 (lr=1.4371e-03) (hash(x)=22112669)
145
+ 137 train 7.728633 (lr=1.4476e-03) (hash(x)=24013079)
146
+ 138 train 7.748622 (lr=1.4580e-03) (hash(x)=22537194)
147
+ 139 train 7.935742 (lr=1.4685e-03) (hash(x)=25322001)
148
+ 140 train 7.958761 (lr=1.4790e-03) (hash(x)=27028126)
149
+ 141 train 7.944808 (lr=1.4895e-03) (hash(x)=26690440)
150
+ 142 train 7.886997 (lr=1.5000e-03) (hash(x)=26967331)
151
+ 143 train 7.717066 (lr=1.5105e-03) (hash(x)=22598512)
152
+ 144 train 7.846643 (lr=1.5210e-03) (hash(x)=24697985)
153
+ 145 train 7.613437 (lr=1.5315e-03) (hash(x)=18473911)
154
+ 146 train 7.716681 (lr=1.5420e-03) (hash(x)=23172124)
155
+ 147 train 8.294428 (lr=1.5524e-03) (hash(x)=22322442)
156
+ 148 train 8.087004 (lr=1.5629e-03) (hash(x)=29659121)
157
+ 149 train 8.092625 (lr=1.5734e-03) (hash(x)=26324153)
158
+ 150 val loss 8.1299
159
+ 150 val perplexity 3394.4277
160
+ 150 train 8.172068 (lr=1.5839e-03) (hash(x)=26939970)
161
+ 151 train 8.349751 (lr=1.5944e-03) (hash(x)=23446058)
162
+ 152 train 8.211806 (lr=1.6049e-03) (hash(x)=29981801)
163
+ 153 train 8.045858 (lr=1.6154e-03) (hash(x)=24843741)
164
+ 154 train 7.915380 (lr=1.6259e-03) (hash(x)=22855363)
165
+ 155 train 7.741583 (lr=1.6364e-03) (hash(x)=23197102)
166
+ 156 train 7.904635 (lr=1.6469e-03) (hash(x)=25704919)
167
+ 157 train 8.231151 (lr=1.6573e-03) (hash(x)=23054940)
168
+ 158 train 8.186505 (lr=1.6678e-03) (hash(x)=24909904)
169
+ 159 train 7.895777 (lr=1.6783e-03) (hash(x)=27140757)
170
+ 160 train 7.758314 (lr=1.6888e-03) (hash(x)=25993110)
171
+ 161 train 7.779690 (lr=1.6993e-03) (hash(x)=25149419)
172
+ 162 train 7.933845 (lr=1.7098e-03) (hash(x)=27623643)
173
+ 163 train 7.906970 (lr=1.7203e-03) (hash(x)=24727764)
174
+ 164 train 7.364115 (lr=1.7308e-03) (hash(x)=23155773)
175
+ 165 train 7.699077 (lr=1.7413e-03) (hash(x)=21935931)
176
+ 166 train 7.563475 (lr=1.7517e-03) (hash(x)=20054917)
177
+ 167 train 7.839293 (lr=1.7622e-03) (hash(x)=27606073)
178
+ 168 train 7.946065 (lr=1.7727e-03) (hash(x)=26534487)
179
+ 169 train 7.756315 (lr=1.7832e-03) (hash(x)=22450341)
180
+ 170 train 7.917379 (lr=1.7937e-03) (hash(x)=25013073)
181
+ 171 train 8.064950 (lr=1.8042e-03) (hash(x)=25945859)
182
+ 172 train 7.798339 (lr=1.8147e-03) (hash(x)=24418091)
183
+ 173 train 7.756101 (lr=1.8252e-03) (hash(x)=21613707)
184
+ 174 train 7.872874 (lr=1.8357e-03) (hash(x)=26386157)
185
+ 175 train 7.787004 (lr=1.8462e-03) (hash(x)=24396519)
186
+ 176 train 7.872267 (lr=1.8566e-03) (hash(x)=23265507)
187
+ 177 train 7.893404 (lr=1.8671e-03) (hash(x)=29138828)
188
+ 178 train 7.902524 (lr=1.8776e-03) (hash(x)=24265353)
189
+ 179 train 7.871665 (lr=1.8881e-03) (hash(x)=24685515)
190
+ 180 train 7.759713 (lr=1.8986e-03) (hash(x)=24876269)
191
+ 181 train 7.831167 (lr=1.9091e-03) (hash(x)=22825749)
192
+ 182 train 7.863775 (lr=1.9196e-03) (hash(x)=21239652)
193
+ 183 train 7.846412 (lr=1.9301e-03) (hash(x)=28767869)
194
+ 184 train 7.877602 (lr=1.9406e-03) (hash(x)=25614301)
195
+ 185 train 7.856432 (lr=1.9510e-03) (hash(x)=21159060)
196
+ 186 train 7.769205 (lr=1.9615e-03) (hash(x)=26751788)
197
+ 187 train 7.740848 (lr=1.9720e-03) (hash(x)=25045397)
198
+ 188 train 7.838425 (lr=1.9825e-03) (hash(x)=29989709)
199
+ 189 train 7.930346 (lr=1.9930e-03) (hash(x)=25560928)
200
+ 190 train 7.859947 (lr=2.0035e-03) (hash(x)=26062687)
201
+ 191 train 8.006618 (lr=2.0140e-03) (hash(x)=28959222)
202
+ 192 train 7.814004 (lr=2.0245e-03) (hash(x)=25434884)
203
+ 193 train 7.832300 (lr=2.0350e-03) (hash(x)=21248405)
204
+ 194 train 7.765103 (lr=2.0455e-03) (hash(x)=25044192)
205
+ 195 train 7.787119 (lr=2.0559e-03) (hash(x)=25830182)
206
+ 196 train 7.797311 (lr=2.0664e-03) (hash(x)=26621419)
207
+ 197 train 7.525129 (lr=2.0769e-03) (hash(x)=25173113)
208
+ 198 train 7.728794 (lr=2.0874e-03) (hash(x)=24400133)
209
+ 199 train 7.710655 (lr=2.0979e-03) (hash(x)=23486277)
210
+ 200 val loss 7.7293
211
+ 200 val perplexity 2274.0620
212
+ 200 train 7.715982 (lr=2.1084e-03) (hash(x)=25978130)
213
+ 201 train 7.763310 (lr=2.1189e-03) (hash(x)=28916006)
214
+ 202 train 7.949074 (lr=2.1294e-03) (hash(x)=33195100)
215
+ 203 train 7.882707 (lr=2.1399e-03) (hash(x)=25556929)
216
+ 204 train 7.818490 (lr=2.1503e-03) (hash(x)=29536986)
217
+ 205 train 7.765347 (lr=2.1608e-03) (hash(x)=29648798)
218
+ 206 train 7.681680 (lr=2.1713e-03) (hash(x)=23124767)
219
+ 207 train 7.682129 (lr=2.1818e-03) (hash(x)=24976217)
220
+ 208 train 7.706589 (lr=2.1923e-03) (hash(x)=23632825)
221
+ 209 train 7.844908 (lr=2.2028e-03) (hash(x)=31019606)
222
+ 210 train 7.881783 (lr=2.2133e-03) (hash(x)=28236580)
223
+ 211 train 7.766644 (lr=2.2238e-03) (hash(x)=28269421)
224
+ 212 train 7.556601 (lr=2.2343e-03) (hash(x)=25563279)
225
+ 213 train 7.854095 (lr=2.2448e-03) (hash(x)=24865358)
226
+ 214 train 7.579433 (lr=2.2552e-03) (hash(x)=22155546)
227
+ 215 train 7.752061 (lr=2.2657e-03) (hash(x)=26984272)
228
+ 216 train 7.777660 (lr=2.2762e-03) (hash(x)=23193673)
229
+ 217 train 7.792168 (lr=2.2867e-03) (hash(x)=24544116)
230
+ 218 train 7.745127 (lr=2.2972e-03) (hash(x)=24547533)
231
+ 219 train 7.722248 (lr=2.3077e-03) (hash(x)=22059850)
232
+ 220 train 7.835535 (lr=2.3182e-03) (hash(x)=25994573)
233
+ 221 train 7.792829 (lr=2.3287e-03) (hash(x)=23926632)
234
+ 222 train 7.621464 (lr=2.3392e-03) (hash(x)=27111369)
235
+ 223 train 7.790362 (lr=2.3497e-03) (hash(x)=27025333)
236
+ 224 train 7.581903 (lr=2.3601e-03) (hash(x)=16908068)
237
+ 225 train 7.746314 (lr=2.3706e-03) (hash(x)=29120407)
238
+ 226 train 7.855801 (lr=2.3811e-03) (hash(x)=27092710)
239
+ 227 train 7.684062 (lr=2.3916e-03) (hash(x)=23827393)
240
+ 228 train 7.746810 (lr=2.4021e-03) (hash(x)=32812727)
241
+ 229 train 7.730716 (lr=2.4126e-03) (hash(x)=27230027)
242
+ 230 train 7.740616 (lr=2.4231e-03) (hash(x)=25703381)
243
+ 231 train 7.686068 (lr=2.4336e-03) (hash(x)=24968260)
244
+ 232 train 7.674274 (lr=2.4441e-03) (hash(x)=25676468)
245
+ 233 train 7.785048 (lr=2.4545e-03) (hash(x)=27544803)
246
+ 234 train 7.781841 (lr=2.4650e-03) (hash(x)=24331407)
247
+ 235 train 7.719216 (lr=2.4755e-03) (hash(x)=26540663)
248
+ 236 train 7.747620 (lr=2.4860e-03) (hash(x)=25532657)
249
+ 237 train 7.922350 (lr=2.4965e-03) (hash(x)=25856625)
250
+ 238 train 7.656663 (lr=2.5070e-03) (hash(x)=23437426)
251
+ 239 train 7.714634 (lr=2.5175e-03) (hash(x)=23164356)
252
+ 240 train 7.711842 (lr=2.5280e-03) (hash(x)=25957896)
253
+ 241 train 7.781186 (lr=2.5385e-03) (hash(x)=25646282)
254
+ 242 train 7.816673 (lr=2.5490e-03) (hash(x)=21513584)
255
+ 243 train 7.699444 (lr=2.5594e-03) (hash(x)=23843233)
256
+ 244 train 7.695591 (lr=2.5699e-03) (hash(x)=23226697)
257
+ 245 train 7.591177 (lr=2.5804e-03) (hash(x)=24506028)
258
+ 246 train 7.657896 (lr=2.5909e-03) (hash(x)=26024321)
259
+ 247 train 7.764372 (lr=2.6014e-03) (hash(x)=22781277)
260
+ 248 train 7.763503 (lr=2.6119e-03) (hash(x)=24451843)
261
+ 249 train 7.697984 (lr=2.6224e-03) (hash(x)=24948650)
262
+ 250 val loss 7.7126
263
+ 250 val perplexity 2236.3733
264
+ 250 train 7.634035 (lr=2.6329e-03) (hash(x)=25559534)
265
+ 251 train 7.751667 (lr=2.6434e-03) (hash(x)=26373991)
266
+ 252 train 7.793786 (lr=2.6538e-03) (hash(x)=27756673)
267
+ 253 train 7.759771 (lr=2.6643e-03) (hash(x)=23402396)
268
+ 254 train 7.804548 (lr=2.6748e-03) (hash(x)=28250354)
269
+ 255 train 7.687270 (lr=2.6853e-03) (hash(x)=21562133)
270
+ 256 train 7.866685 (lr=2.6958e-03) (hash(x)=27253861)
271
+ 257 train 7.754948 (lr=2.7063e-03) (hash(x)=25009210)
272
+ 258 train 7.636994 (lr=2.7168e-03) (hash(x)=23052577)
273
+ 259 train 7.798805 (lr=2.7273e-03) (hash(x)=27161811)
274
+ 260 train 7.728490 (lr=2.7378e-03) (hash(x)=27302459)
275
+ 261 train 7.845384 (lr=2.7483e-03) (hash(x)=29515100)
276
+ 262 train 7.715311 (lr=2.7587e-03) (hash(x)=25318001)
277
+ 263 train 7.736711 (lr=2.7692e-03) (hash(x)=25305929)
278
+ 264 train 7.748530 (lr=2.7797e-03) (hash(x)=24339043)
279
+ 265 train 7.805601 (lr=2.7902e-03) (hash(x)=26808543)
280
+ 266 train 7.587145 (lr=2.8007e-03) (hash(x)=25763991)
281
+ 267 train 7.798524 (lr=2.8112e-03) (hash(x)=25669509)
282
+ 268 train 7.830450 (lr=2.8217e-03) (hash(x)=27676869)
283
+ 269 train 7.739849 (lr=2.8322e-03) (hash(x)=28435805)
284
+ 270 train 7.683696 (lr=2.8427e-03) (hash(x)=23455211)
285
+ 271 train 7.826981 (lr=2.8531e-03) (hash(x)=26482758)
286
+ 272 train 7.680317 (lr=2.8636e-03) (hash(x)=27068280)
287
+ 273 train 7.738005 (lr=2.8741e-03) (hash(x)=24929929)
288
+ 274 train 7.706072 (lr=2.8846e-03) (hash(x)=25373676)
289
+ 275 train 7.630634 (lr=2.8951e-03) (hash(x)=25043037)
290
+ 276 train 7.722453 (lr=2.9056e-03) (hash(x)=27906976)
291
+ 277 train 8.052113 (lr=2.9161e-03) (hash(x)=29151257)
292
+ 278 train 7.770740 (lr=2.9266e-03) (hash(x)=26405613)
293
+ 279 train 7.578698 (lr=2.9371e-03) (hash(x)=26400365)
294
+ 280 train 7.749694 (lr=2.9476e-03) (hash(x)=23971905)
295
+ 281 train 7.664405 (lr=2.9580e-03) (hash(x)=24098812)
296
+ 282 train 7.809106 (lr=2.9685e-03) (hash(x)=29402976)
297
+ 283 train 7.562268 (lr=2.9790e-03) (hash(x)=20967666)
298
+ 284 train 7.664828 (lr=2.9895e-03) (hash(x)=24287610)
299
+ 285 train 8.031435 (lr=3.0000e-03) (hash(x)=28667963)
300
+ 286 train 7.722690 (lr=3.0000e-03) (hash(x)=25974316)
301
+ 287 train 7.863599 (lr=3.0000e-03) (hash(x)=27083417)
302
+ 288 train 7.645386 (lr=2.9999e-03) (hash(x)=28605205)
303
+ 289 train 7.624612 (lr=2.9999e-03) (hash(x)=23649001)
304
+ 290 train 7.634035 (lr=2.9998e-03) (hash(x)=25921700)
305
+ 291 train 7.743294 (lr=2.9997e-03) (hash(x)=31338300)
306
+ 292 train 7.649884 (lr=2.9995e-03) (hash(x)=23862328)
307
+ 293 train 7.696478 (lr=2.9994e-03) (hash(x)=30603174)
308
+ 294 train 7.745442 (lr=2.9992e-03) (hash(x)=23603806)
309
+ 295 train 7.740461 (lr=2.9989e-03) (hash(x)=24830965)
310
+ 296 train 7.861807 (lr=2.9987e-03) (hash(x)=32488729)
311
+ 297 train 7.973266 (lr=2.9984e-03) (hash(x)=27269893)
312
+ 298 train 7.783968 (lr=2.9981e-03) (hash(x)=25215077)
313
+ 299 train 7.736121 (lr=2.9978e-03) (hash(x)=24825041)
314
+ 300 val loss 7.7915
315
+ 300 val perplexity 2419.9070
316
+ 300 train 7.844444 (lr=2.9974e-03) (hash(x)=25893804)
317
+ 301 train 7.675712 (lr=2.9971e-03) (hash(x)=22724871)
318
+ 302 train 7.698071 (lr=2.9967e-03) (hash(x)=27317554)
319
+ 303 train 7.632149 (lr=2.9962e-03) (hash(x)=24683717)
320
+ 304 train 7.734056 (lr=2.9958e-03) (hash(x)=21584083)
321
+ 305 train 7.910256 (lr=2.9953e-03) (hash(x)=26190337)
322
+ 306 train 7.802503 (lr=2.9948e-03) (hash(x)=22249019)
323
+ 307 train 7.747657 (lr=2.9942e-03) (hash(x)=23086289)
324
+ 308 train 7.906690 (lr=2.9937e-03) (hash(x)=28047044)
325
+ 309 train 7.672680 (lr=2.9931e-03) (hash(x)=23593875)
326
+ 310 train 7.696661 (lr=2.9925e-03) (hash(x)=26059735)
327
+ 311 train 7.682204 (lr=2.9918e-03) (hash(x)=22414190)
328
+ 312 train 7.912416 (lr=2.9912e-03) (hash(x)=24204591)
329
+ 313 train 7.867387 (lr=2.9905e-03) (hash(x)=26758114)
330
+ 314 train 7.944136 (lr=2.9898e-03) (hash(x)=20084233)
331
+ 315 train 7.913074 (lr=2.9890e-03) (hash(x)=21563184)
332
+ 316 train 8.025814 (lr=2.9883e-03) (hash(x)=23861489)
333
+ 317 train 8.042313 (lr=2.9875e-03) (hash(x)=19218470)
334
+ 318 train 7.998761 (lr=2.9866e-03) (hash(x)=28469562)
335
+ 319 train 8.158546 (lr=2.9858e-03) (hash(x)=25588469)
336
+ 320 train 7.951131 (lr=2.9849e-03) (hash(x)=25236367)
337
+ 321 train 7.959385 (lr=2.9840e-03) (hash(x)=30433767)
338
+ 322 train 8.121995 (lr=2.9831e-03) (hash(x)=26302626)
339
+ 323 train 7.819376 (lr=2.9821e-03) (hash(x)=17851621)
340
+ 324 train 8.070551 (lr=2.9812e-03) (hash(x)=27028191)
341
+ 325 train 7.982796 (lr=2.9802e-03) (hash(x)=25183690)
342
+ 326 train 7.979324 (lr=2.9791e-03) (hash(x)=26182243)
343
+ 327 train 8.042911 (lr=2.9781e-03) (hash(x)=27040642)
344
+ 328 train 7.791448 (lr=2.9770e-03) (hash(x)=24489607)
345
+ 329 train 7.713860 (lr=2.9759e-03) (hash(x)=22183303)
346
+ 330 train 7.833599 (lr=2.9748e-03) (hash(x)=28844646)
347
+ 331 train 8.241163 (lr=2.9736e-03) (hash(x)=27537063)
348
+ 332 train 7.791712 (lr=2.9724e-03) (hash(x)=25446686)
349
+ 333 train 7.658485 (lr=2.9712e-03) (hash(x)=19492832)
350
+ 334 train 7.685214 (lr=2.9700e-03) (hash(x)=25484958)
351
+ 335 train 7.723529 (lr=2.9687e-03) (hash(x)=25951629)
352
+ 336 train 7.787832 (lr=2.9675e-03) (hash(x)=23462798)
353
+ 337 train 7.692180 (lr=2.9662e-03) (hash(x)=22408682)
354
+ 338 train 7.791191 (lr=2.9648e-03) (hash(x)=31317970)
355
+ 339 train 7.636254 (lr=2.9635e-03) (hash(x)=27633457)
356
+ 340 train 7.635394 (lr=2.9621e-03) (hash(x)=23729283)
357
+ 341 train 7.850798 (lr=2.9607e-03) (hash(x)=27635080)
358
+ 342 train 7.774888 (lr=2.9592e-03) (hash(x)=25918780)
359
+ 343 train 7.681160 (lr=2.9578e-03) (hash(x)=22287596)
360
+ 344 train 7.637040 (lr=2.9563e-03) (hash(x)=22694623)
361
+ 345 train 8.021022 (lr=2.9548e-03) (hash(x)=35895440)
362
+ 346 train 7.758882 (lr=2.9532e-03) (hash(x)=24298886)
363
+ 347 train 7.753109 (lr=2.9517e-03) (hash(x)=25888847)
364
+ 348 train 7.821885 (lr=2.9501e-03) (hash(x)=30860646)
365
+ 349 train 7.768159 (lr=2.9485e-03) (hash(x)=26518800)
366
+ 350 val loss 7.7251
367
+ 350 val perplexity 2264.5071
368
+ 350 train 7.710551 (lr=2.9468e-03) (hash(x)=25385523)
369
+ 351 train 7.883587 (lr=2.9452e-03) (hash(x)=37519283)
370
+ 352 train 8.170580 (lr=2.9435e-03) (hash(x)=20564397)
371
+ 353 train 7.855562 (lr=2.9418e-03) (hash(x)=25141945)
372
+ 354 train 7.856080 (lr=2.9400e-03) (hash(x)=22717866)
373
+ 355 train 8.176699 (lr=2.9383e-03) (hash(x)=28527939)
374
+ 356 train 7.924567 (lr=2.9365e-03) (hash(x)=32757059)
375
+ 357 train 7.743589 (lr=2.9347e-03) (hash(x)=21942471)
376
+ 358 train 8.029419 (lr=2.9328e-03) (hash(x)=24247339)
377
+ 359 train 7.843024 (lr=2.9310e-03) (hash(x)=24955630)
378
+ 360 train 8.002850 (lr=2.9291e-03) (hash(x)=21368780)
379
+ 361 train 7.908947 (lr=2.9272e-03) (hash(x)=25541230)
380
+ 362 train 7.867422 (lr=2.9252e-03) (hash(x)=23098806)
381
+ 363 train 7.824045 (lr=2.9233e-03) (hash(x)=23501481)
382
+ 364 train 7.841673 (lr=2.9213e-03) (hash(x)=23802681)
383
+ 365 train 7.885201 (lr=2.9193e-03) (hash(x)=22754988)
384
+ 366 train 7.800737 (lr=2.9172e-03) (hash(x)=22183009)
385
+ 367 train 7.790150 (lr=2.9152e-03) (hash(x)=23475070)
386
+ 368 train 7.947305 (lr=2.9131e-03) (hash(x)=23359061)
387
+ 369 train 7.948391 (lr=2.9110e-03) (hash(x)=23609959)
388
+ 370 train 7.904934 (lr=2.9088e-03) (hash(x)=22408449)
389
+ 371 train 7.793416 (lr=2.9067e-03) (hash(x)=23441847)
390
+ 372 train 7.832340 (lr=2.9045e-03) (hash(x)=23676365)
391
+ 373 train 7.978876 (lr=2.9023e-03) (hash(x)=23096292)
392
+ 374 train 8.016122 (lr=2.9001e-03) (hash(x)=27706294)
393
+ 375 train 7.901886 (lr=2.8978e-03) (hash(x)=26171093)
394
+ 376 train 8.060016 (lr=2.8955e-03) (hash(x)=34946941)
395
+ 377 train 7.888116 (lr=2.8932e-03) (hash(x)=20060838)
396
+ 378 train 7.768298 (lr=2.8909e-03) (hash(x)=21591702)
397
+ 379 train 7.852520 (lr=2.8885e-03) (hash(x)=25451055)
398
+ 380 train 7.496850 (lr=2.8862e-03) (hash(x)=18691455)
399
+ 381 train 7.825060 (lr=2.8838e-03) (hash(x)=24934792)
400
+ 382 train 7.783340 (lr=2.8813e-03) (hash(x)=23864361)
401
+ 383 train 7.789541 (lr=2.8789e-03) (hash(x)=25367597)
402
+ 384 train 7.920544 (lr=2.8764e-03) (hash(x)=27016702)
403
+ 385 train 7.914216 (lr=2.8739e-03) (hash(x)=25022148)
404
+ 386 train 7.865674 (lr=2.8714e-03) (hash(x)=24459012)
405
+ 387 train 7.867592 (lr=2.8689e-03) (hash(x)=25431365)
406
+ 388 train 7.821844 (lr=2.8663e-03) (hash(x)=25958785)
407
+ 389 train 7.980306 (lr=2.8637e-03) (hash(x)=29307788)
408
+ 390 train 7.783906 (lr=2.8611e-03) (hash(x)=24965435)
409
+ 391 train 7.761221 (lr=2.8585e-03) (hash(x)=25651929)
410
+ 392 train 7.830959 (lr=2.8558e-03) (hash(x)=28451031)
411
+ 393 train 7.876413 (lr=2.8531e-03) (hash(x)=26312387)
412
+ 394 train 7.842983 (lr=2.8504e-03) (hash(x)=25326650)
413
+ 395 train 7.790881 (lr=2.8477e-03) (hash(x)=23804913)
414
+ 396 train 7.928434 (lr=2.8449e-03) (hash(x)=25537529)
415
+ 397 train 7.775769 (lr=2.8422e-03) (hash(x)=26115519)
416
+ 398 train 7.772520 (lr=2.8394e-03) (hash(x)=25494804)
417
+ 399 train 7.905946 (lr=2.8365e-03) (hash(x)=23210747)
418
+ 400 val loss 7.8397
419
+ 400 val perplexity 2539.4446
420
+ 400 train 7.780272 (lr=2.8337e-03) (hash(x)=15717069)
421
+ 401 train 7.714265 (lr=2.8308e-03) (hash(x)=18463712)
422
+ 402 train 7.795696 (lr=2.8279e-03) (hash(x)=20705573)
423
+ 403 train 7.802354 (lr=2.8250e-03) (hash(x)=23245699)
424
+ 404 train 7.810330 (lr=2.8221e-03) (hash(x)=23444521)
425
+ 405 train 7.718651 (lr=2.8191e-03) (hash(x)=21770300)
426
+ 406 train 8.381855 (lr=2.8162e-03) (hash(x)=23334569)
427
+ 407 train 7.648727 (lr=2.8131e-03) (hash(x)=25387532)
428
+ 408 train 7.797783 (lr=2.8101e-03) (hash(x)=29466743)
429
+ 409 train 7.831183 (lr=2.8071e-03) (hash(x)=27718703)
430
+ 410 train 7.858255 (lr=2.8040e-03) (hash(x)=17631973)
431
+ 411 train 7.876110 (lr=2.8009e-03) (hash(x)=27661052)
432
+ 412 train 7.701838 (lr=2.7978e-03) (hash(x)=22244096)
433
+ 413 train 7.841917 (lr=2.7947e-03) (hash(x)=26538928)
434
+ 414 train 7.841354 (lr=2.7915e-03) (hash(x)=26256802)
435
+ 415 train 7.903691 (lr=2.7883e-03) (hash(x)=21772654)
436
+ 416 train 7.879428 (lr=2.7851e-03) (hash(x)=25026277)
437
+ 417 train 8.014720 (lr=2.7819e-03) (hash(x)=26263254)
438
+ 418 train 7.943405 (lr=2.7786e-03) (hash(x)=24726314)
439
+ 419 train 7.758205 (lr=2.7754e-03) (hash(x)=23552356)
440
+ 420 train 7.881383 (lr=2.7721e-03) (hash(x)=26699819)
441
+ 421 train 8.170203 (lr=2.7688e-03) (hash(x)=22707141)
442
+ 422 train 8.194406 (lr=2.7654e-03) (hash(x)=25676149)
443
+ 423 train 7.984984 (lr=2.7621e-03) (hash(x)=34234481)
444
+ 424 train 7.845109 (lr=2.7587e-03) (hash(x)=25401624)
445
+ 425 train 7.981218 (lr=2.7553e-03) (hash(x)=35008670)
446
+ 426 train 8.044947 (lr=2.7519e-03) (hash(x)=25682280)
447
+ 427 train 7.886195 (lr=2.7484e-03) (hash(x)=25891543)
448
+ 428 train 7.822297 (lr=2.7450e-03) (hash(x)=25962407)
449
+ 429 train 7.717039 (lr=2.7415e-03) (hash(x)=21971224)
450
+ 430 train 7.802053 (lr=2.7380e-03) (hash(x)=30838142)
451
+ 431 train 7.866959 (lr=2.7344e-03) (hash(x)=27312186)
452
+ 432 train 7.937634 (lr=2.7309e-03) (hash(x)=23657081)
453
+ 433 train 7.875461 (lr=2.7273e-03) (hash(x)=22163749)
454
+ 434 train 7.614747 (lr=2.7237e-03) (hash(x)=24829818)
455
+ 435 train 7.861908 (lr=2.7201e-03) (hash(x)=25942888)
456
+ 436 train 7.794519 (lr=2.7165e-03) (hash(x)=23785671)
457
+ 437 train 7.796774 (lr=2.7128e-03) (hash(x)=25493916)
458
+ 438 train 7.956283 (lr=2.7092e-03) (hash(x)=28618702)
459
+ 439 train 7.990834 (lr=2.7055e-03) (hash(x)=24289951)
460
+ 440 train 7.783384 (lr=2.7018e-03) (hash(x)=26331486)
461
+ 441 train 7.752622 (lr=2.6980e-03) (hash(x)=23648671)
462
+ 442 train 7.711807 (lr=2.6943e-03) (hash(x)=27292797)
463
+ 443 train 7.551635 (lr=2.6905e-03) (hash(x)=24992055)
464
+ 444 train 8.032678 (lr=2.6867e-03) (hash(x)=24398330)
465
+ 445 train 7.786927 (lr=2.6829e-03) (hash(x)=26005549)
466
+ 446 train 7.775572 (lr=2.6791e-03) (hash(x)=30259737)
467
+ 447 train 7.777521 (lr=2.6752e-03) (hash(x)=21046554)
468
+ 448 train 7.771657 (lr=2.6713e-03) (hash(x)=18773877)
469
+ 449 train 7.807592 (lr=2.6674e-03) (hash(x)=26321539)
470
+ 450 val loss 7.8002
471
+ 450 val perplexity 2441.1553
472
+ 450 train 7.788194 (lr=2.6635e-03) (hash(x)=22850804)
473
+ 451 train 7.847254 (lr=2.6596e-03) (hash(x)=27807872)
474
+ 452 train 7.809424 (lr=2.6556e-03) (hash(x)=25545137)
475
+ 453 train 7.825150 (lr=2.6517e-03) (hash(x)=21555364)
476
+ 454 train 7.851509 (lr=2.6477e-03) (hash(x)=27243798)
477
+ 455 train 7.713109 (lr=2.6437e-03) (hash(x)=26325324)
478
+ 456 train 7.692580 (lr=2.6396e-03) (hash(x)=26228728)
479
+ 457 train 7.905611 (lr=2.6356e-03) (hash(x)=24956294)
480
+ 458 train 7.776278 (lr=2.6315e-03) (hash(x)=25592289)
481
+ 459 train 7.734691 (lr=2.6274e-03) (hash(x)=24429389)
482
+ 460 train 7.770533 (lr=2.6233e-03) (hash(x)=28733708)
483
+ 461 train 7.906196 (lr=2.6192e-03) (hash(x)=27929846)
484
+ 462 train 7.767630 (lr=2.6150e-03) (hash(x)=22417476)
485
+ 463 train 7.859983 (lr=2.6109e-03) (hash(x)=25282897)
486
+ 464 train 7.645618 (lr=2.6067e-03) (hash(x)=22278062)
487
+ 465 train 7.647744 (lr=2.6025e-03) (hash(x)=26802508)
488
+ 466 train 7.869471 (lr=2.5983e-03) (hash(x)=22094832)
489
+ 467 train 7.829452 (lr=2.5940e-03) (hash(x)=28390416)
490
+ 468 train 7.952666 (lr=2.5898e-03) (hash(x)=26379331)
491
+ 469 train 8.004979 (lr=2.5855e-03) (hash(x)=26292451)
492
+ 470 train 7.872075 (lr=2.5812e-03) (hash(x)=26301866)
493
+ 471 train 7.884239 (lr=2.5769e-03) (hash(x)=22915344)
494
+ 472 train 7.772852 (lr=2.5726e-03) (hash(x)=25448318)
495
+ 473 train 7.828226 (lr=2.5682e-03) (hash(x)=24744383)
496
+ 474 train 7.654905 (lr=2.5639e-03) (hash(x)=22835201)
497
+ 475 train 7.778991 (lr=2.5595e-03) (hash(x)=26284678)
498
+ 476 train 7.788090 (lr=2.5551e-03) (hash(x)=24745936)
499
+ 477 train 7.651496 (lr=2.5507e-03) (hash(x)=21654269)
500
+ 478 train 7.846324 (lr=2.5462e-03) (hash(x)=32387999)
501
+ 479 train 8.457752 (lr=2.5418e-03) (hash(x)=29466268)
502
+ 480 train 8.429527 (lr=2.5373e-03) (hash(x)=30736727)
503
+ 481 train 8.409532 (lr=2.5328e-03) (hash(x)=32616762)
504
+ 482 train 8.399527 (lr=2.5283e-03) (hash(x)=29358578)
505
+ 483 train 8.442266 (lr=2.5238e-03) (hash(x)=32201138)
506
+ 484 train 8.443899 (lr=2.5193e-03) (hash(x)=30041621)
507
+ 485 train 8.399940 (lr=2.5147e-03) (hash(x)=33683468)
508
+ 486 train 8.158892 (lr=2.5102e-03) (hash(x)=26837755)
509
+ 487 train 7.997540 (lr=2.5056e-03) (hash(x)=27498125)
510
+ 488 train 8.113495 (lr=2.5010e-03) (hash(x)=22372507)
511
+ 489 train 7.864640 (lr=2.4963e-03) (hash(x)=25164318)
512
+ 490 train 7.863256 (lr=2.4917e-03) (hash(x)=26562947)
513
+ 491 train 7.807984 (lr=2.4871e-03) (hash(x)=25690221)
514
+ 492 train 7.833689 (lr=2.4824e-03) (hash(x)=21947305)
515
+ 493 train 7.809970 (lr=2.4777e-03) (hash(x)=29031550)
516
+ 494 train 7.958906 (lr=2.4730e-03) (hash(x)=26551589)
517
+ 495 train 7.982282 (lr=2.4683e-03) (hash(x)=24094093)
518
+ 496 train 7.813733 (lr=2.4636e-03) (hash(x)=23639076)
519
+ 497 train 9.512466 (lr=2.4588e-03) (hash(x)=61455960)
520
+ 498 train 11.763443 (lr=2.4540e-03) (hash(x)=65636572)
521
+ 499 train 11.569078 (lr=2.4493e-03) (hash(x)=61104710)
522
+ 500 val loss 7.8332
523
+ 500 val perplexity 2522.9082
524
+ 500 train 8.424707 (lr=2.4445e-03) (hash(x)=24123821)
525
+ 501 train 7.481571 (lr=2.4397e-03) (hash(x)=20881126)
526
+ 502 train 7.883404 (lr=2.4348e-03) (hash(x)=23209969)
527
+ 503 train 7.938554 (lr=2.4300e-03) (hash(x)=26920243)
528
+ 504 train 7.879432 (lr=2.4251e-03) (hash(x)=26127756)
529
+ 505 train 7.897921 (lr=2.4203e-03) (hash(x)=21970343)
530
+ 506 train 7.981218 (lr=2.4154e-03) (hash(x)=24097897)
531
+ 507 train 8.110733 (lr=2.4105e-03) (hash(x)=31676352)
532
+ 508 train 8.118727 (lr=2.4056e-03) (hash(x)=23908053)
533
+ 509 train 8.094950 (lr=2.4006e-03) (hash(x)=25600562)
534
+ 510 train 8.161565 (lr=2.3957e-03) (hash(x)=23549168)
535
+ 511 train 8.082437 (lr=2.3907e-03) (hash(x)=22022924)
536
+ 512 train 8.068858 (lr=2.3858e-03) (hash(x)=23294980)
537
+ 513 train 7.912103 (lr=2.3808e-03) (hash(x)=22777085)
538
+ 514 train 8.117293 (lr=2.3758e-03) (hash(x)=26271115)
539
+ 515 train 7.919766 (lr=2.3708e-03) (hash(x)=21727744)
540
+ 516 train 7.892133 (lr=2.3657e-03) (hash(x)=22729318)
541
+ 517 train 7.951366 (lr=2.3607e-03) (hash(x)=20701998)
542
+ 518 train 7.932710 (lr=2.3556e-03) (hash(x)=27928398)
543
+ 519 train 7.939524 (lr=2.3506e-03) (hash(x)=20863982)
544
+ 520 train 7.841370 (lr=2.3455e-03) (hash(x)=22827054)
545
+ 521 train 7.883169 (lr=2.3404e-03) (hash(x)=25084926)
546
+ 522 train 7.994566 (lr=2.3353e-03) (hash(x)=25540961)
547
+ 523 train 7.800932 (lr=2.3301e-03) (hash(x)=24412897)
548
+ 524 train 8.054195 (lr=2.3250e-03) (hash(x)=26269869)
549
+ 525 train 7.839296 (lr=2.3198e-03) (hash(x)=25434913)
550
+ 526 train 7.997049 (lr=2.3147e-03) (hash(x)=25157653)
551
+ 527 train 7.996036 (lr=2.3095e-03) (hash(x)=26302544)
552
+ 528 train 7.885283 (lr=2.3043e-03) (hash(x)=22010703)
553
+ 529 train 8.107448 (lr=2.2991e-03) (hash(x)=34976197)
554
+ 530 train 7.837827 (lr=2.2939e-03) (hash(x)=28248301)
555
+ 531 train 7.838985 (lr=2.2887e-03) (hash(x)=23870805)
556
+ 532 train 7.819644 (lr=2.2834e-03) (hash(x)=24457252)
557
+ 533 train 7.874501 (lr=2.2782e-03) (hash(x)=25041103)
558
+ 534 train 7.815838 (lr=2.2729e-03) (hash(x)=25998487)
559
+ 535 train 7.895395 (lr=2.2676e-03) (hash(x)=26041745)
560
+ 536 train 7.806741 (lr=2.2624e-03) (hash(x)=27249810)
561
+ 537 train 7.854190 (lr=2.2571e-03) (hash(x)=24068619)
562
+ 538 train 7.879787 (lr=2.2517e-03) (hash(x)=26359528)
563
+ 539 train 7.825990 (lr=2.2464e-03) (hash(x)=25441262)
564
+ 540 train 7.782060 (lr=2.2411e-03) (hash(x)=24413659)
565
+ 541 train 8.123633 (lr=2.2357e-03) (hash(x)=29181867)
566
+ 542 train 7.924997 (lr=2.2304e-03) (hash(x)=31551575)
567
+ 543 train 7.916555 (lr=2.2250e-03) (hash(x)=27093757)
568
+ 544 train 7.866809 (lr=2.2196e-03) (hash(x)=23624298)
569
+ 545 train 7.744355 (lr=2.2142e-03) (hash(x)=25150275)
570
+ 546 train 7.748721 (lr=2.2088e-03) (hash(x)=26239886)
571
+ 547 train 8.082330 (lr=2.2034e-03) (hash(x)=33457102)
572
+ 548 train 7.709172 (lr=2.1980e-03) (hash(x)=17499738)
573
+ 549 train 7.777313 (lr=2.1926e-03) (hash(x)=25755322)
574
+ 550 val loss 7.8008
575
+ 550 val perplexity 2442.6189
576
+ 550 train 7.859449 (lr=2.1871e-03) (hash(x)=23722795)
577
+ 551 train 7.755263 (lr=2.1817e-03) (hash(x)=23630794)
578
+ 552 train 7.723437 (lr=2.1762e-03) (hash(x)=23374168)
579
+ 553 train 7.750372 (lr=2.1707e-03) (hash(x)=25167987)
580
+ 554 train 7.967693 (lr=2.1653e-03) (hash(x)=30941010)
581
+ 555 train 7.731464 (lr=2.1598e-03) (hash(x)=24043224)
582
+ 556 train 7.879988 (lr=2.1542e-03) (hash(x)=29736627)
583
+ 557 train 7.828713 (lr=2.1487e-03) (hash(x)=24665411)
584
+ 558 train 7.856012 (lr=2.1432e-03) (hash(x)=27916043)
585
+ 559 train 7.788914 (lr=2.1377e-03) (hash(x)=17321657)
586
+ 560 train 7.623526 (lr=2.1321e-03) (hash(x)=23829833)
587
+ 561 train 8.000790 (lr=2.1266e-03) (hash(x)=27932591)
588
+ 562 train 8.077153 (lr=2.1210e-03) (hash(x)=28206079)
589
+ 563 train 8.141382 (lr=2.1154e-03) (hash(x)=29447954)
590
+ 564 train 7.817850 (lr=2.1099e-03) (hash(x)=25935364)
591
+ 565 train 7.864541 (lr=2.1043e-03) (hash(x)=24621816)
592
+ 566 train 7.768876 (lr=2.0987e-03) (hash(x)=24070405)
593
+ 567 train 7.862611 (lr=2.0931e-03) (hash(x)=24682294)
594
+ 568 train 7.868080 (lr=2.0875e-03) (hash(x)=24050209)
595
+ 569 train 7.820239 (lr=2.0818e-03) (hash(x)=28057095)
596
+ 570 train 7.904572 (lr=2.0762e-03) (hash(x)=22523232)
597
+ 571 train 7.833690 (lr=2.0706e-03) (hash(x)=24547536)
598
+ 572 train 7.742763 (lr=2.0649e-03) (hash(x)=24978712)
599
+ 573 train 7.708252 (lr=2.0593e-03) (hash(x)=18987545)
600
+ 574 train 7.798075 (lr=2.0536e-03) (hash(x)=24092784)
601
+ 575 train 7.780528 (lr=2.0479e-03) (hash(x)=26745063)
602
+ 576 train 7.779048 (lr=2.0422e-03) (hash(x)=22539112)
603
+ 577 train 7.572876 (lr=2.0366e-03) (hash(x)=21447406)
604
+ 578 train 7.597929 (lr=2.0309e-03) (hash(x)=21082139)
605
+ 579 train 7.635394 (lr=2.0252e-03) (hash(x)=24650655)
606
+ 580 train 7.821496 (lr=2.0194e-03) (hash(x)=22130721)
607
+ 581 train 8.105957 (lr=2.0137e-03) (hash(x)=26519976)
608
+ 582 train 8.632696 (lr=2.0080e-03) (hash(x)=25582440)
609
+ 583 train 7.883629 (lr=2.0023e-03) (hash(x)=20106514)
610
+ 584 train 7.701979 (lr=1.9965e-03) (hash(x)=25695510)
611
+ 585 train 7.723776 (lr=1.9908e-03) (hash(x)=20117808)
612
+ 586 train 7.804513 (lr=1.9850e-03) (hash(x)=24501900)
613
+ 587 train 7.896025 (lr=1.9793e-03) (hash(x)=25510334)
614
+ 588 train 7.742244 (lr=1.9735e-03) (hash(x)=22099158)
615
+ 589 train 7.847929 (lr=1.9677e-03) (hash(x)=25749120)
616
+ 590 train 8.105256 (lr=1.9620e-03) (hash(x)=21720691)
617
+ 591 train 7.694623 (lr=1.9562e-03) (hash(x)=24357241)
618
+ 592 train 7.981023 (lr=1.9504e-03) (hash(x)=28911272)
619
+ 593 train 7.835213 (lr=1.9446e-03) (hash(x)=25584945)
620
+ 594 train 7.718126 (lr=1.9388e-03) (hash(x)=21847287)
621
+ 595 train 7.772123 (lr=1.9330e-03) (hash(x)=25198897)
622
+ 596 train 7.684315 (lr=1.9272e-03) (hash(x)=22436833)
623
+ 597 train 7.974393 (lr=1.9214e-03) (hash(x)=23475891)
624
+ 598 train 7.628492 (lr=1.9156e-03) (hash(x)=25083992)
625
+ 599 train 7.868316 (lr=1.9097e-03) (hash(x)=23644804)
626
+ 600 val loss 7.8109
627
+ 600 val perplexity 2467.3823
628
+ 600 train 7.768964 (lr=1.9039e-03) (hash(x)=31377168)
629
+ 601 train 8.139249 (lr=1.8981e-03) (hash(x)=25629584)
630
+ 602 train 7.955036 (lr=1.8922e-03) (hash(x)=29285135)
631
+ 603 train 8.149180 (lr=1.8864e-03) (hash(x)=29797280)
632
+ 604 train 7.965142 (lr=1.8805e-03) (hash(x)=33122326)
633
+ 605 train 8.241395 (lr=1.8747e-03) (hash(x)=30026394)
634
+ 606 train 8.521984 (lr=1.8688e-03) (hash(x)=34353164)
635
+ 607 train 7.761221 (lr=1.8629e-03) (hash(x)=26594177)
636
+ 608 train 7.833024 (lr=1.8571e-03) (hash(x)=25554738)
637
+ 609 train 7.922995 (lr=1.8512e-03) (hash(x)=23766256)
638
+ 610 train 7.798311 (lr=1.8453e-03) (hash(x)=26853444)
639
+ 611 train 7.802524 (lr=1.8395e-03) (hash(x)=24250150)
640
+ 612 train 7.750558 (lr=1.8336e-03) (hash(x)=26289588)
641
+ 613 train 7.808858 (lr=1.8277e-03) (hash(x)=26108817)
642
+ 614 train 7.734978 (lr=1.8218e-03) (hash(x)=23814987)
643
+ 615 train 7.740031 (lr=1.8159e-03) (hash(x)=18582688)
644
+ 616 train 7.739956 (lr=1.8100e-03) (hash(x)=25558792)
645
+ 617 train 7.698025 (lr=1.8041e-03) (hash(x)=23272705)
646
+ 618 train 7.709919 (lr=1.7982e-03) (hash(x)=24709241)
647
+ 619 train 7.749500 (lr=1.7923e-03) (hash(x)=24793412)
648
+ 620 train 7.763094 (lr=1.7864e-03) (hash(x)=27256707)
649
+ 621 train 7.711329 (lr=1.7805e-03) (hash(x)=19280989)
650
+ 622 train 7.786348 (lr=1.7746e-03) (hash(x)=26508642)
651
+ 623 train 7.710653 (lr=1.7686e-03) (hash(x)=22751150)
652
+ 624 train 7.943564 (lr=1.7627e-03) (hash(x)=23880160)
653
+ 625 train 7.731124 (lr=1.7568e-03) (hash(x)=23537942)
654
+ 626 train 7.757173 (lr=1.7509e-03) (hash(x)=26975313)
655
+ 627 train 7.622396 (lr=1.7450e-03) (hash(x)=23748375)
656
+ 628 train 7.695151 (lr=1.7390e-03) (hash(x)=25541754)
657
+ 629 train 8.107981 (lr=1.7331e-03) (hash(x)=27413825)
658
+ 630 train 8.069094 (lr=1.7272e-03) (hash(x)=27075951)
659
+ 631 train 7.919183 (lr=1.7212e-03) (hash(x)=21682445)
660
+ 632 train 7.825186 (lr=1.7153e-03) (hash(x)=23361504)
661
+ 633 train 7.750748 (lr=1.7094e-03) (hash(x)=22040184)
662
+ 634 train 7.796831 (lr=1.7034e-03) (hash(x)=23727731)
663
+ 635 train 7.776978 (lr=1.6975e-03) (hash(x)=24463229)
664
+ 636 train 7.967410 (lr=1.6916e-03) (hash(x)=32663792)
665
+ 637 train 7.792068 (lr=1.6856e-03) (hash(x)=24091954)
666
+ 638 train 7.728922 (lr=1.6797e-03) (hash(x)=23599013)
667
+ 639 train 7.768858 (lr=1.6738e-03) (hash(x)=26670219)
668
+ 640 train 7.777214 (lr=1.6678e-03) (hash(x)=24380893)
669
+ 641 train 7.832374 (lr=1.6619e-03) (hash(x)=25558986)
670
+ 642 train 7.934705 (lr=1.6559e-03) (hash(x)=25729492)
671
+ 643 train 7.677977 (lr=1.6500e-03) (hash(x)=19218943)
672
+ 644 train 7.877739 (lr=1.6441e-03) (hash(x)=23952206)
673
+ 645 train 7.860806 (lr=1.6381e-03) (hash(x)=24551402)
674
+ 646 train 7.735737 (lr=1.6322e-03) (hash(x)=24007642)
675
+ 647 train 7.794797 (lr=1.6262e-03) (hash(x)=25229281)
676
+ 648 train 7.558455 (lr=1.6203e-03) (hash(x)=18214283)
677
+ 649 train 7.530916 (lr=1.6144e-03) (hash(x)=21251127)
678
+ 650 val loss 7.8223
679
+ 650 val perplexity 2495.5376
680
+ 650 train 8.019417 (lr=1.6084e-03) (hash(x)=25097619)
681
+ 651 train 7.695148 (lr=1.6025e-03) (hash(x)=25800857)
682
+ 652 train 7.695196 (lr=1.5966e-03) (hash(x)=22918428)
683
+ 653 train 7.778236 (lr=1.5906e-03) (hash(x)=25836189)
684
+ 654 train 7.730285 (lr=1.5847e-03) (hash(x)=24504567)
685
+ 655 train 7.731892 (lr=1.5788e-03) (hash(x)=25999255)
686
+ 656 train 7.728394 (lr=1.5728e-03) (hash(x)=27308968)
687
+ 657 train 7.813442 (lr=1.5669e-03) (hash(x)=20149394)
688
+ 658 train 7.783248 (lr=1.5610e-03) (hash(x)=24498556)
689
+ 659 train 7.641314 (lr=1.5550e-03) (hash(x)=21999338)
690
+ 660 train 7.647607 (lr=1.5491e-03) (hash(x)=23280568)
691
+ 661 train 7.930717 (lr=1.5432e-03) (hash(x)=27611302)
692
+ 662 train 7.764288 (lr=1.5373e-03) (hash(x)=23982308)
693
+ 663 train 7.728679 (lr=1.5314e-03) (hash(x)=22778356)
694
+ 664 train 7.653712 (lr=1.5254e-03) (hash(x)=23467046)
695
+ 665 train 7.826267 (lr=1.5195e-03) (hash(x)=24195959)
696
+ 666 train 8.109864 (lr=1.5136e-03) (hash(x)=32710438)
697
+ 667 train 7.847405 (lr=1.5077e-03) (hash(x)=19127465)
698
+ 668 train 7.960423 (lr=1.5018e-03) (hash(x)=28962580)
699
+ 669 train 7.707351 (lr=1.4959e-03) (hash(x)=22426274)
700
+ 670 train 7.738353 (lr=1.4900e-03) (hash(x)=25606512)
701
+ 671 train 7.600817 (lr=1.4841e-03) (hash(x)=23785205)
702
+ 672 train 7.733338 (lr=1.4782e-03) (hash(x)=24779414)
703
+ 673 train 7.845780 (lr=1.4723e-03) (hash(x)=28505676)
704
+ 674 train 7.933722 (lr=1.4664e-03) (hash(x)=29156288)
705
+ 675 train 7.628394 (lr=1.4605e-03) (hash(x)=27706477)
706
+ 676 train 7.755944 (lr=1.4547e-03) (hash(x)=23711800)
707
+ 677 train 7.855812 (lr=1.4488e-03) (hash(x)=28684175)
708
+ 678 train 7.708941 (lr=1.4429e-03) (hash(x)=25225466)
709
+ 679 train 7.847448 (lr=1.4371e-03) (hash(x)=28003600)
710
+ 680 train 7.769994 (lr=1.4312e-03) (hash(x)=27989890)
711
+ 681 train 7.804187 (lr=1.4253e-03) (hash(x)=27586845)
712
+ 682 train 7.768234 (lr=1.4195e-03) (hash(x)=22373927)
713
+ 683 train 7.643310 (lr=1.4136e-03) (hash(x)=21176405)
714
+ 684 train 7.619092 (lr=1.4078e-03) (hash(x)=23806052)
715
+ 685 train 7.740448 (lr=1.4019e-03) (hash(x)=25888433)
716
+ 686 train 7.851727 (lr=1.3961e-03) (hash(x)=27797727)
717
+ 687 train 7.737381 (lr=1.3903e-03) (hash(x)=27916982)
718
+ 688 train 7.711718 (lr=1.3844e-03) (hash(x)=24621793)
719
+ 689 train 7.623215 (lr=1.3786e-03) (hash(x)=24334708)
720
+ 690 train 7.793517 (lr=1.3728e-03) (hash(x)=24419143)
721
+ 691 train 7.689885 (lr=1.3670e-03) (hash(x)=25617709)
722
+ 692 train 7.658664 (lr=1.3612e-03) (hash(x)=24556034)
723
+ 693 train 8.096015 (lr=1.3554e-03) (hash(x)=25152362)
724
+ 694 train 7.784501 (lr=1.3496e-03) (hash(x)=24326376)
725
+ 695 train 7.688344 (lr=1.3438e-03) (hash(x)=24276512)
726
+ 696 train 7.540021 (lr=1.3380e-03) (hash(x)=24344695)
727
+ 697 train 7.777508 (lr=1.3323e-03) (hash(x)=26776133)
728
+ 698 train 7.628503 (lr=1.3265e-03) (hash(x)=21847282)
729
+ 699 train 7.488294 (lr=1.3207e-03) (hash(x)=20254159)
730
+ 700 val loss 7.7531
731
+ 700 val perplexity 2328.8625
732
+ 700 train 7.606972 (lr=1.3150e-03) (hash(x)=25044762)
733
+ 701 train 7.675300 (lr=1.3092e-03) (hash(x)=27873855)
734
+ 702 train 7.760502 (lr=1.3035e-03) (hash(x)=22575615)
735
+ 703 train 7.707285 (lr=1.2977e-03) (hash(x)=25113614)
736
+ 704 train 7.807489 (lr=1.2920e-03) (hash(x)=28062905)
737
+ 705 train 7.723654 (lr=1.2863e-03) (hash(x)=26192886)
738
+ 706 train 7.707694 (lr=1.2806e-03) (hash(x)=25770338)
739
+ 707 train 7.612617 (lr=1.2748e-03) (hash(x)=24614912)
740
+ 708 train 7.673173 (lr=1.2691e-03) (hash(x)=26934071)
741
+ 709 train 7.657949 (lr=1.2634e-03) (hash(x)=25397093)
742
+ 710 train 7.762601 (lr=1.2578e-03) (hash(x)=25925963)
743
+ 711 train 7.587402 (lr=1.2521e-03) (hash(x)=19811802)
744
+ 712 train 7.705266 (lr=1.2464e-03) (hash(x)=23248423)
745
+ 713 train 7.654755 (lr=1.2407e-03) (hash(x)=25579655)
746
+ 714 train 7.849772 (lr=1.2351e-03) (hash(x)=25537603)
747
+ 715 train 7.750995 (lr=1.2294e-03) (hash(x)=26096514)
748
+ 716 train 7.783896 (lr=1.2238e-03) (hash(x)=24132888)
749
+ 717 train 7.627257 (lr=1.2182e-03) (hash(x)=22589633)
750
+ 718 train 7.617990 (lr=1.2125e-03) (hash(x)=27781566)
751
+ 719 train 7.850271 (lr=1.2069e-03) (hash(x)=26757147)
752
+ 720 train 7.854634 (lr=1.2013e-03) (hash(x)=26649864)
753
+ 721 train 7.742636 (lr=1.1957e-03) (hash(x)=24963730)
754
+ 722 train 7.727001 (lr=1.1901e-03) (hash(x)=25249959)
755
+ 723 train 7.592384 (lr=1.1846e-03) (hash(x)=27703124)
756
+ 724 train 7.622954 (lr=1.1790e-03) (hash(x)=26459082)
757
+ 725 train 7.557917 (lr=1.1734e-03) (hash(x)=24781713)
758
+ 726 train 7.841195 (lr=1.1679e-03) (hash(x)=26597693)
759
+ 727 train 7.663308 (lr=1.1623e-03) (hash(x)=24720476)
760
+ 728 train 7.690596 (lr=1.1568e-03) (hash(x)=23691295)
761
+ 729 train 7.790357 (lr=1.1513e-03) (hash(x)=23481301)
762
+ 730 train 7.667342 (lr=1.1458e-03) (hash(x)=23295676)
763
+ 731 train 7.636259 (lr=1.1402e-03) (hash(x)=25641033)
764
+ 732 train 7.765248 (lr=1.1347e-03) (hash(x)=22441379)
765
+ 733 train 7.647128 (lr=1.1293e-03) (hash(x)=27369437)
766
+ 734 train 7.696704 (lr=1.1238e-03) (hash(x)=24482272)
767
+ 735 train 7.820377 (lr=1.1183e-03) (hash(x)=24760900)
768
+ 736 train 7.717622 (lr=1.1129e-03) (hash(x)=23919156)
769
+ 737 train 7.598493 (lr=1.1074e-03) (hash(x)=21692546)
770
+ 738 train 7.631771 (lr=1.1020e-03) (hash(x)=27692087)
771
+ 739 train 7.755625 (lr=1.0966e-03) (hash(x)=22720391)
772
+ 740 train 7.571711 (lr=1.0912e-03) (hash(x)=26146560)
773
+ 741 train 7.683884 (lr=1.0858e-03) (hash(x)=24376979)
774
+ 742 train 7.721035 (lr=1.0804e-03) (hash(x)=22917712)
775
+ 743 train 7.618565 (lr=1.0750e-03) (hash(x)=22244509)
776
+ 744 train 7.588663 (lr=1.0696e-03) (hash(x)=25243385)
777
+ 745 train 7.625022 (lr=1.0643e-03) (hash(x)=21831960)
778
+ 746 train 7.629993 (lr=1.0589e-03) (hash(x)=24842373)
779
+ 747 train 7.562655 (lr=1.0536e-03) (hash(x)=24921535)
780
+ 748 train 7.649724 (lr=1.0483e-03) (hash(x)=24344615)
781
+ 749 train 7.647305 (lr=1.0429e-03) (hash(x)=25381251)
782
+ 750 val loss 7.7370
783
+ 750 val perplexity 2291.6387
784
+ 750 train 7.525632 (lr=1.0376e-03) (hash(x)=22887303)
785
+ 751 train 7.744324 (lr=1.0324e-03) (hash(x)=25687352)
786
+ 752 train 7.709785 (lr=1.0271e-03) (hash(x)=27791412)
787
+ 753 train 7.613762 (lr=1.0218e-03) (hash(x)=25221654)
788
+ 754 train 7.714863 (lr=1.0166e-03) (hash(x)=23243731)
789
+ 755 train 7.698258 (lr=1.0113e-03) (hash(x)=24469863)
790
+ 756 train 7.792186 (lr=1.0061e-03) (hash(x)=22484936)
791
+ 757 train 7.729693 (lr=1.0009e-03) (hash(x)=25291413)
792
+ 758 train 7.612449 (lr=9.9568e-04) (hash(x)=23952225)
793
+ 759 train 7.706535 (lr=9.9049e-04) (hash(x)=25744274)
794
+ 760 train 7.621972 (lr=9.8531e-04) (hash(x)=22616739)
795
+ 761 train 7.583839 (lr=9.8015e-04) (hash(x)=24246963)
796
+ 762 train 7.588521 (lr=9.7500e-04) (hash(x)=19407094)
797
+ 763 train 7.680372 (lr=9.6986e-04) (hash(x)=19724058)
798
+ 764 train 7.518268 (lr=9.6474e-04) (hash(x)=18836491)
799
+ 765 train 7.569905 (lr=9.5963e-04) (hash(x)=20705649)
800
+ 766 train 7.783160 (lr=9.5453e-04) (hash(x)=21325875)
801
+ 767 train 7.817092 (lr=9.4944e-04) (hash(x)=24938685)
802
+ 768 train 7.749249 (lr=9.4437e-04) (hash(x)=25152403)
803
+ 769 train 7.893652 (lr=9.3932e-04) (hash(x)=21707000)
804
+ 770 train 7.578857 (lr=9.3427e-04) (hash(x)=17952018)
805
+ 771 train 7.621085 (lr=9.2924e-04) (hash(x)=24948980)
806
+ 772 train 8.082137 (lr=9.2423e-04) (hash(x)=28012110)
807
+ 773 train 7.726658 (lr=9.1923e-04) (hash(x)=30773958)
808
+ 774 train 8.291218 (lr=9.1424e-04) (hash(x)=29481068)
809
+ 775 train 7.720868 (lr=9.0927e-04) (hash(x)=26769046)
810
+ 776 train 7.645037 (lr=9.0431e-04) (hash(x)=26755048)
811
+ 777 train 7.595048 (lr=8.9936e-04) (hash(x)=18849656)
812
+ 778 train 7.891797 (lr=8.9443e-04) (hash(x)=25449800)
813
+ 779 train 8.378242 (lr=8.8952e-04) (hash(x)=24624950)
814
+ 780 train 7.659480 (lr=8.8462e-04) (hash(x)=26174069)
815
+ 781 train 7.848724 (lr=8.7973e-04) (hash(x)=27299605)
816
+ 782 train 7.780833 (lr=8.7486e-04) (hash(x)=24777273)
817
+ 783 train 7.780715 (lr=8.7001e-04) (hash(x)=24113445)
818
+ 784 train 7.639443 (lr=8.6516e-04) (hash(x)=23970384)
819
+ 785 train 7.877852 (lr=8.6034e-04) (hash(x)=32644465)
820
+ 786 train 8.018545 (lr=8.5553e-04) (hash(x)=29951548)
821
+ 787 train 7.730916 (lr=8.5073e-04) (hash(x)=25829219)
822
+ 788 train 7.778768 (lr=8.4596e-04) (hash(x)=20556094)
823
+ 789 train 7.625814 (lr=8.4119e-04) (hash(x)=25194550)
824
+ 790 train 7.743384 (lr=8.3644e-04) (hash(x)=28482186)
825
+ 791 train 7.727232 (lr=8.3171e-04) (hash(x)=25941804)
826
+ 792 train 7.717225 (lr=8.2699e-04) (hash(x)=27049208)
827
+ 793 train 7.741306 (lr=8.2229e-04) (hash(x)=24075727)
828
+ 794 train 7.799503 (lr=8.1761e-04) (hash(x)=25082806)
829
+ 795 train 7.768909 (lr=8.1294e-04) (hash(x)=23104539)
830
+ 796 train 7.676000 (lr=8.0829e-04) (hash(x)=25351113)
831
+ 797 train 7.768722 (lr=8.0365e-04) (hash(x)=23018983)
832
+ 798 train 7.713368 (lr=7.9903e-04) (hash(x)=25302106)
833
+ 799 train 7.641840 (lr=7.9443e-04) (hash(x)=24422929)
834
+ 800 val loss 7.7316
835
+ 800 val perplexity 2279.3098
836
+ 800 train 7.700654 (lr=7.8984e-04) (hash(x)=29052117)
837
+ 801 train 7.710088 (lr=7.8527e-04) (hash(x)=23389696)
838
+ 802 train 7.616472 (lr=7.8072e-04) (hash(x)=23962503)
839
+ 803 train 7.736886 (lr=7.7618e-04) (hash(x)=29102969)
840
+ 804 train 7.842134 (lr=7.7167e-04) (hash(x)=24948166)
841
+ 805 train 7.813540 (lr=7.6716e-04) (hash(x)=28359859)
842
+ 806 train 7.771519 (lr=7.6268e-04) (hash(x)=22360298)
843
+ 807 train 7.902028 (lr=7.5821e-04) (hash(x)=24630955)
844
+ 808 train 7.889272 (lr=7.5376e-04) (hash(x)=26897402)
845
+ 809 train 7.831929 (lr=7.4933e-04) (hash(x)=22969113)
846
+ 810 train 7.650833 (lr=7.4491e-04) (hash(x)=25205781)
847
+ 811 train 7.693582 (lr=7.4051e-04) (hash(x)=27004415)
848
+ 812 train 7.551185 (lr=7.3613e-04) (hash(x)=22974875)
849
+ 813 train 7.660833 (lr=7.3177e-04) (hash(x)=22265063)
850
+ 814 train 7.587789 (lr=7.2742e-04) (hash(x)=23312114)
851
+ 815 train 7.706223 (lr=7.2309e-04) (hash(x)=26905990)
852
+ 816 train 7.698508 (lr=7.1878e-04) (hash(x)=27089280)
853
+ 817 train 7.657216 (lr=7.1449e-04) (hash(x)=24047679)
854
+ 818 train 7.730517 (lr=7.1022e-04) (hash(x)=25034684)
855
+ 819 train 7.899221 (lr=7.0596e-04) (hash(x)=26905582)
856
+ 820 train 7.570920 (lr=7.0173e-04) (hash(x)=21406950)
857
+ 821 train 7.593391 (lr=6.9751e-04) (hash(x)=29018970)
858
+ 822 train 7.571058 (lr=6.9331e-04) (hash(x)=20853530)
859
+ 823 train 7.651440 (lr=6.8913e-04) (hash(x)=24503315)
860
+ 824 train 7.665113 (lr=6.8496e-04) (hash(x)=23080074)
861
+ 825 train 7.521839 (lr=6.8082e-04) (hash(x)=22607537)
862
+ 826 train 7.899835 (lr=6.7669e-04) (hash(x)=32179773)
863
+ 827 train 9.006631 (lr=6.7259e-04) (hash(x)=26690225)
864
+ 828 train 7.695398 (lr=6.6850e-04) (hash(x)=22109609)
865
+ 829 train 7.670384 (lr=6.6443e-04) (hash(x)=24226255)
866
+ 830 train 7.712023 (lr=6.6038e-04) (hash(x)=23013573)
867
+ 831 train 7.518550 (lr=6.5635e-04) (hash(x)=23622702)
868
+ 832 train 7.724260 (lr=6.5234e-04) (hash(x)=21517429)
869
+ 833 train 7.634324 (lr=6.4835e-04) (hash(x)=23910425)
870
+ 834 train 7.770034 (lr=6.4437e-04) (hash(x)=24407484)
871
+ 835 train 8.068078 (lr=6.4042e-04) (hash(x)=27213318)
872
+ 836 train 7.683390 (lr=6.3649e-04) (hash(x)=25524191)
873
+ 837 train 7.731456 (lr=6.3257e-04) (hash(x)=26701577)
874
+ 838 train 7.553108 (lr=6.2868e-04) (hash(x)=22091266)
875
+ 839 train 7.680394 (lr=6.2480e-04) (hash(x)=26216794)
876
+ 840 train 7.616412 (lr=6.2095e-04) (hash(x)=20817340)
877
+ 841 train 8.045223 (lr=6.1711e-04) (hash(x)=27579623)
878
+ 842 train 7.565259 (lr=6.1330e-04) (hash(x)=21532187)
879
+ 843 train 7.835340 (lr=6.0950e-04) (hash(x)=26086198)
880
+ 844 train 7.682477 (lr=6.0573e-04) (hash(x)=23911729)
881
+ 845 train 7.660575 (lr=6.0198e-04) (hash(x)=24863441)
882
+ 846 train 8.091989 (lr=5.9824e-04) (hash(x)=28856978)
883
+ 847 train 7.511626 (lr=5.9453e-04) (hash(x)=25236385)
884
+ 848 train 7.804497 (lr=5.9083e-04) (hash(x)=24278687)
885
+ 849 train 7.928133 (lr=5.8716e-04) (hash(x)=25330803)
886
+ 850 val loss 7.7240
887
+ 850 val perplexity 2262.0801
888
+ 850 train 7.782387 (lr=5.8351e-04) (hash(x)=21568545)
889
+ 851 train 7.679968 (lr=5.7988e-04) (hash(x)=22508532)
890
+ 852 train 7.740410 (lr=5.7627e-04) (hash(x)=24850662)
891
+ 853 train 7.713260 (lr=5.7268e-04) (hash(x)=21799261)
892
+ 854 train 7.862616 (lr=5.6911e-04) (hash(x)=24884891)
893
+ 855 train 7.715223 (lr=5.6556e-04) (hash(x)=26311168)
894
+ 856 train 7.762141 (lr=5.6203e-04) (hash(x)=27549895)
895
+ 857 train 7.463682 (lr=5.5853e-04) (hash(x)=20977430)
896
+ 858 train 7.577446 (lr=5.5504e-04) (hash(x)=23448855)
897
+ 859 train 7.779460 (lr=5.5158e-04) (hash(x)=26941617)
898
+ 860 train 7.986927 (lr=5.4813e-04) (hash(x)=26953192)
899
+ 861 train 7.860533 (lr=5.4471e-04) (hash(x)=30008957)
900
+ 862 train 7.801241 (lr=5.4131e-04) (hash(x)=27069893)
901
+ 863 train 7.733221 (lr=5.3793e-04) (hash(x)=27125962)
902
+ 864 train 7.537300 (lr=5.3458e-04) (hash(x)=23341415)
903
+ 865 train 7.639851 (lr=5.3124e-04) (hash(x)=25888358)
904
+ 866 train 7.815698 (lr=5.2793e-04) (hash(x)=26761645)
905
+ 867 train 7.719583 (lr=5.2464e-04) (hash(x)=21725719)
906
+ 868 train 7.616269 (lr=5.2137e-04) (hash(x)=25746493)
907
+ 869 train 7.624632 (lr=5.1812e-04) (hash(x)=21669704)
908
+ 870 train 7.582312 (lr=5.1489e-04) (hash(x)=21167493)
909
+ 871 train 7.684434 (lr=5.1169e-04) (hash(x)=23166993)
910
+ 872 train 7.602552 (lr=5.0851e-04) (hash(x)=23947208)
911
+ 873 train 7.737777 (lr=5.0534e-04) (hash(x)=26235974)
912
+ 874 train 8.015704 (lr=5.0221e-04) (hash(x)=29822604)
913
+ 875 train 7.618483 (lr=4.9909e-04) (hash(x)=25426430)
914
+ 876 train 7.625041 (lr=4.9600e-04) (hash(x)=26043735)
915
+ 877 train 7.551145 (lr=4.9293e-04) (hash(x)=22966314)
916
+ 878 train 7.860537 (lr=4.8988e-04) (hash(x)=27944619)
917
+ 879 train 7.620350 (lr=4.8685e-04) (hash(x)=23448505)
918
+ 880 train 7.644476 (lr=4.8385e-04) (hash(x)=24205942)
919
+ 881 train 7.555134 (lr=4.8087e-04) (hash(x)=26330693)
920
+ 882 train 7.569466 (lr=4.7791e-04) (hash(x)=24844466)
921
+ 883 train 7.540789 (lr=4.7497e-04) (hash(x)=29550596)
922
+ 884 train 7.616760 (lr=4.7206e-04) (hash(x)=20506540)
923
+ 885 train 7.630806 (lr=4.6917e-04) (hash(x)=22124892)
924
+ 886 train 7.618198 (lr=4.6630e-04) (hash(x)=24867036)
925
+ 887 train 7.641492 (lr=4.6346e-04) (hash(x)=25607640)
926
+ 888 train 7.654864 (lr=4.6063e-04) (hash(x)=26176930)
927
+ 889 train 7.413467 (lr=4.5783e-04) (hash(x)=23795894)
928
+ 890 train 7.335668 (lr=4.5506e-04) (hash(x)=23756298)
929
+ 891 train 7.539684 (lr=4.5231e-04) (hash(x)=24165449)
930
+ 892 train 7.570135 (lr=4.4958e-04) (hash(x)=26643739)
931
+ 893 train 7.740570 (lr=4.4687e-04) (hash(x)=24697756)
932
+ 894 train 7.646945 (lr=4.4419e-04) (hash(x)=24431101)
933
+ 895 train 7.791189 (lr=4.4153e-04) (hash(x)=27218899)
934
+ 896 train 7.601148 (lr=4.3889e-04) (hash(x)=29462219)
935
+ 897 train 7.640650 (lr=4.3628e-04) (hash(x)=24748522)
936
+ 898 train 7.645867 (lr=4.3369e-04) (hash(x)=24843486)
937
+ 899 train 7.770393 (lr=4.3113e-04) (hash(x)=24965317)
938
+ 900 val loss 7.7075
939
+ 900 val perplexity 2224.9780
940
+ 900 train 7.505048 (lr=4.2858e-04) (hash(x)=26584754)
941
+ 901 train 7.450182 (lr=4.2607e-04) (hash(x)=22031210)
942
+ 902 train 7.636608 (lr=4.2357e-04) (hash(x)=24860582)
943
+ 903 train 7.584274 (lr=4.2110e-04) (hash(x)=25877990)
944
+ 904 train 7.651299 (lr=4.1865e-04) (hash(x)=24970646)
945
+ 905 train 7.684777 (lr=4.1623e-04) (hash(x)=24382381)
946
+ 906 train 7.583729 (lr=4.1383e-04) (hash(x)=25221746)
947
+ 907 train 7.581177 (lr=4.1146e-04) (hash(x)=24551999)
948
+ 908 train 7.592172 (lr=4.0910e-04) (hash(x)=28151378)
949
+ 909 train 7.721892 (lr=4.0678e-04) (hash(x)=25765516)
950
+ 910 train 7.704419 (lr=4.0447e-04) (hash(x)=25630696)
951
+ 911 train 7.666819 (lr=4.0220e-04) (hash(x)=25204247)
952
+ 912 train 7.748660 (lr=3.9994e-04) (hash(x)=26326447)
953
+ 913 train 7.735210 (lr=3.9771e-04) (hash(x)=26321813)
954
+ 914 train 7.671812 (lr=3.9550e-04) (hash(x)=25400396)
955
+ 915 train 7.663969 (lr=3.9332e-04) (hash(x)=25528323)
956
+ 916 train 7.655039 (lr=3.9116e-04) (hash(x)=24035353)
957
+ 917 train 7.677423 (lr=3.8903e-04) (hash(x)=25114729)
958
+ 918 train 7.558541 (lr=3.8692e-04) (hash(x)=21634962)
959
+ 919 train 7.685916 (lr=3.8484e-04) (hash(x)=31341964)
960
+ 920 train 7.632485 (lr=3.8277e-04) (hash(x)=29172813)
961
+ 921 train 7.810055 (lr=3.8074e-04) (hash(x)=26326024)
962
+ 922 train 7.541274 (lr=3.7873e-04) (hash(x)=20140071)
963
+ 923 train 7.658505 (lr=3.7674e-04) (hash(x)=27273311)
964
+ 924 train 7.732335 (lr=3.7478e-04) (hash(x)=31276487)
965
+ 925 train 7.442649 (lr=3.7284e-04) (hash(x)=17819313)
966
+ 926 train 7.814517 (lr=3.7093e-04) (hash(x)=25843618)
967
+ 927 train 7.729936 (lr=3.6904e-04) (hash(x)=23769243)
968
+ 928 train 7.648691 (lr=3.6718e-04) (hash(x)=28731298)
969
+ 929 train 7.781154 (lr=3.6534e-04) (hash(x)=23788738)
970
+ 930 train 7.605267 (lr=3.6353e-04) (hash(x)=26599031)
971
+ 931 train 7.717694 (lr=3.6174e-04) (hash(x)=24818744)
972
+ 932 train 7.625623 (lr=3.5998e-04) (hash(x)=23053871)
973
+ 933 train 7.560546 (lr=3.5824e-04) (hash(x)=23472795)
974
+ 934 train 8.132368 (lr=3.5653e-04) (hash(x)=29748225)
975
+ 935 train 7.840744 (lr=3.5484e-04) (hash(x)=24991832)
976
+ 936 train 7.838955 (lr=3.5317e-04) (hash(x)=23681545)
977
+ 937 train 7.631949 (lr=3.5154e-04) (hash(x)=22424066)
978
+ 938 train 7.650187 (lr=3.4992e-04) (hash(x)=25262748)
979
+ 939 train 7.630650 (lr=3.4833e-04) (hash(x)=26474878)
980
+ 940 train 7.807342 (lr=3.4677e-04) (hash(x)=23930593)
981
+ 941 train 7.611388 (lr=3.4523e-04) (hash(x)=22704349)
982
+ 942 train 7.717277 (lr=3.4372e-04) (hash(x)=24801048)
983
+ 943 train 7.626333 (lr=3.4224e-04) (hash(x)=22293673)
984
+ 944 train 7.689445 (lr=3.4077e-04) (hash(x)=23311934)
985
+ 945 train 7.666940 (lr=3.3934e-04) (hash(x)=21785487)
986
+ 946 train 7.573162 (lr=3.3793e-04) (hash(x)=22981456)
987
+ 947 train 7.822468 (lr=3.3654e-04) (hash(x)=25251063)
988
+ 948 train 7.648557 (lr=3.3518e-04) (hash(x)=25796422)
989
+ 949 train 7.573318 (lr=3.3385e-04) (hash(x)=22722211)
990
+ 950 val loss 7.7103
991
+ 950 val perplexity 2231.3188
992
+ 950 train 7.653883 (lr=3.3254e-04) (hash(x)=24296755)
993
+ 951 train 7.667833 (lr=3.3125e-04) (hash(x)=24522269)
994
+ 952 train 7.680471 (lr=3.3000e-04) (hash(x)=25617184)
995
+ 953 train 7.566549 (lr=3.2876e-04) (hash(x)=23329475)
996
+ 954 train 7.682559 (lr=3.2756e-04) (hash(x)=30652062)
997
+ 955 train 7.636557 (lr=3.2638e-04) (hash(x)=24735165)
998
+ 956 train 7.438180 (lr=3.2522e-04) (hash(x)=23795686)
999
+ 957 train 7.495013 (lr=3.2409e-04) (hash(x)=27002892)
1000
+ 958 train 7.555714 (lr=3.2299e-04) (hash(x)=21189101)
1001
+ 959 train 7.481996 (lr=3.2191e-04) (hash(x)=23499349)
1002
+ 960 train 8.006815 (lr=3.2085e-04) (hash(x)=26807896)
1003
+ 961 train 7.635782 (lr=3.1983e-04) (hash(x)=22461589)
1004
+ 962 train 7.693501 (lr=3.1883e-04) (hash(x)=26904123)
1005
+ 963 train 7.534853 (lr=3.1785e-04) (hash(x)=26415040)
1006
+ 964 train 7.798769 (lr=3.1690e-04) (hash(x)=24463181)
1007
+ 965 train 7.842212 (lr=3.1598e-04) (hash(x)=26349465)
1008
+ 966 train 7.492079 (lr=3.1508e-04) (hash(x)=21205988)
1009
+ 967 train 7.531161 (lr=3.1421e-04) (hash(x)=21596677)
1010
+ 968 train 7.640459 (lr=3.1336e-04) (hash(x)=23575666)
1011
+ 969 train 7.462653 (lr=3.1254e-04) (hash(x)=24117012)
1012
+ 970 train 7.555417 (lr=3.1174e-04) (hash(x)=26013014)
1013
+ 971 train 7.692890 (lr=3.1098e-04) (hash(x)=24105761)
1014
+ 972 train 7.546981 (lr=3.1023e-04) (hash(x)=20369133)
1015
+ 973 train 7.580348 (lr=3.0952e-04) (hash(x)=25725049)
1016
+ 974 train 7.421004 (lr=3.0882e-04) (hash(x)=19808118)
1017
+ 975 train 7.377831 (lr=3.0816e-04) (hash(x)=22892776)
1018
+ 976 train 7.563002 (lr=3.0752e-04) (hash(x)=23736780)
1019
+ 977 train 7.510027 (lr=3.0691e-04) (hash(x)=21630383)
1020
+ 978 train 7.713317 (lr=3.0632e-04) (hash(x)=25397467)
1021
+ 979 train 7.491992 (lr=3.0576e-04) (hash(x)=24727094)
1022
+ 980 train 7.556301 (lr=3.0522e-04) (hash(x)=23951182)
1023
+ 981 train 7.503581 (lr=3.0471e-04) (hash(x)=22835000)
1024
+ 982 train 7.682641 (lr=3.0423e-04) (hash(x)=28767849)
1025
+ 983 train 7.679814 (lr=3.0377e-04) (hash(x)=22839049)
1026
+ 984 train 7.580838 (lr=3.0334e-04) (hash(x)=19930900)
1027
+ 985 train 7.472993 (lr=3.0294e-04) (hash(x)=20654136)
1028
+ 986 train 7.678939 (lr=3.0256e-04) (hash(x)=28495621)
1029
+ 987 train 7.816905 (lr=3.0221e-04) (hash(x)=26776000)
1030
+ 988 train 7.504190 (lr=3.0188e-04) (hash(x)=21584429)
1031
+ 989 train 7.291087 (lr=3.0158e-04) (hash(x)=22560658)
1032
+ 990 train 7.888208 (lr=3.0131e-04) (hash(x)=25276565)
1033
+ 991 train 7.678051 (lr=3.0106e-04) (hash(x)=25995933)
1034
+ 992 train 7.751837 (lr=3.0084e-04) (hash(x)=24163910)
1035
+ 993 train 7.708351 (lr=3.0064e-04) (hash(x)=24635394)
1036
+ 994 train 7.807448 (lr=3.0047e-04) (hash(x)=22556381)
1037
+ 995 train 7.858635 (lr=3.0033e-04) (hash(x)=27537228)
1038
+ 996 train 7.661697 (lr=3.0021e-04) (hash(x)=20883633)
1039
+ 997 train 7.846888 (lr=3.0012e-04) (hash(x)=26496730)
1040
+ 998 train 7.455209 (lr=3.0005e-04) (hash(x)=20342150)
1041
+ 999 val loss 7.6994
1042
+ 999 val perplexity 2206.9727
1043
+ 999 train 7.462136 (lr=3.0001e-04) (hash(x)=16533310)
lr3e-3_total_batch_size20480_seq_len512/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c758b2128296c7a7ca867a9a503acb5893e6a3ff8faa3161eb84e1cf65f46ca6
3
+ size 48156226
lr3e-3_total_batch_size20480_seq_len512/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35c93ecd484616bedda7ef4dc9ba66ea80a273a574b5536a69f85a156cddcbaa
3
+ size 71157574