andrew-healey commited on
Commit
3852543
·
verified ·
1 Parent(s): 595416c

Upload folder using huggingface_hub

Browse files
lr1.5e-3_total_batch_size20480_seq_len512/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep_2/lr1.5e-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": 0.0015, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr1.5e-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
lr1.5e-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=5.2448e-06) (hash(x)=27234506)
5
+ 1 train 11.748632 (lr=1.0490e-05) (hash(x)=20782690)
6
+ 2 train 11.751670 (lr=1.5734e-05) (hash(x)=26532095)
7
+ 3 train 11.726258 (lr=2.0979e-05) (hash(x)=23890908)
8
+ 4 train 11.682303 (lr=2.6224e-05) (hash(x)=24011372)
9
+ 5 train 11.633446 (lr=3.1469e-05) (hash(x)=24611628)
10
+ 6 train 11.577148 (lr=3.6713e-05) (hash(x)=19245352)
11
+ 7 train 11.494610 (lr=4.1958e-05) (hash(x)=25767553)
12
+ 8 train 11.430746 (lr=4.7203e-05) (hash(x)=24809873)
13
+ 9 train 11.362621 (lr=5.2448e-05) (hash(x)=22512166)
14
+ 10 train 11.283493 (lr=5.7692e-05) (hash(x)=28092148)
15
+ 11 train 11.198224 (lr=6.2937e-05) (hash(x)=24732833)
16
+ 12 train 11.104171 (lr=6.8182e-05) (hash(x)=22694718)
17
+ 13 train 10.986423 (lr=7.3427e-05) (hash(x)=25129410)
18
+ 14 train 10.908207 (lr=7.8671e-05) (hash(x)=27858570)
19
+ 15 train 10.748894 (lr=8.3916e-05) (hash(x)=26582391)
20
+ 16 train 10.675738 (lr=8.9161e-05) (hash(x)=27279806)
21
+ 17 train 10.589563 (lr=9.4406e-05) (hash(x)=26472336)
22
+ 18 train 10.539284 (lr=9.9650e-05) (hash(x)=28301589)
23
+ 19 train 10.487341 (lr=1.0490e-04) (hash(x)=27654289)
24
+ 20 train 10.435303 (lr=1.1014e-04) (hash(x)=24995715)
25
+ 21 train 10.389433 (lr=1.1538e-04) (hash(x)=26559876)
26
+ 22 train 10.324065 (lr=1.2063e-04) (hash(x)=28364895)
27
+ 23 train 10.346321 (lr=1.2587e-04) (hash(x)=21770329)
28
+ 24 train 10.401286 (lr=1.3112e-04) (hash(x)=24628606)
29
+ 25 train 10.295366 (lr=1.3636e-04) (hash(x)=22723459)
30
+ 26 train 10.182430 (lr=1.4161e-04) (hash(x)=22897919)
31
+ 27 train 10.362915 (lr=1.4685e-04) (hash(x)=30338342)
32
+ 28 train 10.234152 (lr=1.5210e-04) (hash(x)=27063280)
33
+ 29 train 10.190245 (lr=1.5734e-04) (hash(x)=25678059)
34
+ 30 train 10.083306 (lr=1.6259e-04) (hash(x)=21853028)
35
+ 31 train 10.097944 (lr=1.6783e-04) (hash(x)=26109335)
36
+ 32 train 10.126865 (lr=1.7308e-04) (hash(x)=25334848)
37
+ 33 train 10.034185 (lr=1.7832e-04) (hash(x)=25044885)
38
+ 34 train 10.047920 (lr=1.8357e-04) (hash(x)=23678349)
39
+ 35 train 10.080820 (lr=1.8881e-04) (hash(x)=28870690)
40
+ 36 train 10.043217 (lr=1.9406e-04) (hash(x)=32982615)
41
+ 37 train 10.056829 (lr=1.9930e-04) (hash(x)=35279941)
42
+ 38 train 9.962542 (lr=2.0455e-04) (hash(x)=26400041)
43
+ 39 train 9.813856 (lr=2.0979e-04) (hash(x)=25015923)
44
+ 40 train 9.750528 (lr=2.1503e-04) (hash(x)=24853586)
45
+ 41 train 9.746244 (lr=2.2028e-04) (hash(x)=26740855)
46
+ 42 train 9.583813 (lr=2.2552e-04) (hash(x)=24584116)
47
+ 43 train 9.631033 (lr=2.3077e-04) (hash(x)=22262151)
48
+ 44 train 9.557138 (lr=2.3601e-04) (hash(x)=26627860)
49
+ 45 train 9.421465 (lr=2.4126e-04) (hash(x)=26621834)
50
+ 46 train 9.457384 (lr=2.4650e-04) (hash(x)=26930630)
51
+ 47 train 9.391595 (lr=2.5175e-04) (hash(x)=23234741)
52
+ 48 train 9.382905 (lr=2.5699e-04) (hash(x)=28102443)
53
+ 49 train 9.463277 (lr=2.6224e-04) (hash(x)=25470563)
54
+ 50 val loss 9.2689
55
+ 50 val perplexity 10603.3037
56
+ 50 train 9.295059 (lr=2.6748e-04) (hash(x)=28282861)
57
+ 51 train 9.260271 (lr=2.7273e-04) (hash(x)=21799102)
58
+ 52 train 9.133932 (lr=2.7797e-04) (hash(x)=23471769)
59
+ 53 train 9.084118 (lr=2.8322e-04) (hash(x)=23275479)
60
+ 54 train 9.110880 (lr=2.8846e-04) (hash(x)=24473581)
61
+ 55 train 9.068999 (lr=2.9371e-04) (hash(x)=22091862)
62
+ 56 train 8.919904 (lr=2.9895e-04) (hash(x)=26018202)
63
+ 57 train 8.846030 (lr=3.0420e-04) (hash(x)=17430373)
64
+ 58 train 8.648209 (lr=3.0944e-04) (hash(x)=24078083)
65
+ 59 train 8.729909 (lr=3.1469e-04) (hash(x)=22954861)
66
+ 60 train 8.798223 (lr=3.1993e-04) (hash(x)=28695016)
67
+ 61 train 8.743580 (lr=3.2517e-04) (hash(x)=26671799)
68
+ 62 train 8.727453 (lr=3.3042e-04) (hash(x)=23166092)
69
+ 63 train 8.520947 (lr=3.3566e-04) (hash(x)=26924723)
70
+ 64 train 8.565569 (lr=3.4091e-04) (hash(x)=25138659)
71
+ 65 train 8.571987 (lr=3.4615e-04) (hash(x)=23557495)
72
+ 66 train 8.391038 (lr=3.5140e-04) (hash(x)=25288123)
73
+ 67 train 8.237532 (lr=3.5664e-04) (hash(x)=27804380)
74
+ 68 train 8.454062 (lr=3.6189e-04) (hash(x)=26993263)
75
+ 69 train 8.447684 (lr=3.6713e-04) (hash(x)=26654908)
76
+ 70 train 8.257896 (lr=3.7238e-04) (hash(x)=27347722)
77
+ 71 train 8.285466 (lr=3.7762e-04) (hash(x)=23627518)
78
+ 72 train 8.122870 (lr=3.8287e-04) (hash(x)=21768671)
79
+ 73 train 8.160507 (lr=3.8811e-04) (hash(x)=21889990)
80
+ 74 train 8.071836 (lr=3.9336e-04) (hash(x)=23665838)
81
+ 75 train 8.361857 (lr=3.9860e-04) (hash(x)=22528001)
82
+ 76 train 8.167802 (lr=4.0385e-04) (hash(x)=22487328)
83
+ 77 train 8.272122 (lr=4.0909e-04) (hash(x)=29248942)
84
+ 78 train 8.057648 (lr=4.1434e-04) (hash(x)=23543273)
85
+ 79 train 7.996505 (lr=4.1958e-04) (hash(x)=25533417)
86
+ 80 train 8.100206 (lr=4.2483e-04) (hash(x)=25552036)
87
+ 81 train 7.913758 (lr=4.3007e-04) (hash(x)=20612533)
88
+ 82 train 7.694815 (lr=4.3531e-04) (hash(x)=20216795)
89
+ 83 train 7.990070 (lr=4.4056e-04) (hash(x)=24677740)
90
+ 84 train 7.728624 (lr=4.4580e-04) (hash(x)=16722715)
91
+ 85 train 7.927222 (lr=4.5105e-04) (hash(x)=25347203)
92
+ 86 train 7.894966 (lr=4.5629e-04) (hash(x)=24272489)
93
+ 87 train 7.990733 (lr=4.6154e-04) (hash(x)=28922363)
94
+ 88 train 8.003246 (lr=4.6678e-04) (hash(x)=25510798)
95
+ 89 train 7.940214 (lr=4.7203e-04) (hash(x)=25631397)
96
+ 90 train 7.720366 (lr=4.7727e-04) (hash(x)=21333676)
97
+ 91 train 7.993258 (lr=4.8252e-04) (hash(x)=26237983)
98
+ 92 train 7.788907 (lr=4.8776e-04) (hash(x)=25976502)
99
+ 93 train 7.786632 (lr=4.9301e-04) (hash(x)=24742645)
100
+ 94 train 7.895804 (lr=4.9825e-04) (hash(x)=27086333)
101
+ 95 train 7.775636 (lr=5.0350e-04) (hash(x)=26157660)
102
+ 96 train 7.916376 (lr=5.0874e-04) (hash(x)=23616370)
103
+ 97 train 9.522013 (lr=5.1399e-04) (hash(x)=22582169)
104
+ 98 train 8.919479 (lr=5.1923e-04) (hash(x)=26233189)
105
+ 99 train 9.542850 (lr=5.2448e-04) (hash(x)=26431960)
106
+ 100 val loss 9.5885
107
+ 100 val perplexity 14596.4189
108
+ 100 train 9.624008 (lr=5.2972e-04) (hash(x)=30707498)
109
+ 101 train 9.254453 (lr=5.3497e-04) (hash(x)=25716176)
110
+ 102 train 8.185792 (lr=5.4021e-04) (hash(x)=26256420)
111
+ 103 train 8.002180 (lr=5.4545e-04) (hash(x)=28019788)
112
+ 104 train 7.906451 (lr=5.5070e-04) (hash(x)=23510110)
113
+ 105 train 7.838155 (lr=5.5594e-04) (hash(x)=20382963)
114
+ 106 train 7.712401 (lr=5.6119e-04) (hash(x)=25339466)
115
+ 107 train 7.714920 (lr=5.6643e-04) (hash(x)=24489647)
116
+ 108 train 7.715414 (lr=5.7168e-04) (hash(x)=22927699)
117
+ 109 train 7.852539 (lr=5.7692e-04) (hash(x)=25128502)
118
+ 110 train 7.844035 (lr=5.8217e-04) (hash(x)=25808969)
119
+ 111 train 7.801566 (lr=5.8741e-04) (hash(x)=25231390)
120
+ 112 train 7.664445 (lr=5.9266e-04) (hash(x)=21154388)
121
+ 113 train 7.593075 (lr=5.9790e-04) (hash(x)=22708977)
122
+ 114 train 8.022389 (lr=6.0315e-04) (hash(x)=27258353)
123
+ 115 train 7.745034 (lr=6.0839e-04) (hash(x)=24364920)
124
+ 116 train 7.812607 (lr=6.1364e-04) (hash(x)=27243972)
125
+ 117 train 7.888717 (lr=6.1888e-04) (hash(x)=24932925)
126
+ 118 train 7.702134 (lr=6.2413e-04) (hash(x)=26371091)
127
+ 119 train 7.719365 (lr=6.2937e-04) (hash(x)=26913684)
128
+ 120 train 7.864874 (lr=6.3462e-04) (hash(x)=29239611)
129
+ 121 train 7.540283 (lr=6.3986e-04) (hash(x)=18017792)
130
+ 122 train 7.735961 (lr=6.4510e-04) (hash(x)=27574089)
131
+ 123 train 7.806792 (lr=6.5035e-04) (hash(x)=28613882)
132
+ 124 train 7.741596 (lr=6.5559e-04) (hash(x)=26981914)
133
+ 125 train 7.723131 (lr=6.6084e-04) (hash(x)=22652243)
134
+ 126 train 7.483815 (lr=6.6608e-04) (hash(x)=15446025)
135
+ 127 train 7.629873 (lr=6.7133e-04) (hash(x)=27194521)
136
+ 128 train 7.651911 (lr=6.7657e-04) (hash(x)=25264518)
137
+ 129 train 7.837871 (lr=6.8182e-04) (hash(x)=24936836)
138
+ 130 train 7.856664 (lr=6.8706e-04) (hash(x)=28197282)
139
+ 131 train 7.759081 (lr=6.9231e-04) (hash(x)=24765578)
140
+ 132 train 7.817194 (lr=6.9755e-04) (hash(x)=27448185)
141
+ 133 train 7.839071 (lr=7.0280e-04) (hash(x)=23107416)
142
+ 134 train 7.770218 (lr=7.0804e-04) (hash(x)=26755138)
143
+ 135 train 7.741762 (lr=7.1329e-04) (hash(x)=29314255)
144
+ 136 train 7.720432 (lr=7.1853e-04) (hash(x)=22112669)
145
+ 137 train 7.675890 (lr=7.2378e-04) (hash(x)=24013079)
146
+ 138 train 7.690676 (lr=7.2902e-04) (hash(x)=22537194)
147
+ 139 train 7.847948 (lr=7.3427e-04) (hash(x)=25322001)
148
+ 140 train 7.844237 (lr=7.3951e-04) (hash(x)=27028126)
149
+ 141 train 7.866303 (lr=7.4476e-04) (hash(x)=26690440)
150
+ 142 train 7.801899 (lr=7.5000e-04) (hash(x)=26967331)
151
+ 143 train 7.649940 (lr=7.5524e-04) (hash(x)=22598512)
152
+ 144 train 7.722034 (lr=7.6049e-04) (hash(x)=24697985)
153
+ 145 train 7.566094 (lr=7.6573e-04) (hash(x)=18473911)
154
+ 146 train 7.655570 (lr=7.7098e-04) (hash(x)=23172124)
155
+ 147 train 8.176866 (lr=7.7622e-04) (hash(x)=22322442)
156
+ 148 train 8.024118 (lr=7.8147e-04) (hash(x)=29659121)
157
+ 149 train 7.887253 (lr=7.8671e-04) (hash(x)=26324153)
158
+ 150 val loss 7.7609
159
+ 150 val perplexity 2347.0652
160
+ 150 train 7.798794 (lr=7.9196e-04) (hash(x)=26939970)
161
+ 151 train 7.847337 (lr=7.9720e-04) (hash(x)=23446058)
162
+ 152 train 7.886610 (lr=8.0245e-04) (hash(x)=29981801)
163
+ 153 train 7.938977 (lr=8.0769e-04) (hash(x)=24843741)
164
+ 154 train 7.782618 (lr=8.1294e-04) (hash(x)=22855363)
165
+ 155 train 7.689300 (lr=8.1818e-04) (hash(x)=23197102)
166
+ 156 train 7.819600 (lr=8.2343e-04) (hash(x)=25704919)
167
+ 157 train 8.157961 (lr=8.2867e-04) (hash(x)=23054940)
168
+ 158 train 8.132567 (lr=8.3392e-04) (hash(x)=24909904)
169
+ 159 train 7.810771 (lr=8.3916e-04) (hash(x)=27140757)
170
+ 160 train 7.679703 (lr=8.4441e-04) (hash(x)=25993110)
171
+ 161 train 7.709963 (lr=8.4965e-04) (hash(x)=25149419)
172
+ 162 train 7.876023 (lr=8.5490e-04) (hash(x)=27623643)
173
+ 163 train 7.801550 (lr=8.6014e-04) (hash(x)=24727764)
174
+ 164 train 7.378664 (lr=8.6538e-04) (hash(x)=23155773)
175
+ 165 train 7.634564 (lr=8.7063e-04) (hash(x)=21935931)
176
+ 166 train 7.527448 (lr=8.7587e-04) (hash(x)=20054917)
177
+ 167 train 7.789939 (lr=8.8112e-04) (hash(x)=27606073)
178
+ 168 train 7.881632 (lr=8.8636e-04) (hash(x)=26534487)
179
+ 169 train 7.720211 (lr=8.9161e-04) (hash(x)=22450341)
180
+ 170 train 7.883382 (lr=8.9685e-04) (hash(x)=25013073)
181
+ 171 train 8.015116 (lr=9.0210e-04) (hash(x)=25945859)
182
+ 172 train 7.777745 (lr=9.0734e-04) (hash(x)=24418091)
183
+ 173 train 7.727793 (lr=9.1259e-04) (hash(x)=21613707)
184
+ 174 train 7.828713 (lr=9.1783e-04) (hash(x)=26386157)
185
+ 175 train 7.778327 (lr=9.2308e-04) (hash(x)=24396519)
186
+ 176 train 7.815295 (lr=9.2832e-04) (hash(x)=23265507)
187
+ 177 train 7.871771 (lr=9.3357e-04) (hash(x)=29138828)
188
+ 178 train 7.853112 (lr=9.3881e-04) (hash(x)=24265353)
189
+ 179 train 7.885161 (lr=9.4406e-04) (hash(x)=24685515)
190
+ 180 train 7.793088 (lr=9.4930e-04) (hash(x)=24876269)
191
+ 181 train 7.790230 (lr=9.5455e-04) (hash(x)=22825749)
192
+ 182 train 7.813328 (lr=9.5979e-04) (hash(x)=21239652)
193
+ 183 train 7.847096 (lr=9.6503e-04) (hash(x)=28767869)
194
+ 184 train 7.872569 (lr=9.7028e-04) (hash(x)=25614301)
195
+ 185 train 7.847910 (lr=9.7552e-04) (hash(x)=21159060)
196
+ 186 train 7.744434 (lr=9.8077e-04) (hash(x)=26751788)
197
+ 187 train 7.786582 (lr=9.8601e-04) (hash(x)=25045397)
198
+ 188 train 7.837161 (lr=9.9126e-04) (hash(x)=29989709)
199
+ 189 train 7.930603 (lr=9.9650e-04) (hash(x)=25560928)
200
+ 190 train 7.857389 (lr=1.0017e-03) (hash(x)=26062687)
201
+ 191 train 7.966759 (lr=1.0070e-03) (hash(x)=28959222)
202
+ 192 train 7.771972 (lr=1.0122e-03) (hash(x)=25434884)
203
+ 193 train 7.792539 (lr=1.0175e-03) (hash(x)=21248405)
204
+ 194 train 7.718881 (lr=1.0227e-03) (hash(x)=25044192)
205
+ 195 train 7.762788 (lr=1.0280e-03) (hash(x)=25830182)
206
+ 196 train 7.760602 (lr=1.0332e-03) (hash(x)=26621419)
207
+ 197 train 7.560806 (lr=1.0385e-03) (hash(x)=25173113)
208
+ 198 train 7.716091 (lr=1.0437e-03) (hash(x)=24400133)
209
+ 199 train 7.694118 (lr=1.0490e-03) (hash(x)=23486277)
210
+ 200 val loss 7.7135
211
+ 200 val perplexity 2238.4312
212
+ 200 train 7.698823 (lr=1.0542e-03) (hash(x)=25978130)
213
+ 201 train 7.773768 (lr=1.0594e-03) (hash(x)=28916006)
214
+ 202 train 7.974592 (lr=1.0647e-03) (hash(x)=33195100)
215
+ 203 train 7.960951 (lr=1.0699e-03) (hash(x)=25556929)
216
+ 204 train 7.841340 (lr=1.0752e-03) (hash(x)=29536986)
217
+ 205 train 7.785434 (lr=1.0804e-03) (hash(x)=29648798)
218
+ 206 train 7.719274 (lr=1.0857e-03) (hash(x)=23124767)
219
+ 207 train 7.696484 (lr=1.0909e-03) (hash(x)=24976217)
220
+ 208 train 7.721515 (lr=1.0962e-03) (hash(x)=23632825)
221
+ 209 train 7.830029 (lr=1.1014e-03) (hash(x)=31019606)
222
+ 210 train 7.862543 (lr=1.1066e-03) (hash(x)=28236580)
223
+ 211 train 7.744893 (lr=1.1119e-03) (hash(x)=28269421)
224
+ 212 train 7.525059 (lr=1.1171e-03) (hash(x)=25563279)
225
+ 213 train 7.810544 (lr=1.1224e-03) (hash(x)=24865358)
226
+ 214 train 7.561086 (lr=1.1276e-03) (hash(x)=22155546)
227
+ 215 train 7.690288 (lr=1.1329e-03) (hash(x)=26984272)
228
+ 216 train 7.712661 (lr=1.1381e-03) (hash(x)=23193673)
229
+ 217 train 7.785197 (lr=1.1434e-03) (hash(x)=24544116)
230
+ 218 train 7.759787 (lr=1.1486e-03) (hash(x)=24547533)
231
+ 219 train 7.729136 (lr=1.1538e-03) (hash(x)=22059850)
232
+ 220 train 7.789202 (lr=1.1591e-03) (hash(x)=25994573)
233
+ 221 train 7.717119 (lr=1.1643e-03) (hash(x)=23926632)
234
+ 222 train 7.572708 (lr=1.1696e-03) (hash(x)=27111369)
235
+ 223 train 7.750778 (lr=1.1748e-03) (hash(x)=27025333)
236
+ 224 train 7.546805 (lr=1.1801e-03) (hash(x)=16908068)
237
+ 225 train 7.751030 (lr=1.1853e-03) (hash(x)=29120407)
238
+ 226 train 7.845582 (lr=1.1906e-03) (hash(x)=27092710)
239
+ 227 train 7.593475 (lr=1.1958e-03) (hash(x)=23827393)
240
+ 228 train 7.698853 (lr=1.2010e-03) (hash(x)=32812727)
241
+ 229 train 7.703777 (lr=1.2063e-03) (hash(x)=27230027)
242
+ 230 train 7.728557 (lr=1.2115e-03) (hash(x)=25703381)
243
+ 231 train 7.719418 (lr=1.2168e-03) (hash(x)=24968260)
244
+ 232 train 7.700224 (lr=1.2220e-03) (hash(x)=25676468)
245
+ 233 train 7.799487 (lr=1.2273e-03) (hash(x)=27544803)
246
+ 234 train 7.747113 (lr=1.2325e-03) (hash(x)=24331407)
247
+ 235 train 7.689080 (lr=1.2378e-03) (hash(x)=26540663)
248
+ 236 train 7.722171 (lr=1.2430e-03) (hash(x)=25532657)
249
+ 237 train 7.899659 (lr=1.2483e-03) (hash(x)=25856625)
250
+ 238 train 7.622917 (lr=1.2535e-03) (hash(x)=23437426)
251
+ 239 train 7.682978 (lr=1.2587e-03) (hash(x)=23164356)
252
+ 240 train 7.662556 (lr=1.2640e-03) (hash(x)=25957896)
253
+ 241 train 7.721444 (lr=1.2692e-03) (hash(x)=25646282)
254
+ 242 train 7.735603 (lr=1.2745e-03) (hash(x)=21513584)
255
+ 243 train 7.621528 (lr=1.2797e-03) (hash(x)=23843233)
256
+ 244 train 7.634665 (lr=1.2850e-03) (hash(x)=23226697)
257
+ 245 train 7.494020 (lr=1.2902e-03) (hash(x)=24506028)
258
+ 246 train 7.718070 (lr=1.2955e-03) (hash(x)=26024321)
259
+ 247 train 7.715631 (lr=1.3007e-03) (hash(x)=22781277)
260
+ 248 train 7.758938 (lr=1.3059e-03) (hash(x)=24451843)
261
+ 249 train 7.710918 (lr=1.3112e-03) (hash(x)=24948650)
262
+ 250 val loss 7.6961
263
+ 250 val perplexity 2199.8535
264
+ 250 train 7.628471 (lr=1.3164e-03) (hash(x)=25559534)
265
+ 251 train 7.721059 (lr=1.3217e-03) (hash(x)=26373991)
266
+ 252 train 7.789857 (lr=1.3269e-03) (hash(x)=27756673)
267
+ 253 train 7.755773 (lr=1.3322e-03) (hash(x)=23402396)
268
+ 254 train 7.798302 (lr=1.3374e-03) (hash(x)=28250354)
269
+ 255 train 7.708291 (lr=1.3427e-03) (hash(x)=21562133)
270
+ 256 train 7.842077 (lr=1.3479e-03) (hash(x)=27253861)
271
+ 257 train 7.731106 (lr=1.3531e-03) (hash(x)=25009210)
272
+ 258 train 7.557536 (lr=1.3584e-03) (hash(x)=23052577)
273
+ 259 train 7.762327 (lr=1.3636e-03) (hash(x)=27161811)
274
+ 260 train 7.702718 (lr=1.3689e-03) (hash(x)=27302459)
275
+ 261 train 7.800609 (lr=1.3741e-03) (hash(x)=29515100)
276
+ 262 train 7.678573 (lr=1.3794e-03) (hash(x)=25318001)
277
+ 263 train 7.655085 (lr=1.3846e-03) (hash(x)=25305929)
278
+ 264 train 7.642696 (lr=1.3899e-03) (hash(x)=24339043)
279
+ 265 train 7.730836 (lr=1.3951e-03) (hash(x)=26808543)
280
+ 266 train 7.525864 (lr=1.4003e-03) (hash(x)=25763991)
281
+ 267 train 7.715655 (lr=1.4056e-03) (hash(x)=25669509)
282
+ 268 train 7.747199 (lr=1.4108e-03) (hash(x)=27676869)
283
+ 269 train 7.652094 (lr=1.4161e-03) (hash(x)=28435805)
284
+ 270 train 7.619118 (lr=1.4213e-03) (hash(x)=23455211)
285
+ 271 train 7.762983 (lr=1.4266e-03) (hash(x)=26482758)
286
+ 272 train 7.604302 (lr=1.4318e-03) (hash(x)=27068280)
287
+ 273 train 7.638752 (lr=1.4371e-03) (hash(x)=24929929)
288
+ 274 train 7.606762 (lr=1.4423e-03) (hash(x)=25373676)
289
+ 275 train 7.592538 (lr=1.4476e-03) (hash(x)=25043037)
290
+ 276 train 7.656570 (lr=1.4528e-03) (hash(x)=27906976)
291
+ 277 train 8.032547 (lr=1.4580e-03) (hash(x)=29151257)
292
+ 278 train 7.727058 (lr=1.4633e-03) (hash(x)=26405613)
293
+ 279 train 7.477920 (lr=1.4685e-03) (hash(x)=26400365)
294
+ 280 train 7.651788 (lr=1.4738e-03) (hash(x)=23971905)
295
+ 281 train 7.575007 (lr=1.4790e-03) (hash(x)=24098812)
296
+ 282 train 7.758742 (lr=1.4843e-03) (hash(x)=29402976)
297
+ 283 train 7.543567 (lr=1.4895e-03) (hash(x)=20967666)
298
+ 284 train 7.681070 (lr=1.4948e-03) (hash(x)=24287610)
299
+ 285 train 8.061537 (lr=1.5000e-03) (hash(x)=28667963)
300
+ 286 train 7.742517 (lr=1.5000e-03) (hash(x)=25974316)
301
+ 287 train 7.835395 (lr=1.5000e-03) (hash(x)=27083417)
302
+ 288 train 7.609795 (lr=1.5000e-03) (hash(x)=28605205)
303
+ 289 train 7.566931 (lr=1.4999e-03) (hash(x)=23649001)
304
+ 290 train 7.587451 (lr=1.4999e-03) (hash(x)=25921700)
305
+ 291 train 7.726690 (lr=1.4998e-03) (hash(x)=31338300)
306
+ 292 train 7.628217 (lr=1.4998e-03) (hash(x)=23862328)
307
+ 293 train 7.659289 (lr=1.4997e-03) (hash(x)=30603174)
308
+ 294 train 7.728784 (lr=1.4996e-03) (hash(x)=23603806)
309
+ 295 train 7.683750 (lr=1.4995e-03) (hash(x)=24830965)
310
+ 296 train 7.788673 (lr=1.4993e-03) (hash(x)=32488729)
311
+ 297 train 7.936327 (lr=1.4992e-03) (hash(x)=27269893)
312
+ 298 train 7.685667 (lr=1.4991e-03) (hash(x)=25215077)
313
+ 299 train 7.649863 (lr=1.4989e-03) (hash(x)=24825041)
314
+ 300 val loss 7.6852
315
+ 300 val perplexity 2175.9604
316
+ 300 train 7.742365 (lr=1.4987e-03) (hash(x)=25893804)
317
+ 301 train 7.573775 (lr=1.4985e-03) (hash(x)=22724871)
318
+ 302 train 7.630139 (lr=1.4983e-03) (hash(x)=27317554)
319
+ 303 train 7.536047 (lr=1.4981e-03) (hash(x)=24683717)
320
+ 304 train 7.599562 (lr=1.4979e-03) (hash(x)=21584083)
321
+ 305 train 7.776014 (lr=1.4976e-03) (hash(x)=26190337)
322
+ 306 train 7.702585 (lr=1.4974e-03) (hash(x)=22249019)
323
+ 307 train 7.675800 (lr=1.4971e-03) (hash(x)=23086289)
324
+ 308 train 7.836805 (lr=1.4968e-03) (hash(x)=28047044)
325
+ 309 train 7.647109 (lr=1.4965e-03) (hash(x)=23593875)
326
+ 310 train 7.604011 (lr=1.4962e-03) (hash(x)=26059735)
327
+ 311 train 7.508479 (lr=1.4959e-03) (hash(x)=22414190)
328
+ 312 train 7.697688 (lr=1.4956e-03) (hash(x)=24204591)
329
+ 313 train 7.550089 (lr=1.4952e-03) (hash(x)=26758114)
330
+ 314 train 7.598807 (lr=1.4949e-03) (hash(x)=20084233)
331
+ 315 train 7.510942 (lr=1.4945e-03) (hash(x)=21563184)
332
+ 316 train 7.611741 (lr=1.4941e-03) (hash(x)=23861489)
333
+ 317 train 7.617620 (lr=1.4937e-03) (hash(x)=19218470)
334
+ 318 train 7.647524 (lr=1.4933e-03) (hash(x)=28469562)
335
+ 319 train 7.877610 (lr=1.4929e-03) (hash(x)=25588469)
336
+ 320 train 7.726386 (lr=1.4925e-03) (hash(x)=25236367)
337
+ 321 train 7.747544 (lr=1.4920e-03) (hash(x)=30433767)
338
+ 322 train 7.934082 (lr=1.4915e-03) (hash(x)=26302626)
339
+ 323 train 7.613876 (lr=1.4911e-03) (hash(x)=17851621)
340
+ 324 train 7.871740 (lr=1.4906e-03) (hash(x)=27028191)
341
+ 325 train 7.809760 (lr=1.4901e-03) (hash(x)=25183690)
342
+ 326 train 7.793015 (lr=1.4896e-03) (hash(x)=26182243)
343
+ 327 train 7.921977 (lr=1.4890e-03) (hash(x)=27040642)
344
+ 328 train 7.642804 (lr=1.4885e-03) (hash(x)=24489607)
345
+ 329 train 7.572197 (lr=1.4880e-03) (hash(x)=22183303)
346
+ 330 train 7.707925 (lr=1.4874e-03) (hash(x)=28844646)
347
+ 331 train 8.077234 (lr=1.4868e-03) (hash(x)=27537063)
348
+ 332 train 7.791364 (lr=1.4862e-03) (hash(x)=25446686)
349
+ 333 train 7.702782 (lr=1.4856e-03) (hash(x)=19492832)
350
+ 334 train 7.677460 (lr=1.4850e-03) (hash(x)=25484958)
351
+ 335 train 7.697497 (lr=1.4844e-03) (hash(x)=25951629)
352
+ 336 train 7.714735 (lr=1.4837e-03) (hash(x)=23462798)
353
+ 337 train 7.598752 (lr=1.4831e-03) (hash(x)=22408682)
354
+ 338 train 7.722919 (lr=1.4824e-03) (hash(x)=31317970)
355
+ 339 train 7.551249 (lr=1.4817e-03) (hash(x)=27633457)
356
+ 340 train 7.529179 (lr=1.4810e-03) (hash(x)=23729283)
357
+ 341 train 7.719296 (lr=1.4803e-03) (hash(x)=27635080)
358
+ 342 train 7.655334 (lr=1.4796e-03) (hash(x)=25918780)
359
+ 343 train 7.613671 (lr=1.4789e-03) (hash(x)=22287596)
360
+ 344 train 7.549586 (lr=1.4781e-03) (hash(x)=22694623)
361
+ 345 train 7.922359 (lr=1.4774e-03) (hash(x)=35895440)
362
+ 346 train 7.687607 (lr=1.4766e-03) (hash(x)=24298886)
363
+ 347 train 7.690705 (lr=1.4758e-03) (hash(x)=25888847)
364
+ 348 train 7.783933 (lr=1.4750e-03) (hash(x)=30860646)
365
+ 349 train 7.718555 (lr=1.4742e-03) (hash(x)=26518800)
366
+ 350 val loss 7.6714
367
+ 350 val perplexity 2145.9910
368
+ 350 train 7.626570 (lr=1.4734e-03) (hash(x)=25385523)
369
+ 351 train 7.761110 (lr=1.4726e-03) (hash(x)=37519283)
370
+ 352 train 8.139110 (lr=1.4717e-03) (hash(x)=20564397)
371
+ 353 train 7.746608 (lr=1.4709e-03) (hash(x)=25141945)
372
+ 354 train 7.763551 (lr=1.4700e-03) (hash(x)=22717866)
373
+ 355 train 8.055579 (lr=1.4691e-03) (hash(x)=28527939)
374
+ 356 train 7.887047 (lr=1.4682e-03) (hash(x)=32757059)
375
+ 357 train 7.669044 (lr=1.4673e-03) (hash(x)=21942471)
376
+ 358 train 7.925444 (lr=1.4664e-03) (hash(x)=24247339)
377
+ 359 train 7.703249 (lr=1.4655e-03) (hash(x)=24955630)
378
+ 360 train 7.810789 (lr=1.4645e-03) (hash(x)=21368780)
379
+ 361 train 7.714355 (lr=1.4636e-03) (hash(x)=25541230)
380
+ 362 train 7.627358 (lr=1.4626e-03) (hash(x)=23098806)
381
+ 363 train 7.575616 (lr=1.4616e-03) (hash(x)=23501481)
382
+ 364 train 7.639993 (lr=1.4606e-03) (hash(x)=23802681)
383
+ 365 train 7.654021 (lr=1.4596e-03) (hash(x)=22754988)
384
+ 366 train 7.511327 (lr=1.4586e-03) (hash(x)=22183009)
385
+ 367 train 7.568737 (lr=1.4576e-03) (hash(x)=23475070)
386
+ 368 train 7.692790 (lr=1.4565e-03) (hash(x)=23359061)
387
+ 369 train 7.685740 (lr=1.4555e-03) (hash(x)=23609959)
388
+ 370 train 7.622909 (lr=1.4544e-03) (hash(x)=22408449)
389
+ 371 train 7.566359 (lr=1.4533e-03) (hash(x)=23441847)
390
+ 372 train 7.561478 (lr=1.4522e-03) (hash(x)=23676365)
391
+ 373 train 7.731357 (lr=1.4511e-03) (hash(x)=23096292)
392
+ 374 train 7.766334 (lr=1.4500e-03) (hash(x)=27706294)
393
+ 375 train 7.684421 (lr=1.4489e-03) (hash(x)=26171093)
394
+ 376 train 7.852879 (lr=1.4478e-03) (hash(x)=34946941)
395
+ 377 train 7.650363 (lr=1.4466e-03) (hash(x)=20060838)
396
+ 378 train 7.558660 (lr=1.4454e-03) (hash(x)=21591702)
397
+ 379 train 7.635880 (lr=1.4443e-03) (hash(x)=25451055)
398
+ 380 train 7.299196 (lr=1.4431e-03) (hash(x)=18691455)
399
+ 381 train 7.587443 (lr=1.4419e-03) (hash(x)=24934792)
400
+ 382 train 7.539451 (lr=1.4407e-03) (hash(x)=23864361)
401
+ 383 train 7.566133 (lr=1.4394e-03) (hash(x)=25367597)
402
+ 384 train 7.738870 (lr=1.4382e-03) (hash(x)=27016702)
403
+ 385 train 7.734499 (lr=1.4370e-03) (hash(x)=25022148)
404
+ 386 train 7.740314 (lr=1.4357e-03) (hash(x)=24459012)
405
+ 387 train 7.708824 (lr=1.4344e-03) (hash(x)=25431365)
406
+ 388 train 7.633441 (lr=1.4332e-03) (hash(x)=25958785)
407
+ 389 train 7.835214 (lr=1.4319e-03) (hash(x)=29307788)
408
+ 390 train 7.575333 (lr=1.4306e-03) (hash(x)=24965435)
409
+ 391 train 7.600533 (lr=1.4292e-03) (hash(x)=25651929)
410
+ 392 train 7.662337 (lr=1.4279e-03) (hash(x)=28451031)
411
+ 393 train 7.721308 (lr=1.4266e-03) (hash(x)=26312387)
412
+ 394 train 7.762032 (lr=1.4252e-03) (hash(x)=25326650)
413
+ 395 train 7.680292 (lr=1.4238e-03) (hash(x)=23804913)
414
+ 396 train 7.872596 (lr=1.4225e-03) (hash(x)=25537529)
415
+ 397 train 7.687146 (lr=1.4211e-03) (hash(x)=26115519)
416
+ 398 train 7.709513 (lr=1.4197e-03) (hash(x)=25494804)
417
+ 399 train 7.813341 (lr=1.4183e-03) (hash(x)=23210747)
418
+ 400 val loss 7.7681
419
+ 400 val perplexity 2363.9128
420
+ 400 train 7.667335 (lr=1.4169e-03) (hash(x)=15717069)
421
+ 401 train 7.615105 (lr=1.4154e-03) (hash(x)=18463712)
422
+ 402 train 7.687692 (lr=1.4140e-03) (hash(x)=20705573)
423
+ 403 train 7.684934 (lr=1.4125e-03) (hash(x)=23245699)
424
+ 404 train 7.615367 (lr=1.4110e-03) (hash(x)=23444521)
425
+ 405 train 7.533133 (lr=1.4096e-03) (hash(x)=21770300)
426
+ 406 train 8.245329 (lr=1.4081e-03) (hash(x)=23334569)
427
+ 407 train 7.467976 (lr=1.4066e-03) (hash(x)=25387532)
428
+ 408 train 7.680962 (lr=1.4051e-03) (hash(x)=29466743)
429
+ 409 train 7.695983 (lr=1.4035e-03) (hash(x)=27718703)
430
+ 410 train 7.682644 (lr=1.4020e-03) (hash(x)=17631973)
431
+ 411 train 7.724470 (lr=1.4005e-03) (hash(x)=27661052)
432
+ 412 train 7.488457 (lr=1.3989e-03) (hash(x)=22244096)
433
+ 413 train 7.629815 (lr=1.3973e-03) (hash(x)=26538928)
434
+ 414 train 7.625240 (lr=1.3957e-03) (hash(x)=26256802)
435
+ 415 train 7.741020 (lr=1.3942e-03) (hash(x)=21772654)
436
+ 416 train 7.721872 (lr=1.3926e-03) (hash(x)=25026277)
437
+ 417 train 7.814374 (lr=1.3909e-03) (hash(x)=26263254)
438
+ 418 train 7.750460 (lr=1.3893e-03) (hash(x)=24726314)
439
+ 419 train 7.556540 (lr=1.3877e-03) (hash(x)=23552356)
440
+ 420 train 7.685651 (lr=1.3860e-03) (hash(x)=26699819)
441
+ 421 train 8.032447 (lr=1.3844e-03) (hash(x)=22707141)
442
+ 422 train 8.006714 (lr=1.3827e-03) (hash(x)=25676149)
443
+ 423 train 7.780266 (lr=1.3810e-03) (hash(x)=34234481)
444
+ 424 train 7.650903 (lr=1.3793e-03) (hash(x)=25401624)
445
+ 425 train 7.800409 (lr=1.3776e-03) (hash(x)=35008670)
446
+ 426 train 7.895960 (lr=1.3759e-03) (hash(x)=25682280)
447
+ 427 train 7.763099 (lr=1.3742e-03) (hash(x)=25891543)
448
+ 428 train 7.641181 (lr=1.3725e-03) (hash(x)=25962407)
449
+ 429 train 7.539370 (lr=1.3707e-03) (hash(x)=21971224)
450
+ 430 train 7.682867 (lr=1.3690e-03) (hash(x)=30838142)
451
+ 431 train 7.728588 (lr=1.3672e-03) (hash(x)=27312186)
452
+ 432 train 7.810811 (lr=1.3654e-03) (hash(x)=23657081)
453
+ 433 train 7.762174 (lr=1.3637e-03) (hash(x)=22163749)
454
+ 434 train 7.487094 (lr=1.3619e-03) (hash(x)=24829818)
455
+ 435 train 7.746663 (lr=1.3601e-03) (hash(x)=25942888)
456
+ 436 train 7.678055 (lr=1.3582e-03) (hash(x)=23785671)
457
+ 437 train 7.694900 (lr=1.3564e-03) (hash(x)=25493916)
458
+ 438 train 7.852951 (lr=1.3546e-03) (hash(x)=28618702)
459
+ 439 train 7.899300 (lr=1.3527e-03) (hash(x)=24289951)
460
+ 440 train 7.656086 (lr=1.3509e-03) (hash(x)=26331486)
461
+ 441 train 7.640676 (lr=1.3490e-03) (hash(x)=23648671)
462
+ 442 train 7.619852 (lr=1.3471e-03) (hash(x)=27292797)
463
+ 443 train 7.450141 (lr=1.3452e-03) (hash(x)=24992055)
464
+ 444 train 7.950497 (lr=1.3434e-03) (hash(x)=24398330)
465
+ 445 train 7.733499 (lr=1.3414e-03) (hash(x)=26005549)
466
+ 446 train 7.692571 (lr=1.3395e-03) (hash(x)=30259737)
467
+ 447 train 7.695460 (lr=1.3376e-03) (hash(x)=21046554)
468
+ 448 train 7.708872 (lr=1.3357e-03) (hash(x)=18773877)
469
+ 449 train 7.742128 (lr=1.3337e-03) (hash(x)=26321539)
470
+ 450 val loss 7.7286
471
+ 450 val perplexity 2272.4695
472
+ 450 train 7.707807 (lr=1.3318e-03) (hash(x)=22850804)
473
+ 451 train 7.779673 (lr=1.3298e-03) (hash(x)=27807872)
474
+ 452 train 7.770094 (lr=1.3278e-03) (hash(x)=25545137)
475
+ 453 train 7.746851 (lr=1.3258e-03) (hash(x)=21555364)
476
+ 454 train 7.800655 (lr=1.3238e-03) (hash(x)=27243798)
477
+ 455 train 7.690426 (lr=1.3218e-03) (hash(x)=26325324)
478
+ 456 train 7.635402 (lr=1.3198e-03) (hash(x)=26228728)
479
+ 457 train 7.886519 (lr=1.3178e-03) (hash(x)=24956294)
480
+ 458 train 7.724741 (lr=1.3158e-03) (hash(x)=25592289)
481
+ 459 train 7.720500 (lr=1.3137e-03) (hash(x)=24429389)
482
+ 460 train 7.753127 (lr=1.3117e-03) (hash(x)=28733708)
483
+ 461 train 7.931226 (lr=1.3096e-03) (hash(x)=27929846)
484
+ 462 train 7.757050 (lr=1.3075e-03) (hash(x)=22417476)
485
+ 463 train 7.847109 (lr=1.3054e-03) (hash(x)=25282897)
486
+ 464 train 7.653925 (lr=1.3033e-03) (hash(x)=22278062)
487
+ 465 train 7.649493 (lr=1.3012e-03) (hash(x)=26802508)
488
+ 466 train 7.884390 (lr=1.2991e-03) (hash(x)=22094832)
489
+ 467 train 7.811289 (lr=1.2970e-03) (hash(x)=28390416)
490
+ 468 train 7.901832 (lr=1.2949e-03) (hash(x)=26379331)
491
+ 469 train 7.958555 (lr=1.2928e-03) (hash(x)=26292451)
492
+ 470 train 7.851672 (lr=1.2906e-03) (hash(x)=26301866)
493
+ 471 train 7.878325 (lr=1.2885e-03) (hash(x)=22915344)
494
+ 472 train 7.718304 (lr=1.2863e-03) (hash(x)=25448318)
495
+ 473 train 7.807077 (lr=1.2841e-03) (hash(x)=24744383)
496
+ 474 train 7.609165 (lr=1.2819e-03) (hash(x)=22835201)
497
+ 475 train 7.732595 (lr=1.2797e-03) (hash(x)=26284678)
498
+ 476 train 7.720407 (lr=1.2775e-03) (hash(x)=24745936)
499
+ 477 train 7.548991 (lr=1.2753e-03) (hash(x)=21654269)
500
+ 478 train 7.728557 (lr=1.2731e-03) (hash(x)=32387999)
501
+ 479 train 8.360292 (lr=1.2709e-03) (hash(x)=29466268)
502
+ 480 train 8.318131 (lr=1.2687e-03) (hash(x)=30736727)
503
+ 481 train 8.265537 (lr=1.2664e-03) (hash(x)=32616762)
504
+ 482 train 8.218965 (lr=1.2642e-03) (hash(x)=29358578)
505
+ 483 train 8.254589 (lr=1.2619e-03) (hash(x)=32201138)
506
+ 484 train 8.270296 (lr=1.2596e-03) (hash(x)=30041621)
507
+ 485 train 8.243125 (lr=1.2574e-03) (hash(x)=33683468)
508
+ 486 train 8.017868 (lr=1.2551e-03) (hash(x)=26837755)
509
+ 487 train 7.897161 (lr=1.2528e-03) (hash(x)=27498125)
510
+ 488 train 7.995494 (lr=1.2505e-03) (hash(x)=22372507)
511
+ 489 train 7.697562 (lr=1.2482e-03) (hash(x)=25164318)
512
+ 490 train 7.689182 (lr=1.2459e-03) (hash(x)=26562947)
513
+ 491 train 7.634750 (lr=1.2435e-03) (hash(x)=25690221)
514
+ 492 train 7.692049 (lr=1.2412e-03) (hash(x)=21947305)
515
+ 493 train 7.690072 (lr=1.2389e-03) (hash(x)=29031550)
516
+ 494 train 7.830329 (lr=1.2365e-03) (hash(x)=26551589)
517
+ 495 train 7.835774 (lr=1.2341e-03) (hash(x)=24094093)
518
+ 496 train 7.618011 (lr=1.2318e-03) (hash(x)=23639076)
519
+ 497 train 9.262734 (lr=1.2294e-03) (hash(x)=61455960)
520
+ 498 train 11.658505 (lr=1.2270e-03) (hash(x)=65636572)
521
+ 499 train 11.576948 (lr=1.2246e-03) (hash(x)=61104710)
522
+ 500 val loss 7.7299
523
+ 500 val perplexity 2275.3667
524
+ 500 train 8.356382 (lr=1.2222e-03) (hash(x)=24123821)
525
+ 501 train 7.351772 (lr=1.2198e-03) (hash(x)=20881126)
526
+ 502 train 7.727162 (lr=1.2174e-03) (hash(x)=23209969)
527
+ 503 train 7.747780 (lr=1.2150e-03) (hash(x)=26920243)
528
+ 504 train 7.637657 (lr=1.2126e-03) (hash(x)=26127756)
529
+ 505 train 7.581877 (lr=1.2101e-03) (hash(x)=21970343)
530
+ 506 train 7.642146 (lr=1.2077e-03) (hash(x)=24097897)
531
+ 507 train 7.754195 (lr=1.2052e-03) (hash(x)=31676352)
532
+ 508 train 7.706739 (lr=1.2028e-03) (hash(x)=23908053)
533
+ 509 train 7.647904 (lr=1.2003e-03) (hash(x)=25600562)
534
+ 510 train 7.700044 (lr=1.1978e-03) (hash(x)=23549168)
535
+ 511 train 7.629103 (lr=1.1954e-03) (hash(x)=22022924)
536
+ 512 train 7.707285 (lr=1.1929e-03) (hash(x)=23294980)
537
+ 513 train 7.598124 (lr=1.1904e-03) (hash(x)=22777085)
538
+ 514 train 7.876339 (lr=1.1879e-03) (hash(x)=26271115)
539
+ 515 train 7.670959 (lr=1.1854e-03) (hash(x)=21727744)
540
+ 516 train 7.707361 (lr=1.1829e-03) (hash(x)=22729318)
541
+ 517 train 7.889222 (lr=1.1803e-03) (hash(x)=20701998)
542
+ 518 train 7.833536 (lr=1.1778e-03) (hash(x)=27928398)
543
+ 519 train 7.906909 (lr=1.1753e-03) (hash(x)=20863982)
544
+ 520 train 7.776600 (lr=1.1727e-03) (hash(x)=22827054)
545
+ 521 train 7.759516 (lr=1.1702e-03) (hash(x)=25084926)
546
+ 522 train 7.856508 (lr=1.1676e-03) (hash(x)=25540961)
547
+ 523 train 7.591858 (lr=1.1651e-03) (hash(x)=24412897)
548
+ 524 train 7.854822 (lr=1.1625e-03) (hash(x)=26269869)
549
+ 525 train 7.643784 (lr=1.1599e-03) (hash(x)=25434913)
550
+ 526 train 7.821376 (lr=1.1573e-03) (hash(x)=25157653)
551
+ 527 train 7.837183 (lr=1.1548e-03) (hash(x)=26302544)
552
+ 528 train 7.720030 (lr=1.1522e-03) (hash(x)=22010703)
553
+ 529 train 8.022634 (lr=1.1496e-03) (hash(x)=34976197)
554
+ 530 train 7.710197 (lr=1.1470e-03) (hash(x)=28248301)
555
+ 531 train 7.738505 (lr=1.1443e-03) (hash(x)=23870805)
556
+ 532 train 7.734290 (lr=1.1417e-03) (hash(x)=24457252)
557
+ 533 train 7.787120 (lr=1.1391e-03) (hash(x)=25041103)
558
+ 534 train 7.740311 (lr=1.1365e-03) (hash(x)=25998487)
559
+ 535 train 7.851596 (lr=1.1338e-03) (hash(x)=26041745)
560
+ 536 train 7.763353 (lr=1.1312e-03) (hash(x)=27249810)
561
+ 537 train 7.813742 (lr=1.1285e-03) (hash(x)=24068619)
562
+ 538 train 7.833156 (lr=1.1259e-03) (hash(x)=26359528)
563
+ 539 train 7.804993 (lr=1.1232e-03) (hash(x)=25441262)
564
+ 540 train 7.742839 (lr=1.1205e-03) (hash(x)=24413659)
565
+ 541 train 8.161337 (lr=1.1179e-03) (hash(x)=29181867)
566
+ 542 train 7.921298 (lr=1.1152e-03) (hash(x)=31551575)
567
+ 543 train 7.877130 (lr=1.1125e-03) (hash(x)=27093757)
568
+ 544 train 7.854835 (lr=1.1098e-03) (hash(x)=23624298)
569
+ 545 train 7.701891 (lr=1.1071e-03) (hash(x)=25150275)
570
+ 546 train 7.716410 (lr=1.1044e-03) (hash(x)=26239886)
571
+ 547 train 8.084273 (lr=1.1017e-03) (hash(x)=33457102)
572
+ 548 train 7.656469 (lr=1.0990e-03) (hash(x)=17499738)
573
+ 549 train 7.716921 (lr=1.0963e-03) (hash(x)=25755322)
574
+ 550 val loss 7.7539
575
+ 550 val perplexity 2330.5642
576
+ 550 train 7.832466 (lr=1.0936e-03) (hash(x)=23722795)
577
+ 551 train 7.688102 (lr=1.0908e-03) (hash(x)=23630794)
578
+ 552 train 7.654415 (lr=1.0881e-03) (hash(x)=23374168)
579
+ 553 train 7.686577 (lr=1.0854e-03) (hash(x)=25167987)
580
+ 554 train 7.922827 (lr=1.0826e-03) (hash(x)=30941010)
581
+ 555 train 7.670557 (lr=1.0799e-03) (hash(x)=24043224)
582
+ 556 train 7.799884 (lr=1.0771e-03) (hash(x)=29736627)
583
+ 557 train 7.742640 (lr=1.0744e-03) (hash(x)=24665411)
584
+ 558 train 7.767422 (lr=1.0716e-03) (hash(x)=27916043)
585
+ 559 train 7.700399 (lr=1.0688e-03) (hash(x)=17321657)
586
+ 560 train 7.536227 (lr=1.0661e-03) (hash(x)=23829833)
587
+ 561 train 7.907607 (lr=1.0633e-03) (hash(x)=27932591)
588
+ 562 train 7.983245 (lr=1.0605e-03) (hash(x)=28206079)
589
+ 563 train 8.085007 (lr=1.0577e-03) (hash(x)=29447954)
590
+ 564 train 7.734852 (lr=1.0549e-03) (hash(x)=25935364)
591
+ 565 train 7.773105 (lr=1.0521e-03) (hash(x)=24621816)
592
+ 566 train 7.708507 (lr=1.0493e-03) (hash(x)=24070405)
593
+ 567 train 7.758605 (lr=1.0465e-03) (hash(x)=24682294)
594
+ 568 train 7.758308 (lr=1.0437e-03) (hash(x)=24050209)
595
+ 569 train 7.733538 (lr=1.0409e-03) (hash(x)=28057095)
596
+ 570 train 7.799509 (lr=1.0381e-03) (hash(x)=22523232)
597
+ 571 train 7.730530 (lr=1.0353e-03) (hash(x)=24547536)
598
+ 572 train 7.627197 (lr=1.0325e-03) (hash(x)=24978712)
599
+ 573 train 7.601028 (lr=1.0296e-03) (hash(x)=18987545)
600
+ 574 train 7.713584 (lr=1.0268e-03) (hash(x)=24092784)
601
+ 575 train 7.706183 (lr=1.0240e-03) (hash(x)=26745063)
602
+ 576 train 7.711241 (lr=1.0211e-03) (hash(x)=22539112)
603
+ 577 train 7.496004 (lr=1.0183e-03) (hash(x)=21447406)
604
+ 578 train 7.517599 (lr=1.0154e-03) (hash(x)=21082139)
605
+ 579 train 7.576668 (lr=1.0126e-03) (hash(x)=24650655)
606
+ 580 train 7.735999 (lr=1.0097e-03) (hash(x)=22130721)
607
+ 581 train 8.063599 (lr=1.0069e-03) (hash(x)=26519976)
608
+ 582 train 8.615137 (lr=1.0040e-03) (hash(x)=25582440)
609
+ 583 train 7.840294 (lr=1.0011e-03) (hash(x)=20106514)
610
+ 584 train 7.597269 (lr=9.9827e-04) (hash(x)=25695510)
611
+ 585 train 7.635041 (lr=9.9540e-04) (hash(x)=20117808)
612
+ 586 train 7.698525 (lr=9.9252e-04) (hash(x)=24501900)
613
+ 587 train 7.774756 (lr=9.8964e-04) (hash(x)=25510334)
614
+ 588 train 7.623492 (lr=9.8676e-04) (hash(x)=22099158)
615
+ 589 train 7.736458 (lr=9.8387e-04) (hash(x)=25749120)
616
+ 590 train 8.007345 (lr=9.8099e-04) (hash(x)=21720691)
617
+ 591 train 7.595894 (lr=9.7810e-04) (hash(x)=24357241)
618
+ 592 train 7.866467 (lr=9.7520e-04) (hash(x)=28911272)
619
+ 593 train 7.755000 (lr=9.7230e-04) (hash(x)=25584945)
620
+ 594 train 7.649728 (lr=9.6940e-04) (hash(x)=21847287)
621
+ 595 train 7.717608 (lr=9.6650e-04) (hash(x)=25198897)
622
+ 596 train 7.590934 (lr=9.6360e-04) (hash(x)=22436833)
623
+ 597 train 7.854661 (lr=9.6069e-04) (hash(x)=23475891)
624
+ 598 train 7.521270 (lr=9.5778e-04) (hash(x)=25083992)
625
+ 599 train 7.775483 (lr=9.5486e-04) (hash(x)=23644804)
626
+ 600 val loss 7.7063
627
+ 600 val perplexity 2222.2554
628
+ 600 train 7.643409 (lr=9.5195e-04) (hash(x)=31377168)
629
+ 601 train 8.023653 (lr=9.4903e-04) (hash(x)=25629584)
630
+ 602 train 7.859875 (lr=9.4611e-04) (hash(x)=29285135)
631
+ 603 train 8.050512 (lr=9.4319e-04) (hash(x)=29797280)
632
+ 604 train 7.890079 (lr=9.4026e-04) (hash(x)=33122326)
633
+ 605 train 8.135586 (lr=9.3733e-04) (hash(x)=30026394)
634
+ 606 train 8.395484 (lr=9.3440e-04) (hash(x)=34353164)
635
+ 607 train 7.661631 (lr=9.3147e-04) (hash(x)=26594177)
636
+ 608 train 7.758193 (lr=9.2854e-04) (hash(x)=25554738)
637
+ 609 train 7.846495 (lr=9.2560e-04) (hash(x)=23766256)
638
+ 610 train 7.732442 (lr=9.2267e-04) (hash(x)=26853444)
639
+ 611 train 7.744578 (lr=9.1973e-04) (hash(x)=24250150)
640
+ 612 train 7.665815 (lr=9.1678e-04) (hash(x)=26289588)
641
+ 613 train 7.751273 (lr=9.1384e-04) (hash(x)=26108817)
642
+ 614 train 7.680855 (lr=9.1090e-04) (hash(x)=23814987)
643
+ 615 train 7.688418 (lr=9.0795e-04) (hash(x)=18582688)
644
+ 616 train 7.703759 (lr=9.0500e-04) (hash(x)=25558792)
645
+ 617 train 7.658919 (lr=9.0205e-04) (hash(x)=23272705)
646
+ 618 train 7.662617 (lr=8.9910e-04) (hash(x)=24709241)
647
+ 619 train 7.666202 (lr=8.9615e-04) (hash(x)=24793412)
648
+ 620 train 7.687472 (lr=8.9319e-04) (hash(x)=27256707)
649
+ 621 train 7.637940 (lr=8.9024e-04) (hash(x)=19280989)
650
+ 622 train 7.705773 (lr=8.8728e-04) (hash(x)=26508642)
651
+ 623 train 7.613369 (lr=8.8432e-04) (hash(x)=22751150)
652
+ 624 train 7.895378 (lr=8.8136e-04) (hash(x)=23880160)
653
+ 625 train 7.632549 (lr=8.7840e-04) (hash(x)=23537942)
654
+ 626 train 7.647926 (lr=8.7544e-04) (hash(x)=26975313)
655
+ 627 train 7.542642 (lr=8.7248e-04) (hash(x)=23748375)
656
+ 628 train 7.575413 (lr=8.6952e-04) (hash(x)=25541754)
657
+ 629 train 7.959349 (lr=8.6655e-04) (hash(x)=27413825)
658
+ 630 train 7.882129 (lr=8.6359e-04) (hash(x)=27075951)
659
+ 631 train 7.733789 (lr=8.6062e-04) (hash(x)=21682445)
660
+ 632 train 7.691811 (lr=8.5766e-04) (hash(x)=23361504)
661
+ 633 train 7.614659 (lr=8.5469e-04) (hash(x)=22040184)
662
+ 634 train 7.663328 (lr=8.5172e-04) (hash(x)=23727731)
663
+ 635 train 7.649795 (lr=8.4876e-04) (hash(x)=24463229)
664
+ 636 train 7.833916 (lr=8.4579e-04) (hash(x)=32663792)
665
+ 637 train 7.676811 (lr=8.4282e-04) (hash(x)=24091954)
666
+ 638 train 7.644261 (lr=8.3985e-04) (hash(x)=23599013)
667
+ 639 train 7.673655 (lr=8.3688e-04) (hash(x)=26670219)
668
+ 640 train 7.656978 (lr=8.3391e-04) (hash(x)=24380893)
669
+ 641 train 7.710624 (lr=8.3094e-04) (hash(x)=25558986)
670
+ 642 train 7.779457 (lr=8.2797e-04) (hash(x)=25729492)
671
+ 643 train 7.542321 (lr=8.2500e-04) (hash(x)=19218943)
672
+ 644 train 7.759363 (lr=8.2203e-04) (hash(x)=23952206)
673
+ 645 train 7.725920 (lr=8.1906e-04) (hash(x)=24551402)
674
+ 646 train 7.611461 (lr=8.1609e-04) (hash(x)=24007642)
675
+ 647 train 7.649839 (lr=8.1312e-04) (hash(x)=25229281)
676
+ 648 train 7.449235 (lr=8.1015e-04) (hash(x)=18214283)
677
+ 649 train 7.388067 (lr=8.0718e-04) (hash(x)=21251127)
678
+ 650 val loss 7.6927
679
+ 650 val perplexity 2192.2354
680
+ 650 train 7.874477 (lr=8.0421e-04) (hash(x)=25097619)
681
+ 651 train 7.580329 (lr=8.0124e-04) (hash(x)=25800857)
682
+ 652 train 7.565134 (lr=7.9828e-04) (hash(x)=22918428)
683
+ 653 train 7.641159 (lr=7.9531e-04) (hash(x)=25836189)
684
+ 654 train 7.603850 (lr=7.9234e-04) (hash(x)=24504567)
685
+ 655 train 7.598814 (lr=7.8938e-04) (hash(x)=25999255)
686
+ 656 train 7.630563 (lr=7.8641e-04) (hash(x)=27308968)
687
+ 657 train 7.730077 (lr=7.8345e-04) (hash(x)=20149394)
688
+ 658 train 7.655795 (lr=7.8048e-04) (hash(x)=24498556)
689
+ 659 train 7.531661 (lr=7.7752e-04) (hash(x)=21999338)
690
+ 660 train 7.543416 (lr=7.7456e-04) (hash(x)=23280568)
691
+ 661 train 7.835412 (lr=7.7160e-04) (hash(x)=27611302)
692
+ 662 train 7.679036 (lr=7.6864e-04) (hash(x)=23982308)
693
+ 663 train 7.632465 (lr=7.6568e-04) (hash(x)=22778356)
694
+ 664 train 7.555626 (lr=7.6272e-04) (hash(x)=23467046)
695
+ 665 train 7.722482 (lr=7.5976e-04) (hash(x)=24195959)
696
+ 666 train 8.002197 (lr=7.5681e-04) (hash(x)=32710438)
697
+ 667 train 7.738972 (lr=7.5385e-04) (hash(x)=19127465)
698
+ 668 train 7.883466 (lr=7.5090e-04) (hash(x)=28962580)
699
+ 669 train 7.626482 (lr=7.4795e-04) (hash(x)=22426274)
700
+ 670 train 7.645793 (lr=7.4500e-04) (hash(x)=25606512)
701
+ 671 train 7.509284 (lr=7.4205e-04) (hash(x)=23785205)
702
+ 672 train 7.651512 (lr=7.3910e-04) (hash(x)=24779414)
703
+ 673 train 7.776501 (lr=7.3616e-04) (hash(x)=28505676)
704
+ 674 train 7.838102 (lr=7.3322e-04) (hash(x)=29156288)
705
+ 675 train 7.533864 (lr=7.3027e-04) (hash(x)=27706477)
706
+ 676 train 7.668046 (lr=7.2733e-04) (hash(x)=23711800)
707
+ 677 train 7.747730 (lr=7.2440e-04) (hash(x)=28684175)
708
+ 678 train 7.628632 (lr=7.2146e-04) (hash(x)=25225466)
709
+ 679 train 7.745574 (lr=7.1853e-04) (hash(x)=28003600)
710
+ 680 train 7.680043 (lr=7.1560e-04) (hash(x)=27989890)
711
+ 681 train 7.705679 (lr=7.1267e-04) (hash(x)=27586845)
712
+ 682 train 7.682579 (lr=7.0974e-04) (hash(x)=22373927)
713
+ 683 train 7.563611 (lr=7.0681e-04) (hash(x)=21176405)
714
+ 684 train 7.527731 (lr=7.0389e-04) (hash(x)=23806052)
715
+ 685 train 7.629662 (lr=7.0097e-04) (hash(x)=25888433)
716
+ 686 train 7.759294 (lr=6.9805e-04) (hash(x)=27797727)
717
+ 687 train 7.628597 (lr=6.9514e-04) (hash(x)=27916982)
718
+ 688 train 7.578324 (lr=6.9222e-04) (hash(x)=24621793)
719
+ 689 train 7.496653 (lr=6.8931e-04) (hash(x)=24334708)
720
+ 690 train 7.654587 (lr=6.8640e-04) (hash(x)=24419143)
721
+ 691 train 7.557160 (lr=6.8350e-04) (hash(x)=25617709)
722
+ 692 train 7.547180 (lr=6.8060e-04) (hash(x)=24556034)
723
+ 693 train 7.920913 (lr=6.7770e-04) (hash(x)=25152362)
724
+ 694 train 7.656851 (lr=6.7480e-04) (hash(x)=24326376)
725
+ 695 train 7.545552 (lr=6.7190e-04) (hash(x)=24276512)
726
+ 696 train 7.414201 (lr=6.6901e-04) (hash(x)=24344695)
727
+ 697 train 7.650876 (lr=6.6613e-04) (hash(x)=26776133)
728
+ 698 train 7.514264 (lr=6.6324e-04) (hash(x)=21847282)
729
+ 699 train 7.379858 (lr=6.6036e-04) (hash(x)=20254159)
730
+ 700 val loss 7.6545
731
+ 700 val perplexity 2110.1521
732
+ 700 train 7.470055 (lr=6.5748e-04) (hash(x)=25044762)
733
+ 701 train 7.574930 (lr=6.5460e-04) (hash(x)=27873855)
734
+ 702 train 7.657175 (lr=6.5173e-04) (hash(x)=22575615)
735
+ 703 train 7.620751 (lr=6.4886e-04) (hash(x)=25113614)
736
+ 704 train 7.727670 (lr=6.4600e-04) (hash(x)=28062905)
737
+ 705 train 7.646277 (lr=6.4314e-04) (hash(x)=26192886)
738
+ 706 train 7.631146 (lr=6.4028e-04) (hash(x)=25770338)
739
+ 707 train 7.516666 (lr=6.3742e-04) (hash(x)=24614912)
740
+ 708 train 7.594297 (lr=6.3457e-04) (hash(x)=26934071)
741
+ 709 train 7.571792 (lr=6.3172e-04) (hash(x)=25397093)
742
+ 710 train 7.711982 (lr=6.2888e-04) (hash(x)=25925963)
743
+ 711 train 7.564043 (lr=6.2604e-04) (hash(x)=19811802)
744
+ 712 train 7.638424 (lr=6.2320e-04) (hash(x)=23248423)
745
+ 713 train 7.619357 (lr=6.2037e-04) (hash(x)=25579655)
746
+ 714 train 7.802995 (lr=6.1754e-04) (hash(x)=25537603)
747
+ 715 train 7.677469 (lr=6.1472e-04) (hash(x)=26096514)
748
+ 716 train 7.709909 (lr=6.1190e-04) (hash(x)=24132888)
749
+ 717 train 7.540712 (lr=6.0908e-04) (hash(x)=22589633)
750
+ 718 train 7.510253 (lr=6.0627e-04) (hash(x)=27781566)
751
+ 719 train 7.751784 (lr=6.0346e-04) (hash(x)=26757147)
752
+ 720 train 7.791998 (lr=6.0066e-04) (hash(x)=26649864)
753
+ 721 train 7.655463 (lr=5.9786e-04) (hash(x)=24963730)
754
+ 722 train 7.658646 (lr=5.9507e-04) (hash(x)=25249959)
755
+ 723 train 7.509624 (lr=5.9228e-04) (hash(x)=27703124)
756
+ 724 train 7.522089 (lr=5.8949e-04) (hash(x)=26459082)
757
+ 725 train 7.478124 (lr=5.8671e-04) (hash(x)=24781713)
758
+ 726 train 7.758729 (lr=5.8393e-04) (hash(x)=26597693)
759
+ 727 train 7.582549 (lr=5.8116e-04) (hash(x)=24720476)
760
+ 728 train 7.609679 (lr=5.7839e-04) (hash(x)=23691295)
761
+ 729 train 7.725364 (lr=5.7563e-04) (hash(x)=23481301)
762
+ 730 train 7.612714 (lr=5.7288e-04) (hash(x)=23295676)
763
+ 731 train 7.571315 (lr=5.7012e-04) (hash(x)=25641033)
764
+ 732 train 7.677575 (lr=5.6737e-04) (hash(x)=22441379)
765
+ 733 train 7.562910 (lr=5.6463e-04) (hash(x)=27369437)
766
+ 734 train 7.600533 (lr=5.6189e-04) (hash(x)=24482272)
767
+ 735 train 7.717937 (lr=5.5916e-04) (hash(x)=24760900)
768
+ 736 train 7.625099 (lr=5.5643e-04) (hash(x)=23919156)
769
+ 737 train 7.531400 (lr=5.5371e-04) (hash(x)=21692546)
770
+ 738 train 7.540369 (lr=5.5100e-04) (hash(x)=27692087)
771
+ 739 train 7.684278 (lr=5.4828e-04) (hash(x)=22720391)
772
+ 740 train 7.480987 (lr=5.4558e-04) (hash(x)=26146560)
773
+ 741 train 7.598684 (lr=5.4288e-04) (hash(x)=24376979)
774
+ 742 train 7.637449 (lr=5.4018e-04) (hash(x)=22917712)
775
+ 743 train 7.519206 (lr=5.3749e-04) (hash(x)=22244509)
776
+ 744 train 7.525835 (lr=5.3481e-04) (hash(x)=25243385)
777
+ 745 train 7.540352 (lr=5.3213e-04) (hash(x)=21831960)
778
+ 746 train 7.548644 (lr=5.2946e-04) (hash(x)=24842373)
779
+ 747 train 7.459107 (lr=5.2679e-04) (hash(x)=24921535)
780
+ 748 train 7.538298 (lr=5.2413e-04) (hash(x)=24344615)
781
+ 749 train 7.555773 (lr=5.2147e-04) (hash(x)=25381251)
782
+ 750 val loss 7.6435
783
+ 750 val perplexity 2087.0811
784
+ 750 train 7.426806 (lr=5.1882e-04) (hash(x)=22887303)
785
+ 751 train 7.637200 (lr=5.1618e-04) (hash(x)=25687352)
786
+ 752 train 7.603520 (lr=5.1354e-04) (hash(x)=27791412)
787
+ 753 train 7.508053 (lr=5.1091e-04) (hash(x)=25221654)
788
+ 754 train 7.611265 (lr=5.0828e-04) (hash(x)=23243731)
789
+ 755 train 7.585793 (lr=5.0566e-04) (hash(x)=24469863)
790
+ 756 train 7.709322 (lr=5.0305e-04) (hash(x)=22484936)
791
+ 757 train 7.624274 (lr=5.0044e-04) (hash(x)=25291413)
792
+ 758 train 7.514065 (lr=4.9784e-04) (hash(x)=23952225)
793
+ 759 train 7.609661 (lr=4.9525e-04) (hash(x)=25744274)
794
+ 760 train 7.524535 (lr=4.9266e-04) (hash(x)=22616739)
795
+ 761 train 7.491142 (lr=4.9008e-04) (hash(x)=24246963)
796
+ 762 train 7.495816 (lr=4.8750e-04) (hash(x)=19407094)
797
+ 763 train 7.543560 (lr=4.8493e-04) (hash(x)=19724058)
798
+ 764 train 7.409759 (lr=4.8237e-04) (hash(x)=18836491)
799
+ 765 train 7.453597 (lr=4.7981e-04) (hash(x)=20705649)
800
+ 766 train 7.684991 (lr=4.7726e-04) (hash(x)=21325875)
801
+ 767 train 7.725353 (lr=4.7472e-04) (hash(x)=24938685)
802
+ 768 train 7.664145 (lr=4.7219e-04) (hash(x)=25152403)
803
+ 769 train 7.790573 (lr=4.6966e-04) (hash(x)=21707000)
804
+ 770 train 7.489848 (lr=4.6714e-04) (hash(x)=17952018)
805
+ 771 train 7.534187 (lr=4.6462e-04) (hash(x)=24948980)
806
+ 772 train 7.961640 (lr=4.6211e-04) (hash(x)=28012110)
807
+ 773 train 7.616504 (lr=4.5961e-04) (hash(x)=30773958)
808
+ 774 train 8.169166 (lr=4.5712e-04) (hash(x)=29481068)
809
+ 775 train 7.621696 (lr=4.5463e-04) (hash(x)=26769046)
810
+ 776 train 7.558432 (lr=4.5215e-04) (hash(x)=26755048)
811
+ 777 train 7.532619 (lr=4.4968e-04) (hash(x)=18849656)
812
+ 778 train 7.804156 (lr=4.4722e-04) (hash(x)=25449800)
813
+ 779 train 8.266892 (lr=4.4476e-04) (hash(x)=24624950)
814
+ 780 train 7.595702 (lr=4.4231e-04) (hash(x)=26174069)
815
+ 781 train 7.758444 (lr=4.3987e-04) (hash(x)=27299605)
816
+ 782 train 7.701699 (lr=4.3743e-04) (hash(x)=24777273)
817
+ 783 train 7.703703 (lr=4.3500e-04) (hash(x)=24113445)
818
+ 784 train 7.595399 (lr=4.3258e-04) (hash(x)=23970384)
819
+ 785 train 7.794960 (lr=4.3017e-04) (hash(x)=32644465)
820
+ 786 train 7.904031 (lr=4.2776e-04) (hash(x)=29951548)
821
+ 787 train 7.643375 (lr=4.2537e-04) (hash(x)=25829219)
822
+ 788 train 7.690280 (lr=4.2298e-04) (hash(x)=20556094)
823
+ 789 train 7.539969 (lr=4.2060e-04) (hash(x)=25194550)
824
+ 790 train 7.658470 (lr=4.1822e-04) (hash(x)=28482186)
825
+ 791 train 7.636584 (lr=4.1586e-04) (hash(x)=25941804)
826
+ 792 train 7.630209 (lr=4.1350e-04) (hash(x)=27049208)
827
+ 793 train 7.635911 (lr=4.1115e-04) (hash(x)=24075727)
828
+ 794 train 7.726843 (lr=4.0880e-04) (hash(x)=25082806)
829
+ 795 train 7.677123 (lr=4.0647e-04) (hash(x)=23104539)
830
+ 796 train 7.593177 (lr=4.0414e-04) (hash(x)=25351113)
831
+ 797 train 7.696551 (lr=4.0183e-04) (hash(x)=23018983)
832
+ 798 train 7.615673 (lr=3.9952e-04) (hash(x)=25302106)
833
+ 799 train 7.552928 (lr=3.9722e-04) (hash(x)=24422929)
834
+ 800 val loss 7.6394
835
+ 800 val perplexity 2078.4380
836
+ 800 train 7.602225 (lr=3.9492e-04) (hash(x)=29052117)
837
+ 801 train 7.616655 (lr=3.9264e-04) (hash(x)=23389696)
838
+ 802 train 7.530201 (lr=3.9036e-04) (hash(x)=23962503)
839
+ 803 train 7.641126 (lr=3.8809e-04) (hash(x)=29102969)
840
+ 804 train 7.754502 (lr=3.8583e-04) (hash(x)=24948166)
841
+ 805 train 7.708795 (lr=3.8358e-04) (hash(x)=28359859)
842
+ 806 train 7.694516 (lr=3.8134e-04) (hash(x)=22360298)
843
+ 807 train 7.780490 (lr=3.7910e-04) (hash(x)=24630955)
844
+ 808 train 7.793565 (lr=3.7688e-04) (hash(x)=26897402)
845
+ 809 train 7.729742 (lr=3.7466e-04) (hash(x)=22969113)
846
+ 810 train 7.568577 (lr=3.7245e-04) (hash(x)=25205781)
847
+ 811 train 7.600789 (lr=3.7026e-04) (hash(x)=27004415)
848
+ 812 train 7.468259 (lr=3.6806e-04) (hash(x)=22974875)
849
+ 813 train 7.545944 (lr=3.6588e-04) (hash(x)=22265063)
850
+ 814 train 7.473920 (lr=3.6371e-04) (hash(x)=23312114)
851
+ 815 train 7.594315 (lr=3.6155e-04) (hash(x)=26905990)
852
+ 816 train 7.579319 (lr=3.5939e-04) (hash(x)=27089280)
853
+ 817 train 7.558719 (lr=3.5725e-04) (hash(x)=24047679)
854
+ 818 train 7.615869 (lr=3.5511e-04) (hash(x)=25034684)
855
+ 819 train 7.780413 (lr=3.5298e-04) (hash(x)=26905582)
856
+ 820 train 7.461578 (lr=3.5086e-04) (hash(x)=21406950)
857
+ 821 train 7.479541 (lr=3.4875e-04) (hash(x)=29018970)
858
+ 822 train 7.455969 (lr=3.4665e-04) (hash(x)=20853530)
859
+ 823 train 7.533356 (lr=3.4456e-04) (hash(x)=24503315)
860
+ 824 train 7.552749 (lr=3.4248e-04) (hash(x)=23080074)
861
+ 825 train 7.408076 (lr=3.4041e-04) (hash(x)=22607537)
862
+ 826 train 7.813303 (lr=3.3835e-04) (hash(x)=32179773)
863
+ 827 train 8.904925 (lr=3.3629e-04) (hash(x)=26690225)
864
+ 828 train 7.581406 (lr=3.3425e-04) (hash(x)=22109609)
865
+ 829 train 7.552256 (lr=3.3221e-04) (hash(x)=24226255)
866
+ 830 train 7.599922 (lr=3.3019e-04) (hash(x)=23013573)
867
+ 831 train 7.418430 (lr=3.2817e-04) (hash(x)=23622702)
868
+ 832 train 7.602446 (lr=3.2617e-04) (hash(x)=21517429)
869
+ 833 train 7.515487 (lr=3.2417e-04) (hash(x)=23910425)
870
+ 834 train 7.649506 (lr=3.2219e-04) (hash(x)=24407484)
871
+ 835 train 7.959629 (lr=3.2021e-04) (hash(x)=27213318)
872
+ 836 train 7.580740 (lr=3.1824e-04) (hash(x)=25524191)
873
+ 837 train 7.609062 (lr=3.1629e-04) (hash(x)=26701577)
874
+ 838 train 7.449671 (lr=3.1434e-04) (hash(x)=22091266)
875
+ 839 train 7.573025 (lr=3.1240e-04) (hash(x)=26216794)
876
+ 840 train 7.506999 (lr=3.1047e-04) (hash(x)=20817340)
877
+ 841 train 7.908933 (lr=3.0856e-04) (hash(x)=27579623)
878
+ 842 train 7.441338 (lr=3.0665e-04) (hash(x)=21532187)
879
+ 843 train 7.715135 (lr=3.0475e-04) (hash(x)=26086198)
880
+ 844 train 7.559477 (lr=3.0286e-04) (hash(x)=23911729)
881
+ 845 train 7.552556 (lr=3.0099e-04) (hash(x)=24863441)
882
+ 846 train 7.976576 (lr=2.9912e-04) (hash(x)=28856978)
883
+ 847 train 7.396081 (lr=2.9726e-04) (hash(x)=25236385)
884
+ 848 train 7.666218 (lr=2.9542e-04) (hash(x)=24278687)
885
+ 849 train 7.791800 (lr=2.9358e-04) (hash(x)=25330803)
886
+ 850 val loss 7.6027
887
+ 850 val perplexity 2003.6523
888
+ 850 train 7.644930 (lr=2.9175e-04) (hash(x)=21568545)
889
+ 851 train 7.554082 (lr=2.8994e-04) (hash(x)=22508532)
890
+ 852 train 7.597363 (lr=2.8813e-04) (hash(x)=24850662)
891
+ 853 train 7.578736 (lr=2.8634e-04) (hash(x)=21799261)
892
+ 854 train 7.748252 (lr=2.8455e-04) (hash(x)=24884891)
893
+ 855 train 7.563586 (lr=2.8278e-04) (hash(x)=26311168)
894
+ 856 train 7.622894 (lr=2.8102e-04) (hash(x)=27549895)
895
+ 857 train 7.316123 (lr=2.7926e-04) (hash(x)=20977430)
896
+ 858 train 7.454205 (lr=2.7752e-04) (hash(x)=23448855)
897
+ 859 train 7.651011 (lr=2.7579e-04) (hash(x)=26941617)
898
+ 860 train 7.828688 (lr=2.7407e-04) (hash(x)=26953192)
899
+ 861 train 7.772452 (lr=2.7236e-04) (hash(x)=30008957)
900
+ 862 train 7.694398 (lr=2.7066e-04) (hash(x)=27069893)
901
+ 863 train 7.599497 (lr=2.6897e-04) (hash(x)=27125962)
902
+ 864 train 7.435337 (lr=2.6729e-04) (hash(x)=23341415)
903
+ 865 train 7.534457 (lr=2.6562e-04) (hash(x)=25888358)
904
+ 866 train 7.691815 (lr=2.6396e-04) (hash(x)=26761645)
905
+ 867 train 7.599608 (lr=2.6232e-04) (hash(x)=21725719)
906
+ 868 train 7.509883 (lr=2.6068e-04) (hash(x)=25746493)
907
+ 869 train 7.517783 (lr=2.5906e-04) (hash(x)=21669704)
908
+ 870 train 7.454368 (lr=2.5745e-04) (hash(x)=21167493)
909
+ 871 train 7.541224 (lr=2.5584e-04) (hash(x)=23166993)
910
+ 872 train 7.498191 (lr=2.5425e-04) (hash(x)=23947208)
911
+ 873 train 7.617319 (lr=2.5267e-04) (hash(x)=26235974)
912
+ 874 train 7.867727 (lr=2.5110e-04) (hash(x)=29822604)
913
+ 875 train 7.494525 (lr=2.4955e-04) (hash(x)=25426430)
914
+ 876 train 7.494351 (lr=2.4800e-04) (hash(x)=26043735)
915
+ 877 train 7.440683 (lr=2.4646e-04) (hash(x)=22966314)
916
+ 878 train 7.718173 (lr=2.4494e-04) (hash(x)=27944619)
917
+ 879 train 7.501267 (lr=2.4343e-04) (hash(x)=23448505)
918
+ 880 train 7.522300 (lr=2.4192e-04) (hash(x)=24205942)
919
+ 881 train 7.437175 (lr=2.4043e-04) (hash(x)=26330693)
920
+ 882 train 7.437334 (lr=2.3895e-04) (hash(x)=24844466)
921
+ 883 train 7.418519 (lr=2.3749e-04) (hash(x)=29550596)
922
+ 884 train 7.493869 (lr=2.3603e-04) (hash(x)=20506540)
923
+ 885 train 7.511559 (lr=2.3458e-04) (hash(x)=22124892)
924
+ 886 train 7.489201 (lr=2.3315e-04) (hash(x)=24867036)
925
+ 887 train 7.527057 (lr=2.3173e-04) (hash(x)=25607640)
926
+ 888 train 7.531439 (lr=2.3032e-04) (hash(x)=26176930)
927
+ 889 train 7.256721 (lr=2.2892e-04) (hash(x)=23795894)
928
+ 890 train 7.188067 (lr=2.2753e-04) (hash(x)=23756298)
929
+ 891 train 7.410231 (lr=2.2615e-04) (hash(x)=24165449)
930
+ 892 train 7.440681 (lr=2.2479e-04) (hash(x)=26643739)
931
+ 893 train 7.608705 (lr=2.2344e-04) (hash(x)=24697756)
932
+ 894 train 7.523702 (lr=2.2209e-04) (hash(x)=24431101)
933
+ 895 train 7.661815 (lr=2.2076e-04) (hash(x)=27218899)
934
+ 896 train 7.464298 (lr=2.1945e-04) (hash(x)=29462219)
935
+ 897 train 7.510637 (lr=2.1814e-04) (hash(x)=24748522)
936
+ 898 train 7.514422 (lr=2.1685e-04) (hash(x)=24843486)
937
+ 899 train 7.654904 (lr=2.1556e-04) (hash(x)=24965317)
938
+ 900 val loss 7.5896
939
+ 900 val perplexity 1977.5272
940
+ 900 train 7.412416 (lr=2.1429e-04) (hash(x)=26584754)
941
+ 901 train 7.354466 (lr=2.1303e-04) (hash(x)=22031210)
942
+ 902 train 7.544207 (lr=2.1179e-04) (hash(x)=24860582)
943
+ 903 train 7.474760 (lr=2.1055e-04) (hash(x)=25877990)
944
+ 904 train 7.518291 (lr=2.0933e-04) (hash(x)=24970646)
945
+ 905 train 7.561033 (lr=2.0812e-04) (hash(x)=24382381)
946
+ 906 train 7.468290 (lr=2.0692e-04) (hash(x)=25221746)
947
+ 907 train 7.442141 (lr=2.0573e-04) (hash(x)=24551999)
948
+ 908 train 7.471278 (lr=2.0455e-04) (hash(x)=28151378)
949
+ 909 train 7.609212 (lr=2.0339e-04) (hash(x)=25765516)
950
+ 910 train 7.589880 (lr=2.0224e-04) (hash(x)=25630696)
951
+ 911 train 7.567298 (lr=2.0110e-04) (hash(x)=25204247)
952
+ 912 train 7.632396 (lr=1.9997e-04) (hash(x)=26326447)
953
+ 913 train 7.610932 (lr=1.9885e-04) (hash(x)=26321813)
954
+ 914 train 7.556074 (lr=1.9775e-04) (hash(x)=25400396)
955
+ 915 train 7.546350 (lr=1.9666e-04) (hash(x)=25528323)
956
+ 916 train 7.556988 (lr=1.9558e-04) (hash(x)=24035353)
957
+ 917 train 7.555608 (lr=1.9451e-04) (hash(x)=25114729)
958
+ 918 train 7.449489 (lr=1.9346e-04) (hash(x)=21634962)
959
+ 919 train 7.586897 (lr=1.9242e-04) (hash(x)=31341964)
960
+ 920 train 7.489885 (lr=1.9139e-04) (hash(x)=29172813)
961
+ 921 train 7.666150 (lr=1.9037e-04) (hash(x)=26326024)
962
+ 922 train 7.441479 (lr=1.8936e-04) (hash(x)=20140071)
963
+ 923 train 7.534974 (lr=1.8837e-04) (hash(x)=27273311)
964
+ 924 train 7.621279 (lr=1.8739e-04) (hash(x)=31276487)
965
+ 925 train 7.339494 (lr=1.8642e-04) (hash(x)=17819313)
966
+ 926 train 7.679044 (lr=1.8547e-04) (hash(x)=25843618)
967
+ 927 train 7.617510 (lr=1.8452e-04) (hash(x)=23769243)
968
+ 928 train 7.541837 (lr=1.8359e-04) (hash(x)=28731298)
969
+ 929 train 7.643688 (lr=1.8267e-04) (hash(x)=23788738)
970
+ 930 train 7.493644 (lr=1.8176e-04) (hash(x)=26599031)
971
+ 931 train 7.593131 (lr=1.8087e-04) (hash(x)=24818744)
972
+ 932 train 7.499806 (lr=1.7999e-04) (hash(x)=23053871)
973
+ 933 train 7.440404 (lr=1.7912e-04) (hash(x)=23472795)
974
+ 934 train 7.986732 (lr=1.7826e-04) (hash(x)=29748225)
975
+ 935 train 7.711870 (lr=1.7742e-04) (hash(x)=24991832)
976
+ 936 train 7.733427 (lr=1.7659e-04) (hash(x)=23681545)
977
+ 937 train 7.512547 (lr=1.7577e-04) (hash(x)=22424066)
978
+ 938 train 7.523259 (lr=1.7496e-04) (hash(x)=25262748)
979
+ 939 train 7.494957 (lr=1.7417e-04) (hash(x)=26474878)
980
+ 940 train 7.693963 (lr=1.7339e-04) (hash(x)=23930593)
981
+ 941 train 7.477695 (lr=1.7262e-04) (hash(x)=22704349)
982
+ 942 train 7.565039 (lr=1.7186e-04) (hash(x)=24801048)
983
+ 943 train 7.518791 (lr=1.7112e-04) (hash(x)=22293673)
984
+ 944 train 7.566538 (lr=1.7039e-04) (hash(x)=23311934)
985
+ 945 train 7.547297 (lr=1.6967e-04) (hash(x)=21785487)
986
+ 946 train 7.447145 (lr=1.6896e-04) (hash(x)=22981456)
987
+ 947 train 7.701120 (lr=1.6827e-04) (hash(x)=25251063)
988
+ 948 train 7.516181 (lr=1.6759e-04) (hash(x)=25796422)
989
+ 949 train 7.438559 (lr=1.6692e-04) (hash(x)=22722211)
990
+ 950 val loss 7.5807
991
+ 950 val perplexity 1959.9583
992
+ 950 train 7.525630 (lr=1.6627e-04) (hash(x)=24296755)
993
+ 951 train 7.529128 (lr=1.6563e-04) (hash(x)=24522269)
994
+ 952 train 7.542082 (lr=1.6500e-04) (hash(x)=25617184)
995
+ 953 train 7.444283 (lr=1.6438e-04) (hash(x)=23329475)
996
+ 954 train 7.545440 (lr=1.6378e-04) (hash(x)=30652062)
997
+ 955 train 7.501037 (lr=1.6319e-04) (hash(x)=24735165)
998
+ 956 train 7.317701 (lr=1.6261e-04) (hash(x)=23795686)
999
+ 957 train 7.368524 (lr=1.6205e-04) (hash(x)=27002892)
1000
+ 958 train 7.426277 (lr=1.6149e-04) (hash(x)=21189101)
1001
+ 959 train 7.341558 (lr=1.6095e-04) (hash(x)=23499349)
1002
+ 960 train 7.885412 (lr=1.6043e-04) (hash(x)=26807896)
1003
+ 961 train 7.509871 (lr=1.5991e-04) (hash(x)=22461589)
1004
+ 962 train 7.560937 (lr=1.5941e-04) (hash(x)=26904123)
1005
+ 963 train 7.412053 (lr=1.5893e-04) (hash(x)=26415040)
1006
+ 964 train 7.681058 (lr=1.5845e-04) (hash(x)=24463181)
1007
+ 965 train 7.718406 (lr=1.5799e-04) (hash(x)=26349465)
1008
+ 966 train 7.361156 (lr=1.5754e-04) (hash(x)=21205988)
1009
+ 967 train 7.416436 (lr=1.5710e-04) (hash(x)=21596677)
1010
+ 968 train 7.515753 (lr=1.5668e-04) (hash(x)=23575666)
1011
+ 969 train 7.338404 (lr=1.5627e-04) (hash(x)=24117012)
1012
+ 970 train 7.421603 (lr=1.5587e-04) (hash(x)=26013014)
1013
+ 971 train 7.563095 (lr=1.5549e-04) (hash(x)=24105761)
1014
+ 972 train 7.443137 (lr=1.5512e-04) (hash(x)=20369133)
1015
+ 973 train 7.469945 (lr=1.5476e-04) (hash(x)=25725049)
1016
+ 974 train 7.313016 (lr=1.5441e-04) (hash(x)=19808118)
1017
+ 975 train 7.260711 (lr=1.5408e-04) (hash(x)=22892776)
1018
+ 976 train 7.443898 (lr=1.5376e-04) (hash(x)=23736780)
1019
+ 977 train 7.389701 (lr=1.5345e-04) (hash(x)=21630383)
1020
+ 978 train 7.587798 (lr=1.5316e-04) (hash(x)=25397467)
1021
+ 979 train 7.344945 (lr=1.5288e-04) (hash(x)=24727094)
1022
+ 980 train 7.435334 (lr=1.5261e-04) (hash(x)=23951182)
1023
+ 981 train 7.373097 (lr=1.5236e-04) (hash(x)=22835000)
1024
+ 982 train 7.557577 (lr=1.5212e-04) (hash(x)=28767849)
1025
+ 983 train 7.546004 (lr=1.5189e-04) (hash(x)=22839049)
1026
+ 984 train 7.461970 (lr=1.5167e-04) (hash(x)=19930900)
1027
+ 985 train 7.362316 (lr=1.5147e-04) (hash(x)=20654136)
1028
+ 986 train 7.551844 (lr=1.5128e-04) (hash(x)=28495621)
1029
+ 987 train 7.705873 (lr=1.5110e-04) (hash(x)=26776000)
1030
+ 988 train 7.370848 (lr=1.5094e-04) (hash(x)=21584429)
1031
+ 989 train 7.170113 (lr=1.5079e-04) (hash(x)=22560658)
1032
+ 990 train 7.764584 (lr=1.5065e-04) (hash(x)=25276565)
1033
+ 991 train 7.555707 (lr=1.5053e-04) (hash(x)=25995933)
1034
+ 992 train 7.618915 (lr=1.5042e-04) (hash(x)=24163910)
1035
+ 993 train 7.571279 (lr=1.5032e-04) (hash(x)=24635394)
1036
+ 994 train 7.673279 (lr=1.5024e-04) (hash(x)=22556381)
1037
+ 995 train 7.737123 (lr=1.5016e-04) (hash(x)=27537228)
1038
+ 996 train 7.549344 (lr=1.5010e-04) (hash(x)=20883633)
1039
+ 997 train 7.699744 (lr=1.5006e-04) (hash(x)=26496730)
1040
+ 998 train 7.343485 (lr=1.5003e-04) (hash(x)=20342150)
1041
+ 999 val loss 7.5731
1042
+ 999 val perplexity 1945.2418
1043
+ 999 train 7.343350 (lr=1.5001e-04) (hash(x)=16533310)
lr1.5e-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 sha256:039f64b8661e9fe94d6103349581e4b14cf6e740c8219009f2c531e7d9b87ca1
3
+ size 48156226
lr1.5e-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:dd114f5d7b87b1c672eb2282deddb7c2866473ae6b282dfd619d61b563aa66cd
3
+ size 71157574