andrew-healey commited on
Commit
565ae2b
·
verified ·
1 Parent(s): 34c4f15

Upload folder using huggingface_hub

Browse files
lr1e-4_total_batch_size40960_seq_len1024/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep/lr1e-4_total_batch_size40960_seq_len1024", "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", "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": 1024, "batch_size": null, "total_batch_size": 40960, "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": 0.0001, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr1e-4_total_batch_size40960_seq_len1024/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cec289292cf51a5462e2631af59109e64df3eca2d9e62e1fbd1c53a776a0f874
3
+ size 964
lr1e-4_total_batch_size40960_seq_len1024/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.7486
3
+ 0 val perplexity 126579.2500
4
+ 0 train 11.753929 (lr=3.4965e-07) (hash(x)=52769534)
5
+ 1 train 11.745090 (lr=6.9930e-07) (hash(x)=48487754)
6
+ 2 train 11.746176 (lr=1.0490e-06) (hash(x)=48658307)
7
+ 3 train 11.760347 (lr=1.3986e-06) (hash(x)=50106566)
8
+ 4 train 11.740828 (lr=1.7483e-06) (hash(x)=45839663)
9
+ 5 train 11.732976 (lr=2.0979e-06) (hash(x)=50770821)
10
+ 6 train 11.729277 (lr=2.4476e-06) (hash(x)=52323135)
11
+ 7 train 11.726438 (lr=2.7972e-06) (hash(x)=50183213)
12
+ 8 train 11.717405 (lr=3.1469e-06) (hash(x)=56845779)
13
+ 9 train 11.714921 (lr=3.4965e-06) (hash(x)=52671072)
14
+ 10 train 11.717676 (lr=3.8462e-06) (hash(x)=52068996)
15
+ 11 train 11.690260 (lr=4.1958e-06) (hash(x)=47640895)
16
+ 12 train 11.679607 (lr=4.5455e-06) (hash(x)=45438450)
17
+ 13 train 11.686937 (lr=4.8951e-06) (hash(x)=62300690)
18
+ 14 train 11.682680 (lr=5.2448e-06) (hash(x)=54520776)
19
+ 15 train 11.663060 (lr=5.5944e-06) (hash(x)=54982586)
20
+ 16 train 11.658194 (lr=5.9441e-06) (hash(x)=50188334)
21
+ 17 train 11.621685 (lr=6.2937e-06) (hash(x)=60945351)
22
+ 18 train 11.644141 (lr=6.6434e-06) (hash(x)=58412625)
23
+ 19 train 11.634220 (lr=6.9930e-06) (hash(x)=51819337)
24
+ 20 train 11.593446 (lr=7.3427e-06) (hash(x)=52889428)
25
+ 21 train 11.579468 (lr=7.6923e-06) (hash(x)=50686547)
26
+ 22 train 11.565095 (lr=8.0420e-06) (hash(x)=49697510)
27
+ 23 train 11.550033 (lr=8.3916e-06) (hash(x)=46807632)
28
+ 24 train 11.540322 (lr=8.7413e-06) (hash(x)=57723080)
29
+ 25 train 11.508802 (lr=9.0909e-06) (hash(x)=48459663)
30
+ 26 train 11.493458 (lr=9.4406e-06) (hash(x)=47148322)
31
+ 27 train 11.477474 (lr=9.7902e-06) (hash(x)=45337189)
32
+ 28 train 11.441739 (lr=1.0140e-05) (hash(x)=42125704)
33
+ 29 train 11.411876 (lr=1.0490e-05) (hash(x)=50618057)
34
+ 30 train 11.420345 (lr=1.0839e-05) (hash(x)=51754551)
35
+ 31 train 11.396272 (lr=1.1189e-05) (hash(x)=52187435)
36
+ 32 train 11.359236 (lr=1.1538e-05) (hash(x)=51831972)
37
+ 33 train 11.314928 (lr=1.1888e-05) (hash(x)=56443459)
38
+ 34 train 11.336485 (lr=1.2238e-05) (hash(x)=54578790)
39
+ 35 train 11.264068 (lr=1.2587e-05) (hash(x)=45098704)
40
+ 36 train 11.267770 (lr=1.2937e-05) (hash(x)=45655812)
41
+ 37 train 11.246296 (lr=1.3287e-05) (hash(x)=49436419)
42
+ 38 train 11.206903 (lr=1.3636e-05) (hash(x)=56987876)
43
+ 39 train 11.139081 (lr=1.3986e-05) (hash(x)=48339908)
44
+ 40 train 11.079618 (lr=1.4336e-05) (hash(x)=46142231)
45
+ 41 train 11.122161 (lr=1.4685e-05) (hash(x)=46263050)
46
+ 42 train 11.064465 (lr=1.5035e-05) (hash(x)=48285373)
47
+ 43 train 11.032426 (lr=1.5385e-05) (hash(x)=56873965)
48
+ 44 train 10.986260 (lr=1.5734e-05) (hash(x)=50421608)
49
+ 45 train 10.949279 (lr=1.6084e-05) (hash(x)=52802882)
50
+ 46 train 10.940577 (lr=1.6434e-05) (hash(x)=49146692)
51
+ 47 train 10.883380 (lr=1.6783e-05) (hash(x)=45627699)
52
+ 48 train 10.831694 (lr=1.7133e-05) (hash(x)=44771118)
53
+ 49 train 10.807904 (lr=1.7483e-05) (hash(x)=55122837)
54
+ 50 val loss 10.7987
55
+ 50 val perplexity 48959.0000
56
+ 50 train 10.832579 (lr=1.7832e-05) (hash(x)=48891446)
57
+ 51 train 10.768471 (lr=1.8182e-05) (hash(x)=50609171)
58
+ 52 train 10.764984 (lr=1.8531e-05) (hash(x)=45080235)
59
+ 53 train 10.709711 (lr=1.8881e-05) (hash(x)=46674118)
60
+ 54 train 10.652319 (lr=1.9231e-05) (hash(x)=51868493)
61
+ 55 train 10.667731 (lr=1.9580e-05) (hash(x)=50306555)
62
+ 56 train 10.620813 (lr=1.9930e-05) (hash(x)=45930697)
63
+ 57 train 10.639350 (lr=2.0280e-05) (hash(x)=50806347)
64
+ 58 train 10.602842 (lr=2.0629e-05) (hash(x)=50115446)
65
+ 59 train 10.605813 (lr=2.0979e-05) (hash(x)=51243947)
66
+ 60 train 10.555378 (lr=2.1329e-05) (hash(x)=34965283)
67
+ 61 train 10.543650 (lr=2.1678e-05) (hash(x)=53806649)
68
+ 62 train 10.545769 (lr=2.2028e-05) (hash(x)=45810574)
69
+ 63 train 10.455750 (lr=2.2378e-05) (hash(x)=50536970)
70
+ 64 train 10.496509 (lr=2.2727e-05) (hash(x)=56940944)
71
+ 65 train 10.513852 (lr=2.3077e-05) (hash(x)=51669498)
72
+ 66 train 10.483940 (lr=2.3427e-05) (hash(x)=46315245)
73
+ 67 train 10.448788 (lr=2.3776e-05) (hash(x)=50610610)
74
+ 68 train 10.435915 (lr=2.4126e-05) (hash(x)=51478185)
75
+ 69 train 10.405135 (lr=2.4476e-05) (hash(x)=53701929)
76
+ 70 train 10.516146 (lr=2.4825e-05) (hash(x)=51021250)
77
+ 71 train 10.459522 (lr=2.5175e-05) (hash(x)=46930369)
78
+ 72 train 10.412514 (lr=2.5524e-05) (hash(x)=39405431)
79
+ 73 train 10.473856 (lr=2.5874e-05) (hash(x)=57476018)
80
+ 74 train 10.426867 (lr=2.6224e-05) (hash(x)=48315677)
81
+ 75 train 10.438030 (lr=2.6573e-05) (hash(x)=50199187)
82
+ 76 train 10.401720 (lr=2.6923e-05) (hash(x)=49586111)
83
+ 77 train 10.478909 (lr=2.7273e-05) (hash(x)=45949699)
84
+ 78 train 10.493517 (lr=2.7622e-05) (hash(x)=47334388)
85
+ 79 train 10.432835 (lr=2.7972e-05) (hash(x)=53294055)
86
+ 80 train 10.330130 (lr=2.8322e-05) (hash(x)=46499362)
87
+ 81 train 10.415620 (lr=2.8671e-05) (hash(x)=53097392)
88
+ 82 train 10.293704 (lr=2.9021e-05) (hash(x)=46479397)
89
+ 83 train 10.384415 (lr=2.9371e-05) (hash(x)=54479595)
90
+ 84 train 10.421980 (lr=2.9720e-05) (hash(x)=47738093)
91
+ 85 train 10.413005 (lr=3.0070e-05) (hash(x)=59006693)
92
+ 86 train 10.372068 (lr=3.0420e-05) (hash(x)=46312947)
93
+ 87 train 10.362177 (lr=3.0769e-05) (hash(x)=51396394)
94
+ 88 train 10.373586 (lr=3.1119e-05) (hash(x)=55830040)
95
+ 89 train 10.432238 (lr=3.1469e-05) (hash(x)=52098425)
96
+ 90 train 10.364522 (lr=3.1818e-05) (hash(x)=44163258)
97
+ 91 train 10.350318 (lr=3.2168e-05) (hash(x)=58459317)
98
+ 92 train 10.353659 (lr=3.2517e-05) (hash(x)=50402996)
99
+ 93 train 10.327165 (lr=3.2867e-05) (hash(x)=47781307)
100
+ 94 train 10.372005 (lr=3.3217e-05) (hash(x)=55269973)
101
+ 95 train 10.361375 (lr=3.3566e-05) (hash(x)=55088766)
102
+ 96 train 10.324787 (lr=3.3916e-05) (hash(x)=48450358)
103
+ 97 train 10.358734 (lr=3.4266e-05) (hash(x)=50198680)
104
+ 98 train 10.284951 (lr=3.4615e-05) (hash(x)=48650332)
105
+ 99 train 10.342840 (lr=3.4965e-05) (hash(x)=52234688)
106
+ 100 val loss 10.3199
107
+ 100 val perplexity 30331.4551
108
+ 100 train 10.331408 (lr=3.5315e-05) (hash(x)=51346801)
109
+ 101 train 10.395597 (lr=3.5664e-05) (hash(x)=52929403)
110
+ 102 train 10.287546 (lr=3.6014e-05) (hash(x)=56070928)
111
+ 103 train 10.301041 (lr=3.6364e-05) (hash(x)=50526384)
112
+ 104 train 10.290864 (lr=3.6713e-05) (hash(x)=60209461)
113
+ 105 train 10.294998 (lr=3.7063e-05) (hash(x)=52322092)
114
+ 106 train 10.261457 (lr=3.7413e-05) (hash(x)=52916383)
115
+ 107 train 10.211185 (lr=3.7762e-05) (hash(x)=52529702)
116
+ 108 train 10.288001 (lr=3.8112e-05) (hash(x)=51980724)
117
+ 109 train 10.273771 (lr=3.8462e-05) (hash(x)=46826784)
118
+ 110 train 10.257477 (lr=3.8811e-05) (hash(x)=51321857)
119
+ 111 train 10.236631 (lr=3.9161e-05) (hash(x)=53475454)
120
+ 112 train 10.249636 (lr=3.9510e-05) (hash(x)=54787418)
121
+ 113 train 10.308975 (lr=3.9860e-05) (hash(x)=46810389)
122
+ 114 train 10.187868 (lr=4.0210e-05) (hash(x)=53498382)
123
+ 115 train 10.235593 (lr=4.0559e-05) (hash(x)=46323632)
124
+ 116 train 10.238527 (lr=4.0909e-05) (hash(x)=52398798)
125
+ 117 train 10.220640 (lr=4.1259e-05) (hash(x)=53724068)
126
+ 118 train 10.237875 (lr=4.1608e-05) (hash(x)=52768582)
127
+ 119 train 10.225507 (lr=4.1958e-05) (hash(x)=46747022)
128
+ 120 train 10.173197 (lr=4.2308e-05) (hash(x)=50353293)
129
+ 121 train 10.201609 (lr=4.2657e-05) (hash(x)=50822751)
130
+ 122 train 10.194638 (lr=4.3007e-05) (hash(x)=47616170)
131
+ 123 train 10.187654 (lr=4.3357e-05) (hash(x)=48013779)
132
+ 124 train 10.192060 (lr=4.3706e-05) (hash(x)=45598720)
133
+ 125 train 10.168158 (lr=4.4056e-05) (hash(x)=52658193)
134
+ 126 train 10.232330 (lr=4.4406e-05) (hash(x)=48531900)
135
+ 127 train 10.169067 (lr=4.4755e-05) (hash(x)=46134720)
136
+ 128 train 10.224479 (lr=4.5105e-05) (hash(x)=51156418)
137
+ 129 train 10.085089 (lr=4.5455e-05) (hash(x)=52902481)
138
+ 130 train 10.078061 (lr=4.5804e-05) (hash(x)=51604647)
139
+ 131 train 10.110466 (lr=4.6154e-05) (hash(x)=49163526)
140
+ 132 train 10.108348 (lr=4.6503e-05) (hash(x)=53829392)
141
+ 133 train 10.088552 (lr=4.6853e-05) (hash(x)=49028078)
142
+ 134 train 10.103379 (lr=4.7203e-05) (hash(x)=52517672)
143
+ 135 train 10.092919 (lr=4.7552e-05) (hash(x)=59110263)
144
+ 136 train 10.053926 (lr=4.7902e-05) (hash(x)=51195255)
145
+ 137 train 10.103559 (lr=4.8252e-05) (hash(x)=50517780)
146
+ 138 train 10.129488 (lr=4.8601e-05) (hash(x)=62341175)
147
+ 139 train 10.068258 (lr=4.8951e-05) (hash(x)=45890874)
148
+ 140 train 10.024090 (lr=4.9301e-05) (hash(x)=49188067)
149
+ 141 train 10.069763 (lr=4.9650e-05) (hash(x)=47888783)
150
+ 142 train 10.104898 (lr=5.0000e-05) (hash(x)=52775090)
151
+ 143 train 10.076651 (lr=5.0350e-05) (hash(x)=56693767)
152
+ 144 train 10.036687 (lr=5.0699e-05) (hash(x)=45014400)
153
+ 145 train 10.004118 (lr=5.1049e-05) (hash(x)=53102856)
154
+ 146 train 10.013021 (lr=5.1399e-05) (hash(x)=52954310)
155
+ 147 train 10.006205 (lr=5.1748e-05) (hash(x)=47965042)
156
+ 148 train 10.005885 (lr=5.2098e-05) (hash(x)=56016526)
157
+ 149 train 9.987260 (lr=5.2448e-05) (hash(x)=45600303)
158
+ 150 val loss 9.9619
159
+ 150 val perplexity 21203.4434
160
+ 150 train 9.941618 (lr=5.2797e-05) (hash(x)=45764012)
161
+ 151 train 9.917135 (lr=5.3147e-05) (hash(x)=44542942)
162
+ 152 train 9.942680 (lr=5.3497e-05) (hash(x)=51672319)
163
+ 153 train 9.955958 (lr=5.3846e-05) (hash(x)=49315253)
164
+ 154 train 9.944625 (lr=5.4196e-05) (hash(x)=45571167)
165
+ 155 train 9.911818 (lr=5.4545e-05) (hash(x)=45320225)
166
+ 156 train 9.932569 (lr=5.4895e-05) (hash(x)=50184363)
167
+ 157 train 9.901852 (lr=5.5245e-05) (hash(x)=45775751)
168
+ 158 train 9.879288 (lr=5.5594e-05) (hash(x)=42462763)
169
+ 159 train 9.923567 (lr=5.5944e-05) (hash(x)=53985957)
170
+ 160 train 9.842002 (lr=5.6294e-05) (hash(x)=55567006)
171
+ 161 train 9.877817 (lr=5.6643e-05) (hash(x)=41963014)
172
+ 162 train 9.871315 (lr=5.6993e-05) (hash(x)=53862673)
173
+ 163 train 9.893901 (lr=5.7343e-05) (hash(x)=56042923)
174
+ 164 train 9.805661 (lr=5.7692e-05) (hash(x)=43078023)
175
+ 165 train 9.906632 (lr=5.8042e-05) (hash(x)=55023379)
176
+ 166 train 9.810945 (lr=5.8392e-05) (hash(x)=43011460)
177
+ 167 train 9.769646 (lr=5.8741e-05) (hash(x)=52536929)
178
+ 168 train 9.763880 (lr=5.9091e-05) (hash(x)=49761494)
179
+ 169 train 9.797531 (lr=5.9441e-05) (hash(x)=51631508)
180
+ 170 train 9.781748 (lr=5.9790e-05) (hash(x)=50400339)
181
+ 171 train 9.797520 (lr=6.0140e-05) (hash(x)=46560264)
182
+ 172 train 9.782917 (lr=6.0490e-05) (hash(x)=70534997)
183
+ 173 train 9.797155 (lr=6.0839e-05) (hash(x)=52420475)
184
+ 174 train 9.699787 (lr=6.1189e-05) (hash(x)=53194976)
185
+ 175 train 9.757293 (lr=6.1538e-05) (hash(x)=66781899)
186
+ 176 train 9.821731 (lr=6.1888e-05) (hash(x)=51986059)
187
+ 177 train 9.769580 (lr=6.2238e-05) (hash(x)=60275167)
188
+ 178 train 9.758420 (lr=6.2587e-05) (hash(x)=45483557)
189
+ 179 train 9.752421 (lr=6.2937e-05) (hash(x)=49182582)
190
+ 180 train 9.700377 (lr=6.3287e-05) (hash(x)=51446091)
191
+ 181 train 9.618073 (lr=6.3636e-05) (hash(x)=47827767)
192
+ 182 train 9.669464 (lr=6.3986e-05) (hash(x)=46988810)
193
+ 183 train 9.594959 (lr=6.4336e-05) (hash(x)=44918130)
194
+ 184 train 9.638362 (lr=6.4685e-05) (hash(x)=49693485)
195
+ 185 train 9.619593 (lr=6.5035e-05) (hash(x)=40969008)
196
+ 186 train 9.606284 (lr=6.5385e-05) (hash(x)=50344401)
197
+ 187 train 9.649580 (lr=6.5734e-05) (hash(x)=49833636)
198
+ 188 train 9.624366 (lr=6.6084e-05) (hash(x)=41326666)
199
+ 189 train 9.587075 (lr=6.6434e-05) (hash(x)=52346553)
200
+ 190 train 9.532111 (lr=6.6783e-05) (hash(x)=47970930)
201
+ 191 train 9.519365 (lr=6.7133e-05) (hash(x)=47009621)
202
+ 192 train 9.541047 (lr=6.7483e-05) (hash(x)=48602644)
203
+ 193 train 9.567393 (lr=6.7832e-05) (hash(x)=52115732)
204
+ 194 train 9.545091 (lr=6.8182e-05) (hash(x)=56527845)
205
+ 195 train 9.503649 (lr=6.8531e-05) (hash(x)=49459868)
206
+ 196 train 9.547287 (lr=6.8881e-05) (hash(x)=48955151)
207
+ 197 train 9.533555 (lr=6.9231e-05) (hash(x)=48473450)
208
+ 198 train 9.565315 (lr=6.9580e-05) (hash(x)=50306722)
209
+ 199 train 9.452536 (lr=6.9930e-05) (hash(x)=50898037)
210
+ 200 val loss 9.4810
211
+ 200 val perplexity 13107.9121
212
+ 200 train 9.584511 (lr=7.0280e-05) (hash(x)=35015807)
213
+ 201 train 9.462399 (lr=7.0629e-05) (hash(x)=46327297)
214
+ 202 train 9.484540 (lr=7.0979e-05) (hash(x)=44297138)
215
+ 203 train 9.535044 (lr=7.1329e-05) (hash(x)=47776613)
216
+ 204 train 9.453400 (lr=7.1678e-05) (hash(x)=51262732)
217
+ 205 train 9.510057 (lr=7.2028e-05) (hash(x)=51480608)
218
+ 206 train 9.427688 (lr=7.2378e-05) (hash(x)=50929448)
219
+ 207 train 9.427816 (lr=7.2727e-05) (hash(x)=46494961)
220
+ 208 train 9.445492 (lr=7.3077e-05) (hash(x)=59634178)
221
+ 209 train 9.444918 (lr=7.3427e-05) (hash(x)=47517395)
222
+ 210 train 9.494287 (lr=7.3776e-05) (hash(x)=48371589)
223
+ 211 train 9.474874 (lr=7.4126e-05) (hash(x)=61097106)
224
+ 212 train 9.374577 (lr=7.4476e-05) (hash(x)=61405427)
225
+ 213 train 9.368435 (lr=7.4825e-05) (hash(x)=50139108)
226
+ 214 train 9.337618 (lr=7.5175e-05) (hash(x)=45467872)
227
+ 215 train 9.302831 (lr=7.5524e-05) (hash(x)=49686665)
228
+ 216 train 9.375569 (lr=7.5874e-05) (hash(x)=51019411)
229
+ 217 train 9.200969 (lr=7.6224e-05) (hash(x)=49380702)
230
+ 218 train 9.305281 (lr=7.6573e-05) (hash(x)=50156382)
231
+ 219 train 9.386997 (lr=7.6923e-05) (hash(x)=48775872)
232
+ 220 train 9.270819 (lr=7.7273e-05) (hash(x)=49497141)
233
+ 221 train 9.227430 (lr=7.7622e-05) (hash(x)=49456326)
234
+ 222 train 9.310265 (lr=7.7972e-05) (hash(x)=53343959)
235
+ 223 train 9.243369 (lr=7.8322e-05) (hash(x)=44654425)
236
+ 224 train 9.236121 (lr=7.8671e-05) (hash(x)=51367775)
237
+ 225 train 9.231606 (lr=7.9021e-05) (hash(x)=54477325)
238
+ 226 train 9.247385 (lr=7.9371e-05) (hash(x)=45888568)
239
+ 227 train 9.205236 (lr=7.9720e-05) (hash(x)=49384161)
240
+ 228 train 9.188550 (lr=8.0070e-05) (hash(x)=52839038)
241
+ 229 train 9.173941 (lr=8.0420e-05) (hash(x)=49188683)
242
+ 230 train 9.191564 (lr=8.0769e-05) (hash(x)=62547001)
243
+ 231 train 9.187723 (lr=8.1119e-05) (hash(x)=51229664)
244
+ 232 train 9.126299 (lr=8.1469e-05) (hash(x)=48746317)
245
+ 233 train 9.141872 (lr=8.1818e-05) (hash(x)=51961195)
246
+ 234 train 9.253819 (lr=8.2168e-05) (hash(x)=49881442)
247
+ 235 train 9.203301 (lr=8.2517e-05) (hash(x)=48726061)
248
+ 236 train 9.116598 (lr=8.2867e-05) (hash(x)=50492385)
249
+ 237 train 9.036438 (lr=8.3217e-05) (hash(x)=53617083)
250
+ 238 train 9.055651 (lr=8.3566e-05) (hash(x)=45972461)
251
+ 239 train 9.281440 (lr=8.3916e-05) (hash(x)=60830807)
252
+ 240 train 9.413349 (lr=8.4266e-05) (hash(x)=63145389)
253
+ 241 train 9.409700 (lr=8.4615e-05) (hash(x)=64013010)
254
+ 242 train 9.394716 (lr=8.4965e-05) (hash(x)=64418102)
255
+ 243 train 9.171889 (lr=8.5315e-05) (hash(x)=54461658)
256
+ 244 train 9.043411 (lr=8.5664e-05) (hash(x)=49407427)
257
+ 245 train 9.026846 (lr=8.6014e-05) (hash(x)=48800368)
258
+ 246 train 9.049088 (lr=8.6364e-05) (hash(x)=54376909)
259
+ 247 train 9.035469 (lr=8.6713e-05) (hash(x)=49838333)
260
+ 248 train 9.326550 (lr=8.7063e-05) (hash(x)=111351979)
261
+ 249 train 11.010794 (lr=8.7413e-05) (hash(x)=119446440)
262
+ 250 val loss 8.9935
263
+ 250 val perplexity 8050.8823
264
+ 250 train 9.011640 (lr=8.7762e-05) (hash(x)=39507633)
265
+ 251 train 9.021307 (lr=8.8112e-05) (hash(x)=53078397)
266
+ 252 train 8.924015 (lr=8.8462e-05) (hash(x)=47937986)
267
+ 253 train 8.939757 (lr=8.8811e-05) (hash(x)=58893354)
268
+ 254 train 8.945075 (lr=8.9161e-05) (hash(x)=50312697)
269
+ 255 train 8.929859 (lr=8.9510e-05) (hash(x)=45343733)
270
+ 256 train 8.889739 (lr=8.9860e-05) (hash(x)=47270668)
271
+ 257 train 8.947664 (lr=9.0210e-05) (hash(x)=45564330)
272
+ 258 train 8.970765 (lr=9.0559e-05) (hash(x)=54465487)
273
+ 259 train 8.932979 (lr=9.0909e-05) (hash(x)=46808110)
274
+ 260 train 8.893166 (lr=9.1259e-05) (hash(x)=48250815)
275
+ 261 train 8.888174 (lr=9.1608e-05) (hash(x)=47328751)
276
+ 262 train 8.809520 (lr=9.1958e-05) (hash(x)=50483774)
277
+ 263 train 8.819937 (lr=9.2308e-05) (hash(x)=49778011)
278
+ 264 train 8.894714 (lr=9.2657e-05) (hash(x)=69410091)
279
+ 265 train 8.783035 (lr=9.3007e-05) (hash(x)=50448372)
280
+ 266 train 8.813501 (lr=9.3357e-05) (hash(x)=53769931)
281
+ 267 train 8.790530 (lr=9.3706e-05) (hash(x)=53670840)
282
+ 268 train 8.777060 (lr=9.4056e-05) (hash(x)=53844862)
283
+ 269 train 8.765681 (lr=9.4406e-05) (hash(x)=50707781)
284
+ 270 train 8.835920 (lr=9.4755e-05) (hash(x)=54542136)
285
+ 271 train 8.821052 (lr=9.5105e-05) (hash(x)=52796041)
286
+ 272 train 8.719831 (lr=9.5455e-05) (hash(x)=46224031)
287
+ 273 train 8.754648 (lr=9.5804e-05) (hash(x)=63541876)
288
+ 274 train 8.692440 (lr=9.6154e-05) (hash(x)=52401705)
289
+ 275 train 8.678364 (lr=9.6503e-05) (hash(x)=47778722)
290
+ 276 train 8.623478 (lr=9.6853e-05) (hash(x)=53126468)
291
+ 277 train 8.714774 (lr=9.7203e-05) (hash(x)=48019231)
292
+ 278 train 8.698287 (lr=9.7552e-05) (hash(x)=50049909)
293
+ 279 train 8.705967 (lr=9.7902e-05) (hash(x)=39174017)
294
+ 280 train 8.682344 (lr=9.8252e-05) (hash(x)=53181098)
295
+ 281 train 8.747675 (lr=9.8601e-05) (hash(x)=55894374)
296
+ 282 train 8.665668 (lr=9.8951e-05) (hash(x)=49643242)
297
+ 283 train 8.665935 (lr=9.9301e-05) (hash(x)=52422794)
298
+ 284 train 8.639981 (lr=9.9650e-05) (hash(x)=53466863)
299
+ 285 train 8.688701 (lr=1.0000e-04) (hash(x)=48367015)
300
+ 286 train 8.503775 (lr=1.0000e-04) (hash(x)=45469072)
301
+ 287 train 8.565005 (lr=1.0000e-04) (hash(x)=50495673)
302
+ 288 train 8.537472 (lr=9.9998e-05) (hash(x)=42520893)
303
+ 289 train 8.507190 (lr=9.9996e-05) (hash(x)=47779452)
304
+ 290 train 8.668473 (lr=9.9993e-05) (hash(x)=51356106)
305
+ 291 train 8.927492 (lr=9.9989e-05) (hash(x)=46394994)
306
+ 292 train 8.505730 (lr=9.9984e-05) (hash(x)=41504436)
307
+ 293 train 8.566000 (lr=9.9979e-05) (hash(x)=49997685)
308
+ 294 train 8.520734 (lr=9.9972e-05) (hash(x)=53238687)
309
+ 295 train 8.562025 (lr=9.9965e-05) (hash(x)=46877586)
310
+ 296 train 8.528469 (lr=9.9956e-05) (hash(x)=49989026)
311
+ 297 train 8.495488 (lr=9.9947e-05) (hash(x)=46923187)
312
+ 298 train 8.505074 (lr=9.9937e-05) (hash(x)=54278769)
313
+ 299 train 8.493473 (lr=9.9926e-05) (hash(x)=51669015)
314
+ 300 val loss 8.5033
315
+ 300 val perplexity 4930.8296
316
+ 300 train 8.629360 (lr=9.9915e-05) (hash(x)=51463767)
317
+ 301 train 8.695905 (lr=9.9902e-05) (hash(x)=59893724)
318
+ 302 train 8.715659 (lr=9.9889e-05) (hash(x)=59853505)
319
+ 303 train 8.686872 (lr=9.9874e-05) (hash(x)=49259399)
320
+ 304 train 8.512735 (lr=9.9859e-05) (hash(x)=51402337)
321
+ 305 train 8.501915 (lr=9.9843e-05) (hash(x)=51048761)
322
+ 306 train 8.455306 (lr=9.9826e-05) (hash(x)=49666603)
323
+ 307 train 8.411825 (lr=9.9808e-05) (hash(x)=43281885)
324
+ 308 train 8.421158 (lr=9.9789e-05) (hash(x)=45554894)
325
+ 309 train 8.376232 (lr=9.9770e-05) (hash(x)=49755839)
326
+ 310 train 8.394078 (lr=9.9749e-05) (hash(x)=43099820)
327
+ 311 train 8.382740 (lr=9.9728e-05) (hash(x)=43941060)
328
+ 312 train 8.481954 (lr=9.9706e-05) (hash(x)=43560282)
329
+ 313 train 8.322586 (lr=9.9683e-05) (hash(x)=46745578)
330
+ 314 train 8.441370 (lr=9.9659e-05) (hash(x)=53817256)
331
+ 315 train 8.441135 (lr=9.9634e-05) (hash(x)=49733529)
332
+ 316 train 8.366466 (lr=9.9609e-05) (hash(x)=46736835)
333
+ 317 train 8.341794 (lr=9.9582e-05) (hash(x)=50611003)
334
+ 318 train 8.418376 (lr=9.9555e-05) (hash(x)=48670015)
335
+ 319 train 8.350856 (lr=9.9526e-05) (hash(x)=47655500)
336
+ 320 train 8.297417 (lr=9.9497e-05) (hash(x)=48973562)
337
+ 321 train 8.284603 (lr=9.9467e-05) (hash(x)=42548986)
338
+ 322 train 8.399970 (lr=9.9437e-05) (hash(x)=49272586)
339
+ 323 train 8.315205 (lr=9.9405e-05) (hash(x)=50679082)
340
+ 324 train 8.172144 (lr=9.9372e-05) (hash(x)=41312345)
341
+ 325 train 8.300125 (lr=9.9339e-05) (hash(x)=50521445)
342
+ 326 train 8.244252 (lr=9.9305e-05) (hash(x)=48798201)
343
+ 327 train 8.281847 (lr=9.9270e-05) (hash(x)=48888163)
344
+ 328 train 8.314245 (lr=9.9234e-05) (hash(x)=45931709)
345
+ 329 train 8.267549 (lr=9.9197e-05) (hash(x)=44644363)
346
+ 330 train 8.324055 (lr=9.9159e-05) (hash(x)=55945599)
347
+ 331 train 8.282428 (lr=9.9121e-05) (hash(x)=48165003)
348
+ 332 train 8.244661 (lr=9.9081e-05) (hash(x)=49310124)
349
+ 333 train 8.424053 (lr=9.9041e-05) (hash(x)=44013693)
350
+ 334 train 8.322906 (lr=9.9000e-05) (hash(x)=49879848)
351
+ 335 train 8.203651 (lr=9.8958e-05) (hash(x)=45152283)
352
+ 336 train 8.259142 (lr=9.8915e-05) (hash(x)=60540503)
353
+ 337 train 8.211108 (lr=9.8872e-05) (hash(x)=48497343)
354
+ 338 train 8.274627 (lr=9.8827e-05) (hash(x)=54318341)
355
+ 339 train 8.249550 (lr=9.8782e-05) (hash(x)=52297893)
356
+ 340 train 8.250234 (lr=9.8736e-05) (hash(x)=51000121)
357
+ 341 train 8.201115 (lr=9.8689e-05) (hash(x)=45974569)
358
+ 342 train 8.162029 (lr=9.8641e-05) (hash(x)=52735968)
359
+ 343 train 8.246988 (lr=9.8592e-05) (hash(x)=51792713)
360
+ 344 train 8.125650 (lr=9.8543e-05) (hash(x)=45786866)
361
+ 345 train 8.166664 (lr=9.8492e-05) (hash(x)=44500212)
362
+ 346 train 8.288227 (lr=9.8441e-05) (hash(x)=66566677)
363
+ 347 train 8.169264 (lr=9.8389e-05) (hash(x)=48433902)
364
+ 348 train 8.079492 (lr=9.8336e-05) (hash(x)=53700753)
365
+ 349 train 8.037247 (lr=9.8282e-05) (hash(x)=42614965)
366
+ 350 val loss 8.1895
367
+ 350 val perplexity 3603.0779
368
+ 350 train 8.077385 (lr=9.8228e-05) (hash(x)=51631419)
369
+ 351 train 8.144310 (lr=9.8172e-05) (hash(x)=50899313)
370
+ 352 train 8.203080 (lr=9.8116e-05) (hash(x)=50553266)
371
+ 353 train 8.104047 (lr=9.8059e-05) (hash(x)=47548858)
372
+ 354 train 8.121872 (lr=9.8001e-05) (hash(x)=44976927)
373
+ 355 train 8.132928 (lr=9.7942e-05) (hash(x)=40088877)
374
+ 356 train 8.110067 (lr=9.7882e-05) (hash(x)=48104733)
375
+ 357 train 8.206319 (lr=9.7822e-05) (hash(x)=53658356)
376
+ 358 train 8.098846 (lr=9.7761e-05) (hash(x)=49964977)
377
+ 359 train 8.063511 (lr=9.7699e-05) (hash(x)=53925563)
378
+ 360 train 8.168398 (lr=9.7636e-05) (hash(x)=52403880)
379
+ 361 train 8.088741 (lr=9.7572e-05) (hash(x)=48385306)
380
+ 362 train 7.984328 (lr=9.7507e-05) (hash(x)=43581460)
381
+ 363 train 8.122417 (lr=9.7442e-05) (hash(x)=52568018)
382
+ 364 train 8.134740 (lr=9.7376e-05) (hash(x)=50012317)
383
+ 365 train 8.016771 (lr=9.7309e-05) (hash(x)=51336015)
384
+ 366 train 8.075779 (lr=9.7241e-05) (hash(x)=51308604)
385
+ 367 train 8.089677 (lr=9.7172e-05) (hash(x)=52205768)
386
+ 368 train 7.993440 (lr=9.7103e-05) (hash(x)=41696587)
387
+ 369 train 8.061837 (lr=9.7032e-05) (hash(x)=45595459)
388
+ 370 train 8.042090 (lr=9.6961e-05) (hash(x)=47769902)
389
+ 371 train 8.059135 (lr=9.6889e-05) (hash(x)=43562643)
390
+ 372 train 7.990054 (lr=9.6817e-05) (hash(x)=44913267)
391
+ 373 train 7.967371 (lr=9.6743e-05) (hash(x)=49247249)
392
+ 374 train 7.987482 (lr=9.6669e-05) (hash(x)=48475648)
393
+ 375 train 7.962966 (lr=9.6593e-05) (hash(x)=51933643)
394
+ 376 train 7.995642 (lr=9.6518e-05) (hash(x)=46907449)
395
+ 377 train 8.034756 (lr=9.6441e-05) (hash(x)=49807502)
396
+ 378 train 8.070681 (lr=9.6363e-05) (hash(x)=55135176)
397
+ 379 train 7.957666 (lr=9.6285e-05) (hash(x)=51568772)
398
+ 380 train 7.924624 (lr=9.6206e-05) (hash(x)=47804614)
399
+ 381 train 7.949591 (lr=9.6126e-05) (hash(x)=40443372)
400
+ 382 train 7.932366 (lr=9.6045e-05) (hash(x)=45544157)
401
+ 383 train 8.074946 (lr=9.5963e-05) (hash(x)=52906728)
402
+ 384 train 8.119107 (lr=9.5881e-05) (hash(x)=49019570)
403
+ 385 train 7.951980 (lr=9.5798e-05) (hash(x)=50496495)
404
+ 386 train 8.142988 (lr=9.5714e-05) (hash(x)=53872114)
405
+ 387 train 8.200620 (lr=9.5629e-05) (hash(x)=52247792)
406
+ 388 train 7.952759 (lr=9.5544e-05) (hash(x)=40758659)
407
+ 389 train 8.339425 (lr=9.5457e-05) (hash(x)=55069044)
408
+ 390 train 8.026800 (lr=9.5370e-05) (hash(x)=51128293)
409
+ 391 train 8.026655 (lr=9.5282e-05) (hash(x)=50654953)
410
+ 392 train 8.029709 (lr=9.5194e-05) (hash(x)=61276676)
411
+ 393 train 8.095097 (lr=9.5104e-05) (hash(x)=47259878)
412
+ 394 train 7.945934 (lr=9.5014e-05) (hash(x)=50687278)
413
+ 395 train 7.977750 (lr=9.4923e-05) (hash(x)=52045445)
414
+ 396 train 7.968279 (lr=9.4831e-05) (hash(x)=49960644)
415
+ 397 train 8.030515 (lr=9.4739e-05) (hash(x)=49503934)
416
+ 398 train 7.972917 (lr=9.4646e-05) (hash(x)=46047135)
417
+ 399 train 7.918422 (lr=9.4551e-05) (hash(x)=48029368)
418
+ 400 val loss 7.9800
419
+ 400 val perplexity 2921.8518
420
+ 400 train 7.950455 (lr=9.4457e-05) (hash(x)=45374968)
421
+ 401 train 7.909479 (lr=9.4361e-05) (hash(x)=52734397)
422
+ 402 train 8.043240 (lr=9.4265e-05) (hash(x)=52389164)
423
+ 403 train 8.029964 (lr=9.4168e-05) (hash(x)=51521570)
424
+ 404 train 8.074707 (lr=9.4070e-05) (hash(x)=50463113)
425
+ 405 train 7.893105 (lr=9.3971e-05) (hash(x)=54958341)
426
+ 406 train 7.837413 (lr=9.3872e-05) (hash(x)=46153985)
427
+ 407 train 7.853812 (lr=9.3772e-05) (hash(x)=49432790)
428
+ 408 train 7.908645 (lr=9.3671e-05) (hash(x)=48385222)
429
+ 409 train 8.013595 (lr=9.3569e-05) (hash(x)=51630204)
430
+ 410 train 7.803065 (lr=9.3467e-05) (hash(x)=55768152)
431
+ 411 train 7.807565 (lr=9.3364e-05) (hash(x)=50243462)
432
+ 412 train 7.826632 (lr=9.3260e-05) (hash(x)=45951312)
433
+ 413 train 8.426485 (lr=9.3155e-05) (hash(x)=55551835)
434
+ 414 train 7.880831 (lr=9.3050e-05) (hash(x)=47924861)
435
+ 415 train 7.833393 (lr=9.2944e-05) (hash(x)=44943064)
436
+ 416 train 7.878897 (lr=9.2837e-05) (hash(x)=45952924)
437
+ 417 train 8.066874 (lr=9.2729e-05) (hash(x)=52015898)
438
+ 418 train 7.874660 (lr=9.2621e-05) (hash(x)=52449480)
439
+ 419 train 7.825839 (lr=9.2512e-05) (hash(x)=47971635)
440
+ 420 train 8.010769 (lr=9.2402e-05) (hash(x)=59172908)
441
+ 421 train 7.884302 (lr=9.2292e-05) (hash(x)=50097527)
442
+ 422 train 7.852376 (lr=9.2181e-05) (hash(x)=50748027)
443
+ 423 train 7.963111 (lr=9.2069e-05) (hash(x)=46113670)
444
+ 424 train 7.970116 (lr=9.1956e-05) (hash(x)=52184218)
445
+ 425 train 7.873911 (lr=9.1843e-05) (hash(x)=50005550)
446
+ 426 train 7.842113 (lr=9.1729e-05) (hash(x)=47110180)
447
+ 427 train 7.954002 (lr=9.1614e-05) (hash(x)=50198511)
448
+ 428 train 7.752063 (lr=9.1499e-05) (hash(x)=41108331)
449
+ 429 train 7.819736 (lr=9.1382e-05) (hash(x)=51366503)
450
+ 430 train 8.029217 (lr=9.1266e-05) (hash(x)=59593423)
451
+ 431 train 7.899106 (lr=9.1148e-05) (hash(x)=54958512)
452
+ 432 train 7.766047 (lr=9.1030e-05) (hash(x)=50095280)
453
+ 433 train 7.905949 (lr=9.0911e-05) (hash(x)=46982568)
454
+ 434 train 7.765928 (lr=9.0791e-05) (hash(x)=45920340)
455
+ 435 train 7.745382 (lr=9.0671e-05) (hash(x)=48153200)
456
+ 436 train 7.825115 (lr=9.0550e-05) (hash(x)=51503557)
457
+ 437 train 7.931048 (lr=9.0428e-05) (hash(x)=45447171)
458
+ 438 train 7.718319 (lr=9.0306e-05) (hash(x)=43283692)
459
+ 439 train 7.848161 (lr=9.0182e-05) (hash(x)=52017911)
460
+ 440 train 7.728775 (lr=9.0059e-05) (hash(x)=49830063)
461
+ 441 train 7.673191 (lr=8.9934e-05) (hash(x)=56227516)
462
+ 442 train 7.734136 (lr=8.9809e-05) (hash(x)=48017404)
463
+ 443 train 7.773089 (lr=8.9683e-05) (hash(x)=51984978)
464
+ 444 train 7.622454 (lr=8.9557e-05) (hash(x)=47784187)
465
+ 445 train 7.505863 (lr=8.9430e-05) (hash(x)=47512728)
466
+ 446 train 7.764479 (lr=8.9302e-05) (hash(x)=47778087)
467
+ 447 train 7.838997 (lr=8.9173e-05) (hash(x)=53173755)
468
+ 448 train 7.736099 (lr=8.9044e-05) (hash(x)=48702762)
469
+ 449 train 7.830531 (lr=8.8914e-05) (hash(x)=50108146)
470
+ 450 val loss 7.8300
471
+ 450 val perplexity 2514.8296
472
+ 450 train 7.614513 (lr=8.8784e-05) (hash(x)=46030207)
473
+ 451 train 7.742623 (lr=8.8653e-05) (hash(x)=52115953)
474
+ 452 train 7.781646 (lr=8.8521e-05) (hash(x)=47643321)
475
+ 453 train 7.674423 (lr=8.8388e-05) (hash(x)=46586365)
476
+ 454 train 7.777119 (lr=8.8255e-05) (hash(x)=49204978)
477
+ 455 train 7.789364 (lr=8.8122e-05) (hash(x)=51872266)
478
+ 456 train 7.841067 (lr=8.7987e-05) (hash(x)=51204581)
479
+ 457 train 7.768681 (lr=8.7852e-05) (hash(x)=52821094)
480
+ 458 train 7.770690 (lr=8.7717e-05) (hash(x)=53242967)
481
+ 459 train 7.736785 (lr=8.7580e-05) (hash(x)=49251652)
482
+ 460 train 7.764112 (lr=8.7444e-05) (hash(x)=50825785)
483
+ 461 train 7.707139 (lr=8.7306e-05) (hash(x)=54773264)
484
+ 462 train 7.691592 (lr=8.7168e-05) (hash(x)=39155629)
485
+ 463 train 7.828435 (lr=8.7029e-05) (hash(x)=48621984)
486
+ 464 train 7.784278 (lr=8.6890e-05) (hash(x)=49567244)
487
+ 465 train 7.748254 (lr=8.6750e-05) (hash(x)=49633335)
488
+ 466 train 7.663326 (lr=8.6609e-05) (hash(x)=45377574)
489
+ 467 train 8.050900 (lr=8.6468e-05) (hash(x)=56838834)
490
+ 468 train 7.794479 (lr=8.6326e-05) (hash(x)=48343128)
491
+ 469 train 7.678555 (lr=8.6184e-05) (hash(x)=50527481)
492
+ 470 train 7.778000 (lr=8.6041e-05) (hash(x)=44035071)
493
+ 471 train 7.725295 (lr=8.5897e-05) (hash(x)=45518917)
494
+ 472 train 7.761645 (lr=8.5753e-05) (hash(x)=49027591)
495
+ 473 train 7.745924 (lr=8.5608e-05) (hash(x)=48086649)
496
+ 474 train 7.643441 (lr=8.5462e-05) (hash(x)=44730458)
497
+ 475 train 7.724655 (lr=8.5316e-05) (hash(x)=48085406)
498
+ 476 train 7.663602 (lr=8.5170e-05) (hash(x)=49401521)
499
+ 477 train 7.694255 (lr=8.5022e-05) (hash(x)=52347328)
500
+ 478 train 7.520889 (lr=8.4875e-05) (hash(x)=50307408)
501
+ 479 train 7.557461 (lr=8.4726e-05) (hash(x)=43959848)
502
+ 480 train 7.875252 (lr=8.4577e-05) (hash(x)=45469715)
503
+ 481 train 7.652767 (lr=8.4428e-05) (hash(x)=53522067)
504
+ 482 train 7.868656 (lr=8.4278e-05) (hash(x)=58676829)
505
+ 483 train 7.545346 (lr=8.4127e-05) (hash(x)=49691960)
506
+ 484 train 7.583119 (lr=8.3976e-05) (hash(x)=47344591)
507
+ 485 train 7.683199 (lr=8.3824e-05) (hash(x)=50456125)
508
+ 486 train 7.616972 (lr=8.3672e-05) (hash(x)=51037651)
509
+ 487 train 7.478867 (lr=8.3519e-05) (hash(x)=43100867)
510
+ 488 train 7.591134 (lr=8.3366e-05) (hash(x)=43778357)
511
+ 489 train 7.607473 (lr=8.3212e-05) (hash(x)=54476483)
512
+ 490 train 7.541442 (lr=8.3057e-05) (hash(x)=47098615)
513
+ 491 train 7.733136 (lr=8.2902e-05) (hash(x)=49945665)
514
+ 492 train 7.578199 (lr=8.2746e-05) (hash(x)=45656407)
515
+ 493 train 7.795040 (lr=8.2590e-05) (hash(x)=52658605)
516
+ 494 train 7.408205 (lr=8.2434e-05) (hash(x)=47928268)
517
+ 495 train 7.819966 (lr=8.2276e-05) (hash(x)=51979631)
518
+ 496 train 7.748904 (lr=8.2119e-05) (hash(x)=51164625)
519
+ 497 train 7.842674 (lr=8.1960e-05) (hash(x)=57967556)
520
+ 498 train 7.763737 (lr=8.1801e-05) (hash(x)=57145664)
521
+ 499 train 7.496165 (lr=8.1642e-05) (hash(x)=37203059)
522
+ 500 val loss 7.7323
523
+ 500 val perplexity 2280.9297
524
+ 500 train 7.604324 (lr=8.1482e-05) (hash(x)=47984882)
525
+ 501 train 7.808841 (lr=8.1322e-05) (hash(x)=50281972)
526
+ 502 train 7.573428 (lr=8.1161e-05) (hash(x)=42646472)
527
+ 503 train 7.822078 (lr=8.1000e-05) (hash(x)=46152379)
528
+ 504 train 7.705711 (lr=8.0838e-05) (hash(x)=49194603)
529
+ 505 train 7.830586 (lr=8.0676e-05) (hash(x)=51117679)
530
+ 506 train 7.590814 (lr=8.0513e-05) (hash(x)=49614996)
531
+ 507 train 7.740053 (lr=8.0349e-05) (hash(x)=62942518)
532
+ 508 train 7.738645 (lr=8.0186e-05) (hash(x)=48644746)
533
+ 509 train 7.567062 (lr=8.0021e-05) (hash(x)=45751524)
534
+ 510 train 7.705834 (lr=7.9856e-05) (hash(x)=54468682)
535
+ 511 train 7.951861 (lr=7.9691e-05) (hash(x)=50427086)
536
+ 512 train 7.732882 (lr=7.9525e-05) (hash(x)=47363324)
537
+ 513 train 7.693744 (lr=7.9359e-05) (hash(x)=52371471)
538
+ 514 train 7.568422 (lr=7.9192e-05) (hash(x)=42392151)
539
+ 515 train 7.493072 (lr=7.9025e-05) (hash(x)=44756066)
540
+ 516 train 7.611436 (lr=7.8858e-05) (hash(x)=48272019)
541
+ 517 train 7.673866 (lr=7.8689e-05) (hash(x)=48643995)
542
+ 518 train 7.772599 (lr=7.8521e-05) (hash(x)=56761628)
543
+ 519 train 7.753964 (lr=7.8352e-05) (hash(x)=55097568)
544
+ 520 train 7.629611 (lr=7.8182e-05) (hash(x)=52120708)
545
+ 521 train 7.820821 (lr=7.8012e-05) (hash(x)=54223262)
546
+ 522 train 7.633710 (lr=7.7842e-05) (hash(x)=47535338)
547
+ 523 train 7.620634 (lr=7.7671e-05) (hash(x)=52009657)
548
+ 524 train 7.592891 (lr=7.7500e-05) (hash(x)=48521925)
549
+ 525 train 7.578537 (lr=7.7328e-05) (hash(x)=50133238)
550
+ 526 train 7.569391 (lr=7.7156e-05) (hash(x)=49108108)
551
+ 527 train 7.524819 (lr=7.6984e-05) (hash(x)=45702145)
552
+ 528 train 7.710737 (lr=7.6811e-05) (hash(x)=52597986)
553
+ 529 train 7.679502 (lr=7.6637e-05) (hash(x)=54139152)
554
+ 530 train 7.523453 (lr=7.6463e-05) (hash(x)=46444995)
555
+ 531 train 7.690673 (lr=7.6289e-05) (hash(x)=50285792)
556
+ 532 train 7.569821 (lr=7.6115e-05) (hash(x)=49571253)
557
+ 533 train 7.487752 (lr=7.5940e-05) (hash(x)=36544982)
558
+ 534 train 7.592455 (lr=7.5764e-05) (hash(x)=52613692)
559
+ 535 train 7.581751 (lr=7.5588e-05) (hash(x)=48238859)
560
+ 536 train 7.571176 (lr=7.5412e-05) (hash(x)=49886313)
561
+ 537 train 7.631003 (lr=7.5235e-05) (hash(x)=51424104)
562
+ 538 train 7.476130 (lr=7.5058e-05) (hash(x)=47185490)
563
+ 539 train 7.571340 (lr=7.4881e-05) (hash(x)=49586112)
564
+ 540 train 7.541347 (lr=7.4703e-05) (hash(x)=48666312)
565
+ 541 train 7.577102 (lr=7.4525e-05) (hash(x)=45907380)
566
+ 542 train 7.695953 (lr=7.4346e-05) (hash(x)=49064772)
567
+ 543 train 7.555500 (lr=7.4167e-05) (hash(x)=46992549)
568
+ 544 train 7.629718 (lr=7.3988e-05) (hash(x)=54689133)
569
+ 545 train 7.539059 (lr=7.3808e-05) (hash(x)=48901438)
570
+ 546 train 7.562509 (lr=7.3628e-05) (hash(x)=49321281)
571
+ 547 train 7.492568 (lr=7.3448e-05) (hash(x)=40639906)
572
+ 548 train 7.470331 (lr=7.3267e-05) (hash(x)=48380140)
573
+ 549 train 7.539368 (lr=7.3086e-05) (hash(x)=48620828)
574
+ 550 val loss 7.6644
575
+ 550 val perplexity 2131.2075
576
+ 550 train 7.612592 (lr=7.2904e-05) (hash(x)=48019958)
577
+ 551 train 7.611282 (lr=7.2723e-05) (hash(x)=50408031)
578
+ 552 train 7.431355 (lr=7.2540e-05) (hash(x)=49105065)
579
+ 553 train 7.753616 (lr=7.2358e-05) (hash(x)=53413053)
580
+ 554 train 7.533695 (lr=7.2175e-05) (hash(x)=52526702)
581
+ 555 train 7.515712 (lr=7.1992e-05) (hash(x)=47136905)
582
+ 556 train 7.528444 (lr=7.1808e-05) (hash(x)=46757682)
583
+ 557 train 7.618686 (lr=7.1624e-05) (hash(x)=42269456)
584
+ 558 train 7.718602 (lr=7.1440e-05) (hash(x)=52153912)
585
+ 559 train 7.828231 (lr=7.1256e-05) (hash(x)=57982592)
586
+ 560 train 7.621963 (lr=7.1071e-05) (hash(x)=49165336)
587
+ 561 train 7.479816 (lr=7.0886e-05) (hash(x)=34584713)
588
+ 562 train 7.764446 (lr=7.0701e-05) (hash(x)=51652663)
589
+ 563 train 7.560950 (lr=7.0515e-05) (hash(x)=50607950)
590
+ 564 train 7.542317 (lr=7.0329e-05) (hash(x)=48676884)
591
+ 565 train 7.650950 (lr=7.0143e-05) (hash(x)=47888657)
592
+ 566 train 7.621752 (lr=6.9956e-05) (hash(x)=45582208)
593
+ 567 train 7.562435 (lr=6.9769e-05) (hash(x)=44984575)
594
+ 568 train 7.612875 (lr=6.9582e-05) (hash(x)=43110451)
595
+ 569 train 7.531368 (lr=6.9394e-05) (hash(x)=48618029)
596
+ 570 train 7.635777 (lr=6.9207e-05) (hash(x)=48094498)
597
+ 571 train 7.698644 (lr=6.9019e-05) (hash(x)=50663582)
598
+ 572 train 7.613410 (lr=6.8830e-05) (hash(x)=48668321)
599
+ 573 train 7.452733 (lr=6.8642e-05) (hash(x)=50805008)
600
+ 574 train 7.510807 (lr=6.8453e-05) (hash(x)=48716136)
601
+ 575 train 7.514547 (lr=6.8264e-05) (hash(x)=44713794)
602
+ 576 train 7.643648 (lr=6.8075e-05) (hash(x)=52712617)
603
+ 577 train 7.617599 (lr=6.7885e-05) (hash(x)=46510493)
604
+ 578 train 7.710190 (lr=6.7695e-05) (hash(x)=48449239)
605
+ 579 train 7.538903 (lr=6.7505e-05) (hash(x)=48468623)
606
+ 580 train 7.507864 (lr=6.7315e-05) (hash(x)=50563691)
607
+ 581 train 7.784690 (lr=6.7124e-05) (hash(x)=49425907)
608
+ 582 train 7.496567 (lr=6.6933e-05) (hash(x)=47050151)
609
+ 583 train 7.579120 (lr=6.6742e-05) (hash(x)=52642134)
610
+ 584 train 7.694609 (lr=6.6551e-05) (hash(x)=48759265)
611
+ 585 train 7.505916 (lr=6.6360e-05) (hash(x)=53637668)
612
+ 586 train 7.516370 (lr=6.6168e-05) (hash(x)=45678217)
613
+ 587 train 7.690019 (lr=6.5976e-05) (hash(x)=58896478)
614
+ 588 train 7.431674 (lr=6.5784e-05) (hash(x)=43190275)
615
+ 589 train 7.434154 (lr=6.5592e-05) (hash(x)=46112834)
616
+ 590 train 7.390995 (lr=6.5399e-05) (hash(x)=42917232)
617
+ 591 train 7.482100 (lr=6.5206e-05) (hash(x)=47330369)
618
+ 592 train 7.405015 (lr=6.5013e-05) (hash(x)=45177539)
619
+ 593 train 7.545740 (lr=6.4820e-05) (hash(x)=45777708)
620
+ 594 train 7.460274 (lr=6.4627e-05) (hash(x)=53221813)
621
+ 595 train 7.507862 (lr=6.4433e-05) (hash(x)=49047563)
622
+ 596 train 7.497953 (lr=6.4240e-05) (hash(x)=48900735)
623
+ 597 train 7.995917 (lr=6.4046e-05) (hash(x)=52969216)
624
+ 598 train 7.558355 (lr=6.3852e-05) (hash(x)=51116960)
625
+ 599 train 7.678505 (lr=6.3658e-05) (hash(x)=49473785)
626
+ 600 val loss 7.6203
627
+ 600 val perplexity 2039.2434
628
+ 600 train 7.762987 (lr=6.3463e-05) (hash(x)=50739860)
629
+ 601 train 7.791866 (lr=6.3269e-05) (hash(x)=48978103)
630
+ 602 train 7.519073 (lr=6.3074e-05) (hash(x)=50040676)
631
+ 603 train 7.544978 (lr=6.2879e-05) (hash(x)=48245323)
632
+ 604 train 7.585876 (lr=6.2684e-05) (hash(x)=48512694)
633
+ 605 train 7.679477 (lr=6.2489e-05) (hash(x)=51656618)
634
+ 606 train 7.771473 (lr=6.2294e-05) (hash(x)=50124911)
635
+ 607 train 7.588269 (lr=6.2098e-05) (hash(x)=53162165)
636
+ 608 train 7.645240 (lr=6.1903e-05) (hash(x)=55106228)
637
+ 609 train 7.667312 (lr=6.1707e-05) (hash(x)=49745719)
638
+ 610 train 7.589005 (lr=6.1511e-05) (hash(x)=52999037)
639
+ 611 train 7.616910 (lr=6.1315e-05) (hash(x)=43136509)
640
+ 612 train 7.668417 (lr=6.1119e-05) (hash(x)=48103007)
641
+ 613 train 7.569470 (lr=6.0923e-05) (hash(x)=51551441)
642
+ 614 train 7.656134 (lr=6.0726e-05) (hash(x)=51070940)
643
+ 615 train 7.701504 (lr=6.0530e-05) (hash(x)=45691679)
644
+ 616 train 7.666229 (lr=6.0333e-05) (hash(x)=55416593)
645
+ 617 train 7.574953 (lr=6.0137e-05) (hash(x)=47804807)
646
+ 618 train 7.562170 (lr=5.9940e-05) (hash(x)=54133598)
647
+ 619 train 7.548120 (lr=5.9743e-05) (hash(x)=50945937)
648
+ 620 train 7.586721 (lr=5.9546e-05) (hash(x)=51426362)
649
+ 621 train 7.584307 (lr=5.9349e-05) (hash(x)=50499079)
650
+ 622 train 7.879313 (lr=5.9152e-05) (hash(x)=55542741)
651
+ 623 train 7.601506 (lr=5.8955e-05) (hash(x)=46788368)
652
+ 624 train 7.631466 (lr=5.8758e-05) (hash(x)=50140177)
653
+ 625 train 7.638706 (lr=5.8560e-05) (hash(x)=51187898)
654
+ 626 train 7.726056 (lr=5.8363e-05) (hash(x)=43773995)
655
+ 627 train 7.674638 (lr=5.8165e-05) (hash(x)=49942319)
656
+ 628 train 7.439291 (lr=5.7968e-05) (hash(x)=41086061)
657
+ 629 train 7.522415 (lr=5.7770e-05) (hash(x)=49879352)
658
+ 630 train 7.728012 (lr=5.7573e-05) (hash(x)=49365794)
659
+ 631 train 7.487639 (lr=5.7375e-05) (hash(x)=48319660)
660
+ 632 train 7.492344 (lr=5.7177e-05) (hash(x)=49122881)
661
+ 633 train 7.642729 (lr=5.6979e-05) (hash(x)=49909304)
662
+ 634 train 7.545746 (lr=5.6782e-05) (hash(x)=48917440)
663
+ 635 train 7.459764 (lr=5.6584e-05) (hash(x)=54019974)
664
+ 636 train 7.526427 (lr=5.6386e-05) (hash(x)=44808188)
665
+ 637 train 7.630544 (lr=5.6188e-05) (hash(x)=53441724)
666
+ 638 train 7.756757 (lr=5.5990e-05) (hash(x)=45859343)
667
+ 639 train 7.503382 (lr=5.5792e-05) (hash(x)=51204717)
668
+ 640 train 7.592845 (lr=5.5594e-05) (hash(x)=51150390)
669
+ 641 train 7.555241 (lr=5.5396e-05) (hash(x)=52433793)
670
+ 642 train 7.725581 (lr=5.5198e-05) (hash(x)=49329755)
671
+ 643 train 7.797573 (lr=5.5000e-05) (hash(x)=52287233)
672
+ 644 train 7.523757 (lr=5.4802e-05) (hash(x)=49556750)
673
+ 645 train 7.541972 (lr=5.4604e-05) (hash(x)=49458030)
674
+ 646 train 7.616358 (lr=5.4406e-05) (hash(x)=50344153)
675
+ 647 train 7.805715 (lr=5.4208e-05) (hash(x)=54016916)
676
+ 648 train 7.676512 (lr=5.4010e-05) (hash(x)=54062547)
677
+ 649 train 7.606304 (lr=5.3812e-05) (hash(x)=55653116)
678
+ 650 val loss 7.5728
679
+ 650 val perplexity 1944.5360
680
+ 650 train 7.629893 (lr=5.3614e-05) (hash(x)=51250095)
681
+ 651 train 7.621423 (lr=5.3416e-05) (hash(x)=47177229)
682
+ 652 train 7.650302 (lr=5.3218e-05) (hash(x)=51051236)
683
+ 653 train 7.615001 (lr=5.3021e-05) (hash(x)=55260549)
684
+ 654 train 7.651271 (lr=5.2823e-05) (hash(x)=54125192)
685
+ 655 train 7.587038 (lr=5.2625e-05) (hash(x)=43269493)
686
+ 656 train 7.531717 (lr=5.2427e-05) (hash(x)=48540342)
687
+ 657 train 7.692810 (lr=5.2230e-05) (hash(x)=42734512)
688
+ 658 train 7.583550 (lr=5.2032e-05) (hash(x)=54618082)
689
+ 659 train 7.557434 (lr=5.1835e-05) (hash(x)=49394138)
690
+ 660 train 7.603020 (lr=5.1637e-05) (hash(x)=45990832)
691
+ 661 train 7.649940 (lr=5.1440e-05) (hash(x)=52986058)
692
+ 662 train 7.668054 (lr=5.1242e-05) (hash(x)=55242330)
693
+ 663 train 7.411430 (lr=5.1045e-05) (hash(x)=43336676)
694
+ 664 train 7.484661 (lr=5.0848e-05) (hash(x)=37245563)
695
+ 665 train 7.561215 (lr=5.0651e-05) (hash(x)=50645055)
696
+ 666 train 7.642961 (lr=5.0454e-05) (hash(x)=48042559)
697
+ 667 train 7.603570 (lr=5.0257e-05) (hash(x)=51786035)
698
+ 668 train 7.604270 (lr=5.0060e-05) (hash(x)=53181257)
699
+ 669 train 7.584915 (lr=4.9863e-05) (hash(x)=60603860)
700
+ 670 train 7.691325 (lr=4.9667e-05) (hash(x)=49883821)
701
+ 671 train 7.664286 (lr=4.9470e-05) (hash(x)=57298482)
702
+ 672 train 7.605057 (lr=4.9274e-05) (hash(x)=47405200)
703
+ 673 train 7.663365 (lr=4.9077e-05) (hash(x)=55116220)
704
+ 674 train 7.524390 (lr=4.8881e-05) (hash(x)=44918607)
705
+ 675 train 7.509783 (lr=4.8685e-05) (hash(x)=41523799)
706
+ 676 train 7.606831 (lr=4.8489e-05) (hash(x)=49725227)
707
+ 677 train 7.545782 (lr=4.8293e-05) (hash(x)=51589071)
708
+ 678 train 7.606838 (lr=4.8097e-05) (hash(x)=50536573)
709
+ 679 train 7.645326 (lr=4.7902e-05) (hash(x)=47366200)
710
+ 680 train 7.529650 (lr=4.7706e-05) (hash(x)=50406768)
711
+ 681 train 7.634709 (lr=4.7511e-05) (hash(x)=53022008)
712
+ 682 train 7.501176 (lr=4.7316e-05) (hash(x)=43942604)
713
+ 683 train 7.484223 (lr=4.7121e-05) (hash(x)=50507237)
714
+ 684 train 7.570594 (lr=4.6926e-05) (hash(x)=50384951)
715
+ 685 train 7.538752 (lr=4.6731e-05) (hash(x)=52518867)
716
+ 686 train 7.593556 (lr=4.6537e-05) (hash(x)=52261699)
717
+ 687 train 7.568779 (lr=4.6342e-05) (hash(x)=44845666)
718
+ 688 train 7.524303 (lr=4.6148e-05) (hash(x)=45979211)
719
+ 689 train 7.440282 (lr=4.5954e-05) (hash(x)=49705337)
720
+ 690 train 7.494939 (lr=4.5760e-05) (hash(x)=51160936)
721
+ 691 train 7.370516 (lr=4.5567e-05) (hash(x)=37893937)
722
+ 692 train 7.644383 (lr=4.5373e-05) (hash(x)=52735111)
723
+ 693 train 7.433566 (lr=4.5180e-05) (hash(x)=54479657)
724
+ 694 train 7.540282 (lr=4.4987e-05) (hash(x)=48206992)
725
+ 695 train 7.640773 (lr=4.4794e-05) (hash(x)=52920201)
726
+ 696 train 7.618382 (lr=4.4601e-05) (hash(x)=54965547)
727
+ 697 train 7.608090 (lr=4.4408e-05) (hash(x)=46946305)
728
+ 698 train 7.438077 (lr=4.4216e-05) (hash(x)=37814265)
729
+ 699 train 7.753296 (lr=4.4024e-05) (hash(x)=47469055)
730
+ 700 val loss 7.5378
731
+ 700 val perplexity 1877.7775
732
+ 700 train 7.594511 (lr=4.3832e-05) (hash(x)=48282280)
733
+ 701 train 7.876435 (lr=4.3640e-05) (hash(x)=48656681)
734
+ 702 train 7.677108 (lr=4.3449e-05) (hash(x)=52997901)
735
+ 703 train 7.497166 (lr=4.3258e-05) (hash(x)=47620343)
736
+ 704 train 7.563179 (lr=4.3067e-05) (hash(x)=47038936)
737
+ 705 train 7.510221 (lr=4.2876e-05) (hash(x)=46736914)
738
+ 706 train 7.755523 (lr=4.2685e-05) (hash(x)=54497460)
739
+ 707 train 7.449049 (lr=4.2495e-05) (hash(x)=43589469)
740
+ 708 train 7.513351 (lr=4.2305e-05) (hash(x)=49715257)
741
+ 709 train 8.002102 (lr=4.2115e-05) (hash(x)=45856317)
742
+ 710 train 7.518021 (lr=4.1925e-05) (hash(x)=50243508)
743
+ 711 train 7.595821 (lr=4.1736e-05) (hash(x)=51533352)
744
+ 712 train 7.525845 (lr=4.1547e-05) (hash(x)=43093930)
745
+ 713 train 7.500087 (lr=4.1358e-05) (hash(x)=42449079)
746
+ 714 train 7.680701 (lr=4.1170e-05) (hash(x)=54047656)
747
+ 715 train 7.609549 (lr=4.0981e-05) (hash(x)=49191723)
748
+ 716 train 7.553832 (lr=4.0793e-05) (hash(x)=44007317)
749
+ 717 train 7.481680 (lr=4.0606e-05) (hash(x)=54086880)
750
+ 718 train 7.541256 (lr=4.0418e-05) (hash(x)=56741903)
751
+ 719 train 7.591719 (lr=4.0231e-05) (hash(x)=47210623)
752
+ 720 train 7.641633 (lr=4.0044e-05) (hash(x)=51394805)
753
+ 721 train 7.515373 (lr=3.9857e-05) (hash(x)=50935538)
754
+ 722 train 7.537125 (lr=3.9671e-05) (hash(x)=44587058)
755
+ 723 train 7.657060 (lr=3.9485e-05) (hash(x)=51394944)
756
+ 724 train 7.665120 (lr=3.9299e-05) (hash(x)=55439212)
757
+ 725 train 7.513854 (lr=3.9114e-05) (hash(x)=53386161)
758
+ 726 train 7.565424 (lr=3.8929e-05) (hash(x)=50629459)
759
+ 727 train 7.618294 (lr=3.8744e-05) (hash(x)=50661409)
760
+ 728 train 7.507844 (lr=3.8560e-05) (hash(x)=50997400)
761
+ 729 train 7.681775 (lr=3.8376e-05) (hash(x)=47678317)
762
+ 730 train 7.451954 (lr=3.8192e-05) (hash(x)=49990709)
763
+ 731 train 7.645921 (lr=3.8008e-05) (hash(x)=56179487)
764
+ 732 train 7.711086 (lr=3.7825e-05) (hash(x)=53558434)
765
+ 733 train 7.593603 (lr=3.7642e-05) (hash(x)=49531250)
766
+ 734 train 7.667074 (lr=3.7460e-05) (hash(x)=55246569)
767
+ 735 train 7.537549 (lr=3.7277e-05) (hash(x)=52833764)
768
+ 736 train 7.610781 (lr=3.7096e-05) (hash(x)=54901121)
769
+ 737 train 7.587402 (lr=3.6914e-05) (hash(x)=53894387)
770
+ 738 train 7.670984 (lr=3.6733e-05) (hash(x)=48307417)
771
+ 739 train 7.550446 (lr=3.6552e-05) (hash(x)=50490611)
772
+ 740 train 7.495819 (lr=3.6372e-05) (hash(x)=52155436)
773
+ 741 train 7.539612 (lr=3.6192e-05) (hash(x)=45552853)
774
+ 742 train 7.674268 (lr=3.6012e-05) (hash(x)=49739024)
775
+ 743 train 7.497568 (lr=3.5833e-05) (hash(x)=49288247)
776
+ 744 train 7.455999 (lr=3.5654e-05) (hash(x)=50328652)
777
+ 745 train 7.490513 (lr=3.5475e-05) (hash(x)=51558451)
778
+ 746 train 7.620928 (lr=3.5297e-05) (hash(x)=58478863)
779
+ 747 train 7.606379 (lr=3.5119e-05) (hash(x)=46811775)
780
+ 748 train 7.510563 (lr=3.4942e-05) (hash(x)=44132843)
781
+ 749 train 7.382246 (lr=3.4765e-05) (hash(x)=54962889)
782
+ 750 val loss 7.5229
783
+ 750 val perplexity 1849.8328
784
+ 750 train 7.390863 (lr=3.4588e-05) (hash(x)=49974197)
785
+ 751 train 7.535581 (lr=3.4412e-05) (hash(x)=58577985)
786
+ 752 train 7.523450 (lr=3.4236e-05) (hash(x)=50134139)
787
+ 753 train 7.567155 (lr=3.4060e-05) (hash(x)=47649871)
788
+ 754 train 7.601573 (lr=3.3885e-05) (hash(x)=52291145)
789
+ 755 train 7.413115 (lr=3.3711e-05) (hash(x)=45593386)
790
+ 756 train 7.492823 (lr=3.3537e-05) (hash(x)=52038315)
791
+ 757 train 7.427456 (lr=3.3363e-05) (hash(x)=45848371)
792
+ 758 train 7.468984 (lr=3.3189e-05) (hash(x)=50207699)
793
+ 759 train 7.554396 (lr=3.3016e-05) (hash(x)=51149878)
794
+ 760 train 7.650524 (lr=3.2844e-05) (hash(x)=59871083)
795
+ 761 train 7.568895 (lr=3.2672e-05) (hash(x)=51697792)
796
+ 762 train 7.467728 (lr=3.2500e-05) (hash(x)=47945841)
797
+ 763 train 7.545013 (lr=3.2329e-05) (hash(x)=55613373)
798
+ 764 train 7.470401 (lr=3.2158e-05) (hash(x)=47596112)
799
+ 765 train 7.421134 (lr=3.1988e-05) (hash(x)=46794155)
800
+ 766 train 7.650257 (lr=3.1818e-05) (hash(x)=53279481)
801
+ 767 train 7.572445 (lr=3.1648e-05) (hash(x)=61637650)
802
+ 768 train 7.561564 (lr=3.1479e-05) (hash(x)=49431797)
803
+ 769 train 7.746759 (lr=3.1311e-05) (hash(x)=49407940)
804
+ 770 train 7.565355 (lr=3.1142e-05) (hash(x)=46673035)
805
+ 771 train 7.512336 (lr=3.0975e-05) (hash(x)=50650755)
806
+ 772 train 7.558628 (lr=3.0808e-05) (hash(x)=42476007)
807
+ 773 train 7.617481 (lr=3.0641e-05) (hash(x)=48538353)
808
+ 774 train 7.467067 (lr=3.0475e-05) (hash(x)=49575493)
809
+ 775 train 7.596929 (lr=3.0309e-05) (hash(x)=49218892)
810
+ 776 train 7.713521 (lr=3.0144e-05) (hash(x)=51084108)
811
+ 777 train 7.438439 (lr=2.9979e-05) (hash(x)=53934744)
812
+ 778 train 7.437298 (lr=2.9814e-05) (hash(x)=48057225)
813
+ 779 train 7.396704 (lr=2.9651e-05) (hash(x)=45747335)
814
+ 780 train 7.473419 (lr=2.9487e-05) (hash(x)=53160889)
815
+ 781 train 7.615356 (lr=2.9324e-05) (hash(x)=64327478)
816
+ 782 train 7.458179 (lr=2.9162e-05) (hash(x)=50486283)
817
+ 783 train 7.776649 (lr=2.9000e-05) (hash(x)=49031585)
818
+ 784 train 7.738771 (lr=2.8839e-05) (hash(x)=49394132)
819
+ 785 train 7.407112 (lr=2.8678e-05) (hash(x)=46800473)
820
+ 786 train 7.282935 (lr=2.8518e-05) (hash(x)=49181166)
821
+ 787 train 7.444497 (lr=2.8358e-05) (hash(x)=50935471)
822
+ 788 train 7.702753 (lr=2.8199e-05) (hash(x)=55388862)
823
+ 789 train 7.649736 (lr=2.8040e-05) (hash(x)=53344957)
824
+ 790 train 7.726390 (lr=2.7881e-05) (hash(x)=50636572)
825
+ 791 train 7.656892 (lr=2.7724e-05) (hash(x)=50906709)
826
+ 792 train 7.579619 (lr=2.7566e-05) (hash(x)=53337398)
827
+ 793 train 7.440506 (lr=2.7410e-05) (hash(x)=46187511)
828
+ 794 train 7.633956 (lr=2.7254e-05) (hash(x)=49433426)
829
+ 795 train 7.449391 (lr=2.7098e-05) (hash(x)=51641428)
830
+ 796 train 7.489309 (lr=2.6943e-05) (hash(x)=49157344)
831
+ 797 train 7.328343 (lr=2.6788e-05) (hash(x)=51616159)
832
+ 798 train 7.551970 (lr=2.6634e-05) (hash(x)=47462943)
833
+ 799 train 7.566297 (lr=2.6481e-05) (hash(x)=49211707)
834
+ 800 val loss 7.5043
835
+ 800 val perplexity 1815.9023
836
+ 800 train 7.612478 (lr=2.6328e-05) (hash(x)=50417541)
837
+ 801 train 7.502779 (lr=2.6176e-05) (hash(x)=44250219)
838
+ 802 train 7.462137 (lr=2.6024e-05) (hash(x)=50210317)
839
+ 803 train 7.532291 (lr=2.5873e-05) (hash(x)=50609821)
840
+ 804 train 7.465837 (lr=2.5722e-05) (hash(x)=40910622)
841
+ 805 train 7.377593 (lr=2.5572e-05) (hash(x)=49837063)
842
+ 806 train 7.499692 (lr=2.5423e-05) (hash(x)=50596040)
843
+ 807 train 7.678557 (lr=2.5274e-05) (hash(x)=54582969)
844
+ 808 train 7.532219 (lr=2.5125e-05) (hash(x)=51471007)
845
+ 809 train 7.645277 (lr=2.4978e-05) (hash(x)=55788551)
846
+ 810 train 7.385769 (lr=2.4830e-05) (hash(x)=45935856)
847
+ 811 train 7.514287 (lr=2.4684e-05) (hash(x)=47818076)
848
+ 812 train 7.407878 (lr=2.4538e-05) (hash(x)=50632171)
849
+ 813 train 7.378843 (lr=2.4392e-05) (hash(x)=46667980)
850
+ 814 train 7.482218 (lr=2.4247e-05) (hash(x)=51277169)
851
+ 815 train 7.410571 (lr=2.4103e-05) (hash(x)=47115974)
852
+ 816 train 7.313759 (lr=2.3959e-05) (hash(x)=43878503)
853
+ 817 train 7.448693 (lr=2.3816e-05) (hash(x)=52399224)
854
+ 818 train 7.600151 (lr=2.3674e-05) (hash(x)=56384241)
855
+ 819 train 7.467856 (lr=2.3532e-05) (hash(x)=52516612)
856
+ 820 train 7.460621 (lr=2.3391e-05) (hash(x)=47474067)
857
+ 821 train 7.502143 (lr=2.3250e-05) (hash(x)=49236383)
858
+ 822 train 7.478827 (lr=2.3110e-05) (hash(x)=54447373)
859
+ 823 train 7.585988 (lr=2.2971e-05) (hash(x)=60697865)
860
+ 824 train 7.573822 (lr=2.2832e-05) (hash(x)=53137776)
861
+ 825 train 7.333549 (lr=2.2694e-05) (hash(x)=49192482)
862
+ 826 train 7.456583 (lr=2.2556e-05) (hash(x)=53732034)
863
+ 827 train 7.524291 (lr=2.2420e-05) (hash(x)=54504436)
864
+ 828 train 7.474274 (lr=2.2283e-05) (hash(x)=47738508)
865
+ 829 train 7.555201 (lr=2.2148e-05) (hash(x)=51623214)
866
+ 830 train 7.441997 (lr=2.2013e-05) (hash(x)=42124171)
867
+ 831 train 7.406953 (lr=2.1878e-05) (hash(x)=51608067)
868
+ 832 train 7.500198 (lr=2.1745e-05) (hash(x)=51210730)
869
+ 833 train 7.767601 (lr=2.1612e-05) (hash(x)=60111943)
870
+ 834 train 7.414913 (lr=2.1479e-05) (hash(x)=43793972)
871
+ 835 train 7.508626 (lr=2.1347e-05) (hash(x)=44616060)
872
+ 836 train 7.577850 (lr=2.1216e-05) (hash(x)=46527331)
873
+ 837 train 7.504253 (lr=2.1086e-05) (hash(x)=50528250)
874
+ 838 train 7.433561 (lr=2.0956e-05) (hash(x)=52256167)
875
+ 839 train 7.365214 (lr=2.0827e-05) (hash(x)=46388895)
876
+ 840 train 7.815643 (lr=2.0698e-05) (hash(x)=71242124)
877
+ 841 train 7.454727 (lr=2.0570e-05) (hash(x)=49826941)
878
+ 842 train 7.292396 (lr=2.0443e-05) (hash(x)=51315458)
879
+ 843 train 7.771891 (lr=2.0317e-05) (hash(x)=57335639)
880
+ 844 train 7.548403 (lr=2.0191e-05) (hash(x)=50263196)
881
+ 845 train 7.524794 (lr=2.0066e-05) (hash(x)=56050290)
882
+ 846 train 7.565207 (lr=1.9941e-05) (hash(x)=47723258)
883
+ 847 train 7.372895 (lr=1.9818e-05) (hash(x)=49324402)
884
+ 848 train 7.531723 (lr=1.9694e-05) (hash(x)=52128369)
885
+ 849 train 7.413733 (lr=1.9572e-05) (hash(x)=51018209)
886
+ 850 val loss 7.4998
887
+ 850 val perplexity 1807.7416
888
+ 850 train 7.532466 (lr=1.9450e-05) (hash(x)=53105296)
889
+ 851 train 7.448638 (lr=1.9329e-05) (hash(x)=50430474)
890
+ 852 train 7.568802 (lr=1.9209e-05) (hash(x)=47607194)
891
+ 853 train 7.444491 (lr=1.9089e-05) (hash(x)=49326441)
892
+ 854 train 7.391173 (lr=1.8970e-05) (hash(x)=55614122)
893
+ 855 train 7.491209 (lr=1.8852e-05) (hash(x)=48628017)
894
+ 856 train 7.387217 (lr=1.8734e-05) (hash(x)=54346909)
895
+ 857 train 7.541063 (lr=1.8618e-05) (hash(x)=48223880)
896
+ 858 train 7.488844 (lr=1.8501e-05) (hash(x)=45388401)
897
+ 859 train 7.585262 (lr=1.8386e-05) (hash(x)=46001213)
898
+ 860 train 7.544141 (lr=1.8271e-05) (hash(x)=45084714)
899
+ 861 train 7.573071 (lr=1.8157e-05) (hash(x)=57009289)
900
+ 862 train 7.464751 (lr=1.8044e-05) (hash(x)=50637087)
901
+ 863 train 7.499862 (lr=1.7931e-05) (hash(x)=57751312)
902
+ 864 train 7.681859 (lr=1.7819e-05) (hash(x)=56777261)
903
+ 865 train 7.530841 (lr=1.7708e-05) (hash(x)=51606280)
904
+ 866 train 7.520597 (lr=1.7598e-05) (hash(x)=47327839)
905
+ 867 train 7.481063 (lr=1.7488e-05) (hash(x)=36722344)
906
+ 868 train 7.363868 (lr=1.7379e-05) (hash(x)=48753434)
907
+ 869 train 7.475605 (lr=1.7271e-05) (hash(x)=50288290)
908
+ 870 train 7.562962 (lr=1.7163e-05) (hash(x)=49893422)
909
+ 871 train 7.423018 (lr=1.7056e-05) (hash(x)=52490651)
910
+ 872 train 7.486588 (lr=1.6950e-05) (hash(x)=49409003)
911
+ 873 train 7.435451 (lr=1.6845e-05) (hash(x)=53532287)
912
+ 874 train 7.701715 (lr=1.6740e-05) (hash(x)=49447061)
913
+ 875 train 7.857010 (lr=1.6636e-05) (hash(x)=48609404)
914
+ 876 train 7.369622 (lr=1.6533e-05) (hash(x)=48189258)
915
+ 877 train 7.420664 (lr=1.6431e-05) (hash(x)=52319696)
916
+ 878 train 7.558624 (lr=1.6329e-05) (hash(x)=43816075)
917
+ 879 train 7.551716 (lr=1.6228e-05) (hash(x)=49378339)
918
+ 880 train 7.497963 (lr=1.6128e-05) (hash(x)=44838280)
919
+ 881 train 7.480447 (lr=1.6029e-05) (hash(x)=56110408)
920
+ 882 train 7.514180 (lr=1.5930e-05) (hash(x)=45608194)
921
+ 883 train 7.380791 (lr=1.5832e-05) (hash(x)=51307401)
922
+ 884 train 7.517581 (lr=1.5735e-05) (hash(x)=54827224)
923
+ 885 train 7.541292 (lr=1.5639e-05) (hash(x)=51807221)
924
+ 886 train 7.455366 (lr=1.5543e-05) (hash(x)=47566484)
925
+ 887 train 7.444821 (lr=1.5449e-05) (hash(x)=43527966)
926
+ 888 train 7.213089 (lr=1.5354e-05) (hash(x)=44965744)
927
+ 889 train 7.467502 (lr=1.5261e-05) (hash(x)=53260546)
928
+ 890 train 7.510655 (lr=1.5169e-05) (hash(x)=50190502)
929
+ 891 train 7.334621 (lr=1.5077e-05) (hash(x)=42265391)
930
+ 892 train 7.479854 (lr=1.4986e-05) (hash(x)=50082378)
931
+ 893 train 7.504585 (lr=1.4896e-05) (hash(x)=53589270)
932
+ 894 train 7.402186 (lr=1.4806e-05) (hash(x)=50017735)
933
+ 895 train 7.351293 (lr=1.4718e-05) (hash(x)=50193316)
934
+ 896 train 7.363407 (lr=1.4630e-05) (hash(x)=48318967)
935
+ 897 train 7.382689 (lr=1.4543e-05) (hash(x)=48446905)
936
+ 898 train 7.823289 (lr=1.4456e-05) (hash(x)=62919547)
937
+ 899 train 7.615025 (lr=1.4371e-05) (hash(x)=49894505)
938
+ 900 val loss 7.4870
939
+ 900 val perplexity 1784.7444
940
+ 900 train 7.458378 (lr=1.4286e-05) (hash(x)=53945628)
941
+ 901 train 7.335810 (lr=1.4202e-05) (hash(x)=44497610)
942
+ 902 train 7.337255 (lr=1.4119e-05) (hash(x)=50504601)
943
+ 903 train 7.324965 (lr=1.4037e-05) (hash(x)=47215978)
944
+ 904 train 7.406442 (lr=1.3955e-05) (hash(x)=47677269)
945
+ 905 train 7.448306 (lr=1.3874e-05) (hash(x)=49318343)
946
+ 906 train 7.450541 (lr=1.3794e-05) (hash(x)=49903709)
947
+ 907 train 7.388296 (lr=1.3715e-05) (hash(x)=50021640)
948
+ 908 train 7.256376 (lr=1.3637e-05) (hash(x)=50078912)
949
+ 909 train 7.676193 (lr=1.3559e-05) (hash(x)=52643396)
950
+ 910 train 7.401347 (lr=1.3482e-05) (hash(x)=49583258)
951
+ 911 train 7.384390 (lr=1.3407e-05) (hash(x)=47815754)
952
+ 912 train 7.268900 (lr=1.3331e-05) (hash(x)=45502665)
953
+ 913 train 7.571783 (lr=1.3257e-05) (hash(x)=53581822)
954
+ 914 train 7.628668 (lr=1.3183e-05) (hash(x)=49730972)
955
+ 915 train 7.414804 (lr=1.3111e-05) (hash(x)=47030561)
956
+ 916 train 7.531391 (lr=1.3039e-05) (hash(x)=49950830)
957
+ 917 train 7.546016 (lr=1.2968e-05) (hash(x)=52654669)
958
+ 918 train 7.441311 (lr=1.2897e-05) (hash(x)=55636426)
959
+ 919 train 7.482813 (lr=1.2828e-05) (hash(x)=53107904)
960
+ 920 train 7.495267 (lr=1.2759e-05) (hash(x)=50409352)
961
+ 921 train 7.437762 (lr=1.2691e-05) (hash(x)=50440529)
962
+ 922 train 7.456680 (lr=1.2624e-05) (hash(x)=49935114)
963
+ 923 train 7.600952 (lr=1.2558e-05) (hash(x)=52060422)
964
+ 924 train 7.481233 (lr=1.2493e-05) (hash(x)=48103244)
965
+ 925 train 7.516505 (lr=1.2428e-05) (hash(x)=47489666)
966
+ 926 train 7.559579 (lr=1.2364e-05) (hash(x)=51275986)
967
+ 927 train 7.366834 (lr=1.2301e-05) (hash(x)=43893186)
968
+ 928 train 7.376631 (lr=1.2239e-05) (hash(x)=49710654)
969
+ 929 train 7.493164 (lr=1.2178e-05) (hash(x)=57920651)
970
+ 930 train 7.447710 (lr=1.2118e-05) (hash(x)=53051959)
971
+ 931 train 7.462317 (lr=1.2058e-05) (hash(x)=53255049)
972
+ 932 train 7.347263 (lr=1.1999e-05) (hash(x)=43037835)
973
+ 933 train 7.391015 (lr=1.1941e-05) (hash(x)=47189747)
974
+ 934 train 7.330236 (lr=1.1884e-05) (hash(x)=50412552)
975
+ 935 train 7.329170 (lr=1.1828e-05) (hash(x)=47566671)
976
+ 936 train 7.449103 (lr=1.1772e-05) (hash(x)=54189181)
977
+ 937 train 7.555171 (lr=1.1718e-05) (hash(x)=58014227)
978
+ 938 train 7.483793 (lr=1.1664e-05) (hash(x)=52672919)
979
+ 939 train 7.370131 (lr=1.1611e-05) (hash(x)=46462188)
980
+ 940 train 7.438708 (lr=1.1559e-05) (hash(x)=45953229)
981
+ 941 train 7.464208 (lr=1.1508e-05) (hash(x)=47535577)
982
+ 942 train 7.465387 (lr=1.1457e-05) (hash(x)=50126890)
983
+ 943 train 7.418552 (lr=1.1408e-05) (hash(x)=49178186)
984
+ 944 train 7.585033 (lr=1.1359e-05) (hash(x)=50756892)
985
+ 945 train 7.380239 (lr=1.1311e-05) (hash(x)=50555757)
986
+ 946 train 7.508782 (lr=1.1264e-05) (hash(x)=50734447)
987
+ 947 train 7.215734 (lr=1.1218e-05) (hash(x)=46557241)
988
+ 948 train 7.351797 (lr=1.1173e-05) (hash(x)=46234844)
989
+ 949 train 7.379083 (lr=1.1128e-05) (hash(x)=50626064)
990
+ 950 val loss 7.4834
991
+ 950 val perplexity 1778.2892
992
+ 950 train 7.445689 (lr=1.1085e-05) (hash(x)=45428767)
993
+ 951 train 7.384057 (lr=1.1042e-05) (hash(x)=50016505)
994
+ 952 train 7.254467 (lr=1.1000e-05) (hash(x)=38316045)
995
+ 953 train 7.430998 (lr=1.0959e-05) (hash(x)=53588337)
996
+ 954 train 7.515052 (lr=1.0919e-05) (hash(x)=49172718)
997
+ 955 train 7.374635 (lr=1.0879e-05) (hash(x)=48675645)
998
+ 956 train 7.439677 (lr=1.0841e-05) (hash(x)=49197427)
999
+ 957 train 7.401544 (lr=1.0803e-05) (hash(x)=48397771)
1000
+ 958 train 7.588004 (lr=1.0766e-05) (hash(x)=54923539)
1001
+ 959 train 7.339106 (lr=1.0730e-05) (hash(x)=50047979)
1002
+ 960 train 7.350917 (lr=1.0695e-05) (hash(x)=52636339)
1003
+ 961 train 7.486341 (lr=1.0661e-05) (hash(x)=53304318)
1004
+ 962 train 7.367694 (lr=1.0628e-05) (hash(x)=51646229)
1005
+ 963 train 7.494512 (lr=1.0595e-05) (hash(x)=51670794)
1006
+ 964 train 7.684775 (lr=1.0563e-05) (hash(x)=56438447)
1007
+ 965 train 7.566239 (lr=1.0533e-05) (hash(x)=50279366)
1008
+ 966 train 7.539135 (lr=1.0503e-05) (hash(x)=44091655)
1009
+ 967 train 7.441432 (lr=1.0474e-05) (hash(x)=45522017)
1010
+ 968 train 7.392604 (lr=1.0445e-05) (hash(x)=54442058)
1011
+ 969 train 7.422410 (lr=1.0418e-05) (hash(x)=54042710)
1012
+ 970 train 7.425647 (lr=1.0391e-05) (hash(x)=50890161)
1013
+ 971 train 7.488587 (lr=1.0366e-05) (hash(x)=48113844)
1014
+ 972 train 7.370950 (lr=1.0341e-05) (hash(x)=45677997)
1015
+ 973 train 7.444919 (lr=1.0317e-05) (hash(x)=55529403)
1016
+ 974 train 7.224048 (lr=1.0294e-05) (hash(x)=47204834)
1017
+ 975 train 7.590627 (lr=1.0272e-05) (hash(x)=49734327)
1018
+ 976 train 7.416435 (lr=1.0251e-05) (hash(x)=48445241)
1019
+ 977 train 7.315228 (lr=1.0230e-05) (hash(x)=47311591)
1020
+ 978 train 7.319247 (lr=1.0211e-05) (hash(x)=51019082)
1021
+ 979 train 7.304410 (lr=1.0192e-05) (hash(x)=44680866)
1022
+ 980 train 7.437308 (lr=1.0174e-05) (hash(x)=47479531)
1023
+ 981 train 7.225320 (lr=1.0157e-05) (hash(x)=46784017)
1024
+ 982 train 7.384959 (lr=1.0141e-05) (hash(x)=48550320)
1025
+ 983 train 7.394248 (lr=1.0126e-05) (hash(x)=48537238)
1026
+ 984 train 7.630459 (lr=1.0111e-05) (hash(x)=56461945)
1027
+ 985 train 7.744573 (lr=1.0098e-05) (hash(x)=49899383)
1028
+ 986 train 7.441541 (lr=1.0085e-05) (hash(x)=52053967)
1029
+ 987 train 7.389830 (lr=1.0074e-05) (hash(x)=43607110)
1030
+ 988 train 7.380809 (lr=1.0063e-05) (hash(x)=48511684)
1031
+ 989 train 7.442892 (lr=1.0053e-05) (hash(x)=48874947)
1032
+ 990 train 7.489177 (lr=1.0044e-05) (hash(x)=49036386)
1033
+ 991 train 7.410820 (lr=1.0035e-05) (hash(x)=49203903)
1034
+ 992 train 7.320305 (lr=1.0028e-05) (hash(x)=45134836)
1035
+ 993 train 7.496284 (lr=1.0021e-05) (hash(x)=51416994)
1036
+ 994 train 7.738585 (lr=1.0016e-05) (hash(x)=65320726)
1037
+ 995 train 7.339327 (lr=1.0011e-05) (hash(x)=46183046)
1038
+ 996 train 7.592594 (lr=1.0007e-05) (hash(x)=52360896)
1039
+ 997 train 7.519143 (lr=1.0004e-05) (hash(x)=56411717)
1040
+ 998 train 7.443420 (lr=1.0002e-05) (hash(x)=46618013)
1041
+ 999 val loss 7.4830
1042
+ 999 val perplexity 1777.5558
1043
+ 999 train 7.821465 (lr=1.0000e-05) (hash(x)=71777091)
lr1e-4_total_batch_size40960_seq_len1024/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7391292f356b3c2dd9c3c6d0e7e7122abb8316e710019b75afc6ad724276fa61
3
+ size 86167106
lr1e-4_total_batch_size40960_seq_len1024/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbd0cf45a543fe795b6aed2526d2395fee2dfad27110867d5a19b817c48ce155
3
+ size 71681862