andrew-healey commited on
Commit
1933a60
·
verified ·
1 Parent(s): 6701ffd

Upload folder using huggingface_hub

Browse files
lr1e-4_total_batch_size10240_seq_len256/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep/lr1e-4_total_batch_size10240_seq_len256", "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": 256, "batch_size": null, "total_batch_size": 10240, "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_size10240_seq_len256/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b9fd0bfa27ab267155e5ae2bb7e7ccf3640df9cfec6890472ce93f6d886a52e
3
+ size 964
lr1e-4_total_batch_size10240_seq_len256/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.7488
3
+ 0 val perplexity 126607.6172
4
+ 0 train 11.749471 (lr=3.4965e-07) (hash(x)=12683395)
5
+ 1 train 11.747282 (lr=6.9930e-07) (hash(x)=12136211)
6
+ 2 train 11.715776 (lr=1.0490e-06) (hash(x)=13068552)
7
+ 3 train 11.761577 (lr=1.3986e-06) (hash(x)=9915468)
8
+ 4 train 11.756632 (lr=1.7483e-06) (hash(x)=12164183)
9
+ 5 train 11.743533 (lr=2.0979e-06) (hash(x)=13629277)
10
+ 6 train 11.725996 (lr=2.4476e-06) (hash(x)=13413021)
11
+ 7 train 11.750143 (lr=2.7972e-06) (hash(x)=12684830)
12
+ 8 train 11.707050 (lr=3.1469e-06) (hash(x)=13262550)
13
+ 9 train 11.740360 (lr=3.4965e-06) (hash(x)=12003126)
14
+ 10 train 11.716045 (lr=3.8462e-06) (hash(x)=14144015)
15
+ 11 train 11.705021 (lr=4.1958e-06) (hash(x)=12014715)
16
+ 12 train 11.683252 (lr=4.5455e-06) (hash(x)=14912984)
17
+ 13 train 11.655256 (lr=4.8951e-06) (hash(x)=9492167)
18
+ 14 train 11.677439 (lr=5.2448e-06) (hash(x)=15241928)
19
+ 15 train 11.614085 (lr=5.5944e-06) (hash(x)=12595649)
20
+ 16 train 11.615525 (lr=5.9441e-06) (hash(x)=15162520)
21
+ 17 train 11.623725 (lr=6.2937e-06) (hash(x)=11980631)
22
+ 18 train 11.645550 (lr=6.6434e-06) (hash(x)=13172222)
23
+ 19 train 11.608740 (lr=6.9930e-06) (hash(x)=11101031)
24
+ 20 train 11.599634 (lr=7.3427e-06) (hash(x)=13275746)
25
+ 21 train 11.553246 (lr=7.6923e-06) (hash(x)=14316349)
26
+ 22 train 11.544618 (lr=8.0420e-06) (hash(x)=14143687)
27
+ 23 train 11.547494 (lr=8.3916e-06) (hash(x)=12036401)
28
+ 24 train 11.534065 (lr=8.7413e-06) (hash(x)=11143776)
29
+ 25 train 11.500435 (lr=9.0909e-06) (hash(x)=12337125)
30
+ 26 train 11.495814 (lr=9.4406e-06) (hash(x)=10700109)
31
+ 27 train 11.446100 (lr=9.7902e-06) (hash(x)=11823810)
32
+ 28 train 11.443575 (lr=1.0140e-05) (hash(x)=12070591)
33
+ 29 train 11.410051 (lr=1.0490e-05) (hash(x)=13033959)
34
+ 30 train 11.419327 (lr=1.0839e-05) (hash(x)=10786770)
35
+ 31 train 11.376769 (lr=1.1189e-05) (hash(x)=13549925)
36
+ 32 train 11.274728 (lr=1.1538e-05) (hash(x)=14101838)
37
+ 33 train 11.336843 (lr=1.1888e-05) (hash(x)=14220701)
38
+ 34 train 11.270385 (lr=1.2238e-05) (hash(x)=14571180)
39
+ 35 train 11.232204 (lr=1.2587e-05) (hash(x)=13252262)
40
+ 36 train 11.186784 (lr=1.2937e-05) (hash(x)=13676730)
41
+ 37 train 11.188702 (lr=1.3287e-05) (hash(x)=14638145)
42
+ 38 train 11.107222 (lr=1.3636e-05) (hash(x)=14385711)
43
+ 39 train 11.145391 (lr=1.3986e-05) (hash(x)=13749054)
44
+ 40 train 11.132839 (lr=1.4336e-05) (hash(x)=12469703)
45
+ 41 train 11.062656 (lr=1.4685e-05) (hash(x)=12681760)
46
+ 42 train 11.041040 (lr=1.5035e-05) (hash(x)=12423697)
47
+ 43 train 10.977920 (lr=1.5385e-05) (hash(x)=14461706)
48
+ 44 train 10.971835 (lr=1.5734e-05) (hash(x)=13727016)
49
+ 45 train 10.869337 (lr=1.6084e-05) (hash(x)=14235218)
50
+ 46 train 10.883167 (lr=1.6434e-05) (hash(x)=12501568)
51
+ 47 train 10.882954 (lr=1.6783e-05) (hash(x)=11735995)
52
+ 48 train 10.893804 (lr=1.7133e-05) (hash(x)=11569945)
53
+ 49 train 10.767950 (lr=1.7483e-05) (hash(x)=11641112)
54
+ 50 val loss 10.7758
55
+ 50 val perplexity 47847.1719
56
+ 50 train 10.766085 (lr=1.7832e-05) (hash(x)=11536784)
57
+ 51 train 10.807592 (lr=1.8182e-05) (hash(x)=12231701)
58
+ 52 train 10.710636 (lr=1.8531e-05) (hash(x)=13521301)
59
+ 53 train 10.617676 (lr=1.8881e-05) (hash(x)=10175470)
60
+ 54 train 10.694592 (lr=1.9231e-05) (hash(x)=13903066)
61
+ 55 train 10.785358 (lr=1.9580e-05) (hash(x)=14117945)
62
+ 56 train 10.570838 (lr=1.9930e-05) (hash(x)=14851356)
63
+ 57 train 10.582170 (lr=2.0280e-05) (hash(x)=13116544)
64
+ 58 train 10.539407 (lr=2.0629e-05) (hash(x)=16366718)
65
+ 59 train 10.519110 (lr=2.0979e-05) (hash(x)=12678303)
66
+ 60 train 10.429228 (lr=2.1329e-05) (hash(x)=11144768)
67
+ 61 train 10.532296 (lr=2.1678e-05) (hash(x)=11927462)
68
+ 62 train 10.437913 (lr=2.2028e-05) (hash(x)=10418366)
69
+ 63 train 10.490436 (lr=2.2378e-05) (hash(x)=12758461)
70
+ 64 train 10.496836 (lr=2.2727e-05) (hash(x)=11680802)
71
+ 65 train 10.512108 (lr=2.3077e-05) (hash(x)=11946230)
72
+ 66 train 10.403681 (lr=2.3427e-05) (hash(x)=11704519)
73
+ 67 train 10.336603 (lr=2.3776e-05) (hash(x)=13701311)
74
+ 68 train 10.364712 (lr=2.4126e-05) (hash(x)=12751615)
75
+ 69 train 10.376347 (lr=2.4476e-05) (hash(x)=11198382)
76
+ 70 train 10.404160 (lr=2.4825e-05) (hash(x)=16075171)
77
+ 71 train 10.369370 (lr=2.5175e-05) (hash(x)=14346433)
78
+ 72 train 10.362631 (lr=2.5524e-05) (hash(x)=12888806)
79
+ 73 train 10.452332 (lr=2.5874e-05) (hash(x)=17855984)
80
+ 74 train 10.400669 (lr=2.6224e-05) (hash(x)=9616381)
81
+ 75 train 10.530105 (lr=2.6573e-05) (hash(x)=18910052)
82
+ 76 train 10.372682 (lr=2.6923e-05) (hash(x)=12769596)
83
+ 77 train 10.346395 (lr=2.7273e-05) (hash(x)=11885403)
84
+ 78 train 10.346300 (lr=2.7622e-05) (hash(x)=12476514)
85
+ 79 train 10.324388 (lr=2.7972e-05) (hash(x)=12508350)
86
+ 80 train 10.244604 (lr=2.8322e-05) (hash(x)=10745637)
87
+ 81 train 10.281462 (lr=2.8671e-05) (hash(x)=10862804)
88
+ 82 train 10.300925 (lr=2.9021e-05) (hash(x)=15075084)
89
+ 83 train 10.227171 (lr=2.9371e-05) (hash(x)=14043988)
90
+ 84 train 10.114356 (lr=2.9720e-05) (hash(x)=12667924)
91
+ 85 train 10.121610 (lr=3.0070e-05) (hash(x)=11916406)
92
+ 86 train 10.154733 (lr=3.0420e-05) (hash(x)=13406195)
93
+ 87 train 10.179409 (lr=3.0769e-05) (hash(x)=11968227)
94
+ 88 train 10.235752 (lr=3.1119e-05) (hash(x)=11927490)
95
+ 89 train 10.156164 (lr=3.1469e-05) (hash(x)=13290191)
96
+ 90 train 10.153080 (lr=3.1818e-05) (hash(x)=11771253)
97
+ 91 train 10.187243 (lr=3.2168e-05) (hash(x)=15670776)
98
+ 92 train 10.087416 (lr=3.2517e-05) (hash(x)=11306133)
99
+ 93 train 10.183597 (lr=3.2867e-05) (hash(x)=14358384)
100
+ 94 train 10.083723 (lr=3.3217e-05) (hash(x)=10267428)
101
+ 95 train 10.155200 (lr=3.3566e-05) (hash(x)=11289341)
102
+ 96 train 10.078539 (lr=3.3916e-05) (hash(x)=14062293)
103
+ 97 train 10.165846 (lr=3.4266e-05) (hash(x)=14736057)
104
+ 98 train 10.253955 (lr=3.4615e-05) (hash(x)=16036112)
105
+ 99 train 10.239746 (lr=3.4965e-05) (hash(x)=11482477)
106
+ 100 val loss 10.1121
107
+ 100 val perplexity 24638.2129
108
+ 100 train 10.172000 (lr=3.5315e-05) (hash(x)=12742648)
109
+ 101 train 10.065492 (lr=3.5664e-05) (hash(x)=15429973)
110
+ 102 train 10.167632 (lr=3.6014e-05) (hash(x)=13972190)
111
+ 103 train 10.129551 (lr=3.6364e-05) (hash(x)=10345057)
112
+ 104 train 10.108545 (lr=3.6713e-05) (hash(x)=11609727)
113
+ 105 train 10.060003 (lr=3.7063e-05) (hash(x)=10816510)
114
+ 106 train 10.004835 (lr=3.7413e-05) (hash(x)=11801604)
115
+ 107 train 10.101038 (lr=3.7762e-05) (hash(x)=11157124)
116
+ 108 train 10.144393 (lr=3.8112e-05) (hash(x)=11666699)
117
+ 109 train 9.993711 (lr=3.8462e-05) (hash(x)=12349897)
118
+ 110 train 10.040422 (lr=3.8811e-05) (hash(x)=11648826)
119
+ 111 train 10.049404 (lr=3.9161e-05) (hash(x)=10750452)
120
+ 112 train 9.984320 (lr=3.9510e-05) (hash(x)=12114386)
121
+ 113 train 10.059756 (lr=3.9860e-05) (hash(x)=11859950)
122
+ 114 train 9.911646 (lr=4.0210e-05) (hash(x)=12837162)
123
+ 115 train 10.089715 (lr=4.0559e-05) (hash(x)=8921601)
124
+ 116 train 9.872884 (lr=4.0909e-05) (hash(x)=9125058)
125
+ 117 train 9.879999 (lr=4.1259e-05) (hash(x)=11333049)
126
+ 118 train 9.952305 (lr=4.1608e-05) (hash(x)=13807539)
127
+ 119 train 9.949345 (lr=4.1958e-05) (hash(x)=12165500)
128
+ 120 train 10.021984 (lr=4.2308e-05) (hash(x)=14343620)
129
+ 121 train 9.975952 (lr=4.2657e-05) (hash(x)=13153948)
130
+ 122 train 9.955353 (lr=4.3007e-05) (hash(x)=14497853)
131
+ 123 train 9.946017 (lr=4.3357e-05) (hash(x)=13569316)
132
+ 124 train 9.972714 (lr=4.3706e-05) (hash(x)=14212041)
133
+ 125 train 9.982690 (lr=4.4056e-05) (hash(x)=11456278)
134
+ 126 train 9.818647 (lr=4.4406e-05) (hash(x)=13563812)
135
+ 127 train 9.893407 (lr=4.4755e-05) (hash(x)=14043713)
136
+ 128 train 9.916671 (lr=4.5105e-05) (hash(x)=13968786)
137
+ 129 train 9.941666 (lr=4.5455e-05) (hash(x)=11896948)
138
+ 130 train 9.911085 (lr=4.5804e-05) (hash(x)=13298324)
139
+ 131 train 9.903209 (lr=4.6154e-05) (hash(x)=11969650)
140
+ 132 train 9.800587 (lr=4.6503e-05) (hash(x)=12021926)
141
+ 133 train 9.785511 (lr=4.6853e-05) (hash(x)=14387351)
142
+ 134 train 9.791674 (lr=4.7203e-05) (hash(x)=13939547)
143
+ 135 train 9.548393 (lr=4.7552e-05) (hash(x)=13680255)
144
+ 136 train 9.836971 (lr=4.7902e-05) (hash(x)=10982564)
145
+ 137 train 9.907682 (lr=4.8252e-05) (hash(x)=11875534)
146
+ 138 train 9.905413 (lr=4.8601e-05) (hash(x)=13390260)
147
+ 139 train 9.858871 (lr=4.8951e-05) (hash(x)=13871675)
148
+ 140 train 9.728771 (lr=4.9301e-05) (hash(x)=10728444)
149
+ 141 train 9.741235 (lr=4.9650e-05) (hash(x)=13389922)
150
+ 142 train 9.777170 (lr=5.0000e-05) (hash(x)=13458742)
151
+ 143 train 9.768027 (lr=5.0350e-05) (hash(x)=12263431)
152
+ 144 train 9.828722 (lr=5.0699e-05) (hash(x)=12377887)
153
+ 145 train 9.769683 (lr=5.1049e-05) (hash(x)=10141365)
154
+ 146 train 9.740622 (lr=5.1399e-05) (hash(x)=13965689)
155
+ 147 train 9.728474 (lr=5.1748e-05) (hash(x)=12414778)
156
+ 148 train 9.673133 (lr=5.2098e-05) (hash(x)=9727144)
157
+ 149 train 9.691545 (lr=5.2448e-05) (hash(x)=11738134)
158
+ 150 val loss 9.6812
159
+ 150 val perplexity 16014.1143
160
+ 150 train 9.937815 (lr=5.2797e-05) (hash(x)=14924380)
161
+ 151 train 9.645919 (lr=5.3147e-05) (hash(x)=14443665)
162
+ 152 train 9.723100 (lr=5.3497e-05) (hash(x)=12873827)
163
+ 153 train 9.655440 (lr=5.3846e-05) (hash(x)=13435160)
164
+ 154 train 9.705055 (lr=5.4196e-05) (hash(x)=13024409)
165
+ 155 train 9.772352 (lr=5.4545e-05) (hash(x)=13100449)
166
+ 156 train 9.668520 (lr=5.4895e-05) (hash(x)=13304906)
167
+ 157 train 9.558097 (lr=5.5245e-05) (hash(x)=11690244)
168
+ 158 train 9.650925 (lr=5.5594e-05) (hash(x)=12027838)
169
+ 159 train 9.569417 (lr=5.5944e-05) (hash(x)=13953073)
170
+ 160 train 9.516293 (lr=5.6294e-05) (hash(x)=12066345)
171
+ 161 train 9.479020 (lr=5.6643e-05) (hash(x)=12657159)
172
+ 162 train 9.501439 (lr=5.6993e-05) (hash(x)=11335971)
173
+ 163 train 9.593476 (lr=5.7343e-05) (hash(x)=10956041)
174
+ 164 train 9.568235 (lr=5.7692e-05) (hash(x)=9766608)
175
+ 165 train 9.520689 (lr=5.8042e-05) (hash(x)=11060230)
176
+ 166 train 9.604611 (lr=5.8392e-05) (hash(x)=11396268)
177
+ 167 train 9.560225 (lr=5.8741e-05) (hash(x)=12840879)
178
+ 168 train 9.541338 (lr=5.9091e-05) (hash(x)=10987186)
179
+ 169 train 9.538950 (lr=5.9441e-05) (hash(x)=8916599)
180
+ 170 train 9.530159 (lr=5.9790e-05) (hash(x)=13235433)
181
+ 171 train 9.420441 (lr=6.0140e-05) (hash(x)=14428050)
182
+ 172 train 9.415767 (lr=6.0490e-05) (hash(x)=12631329)
183
+ 173 train 9.446746 (lr=6.0839e-05) (hash(x)=12787108)
184
+ 174 train 9.409441 (lr=6.1189e-05) (hash(x)=14499940)
185
+ 175 train 9.463571 (lr=6.1538e-05) (hash(x)=14104576)
186
+ 176 train 9.256237 (lr=6.1888e-05) (hash(x)=13943168)
187
+ 177 train 9.359769 (lr=6.2238e-05) (hash(x)=12623743)
188
+ 178 train 9.525467 (lr=6.2587e-05) (hash(x)=12689741)
189
+ 179 train 9.416092 (lr=6.2937e-05) (hash(x)=14453314)
190
+ 180 train 9.285929 (lr=6.3287e-05) (hash(x)=10824391)
191
+ 181 train 9.239524 (lr=6.3636e-05) (hash(x)=10470424)
192
+ 182 train 9.388404 (lr=6.3986e-05) (hash(x)=15186277)
193
+ 183 train 9.296072 (lr=6.4336e-05) (hash(x)=12918172)
194
+ 184 train 9.286754 (lr=6.4685e-05) (hash(x)=11936107)
195
+ 185 train 9.318821 (lr=6.5035e-05) (hash(x)=13259459)
196
+ 186 train 9.230548 (lr=6.5385e-05) (hash(x)=12918385)
197
+ 187 train 9.423674 (lr=6.5734e-05) (hash(x)=12368572)
198
+ 188 train 9.334161 (lr=6.6084e-05) (hash(x)=13455578)
199
+ 189 train 9.218647 (lr=6.6434e-05) (hash(x)=13505144)
200
+ 190 train 9.266394 (lr=6.6783e-05) (hash(x)=13370946)
201
+ 191 train 9.325062 (lr=6.7133e-05) (hash(x)=13340443)
202
+ 192 train 9.322151 (lr=6.7483e-05) (hash(x)=12806141)
203
+ 193 train 9.254149 (lr=6.7832e-05) (hash(x)=11469900)
204
+ 194 train 9.162863 (lr=6.8182e-05) (hash(x)=11591533)
205
+ 195 train 9.194975 (lr=6.8531e-05) (hash(x)=10742101)
206
+ 196 train 9.181164 (lr=6.8881e-05) (hash(x)=11781196)
207
+ 197 train 9.119824 (lr=6.9231e-05) (hash(x)=13261517)
208
+ 198 train 9.184980 (lr=6.9580e-05) (hash(x)=13592858)
209
+ 199 train 9.250280 (lr=6.9930e-05) (hash(x)=13025742)
210
+ 200 val loss 9.1533
211
+ 200 val perplexity 9445.7783
212
+ 200 train 9.128145 (lr=7.0280e-05) (hash(x)=12152419)
213
+ 201 train 9.359866 (lr=7.0629e-05) (hash(x)=15100390)
214
+ 202 train 9.166225 (lr=7.0979e-05) (hash(x)=10708197)
215
+ 203 train 9.131125 (lr=7.1329e-05) (hash(x)=14351068)
216
+ 204 train 9.210790 (lr=7.1678e-05) (hash(x)=12619936)
217
+ 205 train 9.048228 (lr=7.2028e-05) (hash(x)=13445545)
218
+ 206 train 9.086419 (lr=7.2378e-05) (hash(x)=11375930)
219
+ 207 train 9.123268 (lr=7.2727e-05) (hash(x)=13155663)
220
+ 208 train 9.034597 (lr=7.3077e-05) (hash(x)=13404244)
221
+ 209 train 9.148206 (lr=7.3427e-05) (hash(x)=13309898)
222
+ 210 train 9.095961 (lr=7.3776e-05) (hash(x)=12356526)
223
+ 211 train 9.004326 (lr=7.4126e-05) (hash(x)=9157946)
224
+ 212 train 9.047952 (lr=7.4476e-05) (hash(x)=11837703)
225
+ 213 train 8.920234 (lr=7.4825e-05) (hash(x)=12222186)
226
+ 214 train 9.022038 (lr=7.5175e-05) (hash(x)=12588150)
227
+ 215 train 9.032128 (lr=7.5524e-05) (hash(x)=13872294)
228
+ 216 train 8.879189 (lr=7.5874e-05) (hash(x)=10621792)
229
+ 217 train 8.999229 (lr=7.6224e-05) (hash(x)=11824400)
230
+ 218 train 8.916900 (lr=7.6573e-05) (hash(x)=12495032)
231
+ 219 train 8.963452 (lr=7.6923e-05) (hash(x)=12773612)
232
+ 220 train 8.998839 (lr=7.7273e-05) (hash(x)=10808441)
233
+ 221 train 8.946824 (lr=7.7622e-05) (hash(x)=12794078)
234
+ 222 train 8.854134 (lr=7.7972e-05) (hash(x)=11328897)
235
+ 223 train 8.926605 (lr=7.8322e-05) (hash(x)=12235713)
236
+ 224 train 8.886055 (lr=7.8671e-05) (hash(x)=11512364)
237
+ 225 train 8.876389 (lr=7.9021e-05) (hash(x)=10270693)
238
+ 226 train 8.768729 (lr=7.9371e-05) (hash(x)=12380104)
239
+ 227 train 8.856783 (lr=7.9720e-05) (hash(x)=13686441)
240
+ 228 train 9.082914 (lr=8.0070e-05) (hash(x)=14151557)
241
+ 229 train 8.964374 (lr=8.0420e-05) (hash(x)=13567687)
242
+ 230 train 8.883879 (lr=8.0769e-05) (hash(x)=12718282)
243
+ 231 train 8.832568 (lr=8.1119e-05) (hash(x)=13214962)
244
+ 232 train 8.844346 (lr=8.1469e-05) (hash(x)=10855282)
245
+ 233 train 8.870569 (lr=8.1818e-05) (hash(x)=12404167)
246
+ 234 train 8.878452 (lr=8.2168e-05) (hash(x)=13017256)
247
+ 235 train 8.762455 (lr=8.2517e-05) (hash(x)=12602681)
248
+ 236 train 8.828246 (lr=8.2867e-05) (hash(x)=11916856)
249
+ 237 train 8.841607 (lr=8.3217e-05) (hash(x)=13850551)
250
+ 238 train 8.693924 (lr=8.3566e-05) (hash(x)=11382479)
251
+ 239 train 8.745952 (lr=8.3916e-05) (hash(x)=13829219)
252
+ 240 train 8.892559 (lr=8.4266e-05) (hash(x)=13540459)
253
+ 241 train 8.720128 (lr=8.4615e-05) (hash(x)=15485762)
254
+ 242 train 8.780096 (lr=8.4965e-05) (hash(x)=12268519)
255
+ 243 train 8.660652 (lr=8.5315e-05) (hash(x)=9335220)
256
+ 244 train 8.753222 (lr=8.5664e-05) (hash(x)=13359270)
257
+ 245 train 8.667784 (lr=8.6014e-05) (hash(x)=13162637)
258
+ 246 train 8.744490 (lr=8.6364e-05) (hash(x)=13305408)
259
+ 247 train 8.658298 (lr=8.6713e-05) (hash(x)=15713183)
260
+ 248 train 8.683909 (lr=8.7063e-05) (hash(x)=13536419)
261
+ 249 train 8.657625 (lr=8.7413e-05) (hash(x)=13326615)
262
+ 250 val loss 8.6421
263
+ 250 val perplexity 5665.1299
264
+ 250 train 8.654799 (lr=8.7762e-05) (hash(x)=12112694)
265
+ 251 train 8.727888 (lr=8.8112e-05) (hash(x)=9917297)
266
+ 252 train 8.717404 (lr=8.8462e-05) (hash(x)=12656639)
267
+ 253 train 8.374126 (lr=8.8811e-05) (hash(x)=6829764)
268
+ 254 train 8.550731 (lr=8.9161e-05) (hash(x)=12964320)
269
+ 255 train 8.555574 (lr=8.9510e-05) (hash(x)=12688412)
270
+ 256 train 8.594481 (lr=8.9860e-05) (hash(x)=14906439)
271
+ 257 train 8.283122 (lr=9.0210e-05) (hash(x)=13039934)
272
+ 258 train 8.597650 (lr=9.0559e-05) (hash(x)=12020357)
273
+ 259 train 8.635440 (lr=9.0909e-05) (hash(x)=12108991)
274
+ 260 train 8.611170 (lr=9.1259e-05) (hash(x)=12328329)
275
+ 261 train 8.615200 (lr=9.1608e-05) (hash(x)=14740147)
276
+ 262 train 8.607046 (lr=9.1958e-05) (hash(x)=11559413)
277
+ 263 train 8.544862 (lr=9.2308e-05) (hash(x)=12152216)
278
+ 264 train 8.489405 (lr=9.2657e-05) (hash(x)=12708922)
279
+ 265 train 8.584811 (lr=9.3007e-05) (hash(x)=14469749)
280
+ 266 train 8.570525 (lr=9.3357e-05) (hash(x)=12459368)
281
+ 267 train 8.553911 (lr=9.3706e-05) (hash(x)=11421483)
282
+ 268 train 8.398282 (lr=9.4056e-05) (hash(x)=12939881)
283
+ 269 train 8.572999 (lr=9.4406e-05) (hash(x)=12362234)
284
+ 270 train 8.462978 (lr=9.4755e-05) (hash(x)=12322333)
285
+ 271 train 8.430627 (lr=9.5105e-05) (hash(x)=13448155)
286
+ 272 train 8.439815 (lr=9.5455e-05) (hash(x)=12928941)
287
+ 273 train 8.459054 (lr=9.5804e-05) (hash(x)=10971430)
288
+ 274 train 8.392042 (lr=9.6154e-05) (hash(x)=13627629)
289
+ 275 train 8.429370 (lr=9.6503e-05) (hash(x)=11905081)
290
+ 276 train 8.474653 (lr=9.6853e-05) (hash(x)=14339184)
291
+ 277 train 8.256489 (lr=9.7203e-05) (hash(x)=10917438)
292
+ 278 train 8.317690 (lr=9.7552e-05) (hash(x)=13804603)
293
+ 279 train 8.491489 (lr=9.7902e-05) (hash(x)=12290857)
294
+ 280 train 8.441209 (lr=9.8252e-05) (hash(x)=10211118)
295
+ 281 train 8.609485 (lr=9.8601e-05) (hash(x)=13407076)
296
+ 282 train 8.487457 (lr=9.8951e-05) (hash(x)=12249945)
297
+ 283 train 8.466316 (lr=9.9301e-05) (hash(x)=13234933)
298
+ 284 train 8.357010 (lr=9.9650e-05) (hash(x)=15194855)
299
+ 285 train 8.410402 (lr=1.0000e-04) (hash(x)=12892366)
300
+ 286 train 8.327589 (lr=1.0000e-04) (hash(x)=12609795)
301
+ 287 train 8.370582 (lr=1.0000e-04) (hash(x)=9768085)
302
+ 288 train 8.287266 (lr=9.9998e-05) (hash(x)=12252104)
303
+ 289 train 8.351656 (lr=9.9996e-05) (hash(x)=12692791)
304
+ 290 train 8.328091 (lr=9.9993e-05) (hash(x)=11957245)
305
+ 291 train 8.117153 (lr=9.9989e-05) (hash(x)=9528974)
306
+ 292 train 8.310163 (lr=9.9984e-05) (hash(x)=12231837)
307
+ 293 train 8.247046 (lr=9.9979e-05) (hash(x)=10032842)
308
+ 294 train 8.585519 (lr=9.9972e-05) (hash(x)=17178097)
309
+ 295 train 8.558478 (lr=9.9965e-05) (hash(x)=10720403)
310
+ 296 train 8.258804 (lr=9.9956e-05) (hash(x)=13541352)
311
+ 297 train 8.474586 (lr=9.9947e-05) (hash(x)=15187742)
312
+ 298 train 8.359024 (lr=9.9937e-05) (hash(x)=11971935)
313
+ 299 train 8.303322 (lr=9.9926e-05) (hash(x)=12874592)
314
+ 300 val loss 8.2445
315
+ 300 val perplexity 3806.7012
316
+ 300 train 8.312428 (lr=9.9915e-05) (hash(x)=12462802)
317
+ 301 train 8.306379 (lr=9.9902e-05) (hash(x)=13622545)
318
+ 302 train 8.278027 (lr=9.9889e-05) (hash(x)=14255974)
319
+ 303 train 8.285154 (lr=9.9874e-05) (hash(x)=11025404)
320
+ 304 train 8.372684 (lr=9.9859e-05) (hash(x)=11239636)
321
+ 305 train 8.238194 (lr=9.9843e-05) (hash(x)=15190123)
322
+ 306 train 8.359118 (lr=9.9826e-05) (hash(x)=11464046)
323
+ 307 train 8.336150 (lr=9.9808e-05) (hash(x)=12567779)
324
+ 308 train 8.249514 (lr=9.9789e-05) (hash(x)=16046446)
325
+ 309 train 8.215065 (lr=9.9770e-05) (hash(x)=12439519)
326
+ 310 train 8.055475 (lr=9.9749e-05) (hash(x)=11917658)
327
+ 311 train 8.262974 (lr=9.9728e-05) (hash(x)=11826091)
328
+ 312 train 8.403700 (lr=9.9706e-05) (hash(x)=10807703)
329
+ 313 train 8.210493 (lr=9.9683e-05) (hash(x)=14723866)
330
+ 314 train 8.316089 (lr=9.9659e-05) (hash(x)=11315230)
331
+ 315 train 8.734059 (lr=9.9634e-05) (hash(x)=11259652)
332
+ 316 train 8.689662 (lr=9.9609e-05) (hash(x)=11702582)
333
+ 317 train 8.395004 (lr=9.9582e-05) (hash(x)=11135951)
334
+ 318 train 8.281982 (lr=9.9555e-05) (hash(x)=11570885)
335
+ 319 train 8.054947 (lr=9.9526e-05) (hash(x)=14642293)
336
+ 320 train 7.956946 (lr=9.9497e-05) (hash(x)=13037070)
337
+ 321 train 8.109127 (lr=9.9467e-05) (hash(x)=12485426)
338
+ 322 train 8.172827 (lr=9.9437e-05) (hash(x)=15142358)
339
+ 323 train 8.034015 (lr=9.9405e-05) (hash(x)=12923132)
340
+ 324 train 8.233083 (lr=9.9372e-05) (hash(x)=13055942)
341
+ 325 train 8.147958 (lr=9.9339e-05) (hash(x)=13559280)
342
+ 326 train 8.122777 (lr=9.9305e-05) (hash(x)=15313622)
343
+ 327 train 8.099021 (lr=9.9270e-05) (hash(x)=12169964)
344
+ 328 train 7.780449 (lr=9.9234e-05) (hash(x)=10902629)
345
+ 329 train 7.837113 (lr=9.9197e-05) (hash(x)=12000182)
346
+ 330 train 7.979561 (lr=9.9159e-05) (hash(x)=11345877)
347
+ 331 train 8.069979 (lr=9.9121e-05) (hash(x)=10948640)
348
+ 332 train 7.955867 (lr=9.9081e-05) (hash(x)=13037309)
349
+ 333 train 7.890843 (lr=9.9041e-05) (hash(x)=9666433)
350
+ 334 train 8.022075 (lr=9.9000e-05) (hash(x)=14049895)
351
+ 335 train 8.122494 (lr=9.8958e-05) (hash(x)=14103013)
352
+ 336 train 8.208740 (lr=9.8915e-05) (hash(x)=14058870)
353
+ 337 train 8.148366 (lr=9.8872e-05) (hash(x)=13553506)
354
+ 338 train 8.107164 (lr=9.8827e-05) (hash(x)=12902075)
355
+ 339 train 7.982985 (lr=9.8782e-05) (hash(x)=11135042)
356
+ 340 train 8.263373 (lr=9.8736e-05) (hash(x)=13650023)
357
+ 341 train 8.018551 (lr=9.8689e-05) (hash(x)=14930403)
358
+ 342 train 8.115504 (lr=9.8641e-05) (hash(x)=15485935)
359
+ 343 train 8.335106 (lr=9.8592e-05) (hash(x)=12404200)
360
+ 344 train 7.997009 (lr=9.8543e-05) (hash(x)=12392953)
361
+ 345 train 8.023292 (lr=9.8492e-05) (hash(x)=12280314)
362
+ 346 train 7.974169 (lr=9.8441e-05) (hash(x)=10900559)
363
+ 347 train 8.020061 (lr=9.8389e-05) (hash(x)=11534055)
364
+ 348 train 8.010694 (lr=9.8336e-05) (hash(x)=11997160)
365
+ 349 train 8.066229 (lr=9.8282e-05) (hash(x)=11308531)
366
+ 350 val loss 7.9918
367
+ 350 val perplexity 2956.4854
368
+ 350 train 8.044196 (lr=9.8228e-05) (hash(x)=12759751)
369
+ 351 train 7.994446 (lr=9.8172e-05) (hash(x)=10914085)
370
+ 352 train 8.145109 (lr=9.8116e-05) (hash(x)=12776027)
371
+ 353 train 7.934609 (lr=9.8059e-05) (hash(x)=12357751)
372
+ 354 train 7.963771 (lr=9.8001e-05) (hash(x)=11040798)
373
+ 355 train 8.186275 (lr=9.7942e-05) (hash(x)=14570713)
374
+ 356 train 8.134956 (lr=9.7882e-05) (hash(x)=11492564)
375
+ 357 train 7.998983 (lr=9.7822e-05) (hash(x)=12212797)
376
+ 358 train 8.036196 (lr=9.7761e-05) (hash(x)=13496879)
377
+ 359 train 8.057919 (lr=9.7699e-05) (hash(x)=11680329)
378
+ 360 train 7.917630 (lr=9.7636e-05) (hash(x)=11948575)
379
+ 361 train 7.978311 (lr=9.7572e-05) (hash(x)=13116772)
380
+ 362 train 8.157778 (lr=9.7507e-05) (hash(x)=15527273)
381
+ 363 train 7.760407 (lr=9.7442e-05) (hash(x)=12437096)
382
+ 364 train 8.058022 (lr=9.7376e-05) (hash(x)=12518287)
383
+ 365 train 7.998032 (lr=9.7309e-05) (hash(x)=11220338)
384
+ 366 train 7.820887 (lr=9.7241e-05) (hash(x)=12034192)
385
+ 367 train 8.181369 (lr=9.7172e-05) (hash(x)=12654843)
386
+ 368 train 8.051016 (lr=9.7103e-05) (hash(x)=16116174)
387
+ 369 train 7.986964 (lr=9.7032e-05) (hash(x)=13505076)
388
+ 370 train 7.981853 (lr=9.6961e-05) (hash(x)=11820607)
389
+ 371 train 8.049170 (lr=9.6889e-05) (hash(x)=10606850)
390
+ 372 train 7.887764 (lr=9.6817e-05) (hash(x)=13759635)
391
+ 373 train 7.942155 (lr=9.6743e-05) (hash(x)=15357227)
392
+ 374 train 7.930596 (lr=9.6669e-05) (hash(x)=11597062)
393
+ 375 train 7.926111 (lr=9.6593e-05) (hash(x)=11281537)
394
+ 376 train 7.645357 (lr=9.6518e-05) (hash(x)=13294828)
395
+ 377 train 8.358434 (lr=9.6441e-05) (hash(x)=13253249)
396
+ 378 train 8.031467 (lr=9.6363e-05) (hash(x)=16101250)
397
+ 379 train 8.167181 (lr=9.6285e-05) (hash(x)=12363856)
398
+ 380 train 7.969309 (lr=9.6206e-05) (hash(x)=12206540)
399
+ 381 train 8.106320 (lr=9.6126e-05) (hash(x)=13580623)
400
+ 382 train 8.045162 (lr=9.6045e-05) (hash(x)=12509602)
401
+ 383 train 8.115283 (lr=9.5963e-05) (hash(x)=14001265)
402
+ 384 train 7.875936 (lr=9.5881e-05) (hash(x)=14290431)
403
+ 385 train 7.858977 (lr=9.5798e-05) (hash(x)=13045278)
404
+ 386 train 7.944960 (lr=9.5714e-05) (hash(x)=13371383)
405
+ 387 train 7.959494 (lr=9.5629e-05) (hash(x)=10444420)
406
+ 388 train 7.794241 (lr=9.5544e-05) (hash(x)=12732892)
407
+ 389 train 7.979726 (lr=9.5457e-05) (hash(x)=11832964)
408
+ 390 train 8.000227 (lr=9.5370e-05) (hash(x)=11477958)
409
+ 391 train 7.875288 (lr=9.5282e-05) (hash(x)=12249732)
410
+ 392 train 7.930131 (lr=9.5194e-05) (hash(x)=12132933)
411
+ 393 train 7.912541 (lr=9.5104e-05) (hash(x)=12307821)
412
+ 394 train 7.528294 (lr=9.5014e-05) (hash(x)=11266264)
413
+ 395 train 7.838082 (lr=9.4923e-05) (hash(x)=13803785)
414
+ 396 train 7.891228 (lr=9.4831e-05) (hash(x)=13256773)
415
+ 397 train 7.783422 (lr=9.4739e-05) (hash(x)=12493316)
416
+ 398 train 7.782268 (lr=9.4646e-05) (hash(x)=12991346)
417
+ 399 train 7.894723 (lr=9.4551e-05) (hash(x)=11689978)
418
+ 400 val loss 7.8191
419
+ 400 val perplexity 2487.7292
420
+ 400 train 7.810500 (lr=9.4457e-05) (hash(x)=12056273)
421
+ 401 train 7.893260 (lr=9.4361e-05) (hash(x)=11906763)
422
+ 402 train 7.897350 (lr=9.4265e-05) (hash(x)=12621559)
423
+ 403 train 7.886669 (lr=9.4168e-05) (hash(x)=13865869)
424
+ 404 train 7.896888 (lr=9.4070e-05) (hash(x)=11425874)
425
+ 405 train 8.157940 (lr=9.3971e-05) (hash(x)=16513375)
426
+ 406 train 8.089321 (lr=9.3872e-05) (hash(x)=14492920)
427
+ 407 train 7.961076 (lr=9.3772e-05) (hash(x)=13030618)
428
+ 408 train 7.811200 (lr=9.3671e-05) (hash(x)=12816692)
429
+ 409 train 8.036970 (lr=9.3569e-05) (hash(x)=13701949)
430
+ 410 train 7.846581 (lr=9.3467e-05) (hash(x)=12904618)
431
+ 411 train 7.955222 (lr=9.3364e-05) (hash(x)=14299833)
432
+ 412 train 7.822762 (lr=9.3260e-05) (hash(x)=12586511)
433
+ 413 train 7.842928 (lr=9.3155e-05) (hash(x)=11818854)
434
+ 414 train 7.887273 (lr=9.3050e-05) (hash(x)=13063115)
435
+ 415 train 7.775590 (lr=9.2944e-05) (hash(x)=12908875)
436
+ 416 train 7.960326 (lr=9.2837e-05) (hash(x)=12627550)
437
+ 417 train 7.771889 (lr=9.2729e-05) (hash(x)=12634656)
438
+ 418 train 7.820056 (lr=9.2621e-05) (hash(x)=13466941)
439
+ 419 train 8.015818 (lr=9.2512e-05) (hash(x)=13859304)
440
+ 420 train 7.958475 (lr=9.2402e-05) (hash(x)=9637450)
441
+ 421 train 7.878290 (lr=9.2292e-05) (hash(x)=14940204)
442
+ 422 train 7.806941 (lr=9.2181e-05) (hash(x)=13686717)
443
+ 423 train 7.882282 (lr=9.2069e-05) (hash(x)=14396964)
444
+ 424 train 7.560163 (lr=9.1956e-05) (hash(x)=12949082)
445
+ 425 train 7.748456 (lr=9.1843e-05) (hash(x)=13577749)
446
+ 426 train 7.897635 (lr=9.1729e-05) (hash(x)=13031728)
447
+ 427 train 7.906124 (lr=9.1614e-05) (hash(x)=11269801)
448
+ 428 train 7.694422 (lr=9.1499e-05) (hash(x)=11106556)
449
+ 429 train 7.606047 (lr=9.1382e-05) (hash(x)=11222906)
450
+ 430 train 7.864244 (lr=9.1266e-05) (hash(x)=14052269)
451
+ 431 train 7.661767 (lr=9.1148e-05) (hash(x)=14486071)
452
+ 432 train 7.799129 (lr=9.1030e-05) (hash(x)=13026662)
453
+ 433 train 7.799716 (lr=9.0911e-05) (hash(x)=11065744)
454
+ 434 train 7.848170 (lr=9.0791e-05) (hash(x)=13318947)
455
+ 435 train 7.855151 (lr=9.0671e-05) (hash(x)=11997361)
456
+ 436 train 7.868963 (lr=9.0550e-05) (hash(x)=12437178)
457
+ 437 train 7.727204 (lr=9.0428e-05) (hash(x)=12705537)
458
+ 438 train 7.882643 (lr=9.0306e-05) (hash(x)=15790088)
459
+ 439 train 7.680350 (lr=9.0182e-05) (hash(x)=11010336)
460
+ 440 train 7.876401 (lr=9.0059e-05) (hash(x)=11401216)
461
+ 441 train 7.797796 (lr=8.9934e-05) (hash(x)=15106564)
462
+ 442 train 7.788083 (lr=8.9809e-05) (hash(x)=13019227)
463
+ 443 train 7.793926 (lr=8.9683e-05) (hash(x)=10864248)
464
+ 444 train 7.709727 (lr=8.9557e-05) (hash(x)=12107995)
465
+ 445 train 7.592257 (lr=8.9430e-05) (hash(x)=13284954)
466
+ 446 train 7.844769 (lr=8.9302e-05) (hash(x)=12430135)
467
+ 447 train 7.783175 (lr=8.9173e-05) (hash(x)=11906151)
468
+ 448 train 7.613859 (lr=8.9044e-05) (hash(x)=11697970)
469
+ 449 train 7.608834 (lr=8.8914e-05) (hash(x)=8629295)
470
+ 450 val loss 7.7062
471
+ 450 val perplexity 2222.1143
472
+ 450 train 7.755933 (lr=8.8784e-05) (hash(x)=13977261)
473
+ 451 train 7.785362 (lr=8.8653e-05) (hash(x)=15166897)
474
+ 452 train 7.836470 (lr=8.8521e-05) (hash(x)=13444474)
475
+ 453 train 7.880404 (lr=8.8388e-05) (hash(x)=13108020)
476
+ 454 train 7.701727 (lr=8.8255e-05) (hash(x)=12559086)
477
+ 455 train 7.639854 (lr=8.8122e-05) (hash(x)=12338963)
478
+ 456 train 7.553016 (lr=8.7987e-05) (hash(x)=10411127)
479
+ 457 train 7.871631 (lr=8.7852e-05) (hash(x)=17176736)
480
+ 458 train 7.672147 (lr=8.7717e-05) (hash(x)=13586859)
481
+ 459 train 7.804595 (lr=8.7580e-05) (hash(x)=13235312)
482
+ 460 train 7.661676 (lr=8.7444e-05) (hash(x)=10862031)
483
+ 461 train 7.789411 (lr=8.7306e-05) (hash(x)=12498316)
484
+ 462 train 7.776201 (lr=8.7168e-05) (hash(x)=15109534)
485
+ 463 train 7.602757 (lr=8.7029e-05) (hash(x)=11031343)
486
+ 464 train 7.676269 (lr=8.6890e-05) (hash(x)=13030101)
487
+ 465 train 7.646843 (lr=8.6750e-05) (hash(x)=12787203)
488
+ 466 train 7.778650 (lr=8.6609e-05) (hash(x)=13192715)
489
+ 467 train 7.743580 (lr=8.6468e-05) (hash(x)=14597757)
490
+ 468 train 7.721496 (lr=8.6326e-05) (hash(x)=11548718)
491
+ 469 train 7.728604 (lr=8.6184e-05) (hash(x)=12017158)
492
+ 470 train 7.762467 (lr=8.6041e-05) (hash(x)=10164479)
493
+ 471 train 7.631666 (lr=8.5897e-05) (hash(x)=13571036)
494
+ 472 train 7.674098 (lr=8.5753e-05) (hash(x)=11609529)
495
+ 473 train 7.754757 (lr=8.5608e-05) (hash(x)=12289875)
496
+ 474 train 7.823117 (lr=8.5462e-05) (hash(x)=13086437)
497
+ 475 train 7.947239 (lr=8.5316e-05) (hash(x)=13225100)
498
+ 476 train 7.592534 (lr=8.5170e-05) (hash(x)=12863127)
499
+ 477 train 7.703948 (lr=8.5022e-05) (hash(x)=11162887)
500
+ 478 train 7.827986 (lr=8.4875e-05) (hash(x)=12364743)
501
+ 479 train 7.606346 (lr=8.4726e-05) (hash(x)=12281916)
502
+ 480 train 7.642028 (lr=8.4577e-05) (hash(x)=12249214)
503
+ 481 train 7.757352 (lr=8.4428e-05) (hash(x)=12348477)
504
+ 482 train 7.766273 (lr=8.4278e-05) (hash(x)=12468719)
505
+ 483 train 7.776676 (lr=8.4127e-05) (hash(x)=13634084)
506
+ 484 train 7.732769 (lr=8.3976e-05) (hash(x)=11740722)
507
+ 485 train 7.820585 (lr=8.3824e-05) (hash(x)=11378506)
508
+ 486 train 7.595967 (lr=8.3672e-05) (hash(x)=13417634)
509
+ 487 train 7.762051 (lr=8.3519e-05) (hash(x)=11801649)
510
+ 488 train 7.748933 (lr=8.3366e-05) (hash(x)=10917325)
511
+ 489 train 7.547833 (lr=8.3212e-05) (hash(x)=10552788)
512
+ 490 train 7.337061 (lr=8.3057e-05) (hash(x)=9421716)
513
+ 491 train 7.679709 (lr=8.2902e-05) (hash(x)=11746289)
514
+ 492 train 7.697298 (lr=8.2746e-05) (hash(x)=9589864)
515
+ 493 train 7.647312 (lr=8.2590e-05) (hash(x)=13826293)
516
+ 494 train 7.831108 (lr=8.2434e-05) (hash(x)=12720603)
517
+ 495 train 7.698438 (lr=8.2276e-05) (hash(x)=10783313)
518
+ 496 train 7.733990 (lr=8.2119e-05) (hash(x)=13077539)
519
+ 497 train 7.764796 (lr=8.1960e-05) (hash(x)=12913270)
520
+ 498 train 7.724287 (lr=8.1801e-05) (hash(x)=11228160)
521
+ 499 train 7.653799 (lr=8.1642e-05) (hash(x)=14212467)
522
+ 500 val loss 7.6640
523
+ 500 val perplexity 2130.2698
524
+ 500 train 7.679778 (lr=8.1482e-05) (hash(x)=12714536)
525
+ 501 train 7.602472 (lr=8.1322e-05) (hash(x)=11495162)
526
+ 502 train 7.732669 (lr=8.1161e-05) (hash(x)=13571819)
527
+ 503 train 7.748485 (lr=8.1000e-05) (hash(x)=12994231)
528
+ 504 train 7.684674 (lr=8.0838e-05) (hash(x)=12786058)
529
+ 505 train 7.910808 (lr=8.0676e-05) (hash(x)=15319532)
530
+ 506 train 7.828122 (lr=8.0513e-05) (hash(x)=12432415)
531
+ 507 train 7.677938 (lr=8.0349e-05) (hash(x)=11485137)
532
+ 508 train 7.784925 (lr=8.0186e-05) (hash(x)=13106040)
533
+ 509 train 7.752947 (lr=8.0021e-05) (hash(x)=14010329)
534
+ 510 train 7.748173 (lr=7.9856e-05) (hash(x)=14627424)
535
+ 511 train 7.599130 (lr=7.9691e-05) (hash(x)=10470487)
536
+ 512 train 7.821552 (lr=7.9525e-05) (hash(x)=12735976)
537
+ 513 train 7.873974 (lr=7.9359e-05) (hash(x)=14077502)
538
+ 514 train 7.730788 (lr=7.9192e-05) (hash(x)=13758986)
539
+ 515 train 7.752924 (lr=7.9025e-05) (hash(x)=12205695)
540
+ 516 train 7.658122 (lr=7.8858e-05) (hash(x)=12101077)
541
+ 517 train 7.547831 (lr=7.8689e-05) (hash(x)=12096952)
542
+ 518 train 7.899580 (lr=7.8521e-05) (hash(x)=10823414)
543
+ 519 train 7.686187 (lr=7.8352e-05) (hash(x)=13985657)
544
+ 520 train 7.692110 (lr=7.8182e-05) (hash(x)=13657835)
545
+ 521 train 7.732400 (lr=7.8012e-05) (hash(x)=14868589)
546
+ 522 train 7.893287 (lr=7.7842e-05) (hash(x)=12440659)
547
+ 523 train 7.732920 (lr=7.7671e-05) (hash(x)=12944090)
548
+ 524 train 7.693241 (lr=7.7500e-05) (hash(x)=14152385)
549
+ 525 train 7.637668 (lr=7.7328e-05) (hash(x)=14652164)
550
+ 526 train 7.775796 (lr=7.7156e-05) (hash(x)=11671931)
551
+ 527 train 7.616189 (lr=7.6984e-05) (hash(x)=12831661)
552
+ 528 train 7.829365 (lr=7.6811e-05) (hash(x)=13612890)
553
+ 529 train 7.589914 (lr=7.6637e-05) (hash(x)=13472998)
554
+ 530 train 7.869791 (lr=7.6463e-05) (hash(x)=12912339)
555
+ 531 train 7.665202 (lr=7.6289e-05) (hash(x)=12806095)
556
+ 532 train 7.520229 (lr=7.6115e-05) (hash(x)=10738555)
557
+ 533 train 7.533718 (lr=7.5940e-05) (hash(x)=13410818)
558
+ 534 train 7.863018 (lr=7.5764e-05) (hash(x)=13997116)
559
+ 535 train 7.666398 (lr=7.5588e-05) (hash(x)=14617414)
560
+ 536 train 7.817972 (lr=7.5412e-05) (hash(x)=12788088)
561
+ 537 train 7.755359 (lr=7.5235e-05) (hash(x)=15066658)
562
+ 538 train 7.714177 (lr=7.5058e-05) (hash(x)=13331118)
563
+ 539 train 7.711161 (lr=7.4881e-05) (hash(x)=13221393)
564
+ 540 train 7.697598 (lr=7.4703e-05) (hash(x)=13355413)
565
+ 541 train 7.599565 (lr=7.4525e-05) (hash(x)=11528563)
566
+ 542 train 7.767950 (lr=7.4346e-05) (hash(x)=15327421)
567
+ 543 train 7.848365 (lr=7.4167e-05) (hash(x)=11041023)
568
+ 544 train 7.559992 (lr=7.3988e-05) (hash(x)=13600125)
569
+ 545 train 7.716181 (lr=7.3808e-05) (hash(x)=14885399)
570
+ 546 train 7.625460 (lr=7.3628e-05) (hash(x)=14786577)
571
+ 547 train 7.751172 (lr=7.3448e-05) (hash(x)=11362709)
572
+ 548 train 7.665632 (lr=7.3267e-05) (hash(x)=13426228)
573
+ 549 train 7.614367 (lr=7.3086e-05) (hash(x)=13083412)
574
+ 550 val loss 7.6139
575
+ 550 val perplexity 2026.1960
576
+ 550 train 7.645340 (lr=7.2904e-05) (hash(x)=13873840)
577
+ 551 train 7.626501 (lr=7.2723e-05) (hash(x)=13186073)
578
+ 552 train 7.727996 (lr=7.2540e-05) (hash(x)=13488437)
579
+ 553 train 7.679361 (lr=7.2358e-05) (hash(x)=13988224)
580
+ 554 train 7.678811 (lr=7.2175e-05) (hash(x)=11786791)
581
+ 555 train 8.419575 (lr=7.1992e-05) (hash(x)=13274177)
582
+ 556 train 7.785417 (lr=7.1808e-05) (hash(x)=10204377)
583
+ 557 train 7.715769 (lr=7.1624e-05) (hash(x)=12363265)
584
+ 558 train 7.494042 (lr=7.1440e-05) (hash(x)=12918001)
585
+ 559 train 7.478938 (lr=7.1256e-05) (hash(x)=13507868)
586
+ 560 train 7.751859 (lr=7.1071e-05) (hash(x)=10580307)
587
+ 561 train 7.578580 (lr=7.0886e-05) (hash(x)=12531978)
588
+ 562 train 7.547606 (lr=7.0701e-05) (hash(x)=11570432)
589
+ 563 train 7.691416 (lr=7.0515e-05) (hash(x)=12512404)
590
+ 564 train 7.722275 (lr=7.0329e-05) (hash(x)=12011112)
591
+ 565 train 7.763875 (lr=7.0143e-05) (hash(x)=13439040)
592
+ 566 train 7.389637 (lr=6.9956e-05) (hash(x)=9321383)
593
+ 567 train 7.646101 (lr=6.9769e-05) (hash(x)=11221721)
594
+ 568 train 7.507978 (lr=6.9582e-05) (hash(x)=12048885)
595
+ 569 train 7.720916 (lr=6.9394e-05) (hash(x)=13854047)
596
+ 570 train 7.895522 (lr=6.9207e-05) (hash(x)=13041387)
597
+ 571 train 8.078491 (lr=6.9019e-05) (hash(x)=15372354)
598
+ 572 train 7.797123 (lr=6.8830e-05) (hash(x)=11895595)
599
+ 573 train 7.583886 (lr=6.8642e-05) (hash(x)=13070460)
600
+ 574 train 7.630826 (lr=6.8453e-05) (hash(x)=11507165)
601
+ 575 train 7.936076 (lr=6.8264e-05) (hash(x)=14689778)
602
+ 576 train 7.412996 (lr=6.8075e-05) (hash(x)=10149008)
603
+ 577 train 7.824008 (lr=6.7885e-05) (hash(x)=15364407)
604
+ 578 train 7.584634 (lr=6.7695e-05) (hash(x)=11676391)
605
+ 579 train 7.581218 (lr=6.7505e-05) (hash(x)=13156509)
606
+ 580 train 7.699704 (lr=6.7315e-05) (hash(x)=12911319)
607
+ 581 train 7.464559 (lr=6.7124e-05) (hash(x)=13358322)
608
+ 582 train 7.519526 (lr=6.6933e-05) (hash(x)=11462782)
609
+ 583 train 7.890136 (lr=6.6742e-05) (hash(x)=15734207)
610
+ 584 train 7.565492 (lr=6.6551e-05) (hash(x)=9117677)
611
+ 585 train 7.673103 (lr=6.6360e-05) (hash(x)=12334129)
612
+ 586 train 7.703097 (lr=6.6168e-05) (hash(x)=14097844)
613
+ 587 train 7.578682 (lr=6.5976e-05) (hash(x)=13758140)
614
+ 588 train 7.909241 (lr=6.5784e-05) (hash(x)=15284545)
615
+ 589 train 7.498783 (lr=6.5592e-05) (hash(x)=11809021)
616
+ 590 train 7.671258 (lr=6.5399e-05) (hash(x)=12752849)
617
+ 591 train 7.588071 (lr=6.5206e-05) (hash(x)=12925955)
618
+ 592 train 7.696866 (lr=6.5013e-05) (hash(x)=12026092)
619
+ 593 train 7.779930 (lr=6.4820e-05) (hash(x)=15708918)
620
+ 594 train 8.015055 (lr=6.4627e-05) (hash(x)=13954437)
621
+ 595 train 7.788976 (lr=6.4433e-05) (hash(x)=12750849)
622
+ 596 train 7.576025 (lr=6.4240e-05) (hash(x)=12076716)
623
+ 597 train 7.699966 (lr=6.4046e-05) (hash(x)=12269439)
624
+ 598 train 7.638756 (lr=6.3852e-05) (hash(x)=9706332)
625
+ 599 train 7.574268 (lr=6.3658e-05) (hash(x)=14327624)
626
+ 600 val loss 7.6009
627
+ 600 val perplexity 1999.9515
628
+ 600 train 7.732218 (lr=6.3463e-05) (hash(x)=12001611)
629
+ 601 train 7.645571 (lr=6.3269e-05) (hash(x)=12664301)
630
+ 602 train 7.543394 (lr=6.3074e-05) (hash(x)=10457844)
631
+ 603 train 7.530212 (lr=6.2879e-05) (hash(x)=11819313)
632
+ 604 train 7.549575 (lr=6.2684e-05) (hash(x)=12304915)
633
+ 605 train 7.671464 (lr=6.2489e-05) (hash(x)=13250863)
634
+ 606 train 7.643473 (lr=6.2294e-05) (hash(x)=15885206)
635
+ 607 train 7.390647 (lr=6.2098e-05) (hash(x)=12559422)
636
+ 608 train 7.568728 (lr=6.1903e-05) (hash(x)=12136733)
637
+ 609 train 7.662648 (lr=6.1707e-05) (hash(x)=12183244)
638
+ 610 train 7.775548 (lr=6.1511e-05) (hash(x)=12682752)
639
+ 611 train 7.789057 (lr=6.1315e-05) (hash(x)=12816359)
640
+ 612 train 7.724688 (lr=6.1119e-05) (hash(x)=11628021)
641
+ 613 train 7.608067 (lr=6.0923e-05) (hash(x)=11408330)
642
+ 614 train 7.690547 (lr=6.0726e-05) (hash(x)=12901064)
643
+ 615 train 7.664282 (lr=6.0530e-05) (hash(x)=10162896)
644
+ 616 train 7.771249 (lr=6.0333e-05) (hash(x)=12416355)
645
+ 617 train 7.967935 (lr=6.0137e-05) (hash(x)=12904775)
646
+ 618 train 7.677318 (lr=5.9940e-05) (hash(x)=12697259)
647
+ 619 train 7.526679 (lr=5.9743e-05) (hash(x)=12041552)
648
+ 620 train 7.566912 (lr=5.9546e-05) (hash(x)=13429059)
649
+ 621 train 7.607207 (lr=5.9349e-05) (hash(x)=14176315)
650
+ 622 train 7.575527 (lr=5.9152e-05) (hash(x)=11386804)
651
+ 623 train 7.461130 (lr=5.8955e-05) (hash(x)=10256290)
652
+ 624 train 7.603558 (lr=5.8758e-05) (hash(x)=11669288)
653
+ 625 train 7.792837 (lr=5.8560e-05) (hash(x)=12287950)
654
+ 626 train 7.494392 (lr=5.8363e-05) (hash(x)=12384271)
655
+ 627 train 7.669680 (lr=5.8165e-05) (hash(x)=14312559)
656
+ 628 train 7.547470 (lr=5.7968e-05) (hash(x)=13864960)
657
+ 629 train 7.667377 (lr=5.7770e-05) (hash(x)=10078167)
658
+ 630 train 7.628721 (lr=5.7573e-05) (hash(x)=14121709)
659
+ 631 train 7.439117 (lr=5.7375e-05) (hash(x)=11443472)
660
+ 632 train 7.699893 (lr=5.7177e-05) (hash(x)=11816027)
661
+ 633 train 7.556986 (lr=5.6979e-05) (hash(x)=11630498)
662
+ 634 train 7.787597 (lr=5.6782e-05) (hash(x)=12453204)
663
+ 635 train 7.517763 (lr=5.6584e-05) (hash(x)=10480150)
664
+ 636 train 7.636683 (lr=5.6386e-05) (hash(x)=10854685)
665
+ 637 train 7.687005 (lr=5.6188e-05) (hash(x)=13365961)
666
+ 638 train 7.899031 (lr=5.5990e-05) (hash(x)=13227220)
667
+ 639 train 7.773348 (lr=5.5792e-05) (hash(x)=13927238)
668
+ 640 train 7.611150 (lr=5.5594e-05) (hash(x)=11183563)
669
+ 641 train 7.783179 (lr=5.5396e-05) (hash(x)=12661142)
670
+ 642 train 7.517184 (lr=5.5198e-05) (hash(x)=12116909)
671
+ 643 train 7.921080 (lr=5.5000e-05) (hash(x)=16448258)
672
+ 644 train 8.133081 (lr=5.4802e-05) (hash(x)=11230472)
673
+ 645 train 7.722316 (lr=5.4604e-05) (hash(x)=12689745)
674
+ 646 train 7.883445 (lr=5.4406e-05) (hash(x)=12535866)
675
+ 647 train 7.391381 (lr=5.4208e-05) (hash(x)=10701487)
676
+ 648 train 7.593962 (lr=5.4010e-05) (hash(x)=12572372)
677
+ 649 train 8.168662 (lr=5.3812e-05) (hash(x)=13938367)
678
+ 650 val loss 7.5947
679
+ 650 val perplexity 1987.6559
680
+ 650 train 7.793624 (lr=5.3614e-05) (hash(x)=15954548)
681
+ 651 train 7.903073 (lr=5.3416e-05) (hash(x)=13034568)
682
+ 652 train 7.756711 (lr=5.3218e-05) (hash(x)=17184553)
683
+ 653 train 7.908390 (lr=5.3021e-05) (hash(x)=13398831)
684
+ 654 train 8.057072 (lr=5.2823e-05) (hash(x)=15245462)
685
+ 655 train 7.856922 (lr=5.2625e-05) (hash(x)=14575913)
686
+ 656 train 7.646807 (lr=5.2427e-05) (hash(x)=11389089)
687
+ 657 train 7.684308 (lr=5.2230e-05) (hash(x)=12121790)
688
+ 658 train 7.820537 (lr=5.2032e-05) (hash(x)=11991841)
689
+ 659 train 7.367908 (lr=5.1835e-05) (hash(x)=10875163)
690
+ 660 train 7.709812 (lr=5.1637e-05) (hash(x)=13641028)
691
+ 661 train 7.779671 (lr=5.1440e-05) (hash(x)=13782256)
692
+ 662 train 8.375802 (lr=5.1242e-05) (hash(x)=16715370)
693
+ 663 train 8.014695 (lr=5.1045e-05) (hash(x)=13744329)
694
+ 664 train 7.690497 (lr=5.0848e-05) (hash(x)=12865885)
695
+ 665 train 7.785969 (lr=5.0651e-05) (hash(x)=14443216)
696
+ 666 train 7.802867 (lr=5.0454e-05) (hash(x)=10361766)
697
+ 667 train 7.392825 (lr=5.0257e-05) (hash(x)=8860388)
698
+ 668 train 7.631924 (lr=5.0060e-05) (hash(x)=12188781)
699
+ 669 train 7.607609 (lr=4.9863e-05) (hash(x)=13546558)
700
+ 670 train 7.574569 (lr=4.9667e-05) (hash(x)=14731723)
701
+ 671 train 7.739242 (lr=4.9470e-05) (hash(x)=14097210)
702
+ 672 train 8.021310 (lr=4.9274e-05) (hash(x)=13277664)
703
+ 673 train 7.453959 (lr=4.9077e-05) (hash(x)=12154683)
704
+ 674 train 7.482224 (lr=4.8881e-05) (hash(x)=14244571)
705
+ 675 train 7.744907 (lr=4.8685e-05) (hash(x)=11088127)
706
+ 676 train 7.511981 (lr=4.8489e-05) (hash(x)=10259005)
707
+ 677 train 7.978391 (lr=4.8293e-05) (hash(x)=15853795)
708
+ 678 train 7.487537 (lr=4.8097e-05) (hash(x)=12406215)
709
+ 679 train 7.574934 (lr=4.7902e-05) (hash(x)=14363085)
710
+ 680 train 7.432519 (lr=4.7706e-05) (hash(x)=12303156)
711
+ 681 train 7.612151 (lr=4.7511e-05) (hash(x)=12144036)
712
+ 682 train 7.812609 (lr=4.7316e-05) (hash(x)=14153850)
713
+ 683 train 7.713647 (lr=4.7121e-05) (hash(x)=14133983)
714
+ 684 train 7.676874 (lr=4.6926e-05) (hash(x)=13407720)
715
+ 685 train 7.762440 (lr=4.6731e-05) (hash(x)=13632329)
716
+ 686 train 7.665123 (lr=4.6537e-05) (hash(x)=11013484)
717
+ 687 train 7.499414 (lr=4.6342e-05) (hash(x)=9539498)
718
+ 688 train 7.534472 (lr=4.6148e-05) (hash(x)=11548592)
719
+ 689 train 7.560140 (lr=4.5954e-05) (hash(x)=11883263)
720
+ 690 train 7.736515 (lr=4.5760e-05) (hash(x)=17040815)
721
+ 691 train 8.261700 (lr=4.5567e-05) (hash(x)=17793549)
722
+ 692 train 7.757318 (lr=4.5373e-05) (hash(x)=13566067)
723
+ 693 train 7.627385 (lr=4.5180e-05) (hash(x)=11871877)
724
+ 694 train 7.661992 (lr=4.4987e-05) (hash(x)=13936711)
725
+ 695 train 7.709857 (lr=4.4794e-05) (hash(x)=14163458)
726
+ 696 train 7.567407 (lr=4.4601e-05) (hash(x)=9756608)
727
+ 697 train 8.031370 (lr=4.4408e-05) (hash(x)=15049180)
728
+ 698 train 7.665815 (lr=4.4216e-05) (hash(x)=12741546)
729
+ 699 train 7.740666 (lr=4.4024e-05) (hash(x)=14609724)
730
+ 700 val loss 7.5824
731
+ 700 val perplexity 1963.3566
732
+ 700 train 7.741254 (lr=4.3832e-05) (hash(x)=13165136)
733
+ 701 train 7.501469 (lr=4.3640e-05) (hash(x)=11157433)
734
+ 702 train 7.559573 (lr=4.3449e-05) (hash(x)=10769955)
735
+ 703 train 8.064169 (lr=4.3258e-05) (hash(x)=18154448)
736
+ 704 train 8.089862 (lr=4.3067e-05) (hash(x)=11566844)
737
+ 705 train 8.106384 (lr=4.2876e-05) (hash(x)=10541575)
738
+ 706 train 7.776599 (lr=4.2685e-05) (hash(x)=13400633)
739
+ 707 train 7.658079 (lr=4.2495e-05) (hash(x)=11995482)
740
+ 708 train 7.823750 (lr=4.2305e-05) (hash(x)=12559360)
741
+ 709 train 7.596938 (lr=4.2115e-05) (hash(x)=11153006)
742
+ 710 train 8.187407 (lr=4.1925e-05) (hash(x)=14803712)
743
+ 711 train 7.964265 (lr=4.1736e-05) (hash(x)=13713689)
744
+ 712 train 7.562738 (lr=4.1547e-05) (hash(x)=12883531)
745
+ 713 train 8.095362 (lr=4.1358e-05) (hash(x)=15113554)
746
+ 714 train 7.540861 (lr=4.1170e-05) (hash(x)=11466714)
747
+ 715 train 7.542972 (lr=4.0981e-05) (hash(x)=11555030)
748
+ 716 train 7.662011 (lr=4.0793e-05) (hash(x)=11661705)
749
+ 717 train 8.009615 (lr=4.0606e-05) (hash(x)=13374040)
750
+ 718 train 7.661119 (lr=4.0418e-05) (hash(x)=11884576)
751
+ 719 train 7.596448 (lr=4.0231e-05) (hash(x)=14115772)
752
+ 720 train 7.694456 (lr=4.0044e-05) (hash(x)=14042374)
753
+ 721 train 7.784615 (lr=3.9857e-05) (hash(x)=9720467)
754
+ 722 train 7.686268 (lr=3.9671e-05) (hash(x)=13709799)
755
+ 723 train 7.590731 (lr=3.9485e-05) (hash(x)=12279208)
756
+ 724 train 7.653077 (lr=3.9299e-05) (hash(x)=12168977)
757
+ 725 train 7.460770 (lr=3.9114e-05) (hash(x)=11796701)
758
+ 726 train 7.484088 (lr=3.8929e-05) (hash(x)=12054640)
759
+ 727 train 7.535843 (lr=3.8744e-05) (hash(x)=12021395)
760
+ 728 train 7.732980 (lr=3.8560e-05) (hash(x)=11650114)
761
+ 729 train 7.455564 (lr=3.8376e-05) (hash(x)=10845152)
762
+ 730 train 7.493229 (lr=3.8192e-05) (hash(x)=10962390)
763
+ 731 train 7.687883 (lr=3.8008e-05) (hash(x)=12617648)
764
+ 732 train 7.421487 (lr=3.7825e-05) (hash(x)=9908464)
765
+ 733 train 7.455964 (lr=3.7642e-05) (hash(x)=11035304)
766
+ 734 train 7.608396 (lr=3.7460e-05) (hash(x)=11048281)
767
+ 735 train 7.387291 (lr=3.7277e-05) (hash(x)=12798942)
768
+ 736 train 7.634570 (lr=3.7096e-05) (hash(x)=11858915)
769
+ 737 train 7.680814 (lr=3.6914e-05) (hash(x)=11344910)
770
+ 738 train 7.575029 (lr=3.6733e-05) (hash(x)=11509808)
771
+ 739 train 7.683160 (lr=3.6552e-05) (hash(x)=12366544)
772
+ 740 train 7.678608 (lr=3.6372e-05) (hash(x)=11976631)
773
+ 741 train 7.469357 (lr=3.6192e-05) (hash(x)=11321027)
774
+ 742 train 7.631645 (lr=3.6012e-05) (hash(x)=11893948)
775
+ 743 train 7.329359 (lr=3.5833e-05) (hash(x)=10886073)
776
+ 744 train 7.503840 (lr=3.5654e-05) (hash(x)=9205900)
777
+ 745 train 7.510077 (lr=3.5475e-05) (hash(x)=12334480)
778
+ 746 train 7.716371 (lr=3.5297e-05) (hash(x)=13094093)
779
+ 747 train 7.741260 (lr=3.5119e-05) (hash(x)=12555076)
780
+ 748 train 7.750198 (lr=3.4942e-05) (hash(x)=14919237)
781
+ 749 train 7.665877 (lr=3.4765e-05) (hash(x)=13525646)
782
+ 750 val loss 7.5755
783
+ 750 val perplexity 1949.8712
784
+ 750 train 7.585590 (lr=3.4588e-05) (hash(x)=12350457)
785
+ 751 train 7.584790 (lr=3.4412e-05) (hash(x)=14777879)
786
+ 752 train 7.584960 (lr=3.4236e-05) (hash(x)=9165650)
787
+ 753 train 8.012780 (lr=3.4060e-05) (hash(x)=12138318)
788
+ 754 train 7.791905 (lr=3.3885e-05) (hash(x)=12066198)
789
+ 755 train 7.324716 (lr=3.3711e-05) (hash(x)=9927010)
790
+ 756 train 7.444154 (lr=3.3537e-05) (hash(x)=10963177)
791
+ 757 train 7.435044 (lr=3.3363e-05) (hash(x)=10265465)
792
+ 758 train 7.449288 (lr=3.3189e-05) (hash(x)=10012601)
793
+ 759 train 7.603426 (lr=3.3016e-05) (hash(x)=13073054)
794
+ 760 train 7.358012 (lr=3.2844e-05) (hash(x)=8375643)
795
+ 761 train 6.997710 (lr=3.2672e-05) (hash(x)=10517302)
796
+ 762 train 7.405934 (lr=3.2500e-05) (hash(x)=11617446)
797
+ 763 train 7.626057 (lr=3.2329e-05) (hash(x)=14967464)
798
+ 764 train 7.534800 (lr=3.2158e-05) (hash(x)=11293886)
799
+ 765 train 7.412302 (lr=3.1988e-05) (hash(x)=12190853)
800
+ 766 train 7.415482 (lr=3.1818e-05) (hash(x)=10723695)
801
+ 767 train 7.600772 (lr=3.1648e-05) (hash(x)=13463198)
802
+ 768 train 7.786309 (lr=3.1479e-05) (hash(x)=13293049)
803
+ 769 train 7.572845 (lr=3.1311e-05) (hash(x)=13635928)
804
+ 770 train 7.724860 (lr=3.1142e-05) (hash(x)=11492108)
805
+ 771 train 7.549083 (lr=3.0975e-05) (hash(x)=12209984)
806
+ 772 train 7.694386 (lr=3.0808e-05) (hash(x)=13817436)
807
+ 773 train 7.568054 (lr=3.0641e-05) (hash(x)=11946913)
808
+ 774 train 7.645437 (lr=3.0475e-05) (hash(x)=13760019)
809
+ 775 train 7.597554 (lr=3.0309e-05) (hash(x)=11983929)
810
+ 776 train 7.474491 (lr=3.0144e-05) (hash(x)=12740415)
811
+ 777 train 7.671687 (lr=2.9979e-05) (hash(x)=13044287)
812
+ 778 train 7.576913 (lr=2.9814e-05) (hash(x)=12306640)
813
+ 779 train 7.970211 (lr=2.9651e-05) (hash(x)=13502211)
814
+ 780 train 7.494351 (lr=2.9487e-05) (hash(x)=11651077)
815
+ 781 train 7.560380 (lr=2.9324e-05) (hash(x)=11732764)
816
+ 782 train 7.389981 (lr=2.9162e-05) (hash(x)=12023079)
817
+ 783 train 7.647940 (lr=2.9000e-05) (hash(x)=12563030)
818
+ 784 train 7.584294 (lr=2.8839e-05) (hash(x)=11093401)
819
+ 785 train 7.640553 (lr=2.8678e-05) (hash(x)=13370486)
820
+ 786 train 7.654840 (lr=2.8518e-05) (hash(x)=13064477)
821
+ 787 train 7.648440 (lr=2.8358e-05) (hash(x)=12774663)
822
+ 788 train 7.515447 (lr=2.8199e-05) (hash(x)=10977883)
823
+ 789 train 7.690089 (lr=2.8040e-05) (hash(x)=12581170)
824
+ 790 train 7.574380 (lr=2.7881e-05) (hash(x)=11710735)
825
+ 791 train 7.525580 (lr=2.7724e-05) (hash(x)=12170905)
826
+ 792 train 7.732967 (lr=2.7566e-05) (hash(x)=11984555)
827
+ 793 train 7.720396 (lr=2.7410e-05) (hash(x)=13035845)
828
+ 794 train 7.438868 (lr=2.7254e-05) (hash(x)=13998448)
829
+ 795 train 7.646195 (lr=2.7098e-05) (hash(x)=13348784)
830
+ 796 train 7.551514 (lr=2.6943e-05) (hash(x)=11535637)
831
+ 797 train 7.519851 (lr=2.6788e-05) (hash(x)=14533849)
832
+ 798 train 7.555802 (lr=2.6634e-05) (hash(x)=12593307)
833
+ 799 train 7.871137 (lr=2.6481e-05) (hash(x)=11525433)
834
+ 800 val loss 7.5613
835
+ 800 val perplexity 1922.3682
836
+ 800 train 7.610518 (lr=2.6328e-05) (hash(x)=9198860)
837
+ 801 train 7.558916 (lr=2.6176e-05) (hash(x)=7779807)
838
+ 802 train 7.628307 (lr=2.6024e-05) (hash(x)=8327978)
839
+ 803 train 7.450967 (lr=2.5873e-05) (hash(x)=9800296)
840
+ 804 train 7.621816 (lr=2.5722e-05) (hash(x)=10408800)
841
+ 805 train 7.682235 (lr=2.5572e-05) (hash(x)=11761671)
842
+ 806 train 7.528586 (lr=2.5423e-05) (hash(x)=10137737)
843
+ 807 train 7.603443 (lr=2.5274e-05) (hash(x)=10168268)
844
+ 808 train 7.613646 (lr=2.5125e-05) (hash(x)=15372601)
845
+ 809 train 7.507771 (lr=2.4978e-05) (hash(x)=10272263)
846
+ 810 train 7.427522 (lr=2.4830e-05) (hash(x)=10560999)
847
+ 811 train 7.520030 (lr=2.4684e-05) (hash(x)=10276843)
848
+ 812 train 8.704545 (lr=2.4538e-05) (hash(x)=12452448)
849
+ 813 train 7.772923 (lr=2.4392e-05) (hash(x)=11364751)
850
+ 814 train 7.320310 (lr=2.4247e-05) (hash(x)=11485502)
851
+ 815 train 7.376126 (lr=2.4103e-05) (hash(x)=12497404)
852
+ 816 train 7.282670 (lr=2.3959e-05) (hash(x)=10013569)
853
+ 817 train 7.939144 (lr=2.3816e-05) (hash(x)=14966009)
854
+ 818 train 7.599930 (lr=2.3674e-05) (hash(x)=11562145)
855
+ 819 train 7.653254 (lr=2.3532e-05) (hash(x)=12932339)
856
+ 820 train 7.607935 (lr=2.3391e-05) (hash(x)=11563553)
857
+ 821 train 7.623580 (lr=2.3250e-05) (hash(x)=8563011)
858
+ 822 train 7.724271 (lr=2.3110e-05) (hash(x)=13409258)
859
+ 823 train 7.623437 (lr=2.2971e-05) (hash(x)=14203584)
860
+ 824 train 7.392287 (lr=2.2832e-05) (hash(x)=10360049)
861
+ 825 train 7.479532 (lr=2.2694e-05) (hash(x)=10373543)
862
+ 826 train 7.542297 (lr=2.2556e-05) (hash(x)=12645249)
863
+ 827 train 7.652062 (lr=2.2420e-05) (hash(x)=13921321)
864
+ 828 train 7.742990 (lr=2.2283e-05) (hash(x)=10868669)
865
+ 829 train 7.448482 (lr=2.2148e-05) (hash(x)=12997326)
866
+ 830 train 7.783127 (lr=2.2013e-05) (hash(x)=13553993)
867
+ 831 train 7.581386 (lr=2.1878e-05) (hash(x)=10946961)
868
+ 832 train 7.511065 (lr=2.1745e-05) (hash(x)=13714094)
869
+ 833 train 7.858049 (lr=2.1612e-05) (hash(x)=11317673)
870
+ 834 train 7.773939 (lr=2.1479e-05) (hash(x)=11555387)
871
+ 835 train 7.769975 (lr=2.1347e-05) (hash(x)=12168723)
872
+ 836 train 7.810711 (lr=2.1216e-05) (hash(x)=11310063)
873
+ 837 train 7.583141 (lr=2.1086e-05) (hash(x)=11165585)
874
+ 838 train 7.461092 (lr=2.0956e-05) (hash(x)=10049422)
875
+ 839 train 7.471561 (lr=2.0827e-05) (hash(x)=10942254)
876
+ 840 train 7.629329 (lr=2.0698e-05) (hash(x)=12383062)
877
+ 841 train 7.567740 (lr=2.0570e-05) (hash(x)=12927776)
878
+ 842 train 7.873129 (lr=2.0443e-05) (hash(x)=12835199)
879
+ 843 train 8.062815 (lr=2.0317e-05) (hash(x)=10371632)
880
+ 844 train 7.927230 (lr=2.0191e-05) (hash(x)=15118958)
881
+ 845 train 8.009559 (lr=2.0066e-05) (hash(x)=12057844)
882
+ 846 train 7.538574 (lr=1.9941e-05) (hash(x)=12276438)
883
+ 847 train 7.912656 (lr=1.9818e-05) (hash(x)=16703203)
884
+ 848 train 7.581523 (lr=1.9694e-05) (hash(x)=12033983)
885
+ 849 train 7.580794 (lr=1.9572e-05) (hash(x)=12424204)
886
+ 850 val loss 7.5538
887
+ 850 val perplexity 1907.9677
888
+ 850 train 7.544832 (lr=1.9450e-05) (hash(x)=11830025)
889
+ 851 train 7.934174 (lr=1.9329e-05) (hash(x)=17013185)
890
+ 852 train 7.935362 (lr=1.9209e-05) (hash(x)=16760833)
891
+ 853 train 7.814120 (lr=1.9089e-05) (hash(x)=11881739)
892
+ 854 train 7.776320 (lr=1.8970e-05) (hash(x)=13649855)
893
+ 855 train 7.584765 (lr=1.8852e-05) (hash(x)=15000294)
894
+ 856 train 7.529850 (lr=1.8734e-05) (hash(x)=11412424)
895
+ 857 train 7.583403 (lr=1.8618e-05) (hash(x)=13008387)
896
+ 858 train 7.533602 (lr=1.8501e-05) (hash(x)=10899640)
897
+ 859 train 7.394091 (lr=1.8386e-05) (hash(x)=10834161)
898
+ 860 train 7.330706 (lr=1.8271e-05) (hash(x)=10472237)
899
+ 861 train 7.881645 (lr=1.8157e-05) (hash(x)=15743796)
900
+ 862 train 7.685043 (lr=1.8044e-05) (hash(x)=10842000)
901
+ 863 train 7.650133 (lr=1.7931e-05) (hash(x)=14862449)
902
+ 864 train 8.134257 (lr=1.7819e-05) (hash(x)=15511051)
903
+ 865 train 7.464056 (lr=1.7708e-05) (hash(x)=11257626)
904
+ 866 train 8.021695 (lr=1.7598e-05) (hash(x)=17810791)
905
+ 867 train 7.438634 (lr=1.7488e-05) (hash(x)=10386532)
906
+ 868 train 7.328322 (lr=1.7379e-05) (hash(x)=12466709)
907
+ 869 train 7.500387 (lr=1.7271e-05) (hash(x)=11765736)
908
+ 870 train 7.822093 (lr=1.7163e-05) (hash(x)=12448217)
909
+ 871 train 7.550396 (lr=1.7056e-05) (hash(x)=13806874)
910
+ 872 train 7.690389 (lr=1.6950e-05) (hash(x)=12238204)
911
+ 873 train 7.580042 (lr=1.6845e-05) (hash(x)=12702206)
912
+ 874 train 7.738585 (lr=1.6740e-05) (hash(x)=13884223)
913
+ 875 train 7.543190 (lr=1.6636e-05) (hash(x)=12733894)
914
+ 876 train 7.744803 (lr=1.6533e-05) (hash(x)=7848745)
915
+ 877 train 7.860775 (lr=1.6431e-05) (hash(x)=13860937)
916
+ 878 train 8.041671 (lr=1.6329e-05) (hash(x)=14870569)
917
+ 879 train 7.687901 (lr=1.6228e-05) (hash(x)=12259111)
918
+ 880 train 7.526224 (lr=1.6128e-05) (hash(x)=12747643)
919
+ 881 train 7.656184 (lr=1.6029e-05) (hash(x)=14710730)
920
+ 882 train 7.621186 (lr=1.5930e-05) (hash(x)=12185557)
921
+ 883 train 7.564676 (lr=1.5832e-05) (hash(x)=10003332)
922
+ 884 train 7.367582 (lr=1.5735e-05) (hash(x)=8591840)
923
+ 885 train 7.677679 (lr=1.5639e-05) (hash(x)=13675590)
924
+ 886 train 7.114809 (lr=1.5543e-05) (hash(x)=11265029)
925
+ 887 train 7.531140 (lr=1.5449e-05) (hash(x)=11837867)
926
+ 888 train 8.008759 (lr=1.5354e-05) (hash(x)=11888580)
927
+ 889 train 7.831285 (lr=1.5261e-05) (hash(x)=11296484)
928
+ 890 train 7.545902 (lr=1.5169e-05) (hash(x)=12991870)
929
+ 891 train 7.697248 (lr=1.5077e-05) (hash(x)=13840537)
930
+ 892 train 7.444850 (lr=1.4986e-05) (hash(x)=12800791)
931
+ 893 train 7.766712 (lr=1.4896e-05) (hash(x)=14920737)
932
+ 894 train 7.689218 (lr=1.4806e-05) (hash(x)=13950257)
933
+ 895 train 7.486090 (lr=1.4718e-05) (hash(x)=12159808)
934
+ 896 train 7.696806 (lr=1.4630e-05) (hash(x)=12734109)
935
+ 897 train 7.461656 (lr=1.4543e-05) (hash(x)=9163004)
936
+ 898 train 7.511126 (lr=1.4456e-05) (hash(x)=11771048)
937
+ 899 train 7.730757 (lr=1.4371e-05) (hash(x)=13797226)
938
+ 900 val loss 7.5447
939
+ 900 val perplexity 1890.6975
940
+ 900 train 7.621592 (lr=1.4286e-05) (hash(x)=12813851)
941
+ 901 train 7.536583 (lr=1.4202e-05) (hash(x)=10204611)
942
+ 902 train 7.650610 (lr=1.4119e-05) (hash(x)=11849134)
943
+ 903 train 7.673443 (lr=1.4037e-05) (hash(x)=12372219)
944
+ 904 train 7.685614 (lr=1.3955e-05) (hash(x)=14633473)
945
+ 905 train 7.548129 (lr=1.3874e-05) (hash(x)=12434908)
946
+ 906 train 7.743475 (lr=1.3794e-05) (hash(x)=11743620)
947
+ 907 train 7.475213 (lr=1.3715e-05) (hash(x)=11371435)
948
+ 908 train 7.647787 (lr=1.3637e-05) (hash(x)=12483864)
949
+ 909 train 7.654312 (lr=1.3559e-05) (hash(x)=13523826)
950
+ 910 train 7.522507 (lr=1.3482e-05) (hash(x)=11575838)
951
+ 911 train 7.516685 (lr=1.3407e-05) (hash(x)=10910611)
952
+ 912 train 7.472122 (lr=1.3331e-05) (hash(x)=10322819)
953
+ 913 train 7.521118 (lr=1.3257e-05) (hash(x)=12905087)
954
+ 914 train 7.734788 (lr=1.3183e-05) (hash(x)=11755845)
955
+ 915 train 7.798567 (lr=1.3111e-05) (hash(x)=10985712)
956
+ 916 train 7.369901 (lr=1.3039e-05) (hash(x)=10264591)
957
+ 917 train 7.802237 (lr=1.2968e-05) (hash(x)=13245256)
958
+ 918 train 7.540447 (lr=1.2897e-05) (hash(x)=13080408)
959
+ 919 train 7.579133 (lr=1.2828e-05) (hash(x)=12509276)
960
+ 920 train 7.529182 (lr=1.2759e-05) (hash(x)=12427196)
961
+ 921 train 7.620004 (lr=1.2691e-05) (hash(x)=13538047)
962
+ 922 train 7.714370 (lr=1.2624e-05) (hash(x)=16129945)
963
+ 923 train 7.934924 (lr=1.2558e-05) (hash(x)=12953787)
964
+ 924 train 7.636042 (lr=1.2493e-05) (hash(x)=9473654)
965
+ 925 train 7.569048 (lr=1.2428e-05) (hash(x)=9468319)
966
+ 926 train 7.643005 (lr=1.2364e-05) (hash(x)=13193021)
967
+ 927 train 7.742980 (lr=1.2301e-05) (hash(x)=12720900)
968
+ 928 train 7.486856 (lr=1.2239e-05) (hash(x)=10634942)
969
+ 929 train 7.451910 (lr=1.2178e-05) (hash(x)=12030112)
970
+ 930 train 7.553452 (lr=1.2118e-05) (hash(x)=12212081)
971
+ 931 train 7.460938 (lr=1.2058e-05) (hash(x)=16075757)
972
+ 932 train 8.233881 (lr=1.1999e-05) (hash(x)=16126508)
973
+ 933 train 7.312922 (lr=1.1941e-05) (hash(x)=11540359)
974
+ 934 train 7.563043 (lr=1.1884e-05) (hash(x)=13428693)
975
+ 935 train 7.767107 (lr=1.1828e-05) (hash(x)=14048369)
976
+ 936 train 7.872973 (lr=1.1772e-05) (hash(x)=12554502)
977
+ 937 train 7.769233 (lr=1.1718e-05) (hash(x)=12961504)
978
+ 938 train 7.951363 (lr=1.1664e-05) (hash(x)=12343533)
979
+ 939 train 7.832977 (lr=1.1611e-05) (hash(x)=12602604)
980
+ 940 train 7.659280 (lr=1.1559e-05) (hash(x)=11612970)
981
+ 941 train 7.831597 (lr=1.1508e-05) (hash(x)=14173792)
982
+ 942 train 7.998806 (lr=1.1457e-05) (hash(x)=15456213)
983
+ 943 train 7.485168 (lr=1.1408e-05) (hash(x)=11552959)
984
+ 944 train 7.515705 (lr=1.1359e-05) (hash(x)=9453679)
985
+ 945 train 7.647860 (lr=1.1311e-05) (hash(x)=11986216)
986
+ 946 train 7.558979 (lr=1.1264e-05) (hash(x)=11356627)
987
+ 947 train 7.807800 (lr=1.1218e-05) (hash(x)=9890481)
988
+ 948 train 7.501633 (lr=1.1173e-05) (hash(x)=11347234)
989
+ 949 train 7.432828 (lr=1.1128e-05) (hash(x)=13052540)
990
+ 950 val loss 7.5433
991
+ 950 val perplexity 1887.9705
992
+ 950 train 7.445033 (lr=1.1085e-05) (hash(x)=13065639)
993
+ 951 train 7.709283 (lr=1.1042e-05) (hash(x)=13423367)
994
+ 952 train 7.606824 (lr=1.1000e-05) (hash(x)=13054517)
995
+ 953 train 7.598430 (lr=1.0959e-05) (hash(x)=12494434)
996
+ 954 train 7.405973 (lr=1.0919e-05) (hash(x)=12496694)
997
+ 955 train 7.411688 (lr=1.0879e-05) (hash(x)=11555634)
998
+ 956 train 7.428921 (lr=1.0841e-05) (hash(x)=10983636)
999
+ 957 train 7.855441 (lr=1.0803e-05) (hash(x)=17108448)
1000
+ 958 train 8.344284 (lr=1.0766e-05) (hash(x)=15985570)
1001
+ 959 train 8.328481 (lr=1.0730e-05) (hash(x)=14914789)
1002
+ 960 train 8.285243 (lr=1.0695e-05) (hash(x)=13097897)
1003
+ 961 train 8.310488 (lr=1.0661e-05) (hash(x)=16237319)
1004
+ 962 train 8.252663 (lr=1.0628e-05) (hash(x)=16105534)
1005
+ 963 train 8.302642 (lr=1.0595e-05) (hash(x)=16763637)
1006
+ 964 train 8.251900 (lr=1.0563e-05) (hash(x)=16598501)
1007
+ 965 train 8.279643 (lr=1.0533e-05) (hash(x)=15222896)
1008
+ 966 train 8.301451 (lr=1.0503e-05) (hash(x)=14919621)
1009
+ 967 train 8.366073 (lr=1.0474e-05) (hash(x)=15783017)
1010
+ 968 train 8.255242 (lr=1.0445e-05) (hash(x)=15370617)
1011
+ 969 train 8.472473 (lr=1.0418e-05) (hash(x)=17102921)
1012
+ 970 train 8.245502 (lr=1.0391e-05) (hash(x)=15358790)
1013
+ 971 train 8.435727 (lr=1.0366e-05) (hash(x)=16354494)
1014
+ 972 train 8.313684 (lr=1.0341e-05) (hash(x)=13580270)
1015
+ 973 train 7.656308 (lr=1.0317e-05) (hash(x)=14615340)
1016
+ 974 train 7.627219 (lr=1.0294e-05) (hash(x)=12022804)
1017
+ 975 train 7.756118 (lr=1.0272e-05) (hash(x)=13512761)
1018
+ 976 train 8.073447 (lr=1.0251e-05) (hash(x)=16681653)
1019
+ 977 train 7.654963 (lr=1.0230e-05) (hash(x)=10750782)
1020
+ 978 train 7.595737 (lr=1.0211e-05) (hash(x)=12643825)
1021
+ 979 train 7.448093 (lr=1.0192e-05) (hash(x)=13751537)
1022
+ 980 train 7.640579 (lr=1.0174e-05) (hash(x)=12575386)
1023
+ 981 train 7.570357 (lr=1.0157e-05) (hash(x)=12343566)
1024
+ 982 train 7.574430 (lr=1.0141e-05) (hash(x)=11685246)
1025
+ 983 train 7.486073 (lr=1.0126e-05) (hash(x)=13830852)
1026
+ 984 train 7.794157 (lr=1.0111e-05) (hash(x)=11998502)
1027
+ 985 train 7.478005 (lr=1.0098e-05) (hash(x)=11170832)
1028
+ 986 train 7.504237 (lr=1.0085e-05) (hash(x)=11482404)
1029
+ 987 train 7.663180 (lr=1.0074e-05) (hash(x)=14841133)
1030
+ 988 train 7.762995 (lr=1.0063e-05) (hash(x)=14960072)
1031
+ 989 train 7.690414 (lr=1.0053e-05) (hash(x)=13683107)
1032
+ 990 train 7.643292 (lr=1.0044e-05) (hash(x)=11884964)
1033
+ 991 train 7.709462 (lr=1.0035e-05) (hash(x)=11832803)
1034
+ 992 train 7.442004 (lr=1.0028e-05) (hash(x)=11624359)
1035
+ 993 train 7.522249 (lr=1.0021e-05) (hash(x)=11362245)
1036
+ 994 train 7.657452 (lr=1.0016e-05) (hash(x)=13335953)
1037
+ 995 train 11.158754 (lr=1.0011e-05) (hash(x)=27572323)
1038
+ 996 train 12.319499 (lr=1.0007e-05) (hash(x)=30981287)
1039
+ 997 train 12.357571 (lr=1.0004e-05) (hash(x)=36479466)
1040
+ 998 train 12.973638 (lr=1.0002e-05) (hash(x)=34231765)
1041
+ 999 val loss 7.5451
1042
+ 999 val perplexity 1891.5271
1043
+ 999 train 11.723103 (lr=1.0000e-05) (hash(x)=31819669)
lr1e-4_total_batch_size10240_seq_len256/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1c469f9b61c16a206b7127baa20b7b7ee1e18734f0f62519e00180083f5b5cb
3
+ size 38587970
lr1e-4_total_batch_size10240_seq_len256/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c441404356a5824c1db7af6d4f667edacb0a9e5062b96e21da4957e56c8eab5
3
+ size 70895430