andrew-healey commited on
Commit
7d751b3
·
verified ·
1 Parent(s): 93c7d46

Upload folder using huggingface_hub

Browse files
lr3e-3_total_batch_size40960_seq_len512/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep_2/lr3e-3_total_batch_size40960_seq_len512", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 2, "protect_bos_token": true, "prevent_from_masking_myself": true, "max_steps": null, "group": "proxy_model_sweep_2", "use_wandb": true, "kill_self_after_run": false, "random_seed": 1337, "memory_penalty_epsilon": 0.1, "selection_head_linear_combo": "none", "selection_head_linear_combo_scale": 1.0, "protection_kind": "none", "leaky_relu_alpha": null, "leaky_relu_bias": null, "use_compile": true, "use_mini_model": false, "upload_to_hf": true, "seq_len": 512, "batch_size": null, "total_batch_size": 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": 0.003, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr3e-3_total_batch_size40960_seq_len512/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cec289292cf51a5462e2631af59109e64df3eca2d9e62e1fbd1c53a776a0f874
3
+ size 964
lr3e-3_total_batch_size40960_seq_len512/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.7800
3
+ 0 val perplexity 130608.6406
4
+ 0 train 11.768598 (lr=1.0490e-05) (hash(x)=20782690)
5
+ 1 train 11.754906 (lr=2.0979e-05) (hash(x)=23890908)
6
+ 2 train 11.709332 (lr=3.1469e-05) (hash(x)=24611628)
7
+ 3 train 11.621424 (lr=4.1958e-05) (hash(x)=25767553)
8
+ 4 train 11.541439 (lr=5.2448e-05) (hash(x)=22512166)
9
+ 5 train 11.429130 (lr=6.2937e-05) (hash(x)=24732833)
10
+ 6 train 11.324894 (lr=7.3427e-05) (hash(x)=25129410)
11
+ 7 train 11.179634 (lr=8.3916e-05) (hash(x)=26582391)
12
+ 8 train 10.999023 (lr=9.4406e-05) (hash(x)=26472336)
13
+ 9 train 10.859825 (lr=1.0490e-04) (hash(x)=27654289)
14
+ 10 train 10.721365 (lr=1.1538e-04) (hash(x)=26559876)
15
+ 11 train 10.603133 (lr=1.2587e-04) (hash(x)=21770329)
16
+ 12 train 10.568665 (lr=1.3636e-04) (hash(x)=22723459)
17
+ 13 train 10.471037 (lr=1.4685e-04) (hash(x)=30338342)
18
+ 14 train 10.424945 (lr=1.5734e-04) (hash(x)=25678059)
19
+ 15 train 10.330013 (lr=1.6783e-04) (hash(x)=26109335)
20
+ 16 train 10.348676 (lr=1.7832e-04) (hash(x)=25044885)
21
+ 17 train 10.365445 (lr=1.8881e-04) (hash(x)=28870690)
22
+ 18 train 10.371346 (lr=1.9930e-04) (hash(x)=35279941)
23
+ 19 train 10.256713 (lr=2.0979e-04) (hash(x)=25015923)
24
+ 20 train 10.170407 (lr=2.2028e-04) (hash(x)=26740855)
25
+ 21 train 10.084578 (lr=2.3077e-04) (hash(x)=22262151)
26
+ 22 train 10.019298 (lr=2.4126e-04) (hash(x)=26621834)
27
+ 23 train 9.997997 (lr=2.5175e-04) (hash(x)=23234741)
28
+ 24 train 10.036499 (lr=2.6224e-04) (hash(x)=25470563)
29
+ 25 train 9.930593 (lr=2.7273e-04) (hash(x)=21799102)
30
+ 26 train 9.796321 (lr=2.8322e-04) (hash(x)=23275479)
31
+ 27 train 9.795861 (lr=2.9371e-04) (hash(x)=22091862)
32
+ 28 train 9.596811 (lr=3.0420e-04) (hash(x)=17430373)
33
+ 29 train 9.448159 (lr=3.1469e-04) (hash(x)=22954861)
34
+ 30 train 9.482758 (lr=3.2517e-04) (hash(x)=26671799)
35
+ 31 train 9.340373 (lr=3.3566e-04) (hash(x)=26924723)
36
+ 32 train 9.279879 (lr=3.4615e-04) (hash(x)=23557495)
37
+ 33 train 9.005252 (lr=3.5664e-04) (hash(x)=27804380)
38
+ 34 train 9.112953 (lr=3.6713e-04) (hash(x)=26654908)
39
+ 35 train 8.931460 (lr=3.7762e-04) (hash(x)=23627518)
40
+ 36 train 8.794367 (lr=3.8811e-04) (hash(x)=21889990)
41
+ 37 train 8.792306 (lr=3.9860e-04) (hash(x)=22528001)
42
+ 38 train 8.739619 (lr=4.0909e-04) (hash(x)=29248942)
43
+ 39 train 8.548651 (lr=4.1958e-04) (hash(x)=25533417)
44
+ 40 train 8.430010 (lr=4.3007e-04) (hash(x)=20612533)
45
+ 41 train 8.336095 (lr=4.4056e-04) (hash(x)=24677740)
46
+ 42 train 8.315884 (lr=4.5105e-04) (hash(x)=25347203)
47
+ 43 train 8.299545 (lr=4.6154e-04) (hash(x)=28922363)
48
+ 44 train 8.281492 (lr=4.7203e-04) (hash(x)=25631397)
49
+ 45 train 8.178084 (lr=4.8252e-04) (hash(x)=26237983)
50
+ 46 train 8.094031 (lr=4.9301e-04) (hash(x)=24742645)
51
+ 47 train 8.107033 (lr=5.0350e-04) (hash(x)=26157660)
52
+ 48 train 7.954978 (lr=5.1399e-04) (hash(x)=22582169)
53
+ 49 train 7.947535 (lr=5.2448e-04) (hash(x)=26431960)
54
+ 50 val loss 7.9593
55
+ 50 val perplexity 2861.9529
56
+ 50 train 8.005766 (lr=5.3497e-04) (hash(x)=25716176)
57
+ 51 train 7.957718 (lr=5.4545e-04) (hash(x)=28019788)
58
+ 52 train 7.904066 (lr=5.5594e-04) (hash(x)=20382963)
59
+ 53 train 7.793106 (lr=5.6643e-04) (hash(x)=24489647)
60
+ 54 train 7.822482 (lr=5.7692e-04) (hash(x)=25128502)
61
+ 55 train 7.867463 (lr=5.8741e-04) (hash(x)=25231390)
62
+ 56 train 7.688143 (lr=5.9790e-04) (hash(x)=22708977)
63
+ 57 train 7.933190 (lr=6.0839e-04) (hash(x)=24364920)
64
+ 58 train 7.918108 (lr=6.1888e-04) (hash(x)=24932925)
65
+ 59 train 7.741066 (lr=6.2937e-04) (hash(x)=26913684)
66
+ 60 train 7.691313 (lr=6.3986e-04) (hash(x)=18017792)
67
+ 61 train 7.807190 (lr=6.5035e-04) (hash(x)=28613882)
68
+ 62 train 7.746123 (lr=6.6084e-04) (hash(x)=22652243)
69
+ 63 train 7.553226 (lr=6.7133e-04) (hash(x)=27194521)
70
+ 64 train 7.740228 (lr=6.8182e-04) (hash(x)=24936836)
71
+ 65 train 7.794434 (lr=6.9231e-04) (hash(x)=24765578)
72
+ 66 train 7.809368 (lr=7.0280e-04) (hash(x)=23107416)
73
+ 67 train 7.707591 (lr=7.1329e-04) (hash(x)=29314255)
74
+ 68 train 7.668074 (lr=7.2378e-04) (hash(x)=24013079)
75
+ 69 train 7.717449 (lr=7.3427e-04) (hash(x)=25322001)
76
+ 70 train 7.794168 (lr=7.4476e-04) (hash(x)=26690440)
77
+ 71 train 7.691620 (lr=7.5524e-04) (hash(x)=22598512)
78
+ 72 train 7.615169 (lr=7.6573e-04) (hash(x)=18473911)
79
+ 73 train 7.884169 (lr=7.7622e-04) (hash(x)=22322442)
80
+ 74 train 7.876599 (lr=7.8671e-04) (hash(x)=26324153)
81
+ 75 train 7.775492 (lr=7.9720e-04) (hash(x)=23446058)
82
+ 76 train 7.847318 (lr=8.0769e-04) (hash(x)=24843741)
83
+ 77 train 7.678906 (lr=8.1818e-04) (hash(x)=23197102)
84
+ 78 train 7.984725 (lr=8.2867e-04) (hash(x)=23054940)
85
+ 79 train 7.934973 (lr=8.3916e-04) (hash(x)=27140757)
86
+ 80 train 7.633224 (lr=8.4965e-04) (hash(x)=25149419)
87
+ 81 train 7.769537 (lr=8.6014e-04) (hash(x)=24727764)
88
+ 82 train 7.442677 (lr=8.7063e-04) (hash(x)=21935931)
89
+ 83 train 7.587365 (lr=8.8112e-04) (hash(x)=27606073)
90
+ 84 train 7.730572 (lr=8.9161e-04) (hash(x)=22450341)
91
+ 85 train 7.872334 (lr=9.0210e-04) (hash(x)=25945859)
92
+ 86 train 7.705494 (lr=9.1259e-04) (hash(x)=21613707)
93
+ 87 train 7.720841 (lr=9.2308e-04) (hash(x)=24396519)
94
+ 88 train 7.731224 (lr=9.3357e-04) (hash(x)=29138828)
95
+ 89 train 7.754342 (lr=9.4406e-04) (hash(x)=24685515)
96
+ 90 train 7.683757 (lr=9.5455e-04) (hash(x)=22825749)
97
+ 91 train 7.733871 (lr=9.6503e-04) (hash(x)=28767869)
98
+ 92 train 7.759574 (lr=9.7552e-04) (hash(x)=21159060)
99
+ 93 train 7.663857 (lr=9.8601e-04) (hash(x)=25045397)
100
+ 94 train 7.812467 (lr=9.9650e-04) (hash(x)=25560928)
101
+ 95 train 7.833717 (lr=1.0070e-03) (hash(x)=28959222)
102
+ 96 train 7.715253 (lr=1.0175e-03) (hash(x)=21248405)
103
+ 97 train 7.691627 (lr=1.0280e-03) (hash(x)=25830182)
104
+ 98 train 7.552038 (lr=1.0385e-03) (hash(x)=25173113)
105
+ 99 train 7.643596 (lr=1.0490e-03) (hash(x)=23486277)
106
+ 100 val loss 7.6437
107
+ 100 val perplexity 2087.4246
108
+ 100 train 7.666860 (lr=1.0594e-03) (hash(x)=28916006)
109
+ 101 train 7.850005 (lr=1.0699e-03) (hash(x)=25556929)
110
+ 102 train 7.772370 (lr=1.0804e-03) (hash(x)=29648798)
111
+ 103 train 7.695124 (lr=1.0909e-03) (hash(x)=24976217)
112
+ 104 train 7.784911 (lr=1.1014e-03) (hash(x)=31019606)
113
+ 105 train 7.804643 (lr=1.1119e-03) (hash(x)=28269421)
114
+ 106 train 7.699863 (lr=1.1224e-03) (hash(x)=24865358)
115
+ 107 train 7.625817 (lr=1.1329e-03) (hash(x)=26984272)
116
+ 108 train 7.726004 (lr=1.1434e-03) (hash(x)=24544116)
117
+ 109 train 7.690204 (lr=1.1538e-03) (hash(x)=22059850)
118
+ 110 train 7.713831 (lr=1.1643e-03) (hash(x)=23926632)
119
+ 111 train 7.643813 (lr=1.1748e-03) (hash(x)=27025333)
120
+ 112 train 7.587278 (lr=1.1853e-03) (hash(x)=29120407)
121
+ 113 train 7.730151 (lr=1.1958e-03) (hash(x)=23827393)
122
+ 114 train 7.638643 (lr=1.2063e-03) (hash(x)=27230027)
123
+ 115 train 7.595006 (lr=1.2168e-03) (hash(x)=24968260)
124
+ 116 train 7.604590 (lr=1.2273e-03) (hash(x)=27544803)
125
+ 117 train 7.604594 (lr=1.2378e-03) (hash(x)=26540663)
126
+ 118 train 7.703282 (lr=1.2483e-03) (hash(x)=25856625)
127
+ 119 train 7.574071 (lr=1.2587e-03) (hash(x)=23164356)
128
+ 120 train 7.666543 (lr=1.2692e-03) (hash(x)=25646282)
129
+ 121 train 7.662623 (lr=1.2797e-03) (hash(x)=23843233)
130
+ 122 train 7.511415 (lr=1.2902e-03) (hash(x)=24506028)
131
+ 123 train 7.601475 (lr=1.3007e-03) (hash(x)=22781277)
132
+ 124 train 7.614566 (lr=1.3112e-03) (hash(x)=24948650)
133
+ 125 train 7.586625 (lr=1.3217e-03) (hash(x)=26373991)
134
+ 126 train 7.675754 (lr=1.3322e-03) (hash(x)=23402396)
135
+ 127 train 7.655837 (lr=1.3427e-03) (hash(x)=21562133)
136
+ 128 train 7.696960 (lr=1.3531e-03) (hash(x)=25009210)
137
+ 129 train 7.620250 (lr=1.3636e-03) (hash(x)=27161811)
138
+ 130 train 7.687163 (lr=1.3741e-03) (hash(x)=29515100)
139
+ 131 train 7.602319 (lr=1.3846e-03) (hash(x)=25305929)
140
+ 132 train 7.639222 (lr=1.3951e-03) (hash(x)=26808543)
141
+ 133 train 7.534541 (lr=1.4056e-03) (hash(x)=25669509)
142
+ 134 train 7.665064 (lr=1.4161e-03) (hash(x)=28435805)
143
+ 135 train 7.627543 (lr=1.4266e-03) (hash(x)=26482758)
144
+ 136 train 7.555047 (lr=1.4371e-03) (hash(x)=24929929)
145
+ 137 train 7.545872 (lr=1.4476e-03) (hash(x)=25043037)
146
+ 138 train 7.793445 (lr=1.4580e-03) (hash(x)=29151257)
147
+ 139 train 7.527866 (lr=1.4685e-03) (hash(x)=26400365)
148
+ 140 train 7.544953 (lr=1.4790e-03) (hash(x)=24098812)
149
+ 141 train 7.550236 (lr=1.4895e-03) (hash(x)=20967666)
150
+ 142 train 7.722590 (lr=1.5000e-03) (hash(x)=28667963)
151
+ 143 train 7.682284 (lr=1.5105e-03) (hash(x)=27083417)
152
+ 144 train 7.586846 (lr=1.5210e-03) (hash(x)=23649001)
153
+ 145 train 7.675485 (lr=1.5315e-03) (hash(x)=31338300)
154
+ 146 train 7.638890 (lr=1.5420e-03) (hash(x)=30603174)
155
+ 147 train 7.692130 (lr=1.5524e-03) (hash(x)=24830965)
156
+ 148 train 7.858856 (lr=1.5629e-03) (hash(x)=27269893)
157
+ 149 train 7.751105 (lr=1.5734e-03) (hash(x)=24825041)
158
+ 150 val loss 7.8137
159
+ 150 val perplexity 2474.2771
160
+ 150 train 7.783054 (lr=1.5839e-03) (hash(x)=22724871)
161
+ 151 train 7.743700 (lr=1.5944e-03) (hash(x)=24683717)
162
+ 152 train 7.889342 (lr=1.6049e-03) (hash(x)=26190337)
163
+ 153 train 7.886657 (lr=1.6154e-03) (hash(x)=23086289)
164
+ 154 train 7.936162 (lr=1.6259e-03) (hash(x)=23593875)
165
+ 155 train 7.817874 (lr=1.6364e-03) (hash(x)=22414190)
166
+ 156 train 7.906740 (lr=1.6469e-03) (hash(x)=26758114)
167
+ 157 train 7.854712 (lr=1.6573e-03) (hash(x)=21563184)
168
+ 158 train 7.909463 (lr=1.6678e-03) (hash(x)=19218470)
169
+ 159 train 7.960034 (lr=1.6783e-03) (hash(x)=25588469)
170
+ 160 train 7.916324 (lr=1.6888e-03) (hash(x)=30433767)
171
+ 161 train 7.991632 (lr=1.6993e-03) (hash(x)=17851621)
172
+ 162 train 8.030258 (lr=1.7098e-03) (hash(x)=25183690)
173
+ 163 train 8.061817 (lr=1.7203e-03) (hash(x)=27040642)
174
+ 164 train 7.805053 (lr=1.7308e-03) (hash(x)=22183303)
175
+ 165 train 8.140038 (lr=1.7413e-03) (hash(x)=27537063)
176
+ 166 train 7.807179 (lr=1.7517e-03) (hash(x)=19492832)
177
+ 167 train 7.801187 (lr=1.7622e-03) (hash(x)=25951629)
178
+ 168 train 7.838325 (lr=1.7727e-03) (hash(x)=22408682)
179
+ 169 train 7.774121 (lr=1.7832e-03) (hash(x)=27633457)
180
+ 170 train 7.751040 (lr=1.7937e-03) (hash(x)=27635080)
181
+ 171 train 7.760401 (lr=1.8042e-03) (hash(x)=22287596)
182
+ 172 train 7.855944 (lr=1.8147e-03) (hash(x)=35895440)
183
+ 173 train 7.783485 (lr=1.8252e-03) (hash(x)=25888847)
184
+ 174 train 7.827337 (lr=1.8357e-03) (hash(x)=26518800)
185
+ 175 train 7.793284 (lr=1.8462e-03) (hash(x)=37519283)
186
+ 176 train 7.953126 (lr=1.8566e-03) (hash(x)=25141945)
187
+ 177 train 7.973799 (lr=1.8671e-03) (hash(x)=28527939)
188
+ 178 train 7.778150 (lr=1.8776e-03) (hash(x)=21942471)
189
+ 179 train 7.805886 (lr=1.8881e-03) (hash(x)=24955630)
190
+ 180 train 7.790400 (lr=1.8986e-03) (hash(x)=25541230)
191
+ 181 train 7.664005 (lr=1.9091e-03) (hash(x)=23501481)
192
+ 182 train 7.670851 (lr=1.9196e-03) (hash(x)=22754988)
193
+ 183 train 7.572613 (lr=1.9301e-03) (hash(x)=23475070)
194
+ 184 train 7.726632 (lr=1.9406e-03) (hash(x)=23609959)
195
+ 185 train 7.630090 (lr=1.9510e-03) (hash(x)=23441847)
196
+ 186 train 7.706465 (lr=1.9615e-03) (hash(x)=23096292)
197
+ 187 train 7.737730 (lr=1.9720e-03) (hash(x)=26171093)
198
+ 188 train 7.771799 (lr=1.9825e-03) (hash(x)=20060838)
199
+ 189 train 7.672311 (lr=1.9930e-03) (hash(x)=25451055)
200
+ 190 train 7.593983 (lr=2.0035e-03) (hash(x)=24934792)
201
+ 191 train 7.648713 (lr=2.0140e-03) (hash(x)=25367597)
202
+ 192 train 7.762211 (lr=2.0245e-03) (hash(x)=25022148)
203
+ 193 train 7.717373 (lr=2.0350e-03) (hash(x)=25431365)
204
+ 194 train 7.761075 (lr=2.0455e-03) (hash(x)=29307788)
205
+ 195 train 7.646123 (lr=2.0559e-03) (hash(x)=25651929)
206
+ 196 train 7.724072 (lr=2.0664e-03) (hash(x)=26312387)
207
+ 197 train 7.683270 (lr=2.0769e-03) (hash(x)=23804913)
208
+ 198 train 7.724530 (lr=2.0874e-03) (hash(x)=26115519)
209
+ 199 train 7.700003 (lr=2.0979e-03) (hash(x)=23210747)
210
+ 200 val loss 7.7089
211
+ 200 val perplexity 2228.0920
212
+ 200 train 7.641922 (lr=2.1084e-03) (hash(x)=18463712)
213
+ 201 train 7.682153 (lr=2.1189e-03) (hash(x)=23245699)
214
+ 202 train 7.610382 (lr=2.1294e-03) (hash(x)=21770300)
215
+ 203 train 7.865280 (lr=2.1399e-03) (hash(x)=25387532)
216
+ 204 train 7.675072 (lr=2.1503e-03) (hash(x)=27718703)
217
+ 205 train 7.731622 (lr=2.1608e-03) (hash(x)=27661052)
218
+ 206 train 7.607492 (lr=2.1713e-03) (hash(x)=26538928)
219
+ 207 train 7.701716 (lr=2.1818e-03) (hash(x)=21772654)
220
+ 208 train 7.797286 (lr=2.1923e-03) (hash(x)=26263254)
221
+ 209 train 7.668198 (lr=2.2028e-03) (hash(x)=23552356)
222
+ 210 train 7.863294 (lr=2.2133e-03) (hash(x)=22707141)
223
+ 211 train 7.898393 (lr=2.2238e-03) (hash(x)=34234481)
224
+ 212 train 7.755178 (lr=2.2343e-03) (hash(x)=35008670)
225
+ 213 train 7.876905 (lr=2.2448e-03) (hash(x)=25891543)
226
+ 214 train 7.650317 (lr=2.2552e-03) (hash(x)=21971224)
227
+ 215 train 7.788253 (lr=2.2657e-03) (hash(x)=27312186)
228
+ 216 train 7.860178 (lr=2.2762e-03) (hash(x)=22163749)
229
+ 217 train 7.684685 (lr=2.2867e-03) (hash(x)=25942888)
230
+ 218 train 7.736723 (lr=2.2972e-03) (hash(x)=25493916)
231
+ 219 train 7.964785 (lr=2.3077e-03) (hash(x)=24289951)
232
+ 220 train 7.773687 (lr=2.3182e-03) (hash(x)=23648671)
233
+ 221 train 7.595834 (lr=2.3287e-03) (hash(x)=24992055)
234
+ 222 train 7.923080 (lr=2.3392e-03) (hash(x)=26005549)
235
+ 223 train 7.719576 (lr=2.3497e-03) (hash(x)=21046554)
236
+ 224 train 7.808803 (lr=2.3601e-03) (hash(x)=26321539)
237
+ 225 train 7.830811 (lr=2.3706e-03) (hash(x)=27807872)
238
+ 226 train 7.801832 (lr=2.3811e-03) (hash(x)=21555364)
239
+ 227 train 7.789593 (lr=2.3916e-03) (hash(x)=26325324)
240
+ 228 train 7.788070 (lr=2.4021e-03) (hash(x)=24956294)
241
+ 229 train 7.728257 (lr=2.4126e-03) (hash(x)=24429389)
242
+ 230 train 7.816275 (lr=2.4231e-03) (hash(x)=27929846)
243
+ 231 train 7.770127 (lr=2.4336e-03) (hash(x)=25282897)
244
+ 232 train 7.653672 (lr=2.4441e-03) (hash(x)=26802508)
245
+ 233 train 7.921947 (lr=2.4545e-03) (hash(x)=28390416)
246
+ 234 train 8.141566 (lr=2.4650e-03) (hash(x)=26292451)
247
+ 235 train 7.993920 (lr=2.4755e-03) (hash(x)=22915344)
248
+ 236 train 7.900838 (lr=2.4860e-03) (hash(x)=24744383)
249
+ 237 train 7.757616 (lr=2.4965e-03) (hash(x)=26284678)
250
+ 238 train 7.701021 (lr=2.5070e-03) (hash(x)=21654269)
251
+ 239 train 8.127934 (lr=2.5175e-03) (hash(x)=29466268)
252
+ 240 train 8.333839 (lr=2.5280e-03) (hash(x)=32616762)
253
+ 241 train 8.301709 (lr=2.5385e-03) (hash(x)=32201138)
254
+ 242 train 8.319250 (lr=2.5490e-03) (hash(x)=33683468)
255
+ 243 train 8.061486 (lr=2.5594e-03) (hash(x)=27498125)
256
+ 244 train 8.040285 (lr=2.5699e-03) (hash(x)=25164318)
257
+ 245 train 7.842455 (lr=2.5804e-03) (hash(x)=25690221)
258
+ 246 train 7.897824 (lr=2.5909e-03) (hash(x)=29031550)
259
+ 247 train 8.076825 (lr=2.6014e-03) (hash(x)=24094093)
260
+ 248 train 8.844463 (lr=2.6119e-03) (hash(x)=61455960)
261
+ 249 train 12.673602 (lr=2.6224e-03) (hash(x)=61104710)
262
+ 250 val loss 8.0514
263
+ 250 val perplexity 3138.1499
264
+ 250 train 8.216370 (lr=2.6329e-03) (hash(x)=20881126)
265
+ 251 train 8.069713 (lr=2.6434e-03) (hash(x)=26920243)
266
+ 252 train 7.911462 (lr=2.6538e-03) (hash(x)=21970343)
267
+ 253 train 7.979985 (lr=2.6643e-03) (hash(x)=31676352)
268
+ 254 train 7.899180 (lr=2.6748e-03) (hash(x)=25600562)
269
+ 255 train 7.883381 (lr=2.6853e-03) (hash(x)=22022924)
270
+ 256 train 7.768429 (lr=2.6958e-03) (hash(x)=22777085)
271
+ 257 train 7.870378 (lr=2.7063e-03) (hash(x)=21727744)
272
+ 258 train 7.853065 (lr=2.7168e-03) (hash(x)=20701998)
273
+ 259 train 7.912295 (lr=2.7273e-03) (hash(x)=20863982)
274
+ 260 train 7.820481 (lr=2.7378e-03) (hash(x)=25084926)
275
+ 261 train 7.831124 (lr=2.7483e-03) (hash(x)=24412897)
276
+ 262 train 7.840290 (lr=2.7587e-03) (hash(x)=25434913)
277
+ 263 train 7.822653 (lr=2.7692e-03) (hash(x)=26302544)
278
+ 264 train 7.865217 (lr=2.7797e-03) (hash(x)=34976197)
279
+ 265 train 7.728319 (lr=2.7902e-03) (hash(x)=23870805)
280
+ 266 train 7.762221 (lr=2.8007e-03) (hash(x)=25041103)
281
+ 267 train 7.817769 (lr=2.8112e-03) (hash(x)=26041745)
282
+ 268 train 7.802280 (lr=2.8217e-03) (hash(x)=24068619)
283
+ 269 train 7.831430 (lr=2.8322e-03) (hash(x)=25441262)
284
+ 270 train 7.892151 (lr=2.8427e-03) (hash(x)=29181867)
285
+ 271 train 7.862317 (lr=2.8531e-03) (hash(x)=27093757)
286
+ 272 train 7.769066 (lr=2.8636e-03) (hash(x)=25150275)
287
+ 273 train 7.900351 (lr=2.8741e-03) (hash(x)=33457102)
288
+ 274 train 7.703664 (lr=2.8846e-03) (hash(x)=25755322)
289
+ 275 train 7.782068 (lr=2.8951e-03) (hash(x)=23630794)
290
+ 276 train 7.704752 (lr=2.9056e-03) (hash(x)=25167987)
291
+ 277 train 7.848352 (lr=2.9161e-03) (hash(x)=24043224)
292
+ 278 train 7.832450 (lr=2.9266e-03) (hash(x)=24665411)
293
+ 279 train 7.810645 (lr=2.9371e-03) (hash(x)=17321657)
294
+ 280 train 7.817974 (lr=2.9476e-03) (hash(x)=27932591)
295
+ 281 train 8.104517 (lr=2.9580e-03) (hash(x)=29447954)
296
+ 282 train 7.855990 (lr=2.9685e-03) (hash(x)=24621816)
297
+ 283 train 7.832555 (lr=2.9790e-03) (hash(x)=24682294)
298
+ 284 train 7.840750 (lr=2.9895e-03) (hash(x)=28057095)
299
+ 285 train 7.838946 (lr=3.0000e-03) (hash(x)=24547536)
300
+ 286 train 7.688097 (lr=3.0000e-03) (hash(x)=18987545)
301
+ 287 train 7.747737 (lr=3.0000e-03) (hash(x)=26745063)
302
+ 288 train 7.625693 (lr=2.9999e-03) (hash(x)=21447406)
303
+ 289 train 7.544175 (lr=2.9999e-03) (hash(x)=24650655)
304
+ 290 train 7.923831 (lr=2.9998e-03) (hash(x)=26519976)
305
+ 291 train 8.265354 (lr=2.9997e-03) (hash(x)=20106514)
306
+ 292 train 7.699979 (lr=2.9995e-03) (hash(x)=20117808)
307
+ 293 train 7.848550 (lr=2.9994e-03) (hash(x)=25510334)
308
+ 294 train 7.785791 (lr=2.9992e-03) (hash(x)=25749120)
309
+ 295 train 7.906401 (lr=2.9989e-03) (hash(x)=24357241)
310
+ 296 train 7.915703 (lr=2.9987e-03) (hash(x)=25584945)
311
+ 297 train 7.770537 (lr=2.9984e-03) (hash(x)=25198897)
312
+ 298 train 7.843851 (lr=2.9981e-03) (hash(x)=23475891)
313
+ 299 train 7.778366 (lr=2.9978e-03) (hash(x)=23644804)
314
+ 300 val loss 7.8621
315
+ 300 val perplexity 2597.0713
316
+ 300 train 8.026264 (lr=2.9974e-03) (hash(x)=25629584)
317
+ 301 train 8.069184 (lr=2.9971e-03) (hash(x)=29797280)
318
+ 302 train 8.150885 (lr=2.9967e-03) (hash(x)=30026394)
319
+ 303 train 8.116756 (lr=2.9962e-03) (hash(x)=26594177)
320
+ 304 train 7.944718 (lr=2.9958e-03) (hash(x)=23766256)
321
+ 305 train 7.877314 (lr=2.9953e-03) (hash(x)=24250150)
322
+ 306 train 7.839515 (lr=2.9948e-03) (hash(x)=26108817)
323
+ 307 train 7.783179 (lr=2.9942e-03) (hash(x)=18582688)
324
+ 308 train 7.786873 (lr=2.9937e-03) (hash(x)=23272705)
325
+ 309 train 7.788774 (lr=2.9931e-03) (hash(x)=24793412)
326
+ 310 train 7.774815 (lr=2.9925e-03) (hash(x)=19280989)
327
+ 311 train 7.760036 (lr=2.9918e-03) (hash(x)=22751150)
328
+ 312 train 7.853015 (lr=2.9912e-03) (hash(x)=23537942)
329
+ 313 train 7.700042 (lr=2.9905e-03) (hash(x)=23748375)
330
+ 314 train 7.855317 (lr=2.9898e-03) (hash(x)=27413825)
331
+ 315 train 7.889909 (lr=2.9890e-03) (hash(x)=21682445)
332
+ 316 train 7.753576 (lr=2.9883e-03) (hash(x)=22040184)
333
+ 317 train 7.744501 (lr=2.9875e-03) (hash(x)=24463229)
334
+ 318 train 7.824966 (lr=2.9866e-03) (hash(x)=24091954)
335
+ 319 train 7.716737 (lr=2.9858e-03) (hash(x)=26670219)
336
+ 320 train 7.730119 (lr=2.9849e-03) (hash(x)=25558986)
337
+ 321 train 7.721037 (lr=2.9840e-03) (hash(x)=19218943)
338
+ 322 train 7.795905 (lr=2.9831e-03) (hash(x)=24551402)
339
+ 323 train 7.702512 (lr=2.9821e-03) (hash(x)=25229281)
340
+ 324 train 7.511373 (lr=2.9812e-03) (hash(x)=21251127)
341
+ 325 train 7.800197 (lr=2.9802e-03) (hash(x)=25800857)
342
+ 326 train 7.696661 (lr=2.9791e-03) (hash(x)=25836189)
343
+ 327 train 7.706052 (lr=2.9781e-03) (hash(x)=25999255)
344
+ 328 train 7.754915 (lr=2.9770e-03) (hash(x)=20149394)
345
+ 329 train 7.713728 (lr=2.9759e-03) (hash(x)=21999338)
346
+ 330 train 7.807253 (lr=2.9748e-03) (hash(x)=27611302)
347
+ 331 train 7.759533 (lr=2.9736e-03) (hash(x)=22778356)
348
+ 332 train 7.732005 (lr=2.9724e-03) (hash(x)=24195959)
349
+ 333 train 7.970845 (lr=2.9712e-03) (hash(x)=19127465)
350
+ 334 train 7.837751 (lr=2.9700e-03) (hash(x)=22426274)
351
+ 335 train 7.670108 (lr=2.9687e-03) (hash(x)=23785205)
352
+ 336 train 7.816744 (lr=2.9675e-03) (hash(x)=28505676)
353
+ 337 train 7.802863 (lr=2.9662e-03) (hash(x)=27706477)
354
+ 338 train 7.793706 (lr=2.9648e-03) (hash(x)=28684175)
355
+ 339 train 7.777014 (lr=2.9635e-03) (hash(x)=28003600)
356
+ 340 train 7.806533 (lr=2.9621e-03) (hash(x)=27586845)
357
+ 341 train 7.739265 (lr=2.9607e-03) (hash(x)=21176405)
358
+ 342 train 7.699794 (lr=2.9592e-03) (hash(x)=25888433)
359
+ 343 train 7.821711 (lr=2.9578e-03) (hash(x)=27916982)
360
+ 344 train 7.669634 (lr=2.9563e-03) (hash(x)=24334708)
361
+ 345 train 7.721970 (lr=2.9548e-03) (hash(x)=25617709)
362
+ 346 train 7.881114 (lr=2.9532e-03) (hash(x)=25152362)
363
+ 347 train 7.737413 (lr=2.9517e-03) (hash(x)=24276512)
364
+ 348 train 7.687140 (lr=2.9501e-03) (hash(x)=26776133)
365
+ 349 train 7.618090 (lr=2.9485e-03) (hash(x)=20254159)
366
+ 350 val loss 7.8242
367
+ 350 val perplexity 2500.2637
368
+ 350 train 7.700965 (lr=2.9468e-03) (hash(x)=27873855)
369
+ 351 train 7.808689 (lr=2.9452e-03) (hash(x)=25113614)
370
+ 352 train 7.830647 (lr=2.9435e-03) (hash(x)=26192886)
371
+ 353 train 7.741083 (lr=2.9418e-03) (hash(x)=24614912)
372
+ 354 train 7.743305 (lr=2.9400e-03) (hash(x)=25397093)
373
+ 355 train 7.750391 (lr=2.9383e-03) (hash(x)=19811802)
374
+ 356 train 7.747838 (lr=2.9365e-03) (hash(x)=25579655)
375
+ 357 train 7.887118 (lr=2.9347e-03) (hash(x)=26096514)
376
+ 358 train 7.825898 (lr=2.9328e-03) (hash(x)=22589633)
377
+ 359 train 7.825251 (lr=2.9310e-03) (hash(x)=26757147)
378
+ 360 train 7.862752 (lr=2.9291e-03) (hash(x)=24963730)
379
+ 361 train 7.718019 (lr=2.9272e-03) (hash(x)=27703124)
380
+ 362 train 7.665376 (lr=2.9252e-03) (hash(x)=24781713)
381
+ 363 train 7.812938 (lr=2.9233e-03) (hash(x)=24720476)
382
+ 364 train 7.786627 (lr=2.9213e-03) (hash(x)=23481301)
383
+ 365 train 7.732196 (lr=2.9193e-03) (hash(x)=25641033)
384
+ 366 train 7.808174 (lr=2.9172e-03) (hash(x)=27369437)
385
+ 367 train 7.883943 (lr=2.9152e-03) (hash(x)=24760900)
386
+ 368 train 7.709289 (lr=2.9131e-03) (hash(x)=21692546)
387
+ 369 train 7.802932 (lr=2.9110e-03) (hash(x)=22720391)
388
+ 370 train 7.833300 (lr=2.9088e-03) (hash(x)=24376979)
389
+ 371 train 7.988560 (lr=2.9067e-03) (hash(x)=22244509)
390
+ 372 train 7.951905 (lr=2.9045e-03) (hash(x)=21831960)
391
+ 373 train 7.938270 (lr=2.9023e-03) (hash(x)=24921535)
392
+ 374 train 7.977479 (lr=2.9001e-03) (hash(x)=25381251)
393
+ 375 train 7.943099 (lr=2.8978e-03) (hash(x)=25687352)
394
+ 376 train 7.960665 (lr=2.8955e-03) (hash(x)=25221654)
395
+ 377 train 7.962771 (lr=2.8932e-03) (hash(x)=24469863)
396
+ 378 train 7.955970 (lr=2.8909e-03) (hash(x)=25291413)
397
+ 379 train 7.849071 (lr=2.8885e-03) (hash(x)=25744274)
398
+ 380 train 7.750138 (lr=2.8862e-03) (hash(x)=24246963)
399
+ 381 train 7.737564 (lr=2.8838e-03) (hash(x)=19724058)
400
+ 382 train 7.659513 (lr=2.8813e-03) (hash(x)=20705649)
401
+ 383 train 7.958198 (lr=2.8789e-03) (hash(x)=24938685)
402
+ 384 train 7.977185 (lr=2.8764e-03) (hash(x)=21707000)
403
+ 385 train 7.730520 (lr=2.8739e-03) (hash(x)=24948980)
404
+ 386 train 8.010759 (lr=2.8714e-03) (hash(x)=30773958)
405
+ 387 train 8.115071 (lr=2.8689e-03) (hash(x)=26769046)
406
+ 388 train 7.722587 (lr=2.8663e-03) (hash(x)=18849656)
407
+ 389 train 8.223747 (lr=2.8637e-03) (hash(x)=24624950)
408
+ 390 train 7.816851 (lr=2.8611e-03) (hash(x)=27299605)
409
+ 391 train 7.824122 (lr=2.8585e-03) (hash(x)=24113445)
410
+ 392 train 7.824229 (lr=2.8558e-03) (hash(x)=32644465)
411
+ 393 train 7.964884 (lr=2.8531e-03) (hash(x)=25829219)
412
+ 394 train 7.807649 (lr=2.8504e-03) (hash(x)=25194550)
413
+ 395 train 7.855191 (lr=2.8477e-03) (hash(x)=25941804)
414
+ 396 train 7.824540 (lr=2.8449e-03) (hash(x)=24075727)
415
+ 397 train 7.879422 (lr=2.8422e-03) (hash(x)=23104539)
416
+ 398 train 7.814933 (lr=2.8394e-03) (hash(x)=23018983)
417
+ 399 train 7.738317 (lr=2.8365e-03) (hash(x)=24422929)
418
+ 400 val loss 7.7895
419
+ 400 val perplexity 2415.1045
420
+ 400 train 7.754927 (lr=2.8337e-03) (hash(x)=23389696)
421
+ 401 train 7.717456 (lr=2.8308e-03) (hash(x)=29102969)
422
+ 402 train 7.878465 (lr=2.8279e-03) (hash(x)=28359859)
423
+ 403 train 7.865843 (lr=2.8250e-03) (hash(x)=24630955)
424
+ 404 train 7.918598 (lr=2.8221e-03) (hash(x)=22969113)
425
+ 405 train 7.767677 (lr=2.8191e-03) (hash(x)=27004415)
426
+ 406 train 7.701919 (lr=2.8162e-03) (hash(x)=22265063)
427
+ 407 train 7.735119 (lr=2.8131e-03) (hash(x)=26905990)
428
+ 408 train 7.749963 (lr=2.8101e-03) (hash(x)=24047679)
429
+ 409 train 7.894097 (lr=2.8071e-03) (hash(x)=26905582)
430
+ 410 train 7.662417 (lr=2.8040e-03) (hash(x)=29018970)
431
+ 411 train 7.701310 (lr=2.8009e-03) (hash(x)=24503315)
432
+ 412 train 7.687570 (lr=2.7978e-03) (hash(x)=22607537)
433
+ 413 train 8.411502 (lr=2.7947e-03) (hash(x)=26690225)
434
+ 414 train 7.776808 (lr=2.7915e-03) (hash(x)=24226255)
435
+ 415 train 7.728085 (lr=2.7883e-03) (hash(x)=23622702)
436
+ 416 train 7.809415 (lr=2.7851e-03) (hash(x)=23910425)
437
+ 417 train 8.029179 (lr=2.7819e-03) (hash(x)=27213318)
438
+ 418 train 7.852592 (lr=2.7786e-03) (hash(x)=26701577)
439
+ 419 train 7.759164 (lr=2.7754e-03) (hash(x)=26216794)
440
+ 420 train 7.927636 (lr=2.7721e-03) (hash(x)=27579623)
441
+ 421 train 7.791556 (lr=2.7688e-03) (hash(x)=26086198)
442
+ 422 train 7.794762 (lr=2.7654e-03) (hash(x)=24863441)
443
+ 423 train 8.016621 (lr=2.7621e-03) (hash(x)=25236385)
444
+ 424 train 8.087210 (lr=2.7587e-03) (hash(x)=25330803)
445
+ 425 train 7.956141 (lr=2.7553e-03) (hash(x)=22508532)
446
+ 426 train 7.815234 (lr=2.7519e-03) (hash(x)=21799261)
447
+ 427 train 7.872975 (lr=2.7484e-03) (hash(x)=26311168)
448
+ 428 train 7.756279 (lr=2.7450e-03) (hash(x)=20977430)
449
+ 429 train 7.995820 (lr=2.7415e-03) (hash(x)=26941617)
450
+ 430 train 8.319243 (lr=2.7380e-03) (hash(x)=30008957)
451
+ 431 train 8.370121 (lr=2.7344e-03) (hash(x)=27125962)
452
+ 432 train 8.376110 (lr=2.7309e-03) (hash(x)=25888358)
453
+ 433 train 8.622316 (lr=2.7273e-03) (hash(x)=21725719)
454
+ 434 train 8.493838 (lr=2.7237e-03) (hash(x)=21669704)
455
+ 435 train 8.185451 (lr=2.7201e-03) (hash(x)=23166993)
456
+ 436 train 8.027046 (lr=2.7165e-03) (hash(x)=26235974)
457
+ 437 train 8.070374 (lr=2.7128e-03) (hash(x)=25426430)
458
+ 438 train 7.744142 (lr=2.7092e-03) (hash(x)=22966314)
459
+ 439 train 7.874523 (lr=2.7055e-03) (hash(x)=23448505)
460
+ 440 train 7.767329 (lr=2.7018e-03) (hash(x)=26330693)
461
+ 441 train 7.751914 (lr=2.6980e-03) (hash(x)=29550596)
462
+ 442 train 7.778040 (lr=2.6943e-03) (hash(x)=22124892)
463
+ 443 train 7.707474 (lr=2.6905e-03) (hash(x)=25607640)
464
+ 444 train 7.626467 (lr=2.6867e-03) (hash(x)=23795894)
465
+ 445 train 7.547981 (lr=2.6829e-03) (hash(x)=24165449)
466
+ 446 train 7.876297 (lr=2.6791e-03) (hash(x)=24697756)
467
+ 447 train 7.884721 (lr=2.6752e-03) (hash(x)=27218899)
468
+ 448 train 7.719600 (lr=2.6713e-03) (hash(x)=24748522)
469
+ 449 train 7.786112 (lr=2.6674e-03) (hash(x)=24965317)
470
+ 450 val loss 7.8097
471
+ 450 val perplexity 2464.5073
472
+ 450 train 7.546592 (lr=2.6635e-03) (hash(x)=22031210)
473
+ 451 train 7.734981 (lr=2.6596e-03) (hash(x)=25877990)
474
+ 452 train 7.800375 (lr=2.6556e-03) (hash(x)=24382381)
475
+ 453 train 7.717591 (lr=2.6517e-03) (hash(x)=24551999)
476
+ 454 train 7.774865 (lr=2.6477e-03) (hash(x)=25765516)
477
+ 455 train 7.860291 (lr=2.6437e-03) (hash(x)=25204247)
478
+ 456 train 7.871866 (lr=2.6396e-03) (hash(x)=26321813)
479
+ 457 train 7.764339 (lr=2.6356e-03) (hash(x)=25528323)
480
+ 458 train 7.762071 (lr=2.6315e-03) (hash(x)=25114729)
481
+ 459 train 7.728159 (lr=2.6274e-03) (hash(x)=31341964)
482
+ 460 train 7.845635 (lr=2.6233e-03) (hash(x)=26326024)
483
+ 461 train 7.707335 (lr=2.6192e-03) (hash(x)=27273311)
484
+ 462 train 7.662410 (lr=2.6150e-03) (hash(x)=17819313)
485
+ 463 train 7.903183 (lr=2.6109e-03) (hash(x)=23769243)
486
+ 464 train 7.809059 (lr=2.6067e-03) (hash(x)=23788738)
487
+ 465 train 7.766994 (lr=2.6025e-03) (hash(x)=24818744)
488
+ 466 train 7.689235 (lr=2.5983e-03) (hash(x)=23472795)
489
+ 467 train 8.132747 (lr=2.5940e-03) (hash(x)=24991832)
490
+ 468 train 7.826287 (lr=2.5898e-03) (hash(x)=22424066)
491
+ 469 train 7.743762 (lr=2.5855e-03) (hash(x)=26474878)
492
+ 470 train 7.804330 (lr=2.5812e-03) (hash(x)=22704349)
493
+ 471 train 7.763525 (lr=2.5769e-03) (hash(x)=22293673)
494
+ 472 train 7.734500 (lr=2.5726e-03) (hash(x)=21785487)
495
+ 473 train 7.756430 (lr=2.5682e-03) (hash(x)=25251063)
496
+ 474 train 7.662586 (lr=2.5639e-03) (hash(x)=22722211)
497
+ 475 train 7.729473 (lr=2.5595e-03) (hash(x)=24522269)
498
+ 476 train 7.674975 (lr=2.5551e-03) (hash(x)=23329475)
499
+ 477 train 7.726405 (lr=2.5507e-03) (hash(x)=24735165)
500
+ 478 train 7.571283 (lr=2.5462e-03) (hash(x)=27002892)
501
+ 479 train 7.607370 (lr=2.5418e-03) (hash(x)=23499349)
502
+ 480 train 7.923921 (lr=2.5373e-03) (hash(x)=22461589)
503
+ 481 train 7.694814 (lr=2.5328e-03) (hash(x)=26415040)
504
+ 482 train 7.930052 (lr=2.5283e-03) (hash(x)=26349465)
505
+ 483 train 7.604821 (lr=2.5238e-03) (hash(x)=21596677)
506
+ 484 train 7.672394 (lr=2.5193e-03) (hash(x)=24117012)
507
+ 485 train 7.757414 (lr=2.5147e-03) (hash(x)=24105761)
508
+ 486 train 7.717448 (lr=2.5102e-03) (hash(x)=25725049)
509
+ 487 train 7.556259 (lr=2.5056e-03) (hash(x)=22892776)
510
+ 488 train 7.674599 (lr=2.5010e-03) (hash(x)=21630383)
511
+ 489 train 7.737611 (lr=2.4963e-03) (hash(x)=24727094)
512
+ 490 train 7.648602 (lr=2.4917e-03) (hash(x)=22835000)
513
+ 491 train 7.811471 (lr=2.4871e-03) (hash(x)=22839049)
514
+ 492 train 7.630094 (lr=2.4824e-03) (hash(x)=20654136)
515
+ 493 train 7.838586 (lr=2.4777e-03) (hash(x)=26776000)
516
+ 494 train 7.448941 (lr=2.4730e-03) (hash(x)=22560658)
517
+ 495 train 7.901772 (lr=2.4683e-03) (hash(x)=25995933)
518
+ 496 train 7.832856 (lr=2.4636e-03) (hash(x)=24635394)
519
+ 497 train 7.925488 (lr=2.4588e-03) (hash(x)=27537228)
520
+ 498 train 7.825522 (lr=2.4540e-03) (hash(x)=26496730)
521
+ 499 train 7.520242 (lr=2.4493e-03) (hash(x)=16533310)
522
+ 500 val loss 7.7620
523
+ 500 val perplexity 2349.5042
524
+ 500 train 7.666015 (lr=2.4445e-03) (hash(x)=24881176)
525
+ 501 train 7.859178 (lr=2.4397e-03) (hash(x)=25593613)
526
+ 502 train 7.625164 (lr=2.4348e-03) (hash(x)=22886646)
527
+ 503 train 7.864233 (lr=2.4300e-03) (hash(x)=21407676)
528
+ 504 train 7.751015 (lr=2.4251e-03) (hash(x)=21720304)
529
+ 505 train 7.845598 (lr=2.4203e-03) (hash(x)=27263416)
530
+ 506 train 7.630919 (lr=2.4154e-03) (hash(x)=25840801)
531
+ 507 train 7.779041 (lr=2.4105e-03) (hash(x)=31424568)
532
+ 508 train 7.768766 (lr=2.4056e-03) (hash(x)=23403065)
533
+ 509 train 7.588913 (lr=2.4006e-03) (hash(x)=22656033)
534
+ 510 train 7.756633 (lr=2.3957e-03) (hash(x)=26869295)
535
+ 511 train 7.958341 (lr=2.3907e-03) (hash(x)=25504151)
536
+ 512 train 7.760150 (lr=2.3858e-03) (hash(x)=25781518)
537
+ 513 train 7.717383 (lr=2.3808e-03) (hash(x)=26432850)
538
+ 514 train 7.630435 (lr=2.3758e-03) (hash(x)=20155894)
539
+ 515 train 7.554812 (lr=2.3708e-03) (hash(x)=22270169)
540
+ 516 train 7.681725 (lr=2.3657e-03) (hash(x)=22108461)
541
+ 517 train 7.735168 (lr=2.3607e-03) (hash(x)=24802848)
542
+ 518 train 7.834816 (lr=2.3556e-03) (hash(x)=25086604)
543
+ 519 train 7.834145 (lr=2.3506e-03) (hash(x)=26983186)
544
+ 520 train 7.714084 (lr=2.3455e-03) (hash(x)=27208507)
545
+ 521 train 7.870317 (lr=2.3404e-03) (hash(x)=23862845)
546
+ 522 train 7.689739 (lr=2.3353e-03) (hash(x)=22664494)
547
+ 523 train 7.683052 (lr=2.3301e-03) (hash(x)=24092927)
548
+ 524 train 7.676261 (lr=2.3250e-03) (hash(x)=27213812)
549
+ 525 train 7.624788 (lr=2.3198e-03) (hash(x)=23929955)
550
+ 526 train 7.633423 (lr=2.3147e-03) (hash(x)=23025790)
551
+ 527 train 7.565784 (lr=2.3095e-03) (hash(x)=20452272)
552
+ 528 train 7.773628 (lr=2.3043e-03) (hash(x)=26473464)
553
+ 529 train 7.757082 (lr=2.2991e-03) (hash(x)=28276603)
554
+ 530 train 7.598224 (lr=2.2939e-03) (hash(x)=21480367)
555
+ 531 train 7.781559 (lr=2.2887e-03) (hash(x)=24658642)
556
+ 532 train 7.736247 (lr=2.2834e-03) (hash(x)=26854801)
557
+ 533 train 7.602934 (lr=2.2782e-03) (hash(x)=17862226)
558
+ 534 train 7.701697 (lr=2.2729e-03) (hash(x)=28519182)
559
+ 535 train 7.674885 (lr=2.2676e-03) (hash(x)=25193694)
560
+ 536 train 7.685039 (lr=2.2624e-03) (hash(x)=25642620)
561
+ 537 train 7.760102 (lr=2.2571e-03) (hash(x)=25765909)
562
+ 538 train 7.546041 (lr=2.2517e-03) (hash(x)=20750531)
563
+ 539 train 7.671123 (lr=2.2464e-03) (hash(x)=24917098)
564
+ 540 train 7.662470 (lr=2.2411e-03) (hash(x)=24499766)
565
+ 541 train 7.723074 (lr=2.2357e-03) (hash(x)=25067358)
566
+ 542 train 7.819542 (lr=2.2304e-03) (hash(x)=27479796)
567
+ 543 train 7.674027 (lr=2.2250e-03) (hash(x)=23136080)
568
+ 544 train 7.791669 (lr=2.2196e-03) (hash(x)=29810074)
569
+ 545 train 7.697304 (lr=2.2142e-03) (hash(x)=24765539)
570
+ 546 train 7.716868 (lr=2.2088e-03) (hash(x)=22216309)
571
+ 547 train 7.599475 (lr=2.2034e-03) (hash(x)=20613861)
572
+ 548 train 7.575364 (lr=2.1980e-03) (hash(x)=23462447)
573
+ 549 train 7.640743 (lr=2.1926e-03) (hash(x)=26739455)
574
+ 550 val loss 7.7296
575
+ 550 val perplexity 2274.7952
576
+ 550 train 7.702263 (lr=2.1871e-03) (hash(x)=23222996)
577
+ 551 train 7.687574 (lr=2.1817e-03) (hash(x)=26489285)
578
+ 552 train 7.542457 (lr=2.1762e-03) (hash(x)=25185942)
579
+ 553 train 7.831305 (lr=2.1707e-03) (hash(x)=27476158)
580
+ 554 train 7.632999 (lr=2.1653e-03) (hash(x)=23653175)
581
+ 555 train 7.598640 (lr=2.1598e-03) (hash(x)=22405076)
582
+ 556 train 7.624904 (lr=2.1542e-03) (hash(x)=25673545)
583
+ 557 train 7.688928 (lr=2.1487e-03) (hash(x)=22675428)
584
+ 558 train 7.816274 (lr=2.1432e-03) (hash(x)=28168456)
585
+ 559 train 7.907063 (lr=2.1377e-03) (hash(x)=26072586)
586
+ 560 train 7.686351 (lr=2.1321e-03) (hash(x)=24974331)
587
+ 561 train 7.448319 (lr=2.1266e-03) (hash(x)=21951984)
588
+ 562 train 7.846814 (lr=2.1210e-03) (hash(x)=25403751)
589
+ 563 train 7.664732 (lr=2.1154e-03) (hash(x)=26396012)
590
+ 564 train 7.609475 (lr=2.1099e-03) (hash(x)=24827190)
591
+ 565 train 7.733538 (lr=2.1043e-03) (hash(x)=24847458)
592
+ 566 train 7.706488 (lr=2.0987e-03) (hash(x)=22159088)
593
+ 567 train 7.661698 (lr=2.0931e-03) (hash(x)=22140034)
594
+ 568 train 7.738027 (lr=2.0875e-03) (hash(x)=19044797)
595
+ 569 train 7.646413 (lr=2.0818e-03) (hash(x)=23932794)
596
+ 570 train 7.736259 (lr=2.0762e-03) (hash(x)=25256890)
597
+ 571 train 7.766581 (lr=2.0706e-03) (hash(x)=24161711)
598
+ 572 train 7.694524 (lr=2.0649e-03) (hash(x)=22036817)
599
+ 573 train 7.576969 (lr=2.0593e-03) (hash(x)=26663208)
600
+ 574 train 7.608678 (lr=2.0536e-03) (hash(x)=24072213)
601
+ 575 train 7.588478 (lr=2.0479e-03) (hash(x)=22766044)
602
+ 576 train 7.717205 (lr=2.0422e-03) (hash(x)=27871153)
603
+ 577 train 7.655800 (lr=2.0366e-03) (hash(x)=24037280)
604
+ 578 train 7.773052 (lr=2.0309e-03) (hash(x)=24121850)
605
+ 579 train 7.656820 (lr=2.0252e-03) (hash(x)=24380031)
606
+ 580 train 7.568013 (lr=2.0194e-03) (hash(x)=23055215)
607
+ 581 train 7.824622 (lr=2.0137e-03) (hash(x)=25329724)
608
+ 582 train 7.574950 (lr=2.0080e-03) (hash(x)=25662408)
609
+ 583 train 7.669730 (lr=2.0023e-03) (hash(x)=26346814)
610
+ 584 train 7.763149 (lr=1.9965e-03) (hash(x)=25283256)
611
+ 585 train 7.575156 (lr=1.9908e-03) (hash(x)=24021771)
612
+ 586 train 7.616453 (lr=1.9850e-03) (hash(x)=23576840)
613
+ 587 train 7.746226 (lr=1.9793e-03) (hash(x)=26414858)
614
+ 588 train 7.524205 (lr=1.9735e-03) (hash(x)=20099261)
615
+ 589 train 7.526429 (lr=1.9677e-03) (hash(x)=22453718)
616
+ 590 train 7.476915 (lr=1.9620e-03) (hash(x)=22529445)
617
+ 591 train 7.549730 (lr=1.9562e-03) (hash(x)=24484543)
618
+ 592 train 7.487180 (lr=1.9504e-03) (hash(x)=24043998)
619
+ 593 train 7.624408 (lr=1.9446e-03) (hash(x)=24301906)
620
+ 594 train 7.578555 (lr=1.9388e-03) (hash(x)=26328881)
621
+ 595 train 7.637074 (lr=1.9330e-03) (hash(x)=23101508)
622
+ 596 train 7.598261 (lr=1.9272e-03) (hash(x)=23527176)
623
+ 597 train 8.070280 (lr=1.9214e-03) (hash(x)=22705673)
624
+ 598 train 7.641760 (lr=1.9156e-03) (hash(x)=28664796)
625
+ 599 train 7.756497 (lr=1.9097e-03) (hash(x)=27102890)
626
+ 600 val loss 7.6776
627
+ 600 val perplexity 2159.4553
628
+ 600 train 7.841616 (lr=1.9039e-03) (hash(x)=26787259)
629
+ 601 train 7.874375 (lr=1.8981e-03) (hash(x)=23396088)
630
+ 602 train 7.583198 (lr=1.8922e-03) (hash(x)=24911831)
631
+ 603 train 7.606864 (lr=1.8864e-03) (hash(x)=23453491)
632
+ 604 train 7.665811 (lr=1.8805e-03) (hash(x)=23453396)
633
+ 605 train 7.743348 (lr=1.8747e-03) (hash(x)=23869612)
634
+ 606 train 7.854315 (lr=1.8688e-03) (hash(x)=26581590)
635
+ 607 train 7.661780 (lr=1.8629e-03) (hash(x)=28056342)
636
+ 608 train 7.718257 (lr=1.8571e-03) (hash(x)=25900866)
637
+ 609 train 7.742697 (lr=1.8512e-03) (hash(x)=24549177)
638
+ 610 train 7.661785 (lr=1.8453e-03) (hash(x)=28239583)
639
+ 611 train 7.668579 (lr=1.8395e-03) (hash(x)=23223554)
640
+ 612 train 7.722329 (lr=1.8336e-03) (hash(x)=24793480)
641
+ 613 train 7.658380 (lr=1.8277e-03) (hash(x)=25415570)
642
+ 614 train 7.695836 (lr=1.8218e-03) (hash(x)=24425760)
643
+ 615 train 7.774507 (lr=1.8159e-03) (hash(x)=23707134)
644
+ 616 train 7.689510 (lr=1.8100e-03) (hash(x)=28321697)
645
+ 617 train 7.643762 (lr=1.8041e-03) (hash(x)=23434031)
646
+ 618 train 7.620522 (lr=1.7982e-03) (hash(x)=27003481)
647
+ 619 train 7.584896 (lr=1.7923e-03) (hash(x)=22236893)
648
+ 620 train 7.653390 (lr=1.7864e-03) (hash(x)=26857458)
649
+ 621 train 7.634339 (lr=1.7805e-03) (hash(x)=26413122)
650
+ 622 train 7.927439 (lr=1.7746e-03) (hash(x)=27696537)
651
+ 623 train 7.666698 (lr=1.7686e-03) (hash(x)=21950234)
652
+ 624 train 7.697772 (lr=1.7627e-03) (hash(x)=24051952)
653
+ 625 train 7.710134 (lr=1.7568e-03) (hash(x)=24029939)
654
+ 626 train 7.831930 (lr=1.7509e-03) (hash(x)=24393346)
655
+ 627 train 7.735649 (lr=1.7450e-03) (hash(x)=26985352)
656
+ 628 train 7.442997 (lr=1.7390e-03) (hash(x)=19493265)
657
+ 629 train 7.574813 (lr=1.7331e-03) (hash(x)=23944010)
658
+ 630 train 7.800852 (lr=1.7272e-03) (hash(x)=21517435)
659
+ 631 train 7.555831 (lr=1.7212e-03) (hash(x)=25025868)
660
+ 632 train 7.586493 (lr=1.7153e-03) (hash(x)=22940997)
661
+ 633 train 7.735087 (lr=1.7094e-03) (hash(x)=25864840)
662
+ 634 train 7.632767 (lr=1.7034e-03) (hash(x)=24682151)
663
+ 635 train 7.549878 (lr=1.6975e-03) (hash(x)=27055330)
664
+ 636 train 7.614662 (lr=1.6916e-03) (hash(x)=22705710)
665
+ 637 train 7.687128 (lr=1.6856e-03) (hash(x)=26775238)
666
+ 638 train 7.790126 (lr=1.6797e-03) (hash(x)=21734331)
667
+ 639 train 7.600172 (lr=1.6738e-03) (hash(x)=25693442)
668
+ 640 train 7.679729 (lr=1.6678e-03) (hash(x)=25290454)
669
+ 641 train 7.634040 (lr=1.6619e-03) (hash(x)=25699860)
670
+ 642 train 7.810045 (lr=1.6559e-03) (hash(x)=19919115)
671
+ 643 train 7.852724 (lr=1.6500e-03) (hash(x)=25586967)
672
+ 644 train 7.620392 (lr=1.6441e-03) (hash(x)=25373506)
673
+ 645 train 7.636082 (lr=1.6381e-03) (hash(x)=26496148)
674
+ 646 train 7.692169 (lr=1.6322e-03) (hash(x)=27395083)
675
+ 647 train 7.869017 (lr=1.6262e-03) (hash(x)=27389709)
676
+ 648 train 7.762518 (lr=1.6203e-03) (hash(x)=29172968)
677
+ 649 train 7.688774 (lr=1.6144e-03) (hash(x)=25464920)
678
+ 650 val loss 7.6498
679
+ 650 val perplexity 2100.1680
680
+ 650 train 7.731294 (lr=1.6084e-03) (hash(x)=25992320)
681
+ 651 train 7.687167 (lr=1.6025e-03) (hash(x)=24949509)
682
+ 652 train 7.699178 (lr=1.5966e-03) (hash(x)=25615151)
683
+ 653 train 7.699834 (lr=1.5906e-03) (hash(x)=27923548)
684
+ 654 train 7.739660 (lr=1.5847e-03) (hash(x)=26387848)
685
+ 655 train 7.652199 (lr=1.5788e-03) (hash(x)=22863518)
686
+ 656 train 7.625880 (lr=1.5728e-03) (hash(x)=20869581)
687
+ 657 train 7.772856 (lr=1.5669e-03) (hash(x)=18133606)
688
+ 658 train 7.659211 (lr=1.5610e-03) (hash(x)=28736122)
689
+ 659 train 7.635242 (lr=1.5550e-03) (hash(x)=25528329)
690
+ 660 train 7.693705 (lr=1.5491e-03) (hash(x)=24020765)
691
+ 661 train 7.720052 (lr=1.5432e-03) (hash(x)=27792245)
692
+ 662 train 7.745533 (lr=1.5373e-03) (hash(x)=28287635)
693
+ 663 train 7.492622 (lr=1.5314e-03) (hash(x)=20951206)
694
+ 664 train 7.565140 (lr=1.5254e-03) (hash(x)=22356452)
695
+ 665 train 7.628265 (lr=1.5195e-03) (hash(x)=24385170)
696
+ 666 train 7.702509 (lr=1.5136e-03) (hash(x)=23455854)
697
+ 667 train 7.683649 (lr=1.5077e-03) (hash(x)=21589405)
698
+ 668 train 7.667704 (lr=1.5018e-03) (hash(x)=24698150)
699
+ 669 train 7.654478 (lr=1.4959e-03) (hash(x)=28010836)
700
+ 670 train 7.774923 (lr=1.4900e-03) (hash(x)=25427381)
701
+ 671 train 7.730658 (lr=1.4841e-03) (hash(x)=27478557)
702
+ 672 train 7.691769 (lr=1.4782e-03) (hash(x)=23865569)
703
+ 673 train 7.749096 (lr=1.4723e-03) (hash(x)=23407132)
704
+ 674 train 7.640793 (lr=1.4664e-03) (hash(x)=20253664)
705
+ 675 train 7.604414 (lr=1.4605e-03) (hash(x)=21087928)
706
+ 676 train 7.694075 (lr=1.4547e-03) (hash(x)=25957589)
707
+ 677 train 7.628325 (lr=1.4488e-03) (hash(x)=27008610)
708
+ 678 train 7.715804 (lr=1.4429e-03) (hash(x)=27580265)
709
+ 679 train 7.715343 (lr=1.4371e-03) (hash(x)=24244798)
710
+ 680 train 7.627078 (lr=1.4312e-03) (hash(x)=24535871)
711
+ 681 train 7.727321 (lr=1.4253e-03) (hash(x)=27558681)
712
+ 682 train 7.599652 (lr=1.4195e-03) (hash(x)=21152793)
713
+ 683 train 7.599976 (lr=1.4136e-03) (hash(x)=28147005)
714
+ 684 train 7.675619 (lr=1.4078e-03) (hash(x)=21933180)
715
+ 685 train 7.645042 (lr=1.4019e-03) (hash(x)=24735895)
716
+ 686 train 7.688940 (lr=1.3961e-03) (hash(x)=26209948)
717
+ 687 train 7.675542 (lr=1.3903e-03) (hash(x)=23358313)
718
+ 688 train 7.614634 (lr=1.3844e-03) (hash(x)=22398001)
719
+ 689 train 7.542612 (lr=1.3786e-03) (hash(x)=23724957)
720
+ 690 train 7.602206 (lr=1.3728e-03) (hash(x)=23083514)
721
+ 691 train 7.471515 (lr=1.3670e-03) (hash(x)=16016924)
722
+ 692 train 7.735832 (lr=1.3612e-03) (hash(x)=27173674)
723
+ 693 train 7.535024 (lr=1.3554e-03) (hash(x)=28056250)
724
+ 694 train 7.646617 (lr=1.3496e-03) (hash(x)=22233617)
725
+ 695 train 7.738522 (lr=1.3438e-03) (hash(x)=23777908)
726
+ 696 train 7.698428 (lr=1.3380e-03) (hash(x)=27842849)
727
+ 697 train 7.713118 (lr=1.3323e-03) (hash(x)=23711679)
728
+ 698 train 7.530836 (lr=1.3265e-03) (hash(x)=17204843)
729
+ 699 train 7.857209 (lr=1.3207e-03) (hash(x)=23909800)
730
+ 700 val loss 7.6308
731
+ 700 val perplexity 2060.7917
732
+ 700 train 7.690992 (lr=1.3150e-03) (hash(x)=25033286)
733
+ 701 train 7.964295 (lr=1.3092e-03) (hash(x)=23608371)
734
+ 702 train 7.758265 (lr=1.3035e-03) (hash(x)=23654625)
735
+ 703 train 7.619846 (lr=1.2977e-03) (hash(x)=22633313)
736
+ 704 train 7.662441 (lr=1.2920e-03) (hash(x)=24267011)
737
+ 705 train 7.623178 (lr=1.2863e-03) (hash(x)=23449416)
738
+ 706 train 7.844618 (lr=1.2806e-03) (hash(x)=28527906)
739
+ 707 train 7.559112 (lr=1.2748e-03) (hash(x)=21189075)
740
+ 708 train 7.580860 (lr=1.2691e-03) (hash(x)=25278645)
741
+ 709 train 8.034248 (lr=1.2634e-03) (hash(x)=23654290)
742
+ 710 train 7.611103 (lr=1.2578e-03) (hash(x)=24571139)
743
+ 711 train 7.692595 (lr=1.2521e-03) (hash(x)=26203172)
744
+ 712 train 7.605851 (lr=1.2464e-03) (hash(x)=19623503)
745
+ 713 train 7.620964 (lr=1.2407e-03) (hash(x)=23970545)
746
+ 714 train 7.772737 (lr=1.2351e-03) (hash(x)=25724900)
747
+ 715 train 7.698445 (lr=1.2294e-03) (hash(x)=23245421)
748
+ 716 train 7.659051 (lr=1.2238e-03) (hash(x)=24087537)
749
+ 717 train 7.585750 (lr=1.2182e-03) (hash(x)=25549646)
750
+ 718 train 7.654135 (lr=1.2125e-03) (hash(x)=26406563)
751
+ 719 train 7.701875 (lr=1.2069e-03) (hash(x)=23972429)
752
+ 720 train 7.769635 (lr=1.2013e-03) (hash(x)=26378040)
753
+ 721 train 7.604137 (lr=1.1957e-03) (hash(x)=24618531)
754
+ 722 train 7.630247 (lr=1.1901e-03) (hash(x)=20135714)
755
+ 723 train 7.728841 (lr=1.1846e-03) (hash(x)=26490981)
756
+ 724 train 7.753007 (lr=1.1790e-03) (hash(x)=23927567)
757
+ 725 train 7.620473 (lr=1.1734e-03) (hash(x)=27943609)
758
+ 726 train 7.656029 (lr=1.1679e-03) (hash(x)=22962561)
759
+ 727 train 7.676806 (lr=1.1623e-03) (hash(x)=24772257)
760
+ 728 train 7.625701 (lr=1.1568e-03) (hash(x)=25539761)
761
+ 729 train 7.765790 (lr=1.1513e-03) (hash(x)=23316262)
762
+ 730 train 7.559511 (lr=1.1458e-03) (hash(x)=23857558)
763
+ 731 train 7.730391 (lr=1.1402e-03) (hash(x)=30252690)
764
+ 732 train 7.789316 (lr=1.1347e-03) (hash(x)=30590162)
765
+ 733 train 7.681266 (lr=1.1293e-03) (hash(x)=25015721)
766
+ 734 train 7.745039 (lr=1.1238e-03) (hash(x)=28639679)
767
+ 735 train 7.653567 (lr=1.1183e-03) (hash(x)=26199058)
768
+ 736 train 7.705127 (lr=1.1129e-03) (hash(x)=28907461)
769
+ 737 train 7.692514 (lr=1.1074e-03) (hash(x)=26466915)
770
+ 738 train 7.755908 (lr=1.1020e-03) (hash(x)=20332613)
771
+ 739 train 7.632423 (lr=1.0966e-03) (hash(x)=26766470)
772
+ 740 train 7.592055 (lr=1.0912e-03) (hash(x)=26722145)
773
+ 741 train 7.652725 (lr=1.0858e-03) (hash(x)=21222489)
774
+ 742 train 7.786253 (lr=1.0804e-03) (hash(x)=24662435)
775
+ 743 train 7.586648 (lr=1.0750e-03) (hash(x)=24308927)
776
+ 744 train 7.558877 (lr=1.0696e-03) (hash(x)=28442502)
777
+ 745 train 7.599711 (lr=1.0643e-03) (hash(x)=22810435)
778
+ 746 train 7.712965 (lr=1.0589e-03) (hash(x)=30486518)
779
+ 747 train 7.680569 (lr=1.0536e-03) (hash(x)=22688712)
780
+ 748 train 7.626081 (lr=1.0483e-03) (hash(x)=20712962)
781
+ 749 train 7.526076 (lr=1.0429e-03) (hash(x)=27490086)
782
+ 750 val loss 7.6057
783
+ 750 val perplexity 2009.5970
784
+ 750 train 7.511306 (lr=1.0376e-03) (hash(x)=23973908)
785
+ 751 train 7.656971 (lr=1.0324e-03) (hash(x)=29132401)
786
+ 752 train 7.631149 (lr=1.0271e-03) (hash(x)=27386700)
787
+ 753 train 7.669488 (lr=1.0218e-03) (hash(x)=24268614)
788
+ 754 train 7.689263 (lr=1.0166e-03) (hash(x)=27378387)
789
+ 755 train 7.517143 (lr=1.0113e-03) (hash(x)=22529400)
790
+ 756 train 7.612504 (lr=1.0061e-03) (hash(x)=27586989)
791
+ 757 train 7.536001 (lr=1.0009e-03) (hash(x)=22336349)
792
+ 758 train 7.573464 (lr=9.9568e-04) (hash(x)=26147854)
793
+ 759 train 7.668962 (lr=9.9049e-04) (hash(x)=25346788)
794
+ 760 train 7.727647 (lr=9.8531e-04) (hash(x)=25956855)
795
+ 761 train 7.688026 (lr=9.8015e-04) (hash(x)=24367780)
796
+ 762 train 7.567089 (lr=9.7500e-04) (hash(x)=26084462)
797
+ 763 train 7.639624 (lr=9.6986e-04) (hash(x)=26702459)
798
+ 764 train 7.571278 (lr=9.6474e-04) (hash(x)=25717976)
799
+ 765 train 7.534132 (lr=9.5963e-04) (hash(x)=22141366)
800
+ 766 train 7.747325 (lr=9.5453e-04) (hash(x)=24652423)
801
+ 767 train 7.675252 (lr=9.4944e-04) (hash(x)=30911699)
802
+ 768 train 7.648965 (lr=9.4437e-04) (hash(x)=22076954)
803
+ 769 train 7.844311 (lr=9.3932e-04) (hash(x)=21586769)
804
+ 770 train 7.655581 (lr=9.3427e-04) (hash(x)=21994337)
805
+ 771 train 7.622827 (lr=9.2924e-04) (hash(x)=25454145)
806
+ 772 train 7.695157 (lr=9.2423e-04) (hash(x)=23239550)
807
+ 773 train 7.740240 (lr=9.1923e-04) (hash(x)=23565926)
808
+ 774 train 7.590763 (lr=9.1424e-04) (hash(x)=24705563)
809
+ 775 train 7.662973 (lr=9.0927e-04) (hash(x)=27671095)
810
+ 776 train 7.794932 (lr=9.0431e-04) (hash(x)=26254303)
811
+ 777 train 7.567668 (lr=8.9936e-04) (hash(x)=27454020)
812
+ 778 train 7.559688 (lr=8.9443e-04) (hash(x)=24394565)
813
+ 779 train 7.504857 (lr=8.8952e-04) (hash(x)=21780199)
814
+ 780 train 7.588043 (lr=8.8462e-04) (hash(x)=28296896)
815
+ 781 train 7.700356 (lr=8.7973e-04) (hash(x)=26766303)
816
+ 782 train 7.567606 (lr=8.7486e-04) (hash(x)=21599973)
817
+ 783 train 7.849597 (lr=8.7001e-04) (hash(x)=24969360)
818
+ 784 train 7.913096 (lr=8.6516e-04) (hash(x)=24382006)
819
+ 785 train 7.514079 (lr=8.6034e-04) (hash(x)=22009299)
820
+ 786 train 7.377444 (lr=8.5553e-04) (hash(x)=22376569)
821
+ 787 train 7.542351 (lr=8.5073e-04) (hash(x)=28166843)
822
+ 788 train 7.782198 (lr=8.4596e-04) (hash(x)=28230119)
823
+ 789 train 7.755735 (lr=8.4119e-04) (hash(x)=26207465)
824
+ 790 train 7.800923 (lr=8.3644e-04) (hash(x)=24748118)
825
+ 791 train 7.741440 (lr=8.3171e-04) (hash(x)=27033802)
826
+ 792 train 7.661026 (lr=8.2699e-04) (hash(x)=28788047)
827
+ 793 train 7.566328 (lr=8.2229e-04) (hash(x)=22185099)
828
+ 794 train 7.718136 (lr=8.1761e-04) (hash(x)=21585136)
829
+ 795 train 7.547998 (lr=8.1294e-04) (hash(x)=27691968)
830
+ 796 train 7.581285 (lr=8.0829e-04) (hash(x)=23706103)
831
+ 797 train 7.433530 (lr=8.0365e-04) (hash(x)=27918030)
832
+ 798 train 7.654036 (lr=7.9903e-04) (hash(x)=23127886)
833
+ 799 train 7.675472 (lr=7.9443e-04) (hash(x)=26197070)
834
+ 800 val loss 7.5837
835
+ 800 val perplexity 1965.9525
836
+ 800 train 7.712868 (lr=7.8984e-04) (hash(x)=26463005)
837
+ 801 train 7.618032 (lr=7.8527e-04) (hash(x)=19160134)
838
+ 802 train 7.571336 (lr=7.8072e-04) (hash(x)=23699136)
839
+ 803 train 7.637867 (lr=7.7618e-04) (hash(x)=25353051)
840
+ 804 train 7.562407 (lr=7.7167e-04) (hash(x)=21558484)
841
+ 805 train 7.499489 (lr=7.6716e-04) (hash(x)=22828380)
842
+ 806 train 7.604203 (lr=7.6268e-04) (hash(x)=24623811)
843
+ 807 train 7.752641 (lr=7.5821e-04) (hash(x)=25410299)
844
+ 808 train 7.630766 (lr=7.5376e-04) (hash(x)=26046091)
845
+ 809 train 7.745185 (lr=7.4933e-04) (hash(x)=29887482)
846
+ 810 train 7.502969 (lr=7.4491e-04) (hash(x)=21809706)
847
+ 811 train 7.636881 (lr=7.4051e-04) (hash(x)=23712872)
848
+ 812 train 7.543375 (lr=7.3613e-04) (hash(x)=25875103)
849
+ 813 train 7.498415 (lr=7.3177e-04) (hash(x)=23736441)
850
+ 814 train 7.582377 (lr=7.2742e-04) (hash(x)=25093883)
851
+ 815 train 7.531278 (lr=7.2309e-04) (hash(x)=23692325)
852
+ 816 train 7.404885 (lr=7.1878e-04) (hash(x)=23061082)
853
+ 817 train 7.561911 (lr=7.1449e-04) (hash(x)=26094253)
854
+ 818 train 7.665705 (lr=7.1022e-04) (hash(x)=27327886)
855
+ 819 train 7.545212 (lr=7.0596e-04) (hash(x)=26186143)
856
+ 820 train 7.556739 (lr=7.0173e-04) (hash(x)=25336043)
857
+ 821 train 7.603405 (lr=6.9751e-04) (hash(x)=22811468)
858
+ 822 train 7.599870 (lr=6.9331e-04) (hash(x)=27925831)
859
+ 823 train 7.663346 (lr=6.8913e-04) (hash(x)=35528108)
860
+ 824 train 7.668353 (lr=6.8496e-04) (hash(x)=24949421)
861
+ 825 train 7.444167 (lr=6.8082e-04) (hash(x)=22845735)
862
+ 826 train 7.563560 (lr=6.7669e-04) (hash(x)=28138281)
863
+ 827 train 7.635265 (lr=6.7259e-04) (hash(x)=29038360)
864
+ 828 train 7.584367 (lr=6.6850e-04) (hash(x)=22827903)
865
+ 829 train 7.659996 (lr=6.6443e-04) (hash(x)=24647015)
866
+ 830 train 7.558162 (lr=6.6038e-04) (hash(x)=20477804)
867
+ 831 train 7.528037 (lr=6.5635e-04) (hash(x)=26692640)
868
+ 832 train 7.607703 (lr=6.5234e-04) (hash(x)=27388794)
869
+ 833 train 7.826940 (lr=6.4835e-04) (hash(x)=25265875)
870
+ 834 train 7.535006 (lr=6.4437e-04) (hash(x)=21841888)
871
+ 835 train 7.605297 (lr=6.4042e-04) (hash(x)=19809196)
872
+ 836 train 7.667049 (lr=6.3649e-04) (hash(x)=23007945)
873
+ 837 train 7.623643 (lr=6.3257e-04) (hash(x)=24079409)
874
+ 838 train 7.522236 (lr=6.2868e-04) (hash(x)=25874533)
875
+ 839 train 7.522948 (lr=6.2480e-04) (hash(x)=23660106)
876
+ 840 train 7.907018 (lr=6.2095e-04) (hash(x)=41821329)
877
+ 841 train 7.561615 (lr=6.1711e-04) (hash(x)=25763899)
878
+ 842 train 7.421106 (lr=6.1330e-04) (hash(x)=28584111)
879
+ 843 train 7.856364 (lr=6.0950e-04) (hash(x)=25559887)
880
+ 844 train 7.655566 (lr=6.0573e-04) (hash(x)=26886377)
881
+ 845 train 7.626286 (lr=6.0198e-04) (hash(x)=26461895)
882
+ 846 train 7.669251 (lr=5.9824e-04) (hash(x)=23018827)
883
+ 847 train 7.469522 (lr=5.9453e-04) (hash(x)=24375035)
884
+ 848 train 7.620461 (lr=5.9083e-04) (hash(x)=26431670)
885
+ 849 train 7.517011 (lr=5.8716e-04) (hash(x)=25033511)
886
+ 850 val loss 7.5801
887
+ 850 val perplexity 1958.9175
888
+ 850 train 7.634392 (lr=5.8351e-04) (hash(x)=25602577)
889
+ 851 train 7.568806 (lr=5.7988e-04) (hash(x)=23316582)
890
+ 852 train 7.670450 (lr=5.7627e-04) (hash(x)=22470095)
891
+ 853 train 7.558394 (lr=5.7268e-04) (hash(x)=25089202)
892
+ 854 train 7.511643 (lr=5.6911e-04) (hash(x)=26243247)
893
+ 855 train 7.578367 (lr=5.6556e-04) (hash(x)=22231144)
894
+ 856 train 7.487048 (lr=5.6203e-04) (hash(x)=27646564)
895
+ 857 train 7.642085 (lr=5.5853e-04) (hash(x)=23889875)
896
+ 858 train 7.587790 (lr=5.5504e-04) (hash(x)=25255240)
897
+ 859 train 7.686996 (lr=5.5158e-04) (hash(x)=23883299)
898
+ 860 train 7.666602 (lr=5.4813e-04) (hash(x)=18463841)
899
+ 861 train 7.657323 (lr=5.4471e-04) (hash(x)=30657762)
900
+ 862 train 7.557819 (lr=5.4131e-04) (hash(x)=25288079)
901
+ 863 train 7.590460 (lr=5.3793e-04) (hash(x)=28979310)
902
+ 864 train 7.766407 (lr=5.3458e-04) (hash(x)=30254991)
903
+ 865 train 7.630762 (lr=5.3124e-04) (hash(x)=27274653)
904
+ 866 train 7.613407 (lr=5.2793e-04) (hash(x)=23026066)
905
+ 867 train 7.593307 (lr=5.2464e-04) (hash(x)=14001963)
906
+ 868 train 7.486364 (lr=5.2137e-04) (hash(x)=21988054)
907
+ 869 train 7.575457 (lr=5.1812e-04) (hash(x)=26241099)
908
+ 870 train 7.650391 (lr=5.1489e-04) (hash(x)=26209722)
909
+ 871 train 7.523965 (lr=5.1169e-04) (hash(x)=26928312)
910
+ 872 train 7.604095 (lr=5.0851e-04) (hash(x)=23563272)
911
+ 873 train 7.540094 (lr=5.0534e-04) (hash(x)=25704828)
912
+ 874 train 7.774302 (lr=5.0221e-04) (hash(x)=22498992)
913
+ 875 train 7.929030 (lr=4.9909e-04) (hash(x)=22895918)
914
+ 876 train 7.508719 (lr=4.9600e-04) (hash(x)=24404283)
915
+ 877 train 7.543530 (lr=4.9293e-04) (hash(x)=26663932)
916
+ 878 train 7.675812 (lr=4.8988e-04) (hash(x)=22805316)
917
+ 879 train 7.633501 (lr=4.8685e-04) (hash(x)=23344466)
918
+ 880 train 7.583464 (lr=4.8385e-04) (hash(x)=20966940)
919
+ 881 train 7.575173 (lr=4.8087e-04) (hash(x)=27422664)
920
+ 882 train 7.617881 (lr=4.7791e-04) (hash(x)=22116510)
921
+ 883 train 7.510632 (lr=4.7497e-04) (hash(x)=27269633)
922
+ 884 train 7.618297 (lr=4.7206e-04) (hash(x)=27157711)
923
+ 885 train 7.612289 (lr=4.6917e-04) (hash(x)=22774645)
924
+ 886 train 7.539466 (lr=4.6630e-04) (hash(x)=22143146)
925
+ 887 train 7.556996 (lr=4.6346e-04) (hash(x)=22356438)
926
+ 888 train 7.326814 (lr=4.6063e-04) (hash(x)=23172158)
927
+ 889 train 7.573599 (lr=4.5783e-04) (hash(x)=26971690)
928
+ 890 train 7.617383 (lr=4.5506e-04) (hash(x)=24851922)
929
+ 891 train 7.450786 (lr=4.5231e-04) (hash(x)=21017641)
930
+ 892 train 7.570150 (lr=4.4958e-04) (hash(x)=28186957)
931
+ 893 train 7.609303 (lr=4.4687e-04) (hash(x)=25661082)
932
+ 894 train 7.497657 (lr=4.4419e-04) (hash(x)=25462781)
933
+ 895 train 7.464622 (lr=4.4153e-04) (hash(x)=21774374)
934
+ 896 train 7.458701 (lr=4.3889e-04) (hash(x)=23657055)
935
+ 897 train 7.460801 (lr=4.3628e-04) (hash(x)=23604050)
936
+ 898 train 7.891959 (lr=4.3369e-04) (hash(x)=31534071)
937
+ 899 train 7.717073 (lr=4.3113e-04) (hash(x)=24183732)
938
+ 900 val loss 7.5559
939
+ 900 val perplexity 1911.9567
940
+ 900 train 7.575482 (lr=4.2858e-04) (hash(x)=23384280)
941
+ 901 train 7.449081 (lr=4.2607e-04) (hash(x)=21975928)
942
+ 902 train 7.453537 (lr=4.2357e-04) (hash(x)=25281674)
943
+ 903 train 7.428204 (lr=4.2110e-04) (hash(x)=22947712)
944
+ 904 train 7.520119 (lr=4.1865e-04) (hash(x)=24853977)
945
+ 905 train 7.559927 (lr=4.1623e-04) (hash(x)=23696470)
946
+ 906 train 7.557965 (lr=4.1383e-04) (hash(x)=27489711)
947
+ 907 train 7.490782 (lr=4.1146e-04) (hash(x)=28050163)
948
+ 908 train 7.388462 (lr=4.0910e-04) (hash(x)=26996807)
949
+ 909 train 7.746178 (lr=4.0678e-04) (hash(x)=28626495)
950
+ 910 train 7.473655 (lr=4.0447e-04) (hash(x)=24783242)
951
+ 911 train 7.516463 (lr=4.0220e-04) (hash(x)=25832479)
952
+ 912 train 7.369362 (lr=3.9994e-04) (hash(x)=23522137)
953
+ 913 train 7.667065 (lr=3.9771e-04) (hash(x)=26731973)
954
+ 914 train 7.735716 (lr=3.9550e-04) (hash(x)=26019142)
955
+ 915 train 7.519018 (lr=3.9332e-04) (hash(x)=25081877)
956
+ 916 train 7.622543 (lr=3.9116e-04) (hash(x)=23843571)
957
+ 917 train 7.629456 (lr=3.8903e-04) (hash(x)=26805739)
958
+ 918 train 7.557544 (lr=3.8692e-04) (hash(x)=27169927)
959
+ 919 train 7.578307 (lr=3.8484e-04) (hash(x)=26545491)
960
+ 920 train 7.585835 (lr=3.8277e-04) (hash(x)=25179497)
961
+ 921 train 7.531925 (lr=3.8074e-04) (hash(x)=27085003)
962
+ 922 train 7.545794 (lr=3.7873e-04) (hash(x)=24558202)
963
+ 923 train 7.684460 (lr=3.7674e-04) (hash(x)=25317684)
964
+ 924 train 7.575562 (lr=3.7478e-04) (hash(x)=27104868)
965
+ 925 train 7.608748 (lr=3.7284e-04) (hash(x)=24806274)
966
+ 926 train 7.643869 (lr=3.7093e-04) (hash(x)=24341937)
967
+ 927 train 7.459026 (lr=3.6904e-04) (hash(x)=22736558)
968
+ 928 train 7.473399 (lr=3.6718e-04) (hash(x)=23843744)
969
+ 929 train 7.598492 (lr=3.6534e-04) (hash(x)=27218468)
970
+ 930 train 7.546979 (lr=3.6353e-04) (hash(x)=26042053)
971
+ 931 train 7.545666 (lr=3.6174e-04) (hash(x)=25776082)
972
+ 932 train 7.453916 (lr=3.5998e-04) (hash(x)=21594447)
973
+ 933 train 7.492428 (lr=3.5824e-04) (hash(x)=21674469)
974
+ 934 train 7.440726 (lr=3.5653e-04) (hash(x)=26213966)
975
+ 935 train 7.445779 (lr=3.5484e-04) (hash(x)=23770702)
976
+ 936 train 7.526697 (lr=3.5317e-04) (hash(x)=27829878)
977
+ 937 train 7.658311 (lr=3.5154e-04) (hash(x)=29177278)
978
+ 938 train 7.586535 (lr=3.4992e-04) (hash(x)=26020089)
979
+ 939 train 7.482455 (lr=3.4833e-04) (hash(x)=24676475)
980
+ 940 train 7.547410 (lr=3.4677e-04) (hash(x)=24056174)
981
+ 941 train 7.562239 (lr=3.4523e-04) (hash(x)=21887514)
982
+ 942 train 7.560599 (lr=3.4372e-04) (hash(x)=24196100)
983
+ 943 train 7.528726 (lr=3.4224e-04) (hash(x)=25700093)
984
+ 944 train 7.639983 (lr=3.4077e-04) (hash(x)=24365834)
985
+ 945 train 7.483799 (lr=3.3934e-04) (hash(x)=25631786)
986
+ 946 train 7.597067 (lr=3.3793e-04) (hash(x)=24452345)
987
+ 947 train 7.338262 (lr=3.3654e-04) (hash(x)=21465330)
988
+ 948 train 7.463530 (lr=3.3518e-04) (hash(x)=20427493)
989
+ 949 train 7.478971 (lr=3.3385e-04) (hash(x)=25439720)
990
+ 950 val loss 7.5507
991
+ 950 val perplexity 1902.1067
992
+ 950 train 7.569454 (lr=3.3254e-04) (hash(x)=21652988)
993
+ 951 train 7.487440 (lr=3.3125e-04) (hash(x)=22546779)
994
+ 952 train 7.356591 (lr=3.3000e-04) (hash(x)=20855736)
995
+ 953 train 7.552416 (lr=3.2876e-04) (hash(x)=27725873)
996
+ 954 train 7.596039 (lr=3.2756e-04) (hash(x)=24582989)
997
+ 955 train 7.470160 (lr=3.2638e-04) (hash(x)=24822372)
998
+ 956 train 7.526981 (lr=3.2522e-04) (hash(x)=25624588)
999
+ 957 train 7.484820 (lr=3.2409e-04) (hash(x)=23566489)
1000
+ 958 train 7.676033 (lr=3.2299e-04) (hash(x)=28117387)
1001
+ 959 train 7.465144 (lr=3.2191e-04) (hash(x)=26874283)
1002
+ 960 train 7.459832 (lr=3.2085e-04) (hash(x)=28570123)
1003
+ 961 train 7.585291 (lr=3.1983e-04) (hash(x)=24807017)
1004
+ 962 train 7.483000 (lr=3.1883e-04) (hash(x)=25189993)
1005
+ 963 train 7.610070 (lr=3.1785e-04) (hash(x)=29202226)
1006
+ 964 train 7.767408 (lr=3.1690e-04) (hash(x)=26060713)
1007
+ 965 train 7.656843 (lr=3.1598e-04) (hash(x)=23053285)
1008
+ 966 train 7.611586 (lr=3.1508e-04) (hash(x)=21968481)
1009
+ 967 train 7.568837 (lr=3.1421e-04) (hash(x)=24492698)
1010
+ 968 train 7.495081 (lr=3.1336e-04) (hash(x)=25760806)
1011
+ 969 train 7.526160 (lr=3.1254e-04) (hash(x)=27417392)
1012
+ 970 train 7.519546 (lr=3.1174e-04) (hash(x)=26767105)
1013
+ 971 train 7.582737 (lr=3.1098e-04) (hash(x)=25117678)
1014
+ 972 train 7.474861 (lr=3.1023e-04) (hash(x)=24233758)
1015
+ 973 train 7.535501 (lr=3.0952e-04) (hash(x)=29907634)
1016
+ 974 train 7.345968 (lr=3.0882e-04) (hash(x)=23552436)
1017
+ 975 train 7.648785 (lr=3.0816e-04) (hash(x)=28610406)
1018
+ 976 train 7.514011 (lr=3.0752e-04) (hash(x)=23858212)
1019
+ 977 train 7.408852 (lr=3.0691e-04) (hash(x)=20567764)
1020
+ 978 train 7.444288 (lr=3.0632e-04) (hash(x)=29631106)
1021
+ 979 train 7.406911 (lr=3.0576e-04) (hash(x)=19843564)
1022
+ 980 train 7.523556 (lr=3.0522e-04) (hash(x)=22589891)
1023
+ 981 train 7.322023 (lr=3.0471e-04) (hash(x)=21183943)
1024
+ 982 train 7.495707 (lr=3.0423e-04) (hash(x)=26152241)
1025
+ 983 train 7.508110 (lr=3.0377e-04) (hash(x)=22700785)
1026
+ 984 train 7.718112 (lr=3.0334e-04) (hash(x)=29611280)
1027
+ 985 train 7.834949 (lr=3.0294e-04) (hash(x)=25972823)
1028
+ 986 train 7.513793 (lr=3.0256e-04) (hash(x)=23447782)
1029
+ 987 train 7.493541 (lr=3.0221e-04) (hash(x)=22578453)
1030
+ 988 train 7.471992 (lr=3.0188e-04) (hash(x)=25474622)
1031
+ 989 train 7.565100 (lr=3.0158e-04) (hash(x)=22965762)
1032
+ 990 train 7.571639 (lr=3.0131e-04) (hash(x)=26116321)
1033
+ 991 train 7.477948 (lr=3.0106e-04) (hash(x)=25934549)
1034
+ 992 train 7.412428 (lr=3.0084e-04) (hash(x)=23937495)
1035
+ 993 train 7.553973 (lr=3.0064e-04) (hash(x)=26286442)
1036
+ 994 train 7.786283 (lr=3.0047e-04) (hash(x)=32737711)
1037
+ 995 train 7.436783 (lr=3.0033e-04) (hash(x)=20563613)
1038
+ 996 train 7.675404 (lr=3.0021e-04) (hash(x)=26234770)
1039
+ 997 train 7.592862 (lr=3.0012e-04) (hash(x)=26179422)
1040
+ 998 train 7.542227 (lr=3.0005e-04) (hash(x)=23289651)
1041
+ 999 val loss 7.5441
1042
+ 999 val perplexity 1889.5862
1043
+ 999 train 7.891490 (lr=3.0001e-04) (hash(x)=36436617)
lr3e-3_total_batch_size40960_seq_len512/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee381db29b44f16f8bb9ddc6ef686ad60669169b02f5acda9188464d52d7f9a5
3
+ size 48156226
lr3e-3_total_batch_size40960_seq_len512/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d96d775f617b291833bfeb2f0a7335a7cfa388a0ad2eabc0b473e73fc347e8b
3
+ size 71157574