andrew-healey commited on
Commit
ea95799
·
verified ·
1 Parent(s): f2b5a92

Upload folder using huggingface_hub

Browse files
lr1.5e-3_total_batch_size40960_seq_len512/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep_2/lr1.5e-3_total_batch_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": 0.0015, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr1.5e-3_total_batch_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
lr1.5e-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=5.2448e-06) (hash(x)=20782690)
5
+ 1 train 11.770658 (lr=1.0490e-05) (hash(x)=23890908)
6
+ 2 train 11.746203 (lr=1.5734e-05) (hash(x)=24611628)
7
+ 3 train 11.690123 (lr=2.0979e-05) (hash(x)=25767553)
8
+ 4 train 11.645432 (lr=2.6224e-05) (hash(x)=22512166)
9
+ 5 train 11.599701 (lr=3.1469e-05) (hash(x)=24732833)
10
+ 6 train 11.552396 (lr=3.6713e-05) (hash(x)=25129410)
11
+ 7 train 11.470312 (lr=4.1958e-05) (hash(x)=26582391)
12
+ 8 train 11.376846 (lr=4.7203e-05) (hash(x)=26472336)
13
+ 9 train 11.276885 (lr=5.2448e-05) (hash(x)=27654289)
14
+ 10 train 11.178011 (lr=5.7692e-05) (hash(x)=26559876)
15
+ 11 train 11.047280 (lr=6.2937e-05) (hash(x)=21770329)
16
+ 12 train 10.970761 (lr=6.8182e-05) (hash(x)=22723459)
17
+ 13 train 10.830210 (lr=7.3427e-05) (hash(x)=30338342)
18
+ 14 train 10.750552 (lr=7.8671e-05) (hash(x)=25678059)
19
+ 15 train 10.628061 (lr=8.3916e-05) (hash(x)=26109335)
20
+ 16 train 10.608283 (lr=8.9161e-05) (hash(x)=25044885)
21
+ 17 train 10.588475 (lr=9.4406e-05) (hash(x)=28870690)
22
+ 18 train 10.577917 (lr=9.9650e-05) (hash(x)=35279941)
23
+ 19 train 10.466777 (lr=1.0490e-04) (hash(x)=25015923)
24
+ 20 train 10.401252 (lr=1.1014e-04) (hash(x)=26740855)
25
+ 21 train 10.326657 (lr=1.1538e-04) (hash(x)=22262151)
26
+ 22 train 10.287719 (lr=1.2063e-04) (hash(x)=26621834)
27
+ 23 train 10.301884 (lr=1.2587e-04) (hash(x)=23234741)
28
+ 24 train 10.369473 (lr=1.3112e-04) (hash(x)=25470563)
29
+ 25 train 10.321452 (lr=1.3636e-04) (hash(x)=21799102)
30
+ 26 train 10.246595 (lr=1.4161e-04) (hash(x)=23275479)
31
+ 27 train 10.292705 (lr=1.4685e-04) (hash(x)=22091862)
32
+ 28 train 10.163467 (lr=1.5210e-04) (hash(x)=17430373)
33
+ 29 train 10.091432 (lr=1.5734e-04) (hash(x)=22954861)
34
+ 30 train 10.174285 (lr=1.6259e-04) (hash(x)=26671799)
35
+ 31 train 10.096317 (lr=1.6783e-04) (hash(x)=26924723)
36
+ 32 train 10.095434 (lr=1.7308e-04) (hash(x)=23557495)
37
+ 33 train 9.901454 (lr=1.7832e-04) (hash(x)=27804380)
38
+ 34 train 10.040325 (lr=1.8357e-04) (hash(x)=26654908)
39
+ 35 train 9.942492 (lr=1.8881e-04) (hash(x)=23627518)
40
+ 36 train 9.866219 (lr=1.9406e-04) (hash(x)=21889990)
41
+ 37 train 9.871151 (lr=1.9930e-04) (hash(x)=22528001)
42
+ 38 train 9.835156 (lr=2.0455e-04) (hash(x)=29248942)
43
+ 39 train 9.711053 (lr=2.0979e-04) (hash(x)=25533417)
44
+ 40 train 9.603037 (lr=2.1503e-04) (hash(x)=20612533)
45
+ 41 train 9.561379 (lr=2.2028e-04) (hash(x)=24677740)
46
+ 42 train 9.537317 (lr=2.2552e-04) (hash(x)=25347203)
47
+ 43 train 9.451678 (lr=2.3077e-04) (hash(x)=28922363)
48
+ 44 train 9.395393 (lr=2.3601e-04) (hash(x)=25631397)
49
+ 45 train 9.318373 (lr=2.4126e-04) (hash(x)=26237983)
50
+ 46 train 9.278879 (lr=2.4650e-04) (hash(x)=24742645)
51
+ 47 train 9.244855 (lr=2.5175e-04) (hash(x)=26157660)
52
+ 48 train 9.086165 (lr=2.5699e-04) (hash(x)=22582169)
53
+ 49 train 9.057470 (lr=2.6224e-04) (hash(x)=26431960)
54
+ 50 val loss 9.0651
55
+ 50 val perplexity 8648.1250
56
+ 50 train 9.091146 (lr=2.6748e-04) (hash(x)=25716176)
57
+ 51 train 9.017038 (lr=2.7273e-04) (hash(x)=28019788)
58
+ 52 train 8.955972 (lr=2.7797e-04) (hash(x)=20382963)
59
+ 53 train 8.850412 (lr=2.8322e-04) (hash(x)=24489647)
60
+ 54 train 8.815079 (lr=2.8846e-04) (hash(x)=25128502)
61
+ 55 train 8.791017 (lr=2.9371e-04) (hash(x)=25231390)
62
+ 56 train 8.633046 (lr=2.9895e-04) (hash(x)=22708977)
63
+ 57 train 8.754154 (lr=3.0420e-04) (hash(x)=24364920)
64
+ 58 train 8.682612 (lr=3.0944e-04) (hash(x)=24932925)
65
+ 59 train 8.584327 (lr=3.1469e-04) (hash(x)=26913684)
66
+ 60 train 8.484946 (lr=3.1993e-04) (hash(x)=18017792)
67
+ 61 train 8.503103 (lr=3.2517e-04) (hash(x)=28613882)
68
+ 62 train 8.471970 (lr=3.3042e-04) (hash(x)=22652243)
69
+ 63 train 8.324348 (lr=3.3566e-04) (hash(x)=27194521)
70
+ 64 train 8.353832 (lr=3.4091e-04) (hash(x)=24936836)
71
+ 65 train 8.369169 (lr=3.4615e-04) (hash(x)=24765578)
72
+ 66 train 8.356785 (lr=3.5140e-04) (hash(x)=23107416)
73
+ 67 train 8.257423 (lr=3.5664e-04) (hash(x)=29314255)
74
+ 68 train 8.202351 (lr=3.6189e-04) (hash(x)=24013079)
75
+ 69 train 8.172343 (lr=3.6713e-04) (hash(x)=25322001)
76
+ 70 train 8.240390 (lr=3.7238e-04) (hash(x)=26690440)
77
+ 71 train 8.125504 (lr=3.7762e-04) (hash(x)=22598512)
78
+ 72 train 8.039829 (lr=3.8287e-04) (hash(x)=18473911)
79
+ 73 train 8.227710 (lr=3.8811e-04) (hash(x)=22322442)
80
+ 74 train 8.208019 (lr=3.9336e-04) (hash(x)=26324153)
81
+ 75 train 8.074037 (lr=3.9860e-04) (hash(x)=23446058)
82
+ 76 train 8.138404 (lr=4.0385e-04) (hash(x)=24843741)
83
+ 77 train 7.968390 (lr=4.0909e-04) (hash(x)=23197102)
84
+ 78 train 8.208050 (lr=4.1434e-04) (hash(x)=23054940)
85
+ 79 train 8.141060 (lr=4.1958e-04) (hash(x)=27140757)
86
+ 80 train 7.845527 (lr=4.2483e-04) (hash(x)=25149419)
87
+ 81 train 7.983657 (lr=4.3007e-04) (hash(x)=24727764)
88
+ 82 train 7.654548 (lr=4.3531e-04) (hash(x)=21935931)
89
+ 83 train 7.798020 (lr=4.4056e-04) (hash(x)=27606073)
90
+ 84 train 7.893735 (lr=4.4580e-04) (hash(x)=22450341)
91
+ 85 train 8.048264 (lr=4.5105e-04) (hash(x)=25945859)
92
+ 86 train 7.780974 (lr=4.5629e-04) (hash(x)=21613707)
93
+ 87 train 7.852329 (lr=4.6154e-04) (hash(x)=24396519)
94
+ 88 train 7.875669 (lr=4.6678e-04) (hash(x)=29138828)
95
+ 89 train 7.887273 (lr=4.7203e-04) (hash(x)=24685515)
96
+ 90 train 7.800750 (lr=4.7727e-04) (hash(x)=22825749)
97
+ 91 train 7.837969 (lr=4.8252e-04) (hash(x)=28767869)
98
+ 92 train 7.883712 (lr=4.8776e-04) (hash(x)=21159060)
99
+ 93 train 7.779494 (lr=4.9301e-04) (hash(x)=25045397)
100
+ 94 train 7.935764 (lr=4.9825e-04) (hash(x)=25560928)
101
+ 95 train 7.938442 (lr=5.0350e-04) (hash(x)=28959222)
102
+ 96 train 7.786013 (lr=5.0874e-04) (hash(x)=21248405)
103
+ 97 train 7.801168 (lr=5.1399e-04) (hash(x)=25830182)
104
+ 98 train 7.698813 (lr=5.1923e-04) (hash(x)=25173113)
105
+ 99 train 7.732969 (lr=5.2448e-04) (hash(x)=23486277)
106
+ 100 val loss 7.7323
107
+ 100 val perplexity 2280.9004
108
+ 100 train 7.765555 (lr=5.2972e-04) (hash(x)=28916006)
109
+ 101 train 7.916852 (lr=5.3497e-04) (hash(x)=25556929)
110
+ 102 train 7.812853 (lr=5.4021e-04) (hash(x)=29648798)
111
+ 103 train 7.720037 (lr=5.4545e-04) (hash(x)=24976217)
112
+ 104 train 7.798210 (lr=5.5070e-04) (hash(x)=31019606)
113
+ 105 train 7.809225 (lr=5.5594e-04) (hash(x)=28269421)
114
+ 106 train 7.722915 (lr=5.6119e-04) (hash(x)=24865358)
115
+ 107 train 7.648256 (lr=5.6643e-04) (hash(x)=26984272)
116
+ 108 train 7.780168 (lr=5.7168e-04) (hash(x)=24544116)
117
+ 109 train 7.746596 (lr=5.7692e-04) (hash(x)=22059850)
118
+ 110 train 7.779303 (lr=5.8217e-04) (hash(x)=23926632)
119
+ 111 train 7.696063 (lr=5.8741e-04) (hash(x)=27025333)
120
+ 112 train 7.634594 (lr=5.9266e-04) (hash(x)=29120407)
121
+ 113 train 7.740953 (lr=5.9790e-04) (hash(x)=23827393)
122
+ 114 train 7.691010 (lr=6.0315e-04) (hash(x)=27230027)
123
+ 115 train 7.672576 (lr=6.0839e-04) (hash(x)=24968260)
124
+ 116 train 7.669037 (lr=6.1364e-04) (hash(x)=27544803)
125
+ 117 train 7.686092 (lr=6.1888e-04) (hash(x)=26540663)
126
+ 118 train 7.778410 (lr=6.2413e-04) (hash(x)=25856625)
127
+ 119 train 7.673829 (lr=6.2937e-04) (hash(x)=23164356)
128
+ 120 train 7.718043 (lr=6.3462e-04) (hash(x)=25646282)
129
+ 121 train 7.710485 (lr=6.3986e-04) (hash(x)=23843233)
130
+ 122 train 7.558786 (lr=6.4510e-04) (hash(x)=24506028)
131
+ 123 train 7.680099 (lr=6.5035e-04) (hash(x)=22781277)
132
+ 124 train 7.690490 (lr=6.5559e-04) (hash(x)=24948650)
133
+ 125 train 7.679358 (lr=6.6084e-04) (hash(x)=26373991)
134
+ 126 train 7.766599 (lr=6.6608e-04) (hash(x)=23402396)
135
+ 127 train 7.732694 (lr=6.7133e-04) (hash(x)=21562133)
136
+ 128 train 7.777441 (lr=6.7657e-04) (hash(x)=25009210)
137
+ 129 train 7.701021 (lr=6.8182e-04) (hash(x)=27161811)
138
+ 130 train 7.769850 (lr=6.8706e-04) (hash(x)=29515100)
139
+ 131 train 7.682714 (lr=6.9231e-04) (hash(x)=25305929)
140
+ 132 train 7.744199 (lr=6.9755e-04) (hash(x)=26808543)
141
+ 133 train 7.673382 (lr=7.0280e-04) (hash(x)=25669509)
142
+ 134 train 7.767658 (lr=7.0804e-04) (hash(x)=28435805)
143
+ 135 train 7.734368 (lr=7.1329e-04) (hash(x)=26482758)
144
+ 136 train 7.667161 (lr=7.1853e-04) (hash(x)=24929929)
145
+ 137 train 7.645678 (lr=7.2378e-04) (hash(x)=25043037)
146
+ 138 train 7.875208 (lr=7.2902e-04) (hash(x)=29151257)
147
+ 139 train 7.638447 (lr=7.3427e-04) (hash(x)=26400365)
148
+ 140 train 7.639000 (lr=7.3951e-04) (hash(x)=24098812)
149
+ 141 train 7.628123 (lr=7.4476e-04) (hash(x)=20967666)
150
+ 142 train 7.793742 (lr=7.5000e-04) (hash(x)=28667963)
151
+ 143 train 7.744076 (lr=7.5524e-04) (hash(x)=27083417)
152
+ 144 train 7.622779 (lr=7.6049e-04) (hash(x)=23649001)
153
+ 145 train 7.667747 (lr=7.6573e-04) (hash(x)=31338300)
154
+ 146 train 7.649314 (lr=7.7098e-04) (hash(x)=30603174)
155
+ 147 train 7.673859 (lr=7.7622e-04) (hash(x)=24830965)
156
+ 148 train 7.813186 (lr=7.8147e-04) (hash(x)=27269893)
157
+ 149 train 7.654194 (lr=7.8671e-04) (hash(x)=24825041)
158
+ 150 val loss 7.6721
159
+ 150 val perplexity 2147.5459
160
+ 150 train 7.638474 (lr=7.9196e-04) (hash(x)=22724871)
161
+ 151 train 7.572466 (lr=7.9720e-04) (hash(x)=24683717)
162
+ 152 train 7.696836 (lr=8.0245e-04) (hash(x)=26190337)
163
+ 153 train 7.688428 (lr=8.0769e-04) (hash(x)=23086289)
164
+ 154 train 7.727032 (lr=8.1294e-04) (hash(x)=23593875)
165
+ 155 train 7.566096 (lr=8.1818e-04) (hash(x)=22414190)
166
+ 156 train 7.646628 (lr=8.2343e-04) (hash(x)=26758114)
167
+ 157 train 7.590486 (lr=8.2867e-04) (hash(x)=21563184)
168
+ 158 train 7.644220 (lr=8.3392e-04) (hash(x)=19218470)
169
+ 159 train 7.761006 (lr=8.3916e-04) (hash(x)=25588469)
170
+ 160 train 7.717986 (lr=8.4441e-04) (hash(x)=30433767)
171
+ 161 train 7.785394 (lr=8.4965e-04) (hash(x)=17851621)
172
+ 162 train 7.850897 (lr=8.5490e-04) (hash(x)=25183690)
173
+ 163 train 7.884908 (lr=8.6014e-04) (hash(x)=27040642)
174
+ 164 train 7.649833 (lr=8.6538e-04) (hash(x)=22183303)
175
+ 165 train 7.953856 (lr=8.7063e-04) (hash(x)=27537063)
176
+ 166 train 7.666250 (lr=8.7587e-04) (hash(x)=19492832)
177
+ 167 train 7.661459 (lr=8.8112e-04) (hash(x)=25951629)
178
+ 168 train 7.668222 (lr=8.8636e-04) (hash(x)=22408682)
179
+ 169 train 7.656032 (lr=8.9161e-04) (hash(x)=27633457)
180
+ 170 train 7.754194 (lr=8.9685e-04) (hash(x)=27635080)
181
+ 171 train 7.939156 (lr=9.0210e-04) (hash(x)=22287596)
182
+ 172 train 7.833120 (lr=9.0734e-04) (hash(x)=35895440)
183
+ 173 train 7.731757 (lr=9.1259e-04) (hash(x)=25888847)
184
+ 174 train 7.747379 (lr=9.1783e-04) (hash(x)=26518800)
185
+ 175 train 7.691576 (lr=9.2308e-04) (hash(x)=37519283)
186
+ 176 train 7.911466 (lr=9.2832e-04) (hash(x)=25141945)
187
+ 177 train 7.892797 (lr=9.3357e-04) (hash(x)=28527939)
188
+ 178 train 7.734830 (lr=9.3881e-04) (hash(x)=21942471)
189
+ 179 train 7.736651 (lr=9.4406e-04) (hash(x)=24955630)
190
+ 180 train 7.667367 (lr=9.4930e-04) (hash(x)=25541230)
191
+ 181 train 7.536716 (lr=9.5455e-04) (hash(x)=23501481)
192
+ 182 train 7.590546 (lr=9.5979e-04) (hash(x)=22754988)
193
+ 183 train 7.491921 (lr=9.6503e-04) (hash(x)=23475070)
194
+ 184 train 7.666139 (lr=9.7028e-04) (hash(x)=23609959)
195
+ 185 train 7.537679 (lr=9.7552e-04) (hash(x)=23441847)
196
+ 186 train 7.620502 (lr=9.8077e-04) (hash(x)=23096292)
197
+ 187 train 7.678520 (lr=9.8601e-04) (hash(x)=26171093)
198
+ 188 train 7.684146 (lr=9.9126e-04) (hash(x)=20060838)
199
+ 189 train 7.492192 (lr=9.9650e-04) (hash(x)=25451055)
200
+ 190 train 7.317529 (lr=1.0017e-03) (hash(x)=24934792)
201
+ 191 train 7.457066 (lr=1.0070e-03) (hash(x)=25367597)
202
+ 192 train 7.602320 (lr=1.0122e-03) (hash(x)=25022148)
203
+ 193 train 7.610182 (lr=1.0175e-03) (hash(x)=25431365)
204
+ 194 train 7.656858 (lr=1.0227e-03) (hash(x)=29307788)
205
+ 195 train 7.519479 (lr=1.0280e-03) (hash(x)=25651929)
206
+ 196 train 7.552220 (lr=1.0332e-03) (hash(x)=26312387)
207
+ 197 train 7.505221 (lr=1.0385e-03) (hash(x)=23804913)
208
+ 198 train 7.556361 (lr=1.0437e-03) (hash(x)=26115519)
209
+ 199 train 7.569554 (lr=1.0490e-03) (hash(x)=23210747)
210
+ 200 val loss 7.5793
211
+ 200 val perplexity 1957.1687
212
+ 200 train 7.633120 (lr=1.0542e-03) (hash(x)=18463712)
213
+ 201 train 7.553512 (lr=1.0594e-03) (hash(x)=23245699)
214
+ 202 train 7.443579 (lr=1.0647e-03) (hash(x)=21770300)
215
+ 203 train 7.739912 (lr=1.0699e-03) (hash(x)=25387532)
216
+ 204 train 7.624924 (lr=1.0752e-03) (hash(x)=27718703)
217
+ 205 train 7.699632 (lr=1.0804e-03) (hash(x)=27661052)
218
+ 206 train 7.632099 (lr=1.0857e-03) (hash(x)=26538928)
219
+ 207 train 7.688132 (lr=1.0909e-03) (hash(x)=21772654)
220
+ 208 train 7.776151 (lr=1.0962e-03) (hash(x)=26263254)
221
+ 209 train 7.638481 (lr=1.1014e-03) (hash(x)=23552356)
222
+ 210 train 7.857795 (lr=1.1066e-03) (hash(x)=22707141)
223
+ 211 train 7.910218 (lr=1.1119e-03) (hash(x)=34234481)
224
+ 212 train 7.721779 (lr=1.1171e-03) (hash(x)=35008670)
225
+ 213 train 7.854199 (lr=1.1224e-03) (hash(x)=25891543)
226
+ 214 train 7.546402 (lr=1.1276e-03) (hash(x)=21971224)
227
+ 215 train 7.694109 (lr=1.1329e-03) (hash(x)=27312186)
228
+ 216 train 7.836606 (lr=1.1381e-03) (hash(x)=22163749)
229
+ 217 train 7.540182 (lr=1.1434e-03) (hash(x)=25942888)
230
+ 218 train 7.612833 (lr=1.1486e-03) (hash(x)=25493916)
231
+ 219 train 7.804924 (lr=1.1538e-03) (hash(x)=24289951)
232
+ 220 train 7.575101 (lr=1.1591e-03) (hash(x)=23648671)
233
+ 221 train 7.427884 (lr=1.1643e-03) (hash(x)=24992055)
234
+ 222 train 7.706512 (lr=1.1696e-03) (hash(x)=26005549)
235
+ 223 train 7.590233 (lr=1.1748e-03) (hash(x)=21046554)
236
+ 224 train 7.574853 (lr=1.1801e-03) (hash(x)=26321539)
237
+ 225 train 7.584645 (lr=1.1853e-03) (hash(x)=27807872)
238
+ 226 train 7.565428 (lr=1.1906e-03) (hash(x)=21555364)
239
+ 227 train 7.539972 (lr=1.1958e-03) (hash(x)=26325324)
240
+ 228 train 7.567332 (lr=1.2010e-03) (hash(x)=24956294)
241
+ 229 train 7.503779 (lr=1.2063e-03) (hash(x)=24429389)
242
+ 230 train 7.614590 (lr=1.2115e-03) (hash(x)=27929846)
243
+ 231 train 7.592738 (lr=1.2168e-03) (hash(x)=25282897)
244
+ 232 train 7.420703 (lr=1.2220e-03) (hash(x)=26802508)
245
+ 233 train 7.618175 (lr=1.2273e-03) (hash(x)=28390416)
246
+ 234 train 7.803466 (lr=1.2325e-03) (hash(x)=26292451)
247
+ 235 train 7.671797 (lr=1.2378e-03) (hash(x)=22915344)
248
+ 236 train 7.558874 (lr=1.2430e-03) (hash(x)=24744383)
249
+ 237 train 7.492089 (lr=1.2483e-03) (hash(x)=26284678)
250
+ 238 train 7.486214 (lr=1.2535e-03) (hash(x)=21654269)
251
+ 239 train 7.917832 (lr=1.2587e-03) (hash(x)=29466268)
252
+ 240 train 8.176592 (lr=1.2640e-03) (hash(x)=32616762)
253
+ 241 train 8.115485 (lr=1.2692e-03) (hash(x)=32201138)
254
+ 242 train 8.132523 (lr=1.2745e-03) (hash(x)=33683468)
255
+ 243 train 7.927503 (lr=1.2797e-03) (hash(x)=27498125)
256
+ 244 train 7.854294 (lr=1.2850e-03) (hash(x)=25164318)
257
+ 245 train 7.738535 (lr=1.2902e-03) (hash(x)=25690221)
258
+ 246 train 7.655647 (lr=1.2955e-03) (hash(x)=29031550)
259
+ 247 train 7.739473 (lr=1.3007e-03) (hash(x)=24094093)
260
+ 248 train 8.351751 (lr=1.3059e-03) (hash(x)=61455960)
261
+ 249 train 11.749212 (lr=1.3112e-03) (hash(x)=61104710)
262
+ 250 val loss 7.6573
263
+ 250 val perplexity 2115.9536
264
+ 250 train 7.763918 (lr=1.3164e-03) (hash(x)=20881126)
265
+ 251 train 7.700055 (lr=1.3217e-03) (hash(x)=26920243)
266
+ 252 train 7.610800 (lr=1.3269e-03) (hash(x)=21970343)
267
+ 253 train 7.723437 (lr=1.3322e-03) (hash(x)=31676352)
268
+ 254 train 7.738579 (lr=1.3374e-03) (hash(x)=25600562)
269
+ 255 train 7.726973 (lr=1.3427e-03) (hash(x)=22022924)
270
+ 256 train 7.672328 (lr=1.3479e-03) (hash(x)=22777085)
271
+ 257 train 7.796449 (lr=1.3531e-03) (hash(x)=21727744)
272
+ 258 train 7.771986 (lr=1.3584e-03) (hash(x)=20701998)
273
+ 259 train 7.803832 (lr=1.3636e-03) (hash(x)=20863982)
274
+ 260 train 7.722167 (lr=1.3689e-03) (hash(x)=25084926)
275
+ 261 train 7.766212 (lr=1.3741e-03) (hash(x)=24412897)
276
+ 262 train 7.803134 (lr=1.3794e-03) (hash(x)=25434913)
277
+ 263 train 7.744126 (lr=1.3846e-03) (hash(x)=26302544)
278
+ 264 train 7.801633 (lr=1.3899e-03) (hash(x)=34976197)
279
+ 265 train 7.663665 (lr=1.3951e-03) (hash(x)=23870805)
280
+ 266 train 7.690845 (lr=1.4003e-03) (hash(x)=25041103)
281
+ 267 train 7.684714 (lr=1.4056e-03) (hash(x)=26041745)
282
+ 268 train 7.683321 (lr=1.4108e-03) (hash(x)=24068619)
283
+ 269 train 7.745647 (lr=1.4161e-03) (hash(x)=25441262)
284
+ 270 train 7.739688 (lr=1.4213e-03) (hash(x)=29181867)
285
+ 271 train 7.743477 (lr=1.4266e-03) (hash(x)=27093757)
286
+ 272 train 7.643830 (lr=1.4318e-03) (hash(x)=25150275)
287
+ 273 train 7.761814 (lr=1.4371e-03) (hash(x)=33457102)
288
+ 274 train 7.562437 (lr=1.4423e-03) (hash(x)=25755322)
289
+ 275 train 7.629660 (lr=1.4476e-03) (hash(x)=23630794)
290
+ 276 train 7.559886 (lr=1.4528e-03) (hash(x)=25167987)
291
+ 277 train 7.663910 (lr=1.4580e-03) (hash(x)=24043224)
292
+ 278 train 7.643922 (lr=1.4633e-03) (hash(x)=24665411)
293
+ 279 train 7.606524 (lr=1.4685e-03) (hash(x)=17321657)
294
+ 280 train 7.602100 (lr=1.4738e-03) (hash(x)=27932591)
295
+ 281 train 7.911195 (lr=1.4790e-03) (hash(x)=29447954)
296
+ 282 train 7.671704 (lr=1.4843e-03) (hash(x)=24621816)
297
+ 283 train 7.697078 (lr=1.4895e-03) (hash(x)=24682294)
298
+ 284 train 7.648658 (lr=1.4948e-03) (hash(x)=28057095)
299
+ 285 train 7.683033 (lr=1.5000e-03) (hash(x)=24547536)
300
+ 286 train 7.542697 (lr=1.5000e-03) (hash(x)=18987545)
301
+ 287 train 7.648271 (lr=1.5000e-03) (hash(x)=26745063)
302
+ 288 train 7.549035 (lr=1.5000e-03) (hash(x)=21447406)
303
+ 289 train 7.381471 (lr=1.4999e-03) (hash(x)=24650655)
304
+ 290 train 7.783180 (lr=1.4999e-03) (hash(x)=26519976)
305
+ 291 train 8.176542 (lr=1.4998e-03) (hash(x)=20106514)
306
+ 292 train 7.675783 (lr=1.4998e-03) (hash(x)=20117808)
307
+ 293 train 7.885852 (lr=1.4997e-03) (hash(x)=25510334)
308
+ 294 train 7.811205 (lr=1.4996e-03) (hash(x)=25749120)
309
+ 295 train 7.926776 (lr=1.4995e-03) (hash(x)=24357241)
310
+ 296 train 7.861516 (lr=1.4993e-03) (hash(x)=25584945)
311
+ 297 train 7.765106 (lr=1.4992e-03) (hash(x)=25198897)
312
+ 298 train 7.929477 (lr=1.4991e-03) (hash(x)=23475891)
313
+ 299 train 7.910204 (lr=1.4989e-03) (hash(x)=23644804)
314
+ 300 val loss 7.9891
315
+ 300 val perplexity 2948.5759
316
+ 300 train 8.168637 (lr=1.4987e-03) (hash(x)=25629584)
317
+ 301 train 8.261678 (lr=1.4985e-03) (hash(x)=29797280)
318
+ 302 train 8.238178 (lr=1.4983e-03) (hash(x)=30026394)
319
+ 303 train 8.294258 (lr=1.4981e-03) (hash(x)=26594177)
320
+ 304 train 7.990581 (lr=1.4979e-03) (hash(x)=23766256)
321
+ 305 train 7.858694 (lr=1.4976e-03) (hash(x)=24250150)
322
+ 306 train 7.779681 (lr=1.4974e-03) (hash(x)=26108817)
323
+ 307 train 7.709070 (lr=1.4971e-03) (hash(x)=18582688)
324
+ 308 train 7.678756 (lr=1.4968e-03) (hash(x)=23272705)
325
+ 309 train 7.716678 (lr=1.4965e-03) (hash(x)=24793412)
326
+ 310 train 7.693946 (lr=1.4962e-03) (hash(x)=19280989)
327
+ 311 train 7.678008 (lr=1.4959e-03) (hash(x)=22751150)
328
+ 312 train 7.784314 (lr=1.4956e-03) (hash(x)=23537942)
329
+ 313 train 7.599751 (lr=1.4952e-03) (hash(x)=23748375)
330
+ 314 train 7.765629 (lr=1.4949e-03) (hash(x)=27413825)
331
+ 315 train 7.786789 (lr=1.4945e-03) (hash(x)=21682445)
332
+ 316 train 7.659910 (lr=1.4941e-03) (hash(x)=22040184)
333
+ 317 train 7.665611 (lr=1.4937e-03) (hash(x)=24463229)
334
+ 318 train 7.786479 (lr=1.4933e-03) (hash(x)=24091954)
335
+ 319 train 7.720454 (lr=1.4929e-03) (hash(x)=26670219)
336
+ 320 train 7.754362 (lr=1.4925e-03) (hash(x)=25558986)
337
+ 321 train 7.756287 (lr=1.4920e-03) (hash(x)=19218943)
338
+ 322 train 7.845184 (lr=1.4915e-03) (hash(x)=24551402)
339
+ 323 train 7.767998 (lr=1.4911e-03) (hash(x)=25229281)
340
+ 324 train 7.596134 (lr=1.4906e-03) (hash(x)=21251127)
341
+ 325 train 7.816780 (lr=1.4901e-03) (hash(x)=25800857)
342
+ 326 train 7.737436 (lr=1.4896e-03) (hash(x)=25836189)
343
+ 327 train 7.745217 (lr=1.4890e-03) (hash(x)=25999255)
344
+ 328 train 7.801631 (lr=1.4885e-03) (hash(x)=20149394)
345
+ 329 train 7.734929 (lr=1.4880e-03) (hash(x)=21999338)
346
+ 330 train 7.814485 (lr=1.4874e-03) (hash(x)=27611302)
347
+ 331 train 7.764083 (lr=1.4868e-03) (hash(x)=22778356)
348
+ 332 train 7.748129 (lr=1.4862e-03) (hash(x)=24195959)
349
+ 333 train 7.946766 (lr=1.4856e-03) (hash(x)=19127465)
350
+ 334 train 7.812947 (lr=1.4850e-03) (hash(x)=22426274)
351
+ 335 train 7.641294 (lr=1.4844e-03) (hash(x)=23785205)
352
+ 336 train 7.793352 (lr=1.4837e-03) (hash(x)=28505676)
353
+ 337 train 7.733229 (lr=1.4831e-03) (hash(x)=27706477)
354
+ 338 train 7.768242 (lr=1.4824e-03) (hash(x)=28684175)
355
+ 339 train 7.753277 (lr=1.4817e-03) (hash(x)=28003600)
356
+ 340 train 7.760486 (lr=1.4810e-03) (hash(x)=27586845)
357
+ 341 train 7.696447 (lr=1.4803e-03) (hash(x)=21176405)
358
+ 342 train 7.651995 (lr=1.4796e-03) (hash(x)=25888433)
359
+ 343 train 7.759183 (lr=1.4789e-03) (hash(x)=27916982)
360
+ 344 train 7.615705 (lr=1.4781e-03) (hash(x)=24334708)
361
+ 345 train 7.668167 (lr=1.4774e-03) (hash(x)=25617709)
362
+ 346 train 7.782055 (lr=1.4766e-03) (hash(x)=25152362)
363
+ 347 train 7.660821 (lr=1.4758e-03) (hash(x)=24276512)
364
+ 348 train 7.582575 (lr=1.4750e-03) (hash(x)=26776133)
365
+ 349 train 7.499701 (lr=1.4742e-03) (hash(x)=20254159)
366
+ 350 val loss 7.6922
367
+ 350 val perplexity 2191.2405
368
+ 350 train 7.571799 (lr=1.4734e-03) (hash(x)=27873855)
369
+ 351 train 7.654586 (lr=1.4726e-03) (hash(x)=25113614)
370
+ 352 train 7.705471 (lr=1.4717e-03) (hash(x)=26192886)
371
+ 353 train 7.605590 (lr=1.4709e-03) (hash(x)=24614912)
372
+ 354 train 7.602164 (lr=1.4700e-03) (hash(x)=25397093)
373
+ 355 train 7.621955 (lr=1.4691e-03) (hash(x)=19811802)
374
+ 356 train 7.612000 (lr=1.4682e-03) (hash(x)=25579655)
375
+ 357 train 7.730766 (lr=1.4673e-03) (hash(x)=26096514)
376
+ 358 train 7.633190 (lr=1.4664e-03) (hash(x)=22589633)
377
+ 359 train 7.646388 (lr=1.4655e-03) (hash(x)=26757147)
378
+ 360 train 7.728815 (lr=1.4645e-03) (hash(x)=24963730)
379
+ 361 train 7.589528 (lr=1.4636e-03) (hash(x)=27703124)
380
+ 362 train 7.494028 (lr=1.4626e-03) (hash(x)=24781713)
381
+ 363 train 7.676754 (lr=1.4616e-03) (hash(x)=24720476)
382
+ 364 train 7.654010 (lr=1.4606e-03) (hash(x)=23481301)
383
+ 365 train 7.575866 (lr=1.4596e-03) (hash(x)=25641033)
384
+ 366 train 7.610944 (lr=1.4586e-03) (hash(x)=27369437)
385
+ 367 train 7.659732 (lr=1.4576e-03) (hash(x)=24760900)
386
+ 368 train 7.558006 (lr=1.4565e-03) (hash(x)=21692546)
387
+ 369 train 7.615930 (lr=1.4555e-03) (hash(x)=22720391)
388
+ 370 train 7.567161 (lr=1.4544e-03) (hash(x)=24376979)
389
+ 371 train 7.607064 (lr=1.4533e-03) (hash(x)=22244509)
390
+ 372 train 7.547781 (lr=1.4522e-03) (hash(x)=21831960)
391
+ 373 train 7.542116 (lr=1.4511e-03) (hash(x)=24921535)
392
+ 374 train 7.580855 (lr=1.4500e-03) (hash(x)=25381251)
393
+ 375 train 7.564873 (lr=1.4489e-03) (hash(x)=25687352)
394
+ 376 train 7.586167 (lr=1.4478e-03) (hash(x)=25221654)
395
+ 377 train 7.617587 (lr=1.4466e-03) (hash(x)=24469863)
396
+ 378 train 7.691230 (lr=1.4454e-03) (hash(x)=25291413)
397
+ 379 train 7.588692 (lr=1.4443e-03) (hash(x)=25744274)
398
+ 380 train 7.528543 (lr=1.4431e-03) (hash(x)=24246963)
399
+ 381 train 7.528052 (lr=1.4419e-03) (hash(x)=19724058)
400
+ 382 train 7.421812 (lr=1.4407e-03) (hash(x)=20705649)
401
+ 383 train 7.741775 (lr=1.4394e-03) (hash(x)=24938685)
402
+ 384 train 7.765732 (lr=1.4382e-03) (hash(x)=21707000)
403
+ 385 train 7.546111 (lr=1.4370e-03) (hash(x)=24948980)
404
+ 386 train 7.848774 (lr=1.4357e-03) (hash(x)=30773958)
405
+ 387 train 7.957592 (lr=1.4344e-03) (hash(x)=26769046)
406
+ 388 train 7.631120 (lr=1.4332e-03) (hash(x)=18849656)
407
+ 389 train 8.082865 (lr=1.4319e-03) (hash(x)=24624950)
408
+ 390 train 7.684142 (lr=1.4306e-03) (hash(x)=27299605)
409
+ 391 train 7.721142 (lr=1.4292e-03) (hash(x)=24113445)
410
+ 392 train 7.752143 (lr=1.4279e-03) (hash(x)=32644465)
411
+ 393 train 7.869693 (lr=1.4266e-03) (hash(x)=25829219)
412
+ 394 train 7.697062 (lr=1.4252e-03) (hash(x)=25194550)
413
+ 395 train 7.748827 (lr=1.4238e-03) (hash(x)=25941804)
414
+ 396 train 7.718336 (lr=1.4225e-03) (hash(x)=24075727)
415
+ 397 train 7.757566 (lr=1.4211e-03) (hash(x)=23104539)
416
+ 398 train 7.733985 (lr=1.4197e-03) (hash(x)=23018983)
417
+ 399 train 7.705534 (lr=1.4183e-03) (hash(x)=24422929)
418
+ 400 val loss 7.7451
419
+ 400 val perplexity 2310.1960
420
+ 400 train 7.712281 (lr=1.4169e-03) (hash(x)=23389696)
421
+ 401 train 7.674834 (lr=1.4154e-03) (hash(x)=29102969)
422
+ 402 train 7.797607 (lr=1.4140e-03) (hash(x)=28359859)
423
+ 403 train 7.824946 (lr=1.4125e-03) (hash(x)=24630955)
424
+ 404 train 7.807680 (lr=1.4110e-03) (hash(x)=22969113)
425
+ 405 train 7.610856 (lr=1.4096e-03) (hash(x)=27004415)
426
+ 406 train 7.535142 (lr=1.4081e-03) (hash(x)=22265063)
427
+ 407 train 7.581593 (lr=1.4066e-03) (hash(x)=26905990)
428
+ 408 train 7.646751 (lr=1.4051e-03) (hash(x)=24047679)
429
+ 409 train 7.775494 (lr=1.4035e-03) (hash(x)=26905582)
430
+ 410 train 7.549853 (lr=1.4020e-03) (hash(x)=29018970)
431
+ 411 train 7.571667 (lr=1.4005e-03) (hash(x)=24503315)
432
+ 412 train 7.565040 (lr=1.3989e-03) (hash(x)=22607537)
433
+ 413 train 8.223868 (lr=1.3973e-03) (hash(x)=26690225)
434
+ 414 train 7.604375 (lr=1.3957e-03) (hash(x)=24226255)
435
+ 415 train 7.556463 (lr=1.3942e-03) (hash(x)=23622702)
436
+ 416 train 7.607314 (lr=1.3926e-03) (hash(x)=23910425)
437
+ 417 train 7.858437 (lr=1.3909e-03) (hash(x)=27213318)
438
+ 418 train 7.658178 (lr=1.3893e-03) (hash(x)=26701577)
439
+ 419 train 7.560061 (lr=1.3877e-03) (hash(x)=26216794)
440
+ 420 train 7.817292 (lr=1.3860e-03) (hash(x)=27579623)
441
+ 421 train 7.665510 (lr=1.3844e-03) (hash(x)=26086198)
442
+ 422 train 7.633306 (lr=1.3827e-03) (hash(x)=24863441)
443
+ 423 train 7.765737 (lr=1.3810e-03) (hash(x)=25236385)
444
+ 424 train 7.805654 (lr=1.3793e-03) (hash(x)=25330803)
445
+ 425 train 7.627540 (lr=1.3776e-03) (hash(x)=22508532)
446
+ 426 train 7.630258 (lr=1.3759e-03) (hash(x)=21799261)
447
+ 427 train 7.716634 (lr=1.3742e-03) (hash(x)=26311168)
448
+ 428 train 7.515151 (lr=1.3725e-03) (hash(x)=20977430)
449
+ 429 train 7.601658 (lr=1.3707e-03) (hash(x)=26941617)
450
+ 430 train 7.845469 (lr=1.3690e-03) (hash(x)=30008957)
451
+ 431 train 7.712615 (lr=1.3672e-03) (hash(x)=27125962)
452
+ 432 train 7.558764 (lr=1.3654e-03) (hash(x)=25888358)
453
+ 433 train 7.719246 (lr=1.3637e-03) (hash(x)=21725719)
454
+ 434 train 7.577919 (lr=1.3619e-03) (hash(x)=21669704)
455
+ 435 train 7.527490 (lr=1.3601e-03) (hash(x)=23166993)
456
+ 436 train 7.618073 (lr=1.3582e-03) (hash(x)=26235974)
457
+ 437 train 7.777010 (lr=1.3564e-03) (hash(x)=25426430)
458
+ 438 train 7.520299 (lr=1.3546e-03) (hash(x)=22966314)
459
+ 439 train 7.663430 (lr=1.3527e-03) (hash(x)=23448505)
460
+ 440 train 7.515452 (lr=1.3509e-03) (hash(x)=26330693)
461
+ 441 train 7.470123 (lr=1.3490e-03) (hash(x)=29550596)
462
+ 442 train 7.538545 (lr=1.3471e-03) (hash(x)=22124892)
463
+ 443 train 7.558524 (lr=1.3452e-03) (hash(x)=25607640)
464
+ 444 train 7.446533 (lr=1.3434e-03) (hash(x)=23795894)
465
+ 445 train 7.293677 (lr=1.3414e-03) (hash(x)=24165449)
466
+ 446 train 7.611935 (lr=1.3395e-03) (hash(x)=24697756)
467
+ 447 train 7.700320 (lr=1.3376e-03) (hash(x)=27218899)
468
+ 448 train 7.583130 (lr=1.3357e-03) (hash(x)=24748522)
469
+ 449 train 7.685932 (lr=1.3337e-03) (hash(x)=24965317)
470
+ 450 val loss 7.6971
471
+ 450 val perplexity 2201.9817
472
+ 450 train 7.479536 (lr=1.3318e-03) (hash(x)=22031210)
473
+ 451 train 7.586784 (lr=1.3298e-03) (hash(x)=25877990)
474
+ 452 train 7.598782 (lr=1.3278e-03) (hash(x)=24382381)
475
+ 453 train 7.508734 (lr=1.3258e-03) (hash(x)=24551999)
476
+ 454 train 7.615926 (lr=1.3238e-03) (hash(x)=25765516)
477
+ 455 train 7.650156 (lr=1.3218e-03) (hash(x)=25204247)
478
+ 456 train 7.686562 (lr=1.3198e-03) (hash(x)=26321813)
479
+ 457 train 7.603608 (lr=1.3178e-03) (hash(x)=25528323)
480
+ 458 train 7.597541 (lr=1.3158e-03) (hash(x)=25114729)
481
+ 459 train 7.548402 (lr=1.3137e-03) (hash(x)=31341964)
482
+ 460 train 7.667234 (lr=1.3117e-03) (hash(x)=26326024)
483
+ 461 train 7.534320 (lr=1.3096e-03) (hash(x)=27273311)
484
+ 462 train 7.505772 (lr=1.3075e-03) (hash(x)=17819313)
485
+ 463 train 7.758073 (lr=1.3054e-03) (hash(x)=23769243)
486
+ 464 train 7.698005 (lr=1.3033e-03) (hash(x)=23788738)
487
+ 465 train 7.670488 (lr=1.3012e-03) (hash(x)=24818744)
488
+ 466 train 7.602197 (lr=1.2991e-03) (hash(x)=23472795)
489
+ 467 train 8.034423 (lr=1.2970e-03) (hash(x)=24991832)
490
+ 468 train 7.796598 (lr=1.2949e-03) (hash(x)=22424066)
491
+ 469 train 7.680362 (lr=1.2928e-03) (hash(x)=26474878)
492
+ 470 train 7.735012 (lr=1.2906e-03) (hash(x)=22704349)
493
+ 471 train 7.650825 (lr=1.2885e-03) (hash(x)=22293673)
494
+ 472 train 7.636174 (lr=1.2863e-03) (hash(x)=21785487)
495
+ 473 train 7.662414 (lr=1.2841e-03) (hash(x)=25251063)
496
+ 474 train 7.556462 (lr=1.2819e-03) (hash(x)=22722211)
497
+ 475 train 7.615918 (lr=1.2797e-03) (hash(x)=24522269)
498
+ 476 train 7.553814 (lr=1.2775e-03) (hash(x)=23329475)
499
+ 477 train 7.587357 (lr=1.2753e-03) (hash(x)=24735165)
500
+ 478 train 7.369074 (lr=1.2731e-03) (hash(x)=27002892)
501
+ 479 train 7.441413 (lr=1.2709e-03) (hash(x)=23499349)
502
+ 480 train 7.777852 (lr=1.2687e-03) (hash(x)=22461589)
503
+ 481 train 7.550361 (lr=1.2664e-03) (hash(x)=26415040)
504
+ 482 train 7.766554 (lr=1.2642e-03) (hash(x)=26349465)
505
+ 483 train 7.419320 (lr=1.2619e-03) (hash(x)=21596677)
506
+ 484 train 7.486763 (lr=1.2596e-03) (hash(x)=24117012)
507
+ 485 train 7.576591 (lr=1.2574e-03) (hash(x)=24105761)
508
+ 486 train 7.536746 (lr=1.2551e-03) (hash(x)=25725049)
509
+ 487 train 7.396511 (lr=1.2528e-03) (hash(x)=22892776)
510
+ 488 train 7.497412 (lr=1.2505e-03) (hash(x)=21630383)
511
+ 489 train 7.526893 (lr=1.2482e-03) (hash(x)=24727094)
512
+ 490 train 7.472238 (lr=1.2459e-03) (hash(x)=22835000)
513
+ 491 train 7.576504 (lr=1.2435e-03) (hash(x)=22839049)
514
+ 492 train 7.440145 (lr=1.2412e-03) (hash(x)=20654136)
515
+ 493 train 7.668361 (lr=1.2389e-03) (hash(x)=26776000)
516
+ 494 train 7.295555 (lr=1.2365e-03) (hash(x)=22560658)
517
+ 495 train 7.722666 (lr=1.2341e-03) (hash(x)=25995933)
518
+ 496 train 7.647247 (lr=1.2318e-03) (hash(x)=24635394)
519
+ 497 train 7.757373 (lr=1.2294e-03) (hash(x)=27537228)
520
+ 498 train 7.688357 (lr=1.2270e-03) (hash(x)=26496730)
521
+ 499 train 7.387195 (lr=1.2246e-03) (hash(x)=16533310)
522
+ 500 val loss 7.6304
523
+ 500 val perplexity 2059.8457
524
+ 500 train 7.514225 (lr=1.2222e-03) (hash(x)=24881176)
525
+ 501 train 7.719615 (lr=1.2198e-03) (hash(x)=25593613)
526
+ 502 train 7.482310 (lr=1.2174e-03) (hash(x)=22886646)
527
+ 503 train 7.716130 (lr=1.2150e-03) (hash(x)=21407676)
528
+ 504 train 7.616419 (lr=1.2126e-03) (hash(x)=21720304)
529
+ 505 train 7.742392 (lr=1.2101e-03) (hash(x)=27263416)
530
+ 506 train 7.504223 (lr=1.2077e-03) (hash(x)=25840801)
531
+ 507 train 7.637878 (lr=1.2052e-03) (hash(x)=31424568)
532
+ 508 train 7.641825 (lr=1.2028e-03) (hash(x)=23403065)
533
+ 509 train 7.464435 (lr=1.2003e-03) (hash(x)=22656033)
534
+ 510 train 7.592577 (lr=1.1978e-03) (hash(x)=26869295)
535
+ 511 train 7.830931 (lr=1.1954e-03) (hash(x)=25504151)
536
+ 512 train 7.610022 (lr=1.1929e-03) (hash(x)=25781518)
537
+ 513 train 7.578055 (lr=1.1904e-03) (hash(x)=26432850)
538
+ 514 train 7.472413 (lr=1.1879e-03) (hash(x)=20155894)
539
+ 515 train 7.388950 (lr=1.1854e-03) (hash(x)=22270169)
540
+ 516 train 7.539298 (lr=1.1829e-03) (hash(x)=22108461)
541
+ 517 train 7.568334 (lr=1.1803e-03) (hash(x)=24802848)
542
+ 518 train 7.668356 (lr=1.1778e-03) (hash(x)=25086604)
543
+ 519 train 7.674907 (lr=1.1753e-03) (hash(x)=26983186)
544
+ 520 train 7.543956 (lr=1.1727e-03) (hash(x)=27208507)
545
+ 521 train 7.731969 (lr=1.1702e-03) (hash(x)=23862845)
546
+ 522 train 7.536292 (lr=1.1676e-03) (hash(x)=22664494)
547
+ 523 train 7.525744 (lr=1.1651e-03) (hash(x)=24092927)
548
+ 524 train 7.528088 (lr=1.1625e-03) (hash(x)=27213812)
549
+ 525 train 7.503677 (lr=1.1599e-03) (hash(x)=23929955)
550
+ 526 train 7.503210 (lr=1.1573e-03) (hash(x)=23025790)
551
+ 527 train 7.429008 (lr=1.1548e-03) (hash(x)=20452272)
552
+ 528 train 7.609464 (lr=1.1522e-03) (hash(x)=26473464)
553
+ 529 train 7.603611 (lr=1.1496e-03) (hash(x)=28276603)
554
+ 530 train 7.421005 (lr=1.1470e-03) (hash(x)=21480367)
555
+ 531 train 7.599454 (lr=1.1443e-03) (hash(x)=24658642)
556
+ 532 train 7.524871 (lr=1.1417e-03) (hash(x)=26854801)
557
+ 533 train 7.411304 (lr=1.1391e-03) (hash(x)=17862226)
558
+ 534 train 7.510953 (lr=1.1365e-03) (hash(x)=28519182)
559
+ 535 train 7.507989 (lr=1.1338e-03) (hash(x)=25193694)
560
+ 536 train 7.476536 (lr=1.1312e-03) (hash(x)=25642620)
561
+ 537 train 7.554841 (lr=1.1285e-03) (hash(x)=25765909)
562
+ 538 train 7.387722 (lr=1.1259e-03) (hash(x)=20750531)
563
+ 539 train 7.488647 (lr=1.1232e-03) (hash(x)=24917098)
564
+ 540 train 7.442611 (lr=1.1205e-03) (hash(x)=24499766)
565
+ 541 train 7.503558 (lr=1.1179e-03) (hash(x)=25067358)
566
+ 542 train 7.595992 (lr=1.1152e-03) (hash(x)=27479796)
567
+ 543 train 7.484142 (lr=1.1125e-03) (hash(x)=23136080)
568
+ 544 train 7.574373 (lr=1.1098e-03) (hash(x)=29810074)
569
+ 545 train 7.479912 (lr=1.1071e-03) (hash(x)=24765539)
570
+ 546 train 7.518897 (lr=1.1044e-03) (hash(x)=22216309)
571
+ 547 train 7.450561 (lr=1.1017e-03) (hash(x)=20613861)
572
+ 548 train 7.434764 (lr=1.0990e-03) (hash(x)=23462447)
573
+ 549 train 7.492900 (lr=1.0963e-03) (hash(x)=26739455)
574
+ 550 val loss 7.5767
575
+ 550 val perplexity 1952.2520
576
+ 550 train 7.517932 (lr=1.0936e-03) (hash(x)=23222996)
577
+ 551 train 7.515204 (lr=1.0908e-03) (hash(x)=26489285)
578
+ 552 train 7.373601 (lr=1.0881e-03) (hash(x)=25185942)
579
+ 553 train 7.698931 (lr=1.0854e-03) (hash(x)=27476158)
580
+ 554 train 7.466922 (lr=1.0826e-03) (hash(x)=23653175)
581
+ 555 train 7.443189 (lr=1.0799e-03) (hash(x)=22405076)
582
+ 556 train 7.469342 (lr=1.0771e-03) (hash(x)=25673545)
583
+ 557 train 7.538731 (lr=1.0744e-03) (hash(x)=22675428)
584
+ 558 train 7.651370 (lr=1.0716e-03) (hash(x)=28168456)
585
+ 559 train 7.761604 (lr=1.0688e-03) (hash(x)=26072586)
586
+ 560 train 7.567846 (lr=1.0661e-03) (hash(x)=24974331)
587
+ 561 train 7.474091 (lr=1.0633e-03) (hash(x)=21951984)
588
+ 562 train 7.705449 (lr=1.0605e-03) (hash(x)=25403751)
589
+ 563 train 7.551829 (lr=1.0577e-03) (hash(x)=26396012)
590
+ 564 train 7.561246 (lr=1.0549e-03) (hash(x)=24827190)
591
+ 565 train 7.672754 (lr=1.0521e-03) (hash(x)=24847458)
592
+ 566 train 7.700562 (lr=1.0493e-03) (hash(x)=22159088)
593
+ 567 train 7.648977 (lr=1.0465e-03) (hash(x)=22140034)
594
+ 568 train 7.713692 (lr=1.0437e-03) (hash(x)=19044797)
595
+ 569 train 7.635394 (lr=1.0409e-03) (hash(x)=23932794)
596
+ 570 train 7.645949 (lr=1.0381e-03) (hash(x)=25256890)
597
+ 571 train 7.688557 (lr=1.0353e-03) (hash(x)=24161711)
598
+ 572 train 7.600222 (lr=1.0325e-03) (hash(x)=22036817)
599
+ 573 train 7.457484 (lr=1.0296e-03) (hash(x)=26663208)
600
+ 574 train 7.511261 (lr=1.0268e-03) (hash(x)=24072213)
601
+ 575 train 7.516840 (lr=1.0240e-03) (hash(x)=22766044)
602
+ 576 train 7.658572 (lr=1.0211e-03) (hash(x)=27871153)
603
+ 577 train 7.602333 (lr=1.0183e-03) (hash(x)=24037280)
604
+ 578 train 7.724504 (lr=1.0154e-03) (hash(x)=24121850)
605
+ 579 train 7.571150 (lr=1.0126e-03) (hash(x)=24380031)
606
+ 580 train 7.503993 (lr=1.0097e-03) (hash(x)=23055215)
607
+ 581 train 7.759126 (lr=1.0069e-03) (hash(x)=25329724)
608
+ 582 train 7.485330 (lr=1.0040e-03) (hash(x)=25662408)
609
+ 583 train 7.576025 (lr=1.0011e-03) (hash(x)=26346814)
610
+ 584 train 7.687614 (lr=9.9827e-04) (hash(x)=25283256)
611
+ 585 train 7.508051 (lr=9.9540e-04) (hash(x)=24021771)
612
+ 586 train 7.531367 (lr=9.9252e-04) (hash(x)=23576840)
613
+ 587 train 7.667334 (lr=9.8964e-04) (hash(x)=26414858)
614
+ 588 train 7.481088 (lr=9.8676e-04) (hash(x)=20099261)
615
+ 589 train 7.477084 (lr=9.8387e-04) (hash(x)=22453718)
616
+ 590 train 7.430462 (lr=9.8099e-04) (hash(x)=22529445)
617
+ 591 train 7.497760 (lr=9.7810e-04) (hash(x)=24484543)
618
+ 592 train 7.418398 (lr=9.7520e-04) (hash(x)=24043998)
619
+ 593 train 7.533081 (lr=9.7230e-04) (hash(x)=24301906)
620
+ 594 train 7.473618 (lr=9.6940e-04) (hash(x)=26328881)
621
+ 595 train 7.500732 (lr=9.6650e-04) (hash(x)=23101508)
622
+ 596 train 7.482075 (lr=9.6360e-04) (hash(x)=23527176)
623
+ 597 train 7.968163 (lr=9.6069e-04) (hash(x)=22705673)
624
+ 598 train 7.529497 (lr=9.5778e-04) (hash(x)=28664796)
625
+ 599 train 7.653956 (lr=9.5486e-04) (hash(x)=27102890)
626
+ 600 val loss 7.5918
627
+ 600 val perplexity 1981.9282
628
+ 600 train 7.753281 (lr=9.5195e-04) (hash(x)=26787259)
629
+ 601 train 7.764432 (lr=9.4903e-04) (hash(x)=23396088)
630
+ 602 train 7.507740 (lr=9.4611e-04) (hash(x)=24911831)
631
+ 603 train 7.507074 (lr=9.4319e-04) (hash(x)=23453491)
632
+ 604 train 7.559050 (lr=9.4026e-04) (hash(x)=23453396)
633
+ 605 train 7.638343 (lr=9.3733e-04) (hash(x)=23869612)
634
+ 606 train 7.722578 (lr=9.3440e-04) (hash(x)=26581590)
635
+ 607 train 7.557317 (lr=9.3147e-04) (hash(x)=28056342)
636
+ 608 train 7.612069 (lr=9.2854e-04) (hash(x)=25900866)
637
+ 609 train 7.631943 (lr=9.2560e-04) (hash(x)=24549177)
638
+ 610 train 7.596582 (lr=9.2267e-04) (hash(x)=28239583)
639
+ 611 train 7.602587 (lr=9.1973e-04) (hash(x)=23223554)
640
+ 612 train 7.686102 (lr=9.1678e-04) (hash(x)=24793480)
641
+ 613 train 7.629915 (lr=9.1384e-04) (hash(x)=25415570)
642
+ 614 train 7.674843 (lr=9.1090e-04) (hash(x)=24425760)
643
+ 615 train 7.736821 (lr=9.0795e-04) (hash(x)=23707134)
644
+ 616 train 7.644773 (lr=9.0500e-04) (hash(x)=28321697)
645
+ 617 train 7.578841 (lr=9.0205e-04) (hash(x)=23434031)
646
+ 618 train 7.560505 (lr=8.9910e-04) (hash(x)=27003481)
647
+ 619 train 7.520029 (lr=8.9615e-04) (hash(x)=22236893)
648
+ 620 train 7.557211 (lr=8.9319e-04) (hash(x)=26857458)
649
+ 621 train 7.546520 (lr=8.9024e-04) (hash(x)=26413122)
650
+ 622 train 7.818306 (lr=8.8728e-04) (hash(x)=27696537)
651
+ 623 train 7.589994 (lr=8.8432e-04) (hash(x)=21950234)
652
+ 624 train 7.608563 (lr=8.8136e-04) (hash(x)=24051952)
653
+ 625 train 7.613591 (lr=8.7840e-04) (hash(x)=24029939)
654
+ 626 train 7.735217 (lr=8.7544e-04) (hash(x)=24393346)
655
+ 627 train 7.665747 (lr=8.7248e-04) (hash(x)=26985352)
656
+ 628 train 7.381147 (lr=8.6952e-04) (hash(x)=19493265)
657
+ 629 train 7.491529 (lr=8.6655e-04) (hash(x)=23944010)
658
+ 630 train 7.710825 (lr=8.6359e-04) (hash(x)=21517435)
659
+ 631 train 7.463479 (lr=8.6062e-04) (hash(x)=25025868)
660
+ 632 train 7.498950 (lr=8.5766e-04) (hash(x)=22940997)
661
+ 633 train 7.632817 (lr=8.5469e-04) (hash(x)=25864840)
662
+ 634 train 7.556394 (lr=8.5172e-04) (hash(x)=24682151)
663
+ 635 train 7.465832 (lr=8.4876e-04) (hash(x)=27055330)
664
+ 636 train 7.536569 (lr=8.4579e-04) (hash(x)=22705710)
665
+ 637 train 7.638597 (lr=8.4282e-04) (hash(x)=26775238)
666
+ 638 train 7.740885 (lr=8.3985e-04) (hash(x)=21734331)
667
+ 639 train 7.521539 (lr=8.3688e-04) (hash(x)=25693442)
668
+ 640 train 7.591055 (lr=8.3391e-04) (hash(x)=25290454)
669
+ 641 train 7.547258 (lr=8.3094e-04) (hash(x)=25699860)
670
+ 642 train 7.715824 (lr=8.2797e-04) (hash(x)=19919115)
671
+ 643 train 7.746837 (lr=8.2500e-04) (hash(x)=25586967)
672
+ 644 train 7.523610 (lr=8.2203e-04) (hash(x)=25373506)
673
+ 645 train 7.527848 (lr=8.1906e-04) (hash(x)=26496148)
674
+ 646 train 7.605278 (lr=8.1609e-04) (hash(x)=27395083)
675
+ 647 train 7.747890 (lr=8.1312e-04) (hash(x)=27389709)
676
+ 648 train 7.658159 (lr=8.1015e-04) (hash(x)=29172968)
677
+ 649 train 7.584540 (lr=8.0718e-04) (hash(x)=25464920)
678
+ 650 val loss 7.5500
679
+ 650 val perplexity 1900.6597
680
+ 650 train 7.615782 (lr=8.0421e-04) (hash(x)=25992320)
681
+ 651 train 7.598061 (lr=8.0124e-04) (hash(x)=24949509)
682
+ 652 train 7.634686 (lr=7.9828e-04) (hash(x)=25615151)
683
+ 653 train 7.587420 (lr=7.9531e-04) (hash(x)=27923548)
684
+ 654 train 7.628890 (lr=7.9234e-04) (hash(x)=26387848)
685
+ 655 train 7.577825 (lr=7.8938e-04) (hash(x)=22863518)
686
+ 656 train 7.524947 (lr=7.8641e-04) (hash(x)=20869581)
687
+ 657 train 7.694536 (lr=7.8345e-04) (hash(x)=18133606)
688
+ 658 train 7.577940 (lr=7.8048e-04) (hash(x)=28736122)
689
+ 659 train 7.564063 (lr=7.7752e-04) (hash(x)=25528329)
690
+ 660 train 7.616781 (lr=7.7456e-04) (hash(x)=24020765)
691
+ 661 train 7.639625 (lr=7.7160e-04) (hash(x)=27792245)
692
+ 662 train 7.661139 (lr=7.6864e-04) (hash(x)=28287635)
693
+ 663 train 7.391479 (lr=7.6568e-04) (hash(x)=20951206)
694
+ 664 train 7.482939 (lr=7.6272e-04) (hash(x)=22356452)
695
+ 665 train 7.559048 (lr=7.5976e-04) (hash(x)=24385170)
696
+ 666 train 7.650060 (lr=7.5681e-04) (hash(x)=23455854)
697
+ 667 train 7.620369 (lr=7.5385e-04) (hash(x)=21589405)
698
+ 668 train 7.591280 (lr=7.5090e-04) (hash(x)=24698150)
699
+ 669 train 7.581066 (lr=7.4795e-04) (hash(x)=28010836)
700
+ 670 train 7.726916 (lr=7.4500e-04) (hash(x)=25427381)
701
+ 671 train 7.661282 (lr=7.4205e-04) (hash(x)=27478557)
702
+ 672 train 7.610360 (lr=7.3910e-04) (hash(x)=23865569)
703
+ 673 train 7.674029 (lr=7.3616e-04) (hash(x)=23407132)
704
+ 674 train 7.558631 (lr=7.3322e-04) (hash(x)=20253664)
705
+ 675 train 7.533340 (lr=7.3027e-04) (hash(x)=21087928)
706
+ 676 train 7.614180 (lr=7.2733e-04) (hash(x)=25957589)
707
+ 677 train 7.548669 (lr=7.2440e-04) (hash(x)=27008610)
708
+ 678 train 7.631154 (lr=7.2146e-04) (hash(x)=27580265)
709
+ 679 train 7.625354 (lr=7.1853e-04) (hash(x)=24244798)
710
+ 680 train 7.538243 (lr=7.1560e-04) (hash(x)=24535871)
711
+ 681 train 7.641733 (lr=7.1267e-04) (hash(x)=27558681)
712
+ 682 train 7.521041 (lr=7.0974e-04) (hash(x)=21152793)
713
+ 683 train 7.506861 (lr=7.0681e-04) (hash(x)=28147005)
714
+ 684 train 7.570378 (lr=7.0389e-04) (hash(x)=21933180)
715
+ 685 train 7.554597 (lr=7.0097e-04) (hash(x)=24735895)
716
+ 686 train 7.581790 (lr=6.9805e-04) (hash(x)=26209948)
717
+ 687 train 7.584709 (lr=6.9514e-04) (hash(x)=23358313)
718
+ 688 train 7.521344 (lr=6.9222e-04) (hash(x)=22398001)
719
+ 689 train 7.444656 (lr=6.8931e-04) (hash(x)=23724957)
720
+ 690 train 7.508561 (lr=6.8640e-04) (hash(x)=23083514)
721
+ 691 train 7.383429 (lr=6.8350e-04) (hash(x)=16016924)
722
+ 692 train 7.626807 (lr=6.8060e-04) (hash(x)=27173674)
723
+ 693 train 7.437594 (lr=6.7770e-04) (hash(x)=28056250)
724
+ 694 train 7.559197 (lr=6.7480e-04) (hash(x)=22233617)
725
+ 695 train 7.634691 (lr=6.7190e-04) (hash(x)=23777908)
726
+ 696 train 7.594773 (lr=6.6901e-04) (hash(x)=27842849)
727
+ 697 train 7.601330 (lr=6.6613e-04) (hash(x)=23711679)
728
+ 698 train 7.441089 (lr=6.6324e-04) (hash(x)=17204843)
729
+ 699 train 7.762423 (lr=6.6036e-04) (hash(x)=23909800)
730
+ 700 val loss 7.5392
731
+ 700 val perplexity 1880.2316
732
+ 700 train 7.591136 (lr=6.5748e-04) (hash(x)=25033286)
733
+ 701 train 7.875526 (lr=6.5460e-04) (hash(x)=23608371)
734
+ 702 train 7.671611 (lr=6.5173e-04) (hash(x)=23654625)
735
+ 703 train 7.524208 (lr=6.4886e-04) (hash(x)=22633313)
736
+ 704 train 7.565348 (lr=6.4600e-04) (hash(x)=24267011)
737
+ 705 train 7.511728 (lr=6.4314e-04) (hash(x)=23449416)
738
+ 706 train 7.736560 (lr=6.4028e-04) (hash(x)=28527906)
739
+ 707 train 7.449108 (lr=6.3742e-04) (hash(x)=21189075)
740
+ 708 train 7.468836 (lr=6.3457e-04) (hash(x)=25278645)
741
+ 709 train 7.954392 (lr=6.3172e-04) (hash(x)=23654290)
742
+ 710 train 7.508113 (lr=6.2888e-04) (hash(x)=24571139)
743
+ 711 train 7.597071 (lr=6.2604e-04) (hash(x)=26203172)
744
+ 712 train 7.523687 (lr=6.2320e-04) (hash(x)=19623503)
745
+ 713 train 7.534965 (lr=6.2037e-04) (hash(x)=23970545)
746
+ 714 train 7.689559 (lr=6.1754e-04) (hash(x)=25724900)
747
+ 715 train 7.623696 (lr=6.1472e-04) (hash(x)=23245421)
748
+ 716 train 7.577881 (lr=6.1190e-04) (hash(x)=24087537)
749
+ 717 train 7.500026 (lr=6.0908e-04) (hash(x)=25549646)
750
+ 718 train 7.560578 (lr=6.0627e-04) (hash(x)=26406563)
751
+ 719 train 7.621702 (lr=6.0346e-04) (hash(x)=23972429)
752
+ 720 train 7.662478 (lr=6.0066e-04) (hash(x)=26378040)
753
+ 721 train 7.543540 (lr=5.9786e-04) (hash(x)=24618531)
754
+ 722 train 7.558386 (lr=5.9507e-04) (hash(x)=20135714)
755
+ 723 train 7.650560 (lr=5.9228e-04) (hash(x)=26490981)
756
+ 724 train 7.674230 (lr=5.8949e-04) (hash(x)=23927567)
757
+ 725 train 7.540905 (lr=5.8671e-04) (hash(x)=27943609)
758
+ 726 train 7.582318 (lr=5.8393e-04) (hash(x)=22962561)
759
+ 727 train 7.609220 (lr=5.8116e-04) (hash(x)=24772257)
760
+ 728 train 7.574703 (lr=5.7839e-04) (hash(x)=25539761)
761
+ 729 train 7.677408 (lr=5.7563e-04) (hash(x)=23316262)
762
+ 730 train 7.506718 (lr=5.7288e-04) (hash(x)=23857558)
763
+ 731 train 7.655554 (lr=5.7012e-04) (hash(x)=30252690)
764
+ 732 train 7.707663 (lr=5.6737e-04) (hash(x)=30590162)
765
+ 733 train 7.604349 (lr=5.6463e-04) (hash(x)=25015721)
766
+ 734 train 7.683868 (lr=5.6189e-04) (hash(x)=28639679)
767
+ 735 train 7.585419 (lr=5.5916e-04) (hash(x)=26199058)
768
+ 736 train 7.632648 (lr=5.5643e-04) (hash(x)=28907461)
769
+ 737 train 7.619796 (lr=5.5371e-04) (hash(x)=26466915)
770
+ 738 train 7.684072 (lr=5.5100e-04) (hash(x)=20332613)
771
+ 739 train 7.590016 (lr=5.4828e-04) (hash(x)=26766470)
772
+ 740 train 7.560217 (lr=5.4558e-04) (hash(x)=26722145)
773
+ 741 train 7.595085 (lr=5.4288e-04) (hash(x)=21222489)
774
+ 742 train 7.710942 (lr=5.4018e-04) (hash(x)=24662435)
775
+ 743 train 7.541270 (lr=5.3749e-04) (hash(x)=24308927)
776
+ 744 train 7.517354 (lr=5.3481e-04) (hash(x)=28442502)
777
+ 745 train 7.568849 (lr=5.3213e-04) (hash(x)=22810435)
778
+ 746 train 7.652299 (lr=5.2946e-04) (hash(x)=30486518)
779
+ 747 train 7.637340 (lr=5.2679e-04) (hash(x)=22688712)
780
+ 748 train 7.574488 (lr=5.2413e-04) (hash(x)=20712962)
781
+ 749 train 7.483950 (lr=5.2147e-04) (hash(x)=27490086)
782
+ 750 val loss 7.5612
783
+ 750 val perplexity 1922.1996
784
+ 750 train 7.463133 (lr=5.1882e-04) (hash(x)=23973908)
785
+ 751 train 7.618118 (lr=5.1618e-04) (hash(x)=29132401)
786
+ 752 train 7.569785 (lr=5.1354e-04) (hash(x)=27386700)
787
+ 753 train 7.616271 (lr=5.1091e-04) (hash(x)=24268614)
788
+ 754 train 7.635522 (lr=5.0828e-04) (hash(x)=27378387)
789
+ 755 train 7.483657 (lr=5.0566e-04) (hash(x)=22529400)
790
+ 756 train 7.564470 (lr=5.0305e-04) (hash(x)=27586989)
791
+ 757 train 7.503036 (lr=5.0044e-04) (hash(x)=22336349)
792
+ 758 train 7.536838 (lr=4.9784e-04) (hash(x)=26147854)
793
+ 759 train 7.621642 (lr=4.9525e-04) (hash(x)=25346788)
794
+ 760 train 7.670808 (lr=4.9266e-04) (hash(x)=25956855)
795
+ 761 train 7.627720 (lr=4.9008e-04) (hash(x)=24367780)
796
+ 762 train 7.514072 (lr=4.8750e-04) (hash(x)=26084462)
797
+ 763 train 7.589059 (lr=4.8493e-04) (hash(x)=26702459)
798
+ 764 train 7.524665 (lr=4.8237e-04) (hash(x)=25717976)
799
+ 765 train 7.491129 (lr=4.7981e-04) (hash(x)=22141366)
800
+ 766 train 7.674084 (lr=4.7726e-04) (hash(x)=24652423)
801
+ 767 train 7.604540 (lr=4.7472e-04) (hash(x)=30911699)
802
+ 768 train 7.581646 (lr=4.7219e-04) (hash(x)=22076954)
803
+ 769 train 7.751569 (lr=4.6966e-04) (hash(x)=21586769)
804
+ 770 train 7.587068 (lr=4.6714e-04) (hash(x)=21994337)
805
+ 771 train 7.556141 (lr=4.6462e-04) (hash(x)=25454145)
806
+ 772 train 7.625103 (lr=4.6211e-04) (hash(x)=23239550)
807
+ 773 train 7.657763 (lr=4.5961e-04) (hash(x)=23565926)
808
+ 774 train 7.525883 (lr=4.5712e-04) (hash(x)=24705563)
809
+ 775 train 7.606966 (lr=4.5463e-04) (hash(x)=27671095)
810
+ 776 train 7.725169 (lr=4.5215e-04) (hash(x)=26254303)
811
+ 777 train 7.504974 (lr=4.4968e-04) (hash(x)=27454020)
812
+ 778 train 7.494703 (lr=4.4722e-04) (hash(x)=24394565)
813
+ 779 train 7.446939 (lr=4.4476e-04) (hash(x)=21780199)
814
+ 780 train 7.530329 (lr=4.4231e-04) (hash(x)=28296896)
815
+ 781 train 7.625839 (lr=4.3987e-04) (hash(x)=26766303)
816
+ 782 train 7.516381 (lr=4.3743e-04) (hash(x)=21599973)
817
+ 783 train 7.780479 (lr=4.3500e-04) (hash(x)=24969360)
818
+ 784 train 7.796500 (lr=4.3258e-04) (hash(x)=24382006)
819
+ 785 train 7.469198 (lr=4.3017e-04) (hash(x)=22009299)
820
+ 786 train 7.339222 (lr=4.2776e-04) (hash(x)=22376569)
821
+ 787 train 7.493520 (lr=4.2537e-04) (hash(x)=28166843)
822
+ 788 train 7.731756 (lr=4.2298e-04) (hash(x)=28230119)
823
+ 789 train 7.695768 (lr=4.2060e-04) (hash(x)=26207465)
824
+ 790 train 7.744191 (lr=4.1822e-04) (hash(x)=24748118)
825
+ 791 train 7.674869 (lr=4.1586e-04) (hash(x)=27033802)
826
+ 792 train 7.616435 (lr=4.1350e-04) (hash(x)=28788047)
827
+ 793 train 7.521762 (lr=4.1115e-04) (hash(x)=22185099)
828
+ 794 train 7.658455 (lr=4.0880e-04) (hash(x)=21585136)
829
+ 795 train 7.508509 (lr=4.0647e-04) (hash(x)=27691968)
830
+ 796 train 7.543262 (lr=4.0414e-04) (hash(x)=23706103)
831
+ 797 train 7.413793 (lr=4.0183e-04) (hash(x)=27918030)
832
+ 798 train 7.601877 (lr=3.9952e-04) (hash(x)=23127886)
833
+ 799 train 7.613979 (lr=3.9722e-04) (hash(x)=26197070)
834
+ 800 val loss 7.5410
835
+ 800 val perplexity 1883.7189
836
+ 800 train 7.657513 (lr=3.9492e-04) (hash(x)=26463005)
837
+ 801 train 7.571740 (lr=3.9264e-04) (hash(x)=19160134)
838
+ 802 train 7.519664 (lr=3.9036e-04) (hash(x)=23699136)
839
+ 803 train 7.587438 (lr=3.8809e-04) (hash(x)=25353051)
840
+ 804 train 7.508205 (lr=3.8583e-04) (hash(x)=21558484)
841
+ 805 train 7.448887 (lr=3.8358e-04) (hash(x)=22828380)
842
+ 806 train 7.565960 (lr=3.8134e-04) (hash(x)=24623811)
843
+ 807 train 7.701390 (lr=3.7910e-04) (hash(x)=25410299)
844
+ 808 train 7.576618 (lr=3.7688e-04) (hash(x)=26046091)
845
+ 809 train 7.721347 (lr=3.7466e-04) (hash(x)=29887482)
846
+ 810 train 7.468916 (lr=3.7245e-04) (hash(x)=21809706)
847
+ 811 train 7.587492 (lr=3.7026e-04) (hash(x)=23712872)
848
+ 812 train 7.496187 (lr=3.6806e-04) (hash(x)=25875103)
849
+ 813 train 7.452505 (lr=3.6588e-04) (hash(x)=23736441)
850
+ 814 train 7.526761 (lr=3.6371e-04) (hash(x)=25093883)
851
+ 815 train 7.489984 (lr=3.6155e-04) (hash(x)=23692325)
852
+ 816 train 7.365784 (lr=3.5939e-04) (hash(x)=23061082)
853
+ 817 train 7.540975 (lr=3.5725e-04) (hash(x)=26094253)
854
+ 818 train 7.622797 (lr=3.5511e-04) (hash(x)=27327886)
855
+ 819 train 7.509390 (lr=3.5298e-04) (hash(x)=26186143)
856
+ 820 train 7.509499 (lr=3.5086e-04) (hash(x)=25336043)
857
+ 821 train 7.565090 (lr=3.4875e-04) (hash(x)=22811468)
858
+ 822 train 7.536348 (lr=3.4665e-04) (hash(x)=27925831)
859
+ 823 train 7.616776 (lr=3.4456e-04) (hash(x)=35528108)
860
+ 824 train 7.611670 (lr=3.4248e-04) (hash(x)=24949421)
861
+ 825 train 7.403137 (lr=3.4041e-04) (hash(x)=22845735)
862
+ 826 train 7.506923 (lr=3.3835e-04) (hash(x)=28138281)
863
+ 827 train 7.572747 (lr=3.3629e-04) (hash(x)=29038360)
864
+ 828 train 7.529845 (lr=3.3425e-04) (hash(x)=22827903)
865
+ 829 train 7.604974 (lr=3.3221e-04) (hash(x)=24647015)
866
+ 830 train 7.503493 (lr=3.3019e-04) (hash(x)=20477804)
867
+ 831 train 7.468326 (lr=3.2817e-04) (hash(x)=26692640)
868
+ 832 train 7.544086 (lr=3.2617e-04) (hash(x)=27388794)
869
+ 833 train 7.773536 (lr=3.2417e-04) (hash(x)=25265875)
870
+ 834 train 7.465349 (lr=3.2219e-04) (hash(x)=21841888)
871
+ 835 train 7.541018 (lr=3.2021e-04) (hash(x)=19809196)
872
+ 836 train 7.618773 (lr=3.1824e-04) (hash(x)=23007945)
873
+ 837 train 7.560459 (lr=3.1629e-04) (hash(x)=24079409)
874
+ 838 train 7.473290 (lr=3.1434e-04) (hash(x)=25874533)
875
+ 839 train 7.451907 (lr=3.1240e-04) (hash(x)=23660106)
876
+ 840 train 7.854483 (lr=3.1047e-04) (hash(x)=41821329)
877
+ 841 train 7.524031 (lr=3.0856e-04) (hash(x)=25763899)
878
+ 842 train 7.352577 (lr=3.0665e-04) (hash(x)=28584111)
879
+ 843 train 7.801500 (lr=3.0475e-04) (hash(x)=25559887)
880
+ 844 train 7.613263 (lr=3.0286e-04) (hash(x)=26886377)
881
+ 845 train 7.578320 (lr=3.0099e-04) (hash(x)=26461895)
882
+ 846 train 7.610428 (lr=2.9912e-04) (hash(x)=23018827)
883
+ 847 train 7.438640 (lr=2.9726e-04) (hash(x)=24375035)
884
+ 848 train 7.572320 (lr=2.9542e-04) (hash(x)=26431670)
885
+ 849 train 7.476664 (lr=2.9358e-04) (hash(x)=25033511)
886
+ 850 val loss 7.5310
887
+ 850 val perplexity 1865.0392
888
+ 850 train 7.569345 (lr=2.9175e-04) (hash(x)=25602577)
889
+ 851 train 7.529427 (lr=2.8994e-04) (hash(x)=23316582)
890
+ 852 train 7.600564 (lr=2.8813e-04) (hash(x)=22470095)
891
+ 853 train 7.505586 (lr=2.8634e-04) (hash(x)=25089202)
892
+ 854 train 7.470884 (lr=2.8455e-04) (hash(x)=26243247)
893
+ 855 train 7.535011 (lr=2.8278e-04) (hash(x)=22231144)
894
+ 856 train 7.450801 (lr=2.8102e-04) (hash(x)=27646564)
895
+ 857 train 7.582968 (lr=2.7926e-04) (hash(x)=23889875)
896
+ 858 train 7.542249 (lr=2.7752e-04) (hash(x)=25255240)
897
+ 859 train 7.622897 (lr=2.7579e-04) (hash(x)=23883299)
898
+ 860 train 7.592169 (lr=2.7407e-04) (hash(x)=18463841)
899
+ 861 train 7.601661 (lr=2.7236e-04) (hash(x)=30657762)
900
+ 862 train 7.510190 (lr=2.7066e-04) (hash(x)=25288079)
901
+ 863 train 7.526364 (lr=2.6897e-04) (hash(x)=28979310)
902
+ 864 train 7.727425 (lr=2.6729e-04) (hash(x)=30254991)
903
+ 865 train 7.580310 (lr=2.6562e-04) (hash(x)=27274653)
904
+ 866 train 7.562562 (lr=2.6396e-04) (hash(x)=23026066)
905
+ 867 train 7.542522 (lr=2.6232e-04) (hash(x)=14001963)
906
+ 868 train 7.445827 (lr=2.6068e-04) (hash(x)=21988054)
907
+ 869 train 7.533130 (lr=2.5906e-04) (hash(x)=26241099)
908
+ 870 train 7.598836 (lr=2.5745e-04) (hash(x)=26209722)
909
+ 871 train 7.482797 (lr=2.5584e-04) (hash(x)=26928312)
910
+ 872 train 7.552778 (lr=2.5425e-04) (hash(x)=23563272)
911
+ 873 train 7.480335 (lr=2.5267e-04) (hash(x)=25704828)
912
+ 874 train 7.717438 (lr=2.5110e-04) (hash(x)=22498992)
913
+ 875 train 7.871714 (lr=2.4955e-04) (hash(x)=22895918)
914
+ 876 train 7.478352 (lr=2.4800e-04) (hash(x)=24404283)
915
+ 877 train 7.504698 (lr=2.4646e-04) (hash(x)=26663932)
916
+ 878 train 7.619755 (lr=2.4494e-04) (hash(x)=22805316)
917
+ 879 train 7.578790 (lr=2.4343e-04) (hash(x)=23344466)
918
+ 880 train 7.527226 (lr=2.4192e-04) (hash(x)=20966940)
919
+ 881 train 7.534534 (lr=2.4043e-04) (hash(x)=27422664)
920
+ 882 train 7.592794 (lr=2.3895e-04) (hash(x)=22116510)
921
+ 883 train 7.491427 (lr=2.3749e-04) (hash(x)=27269633)
922
+ 884 train 7.557211 (lr=2.3603e-04) (hash(x)=27157711)
923
+ 885 train 7.578829 (lr=2.3458e-04) (hash(x)=22774645)
924
+ 886 train 7.496717 (lr=2.3315e-04) (hash(x)=22143146)
925
+ 887 train 7.523676 (lr=2.3173e-04) (hash(x)=22356438)
926
+ 888 train 7.299496 (lr=2.3032e-04) (hash(x)=23172158)
927
+ 889 train 7.529963 (lr=2.2892e-04) (hash(x)=26971690)
928
+ 890 train 7.562895 (lr=2.2753e-04) (hash(x)=24851922)
929
+ 891 train 7.423007 (lr=2.2615e-04) (hash(x)=21017641)
930
+ 892 train 7.544005 (lr=2.2479e-04) (hash(x)=28186957)
931
+ 893 train 7.548420 (lr=2.2344e-04) (hash(x)=25661082)
932
+ 894 train 7.461651 (lr=2.2209e-04) (hash(x)=25462781)
933
+ 895 train 7.430133 (lr=2.2076e-04) (hash(x)=21774374)
934
+ 896 train 7.430188 (lr=2.1945e-04) (hash(x)=23657055)
935
+ 897 train 7.424705 (lr=2.1814e-04) (hash(x)=23604050)
936
+ 898 train 7.838993 (lr=2.1685e-04) (hash(x)=31534071)
937
+ 899 train 7.661458 (lr=2.1556e-04) (hash(x)=24183732)
938
+ 900 val loss 7.5207
939
+ 900 val perplexity 1845.8739
940
+ 900 train 7.538605 (lr=2.1429e-04) (hash(x)=23384280)
941
+ 901 train 7.416178 (lr=2.1303e-04) (hash(x)=21975928)
942
+ 902 train 7.418263 (lr=2.1179e-04) (hash(x)=25281674)
943
+ 903 train 7.402103 (lr=2.1055e-04) (hash(x)=22947712)
944
+ 904 train 7.486440 (lr=2.0933e-04) (hash(x)=24853977)
945
+ 905 train 7.514587 (lr=2.0812e-04) (hash(x)=23696470)
946
+ 906 train 7.502135 (lr=2.0692e-04) (hash(x)=27489711)
947
+ 907 train 7.462066 (lr=2.0573e-04) (hash(x)=28050163)
948
+ 908 train 7.359514 (lr=2.0455e-04) (hash(x)=26996807)
949
+ 909 train 7.697375 (lr=2.0339e-04) (hash(x)=28626495)
950
+ 910 train 7.427105 (lr=2.0224e-04) (hash(x)=24783242)
951
+ 911 train 7.491423 (lr=2.0110e-04) (hash(x)=25832479)
952
+ 912 train 7.336651 (lr=1.9997e-04) (hash(x)=23522137)
953
+ 913 train 7.628847 (lr=1.9885e-04) (hash(x)=26731973)
954
+ 914 train 7.671913 (lr=1.9775e-04) (hash(x)=26019142)
955
+ 915 train 7.488277 (lr=1.9666e-04) (hash(x)=25081877)
956
+ 916 train 7.580016 (lr=1.9558e-04) (hash(x)=23843571)
957
+ 917 train 7.592031 (lr=1.9451e-04) (hash(x)=26805739)
958
+ 918 train 7.519039 (lr=1.9346e-04) (hash(x)=27169927)
959
+ 919 train 7.541849 (lr=1.9242e-04) (hash(x)=26545491)
960
+ 920 train 7.529106 (lr=1.9139e-04) (hash(x)=25179497)
961
+ 921 train 7.495086 (lr=1.9037e-04) (hash(x)=27085003)
962
+ 922 train 7.506018 (lr=1.8936e-04) (hash(x)=24558202)
963
+ 923 train 7.625360 (lr=1.8837e-04) (hash(x)=25317684)
964
+ 924 train 7.535580 (lr=1.8739e-04) (hash(x)=27104868)
965
+ 925 train 7.559785 (lr=1.8642e-04) (hash(x)=24806274)
966
+ 926 train 7.598135 (lr=1.8547e-04) (hash(x)=24341937)
967
+ 927 train 7.421186 (lr=1.8452e-04) (hash(x)=22736558)
968
+ 928 train 7.441630 (lr=1.8359e-04) (hash(x)=23843744)
969
+ 929 train 7.531722 (lr=1.8267e-04) (hash(x)=27218468)
970
+ 930 train 7.513059 (lr=1.8176e-04) (hash(x)=26042053)
971
+ 931 train 7.500450 (lr=1.8087e-04) (hash(x)=25776082)
972
+ 932 train 7.427333 (lr=1.7999e-04) (hash(x)=21594447)
973
+ 933 train 7.456558 (lr=1.7912e-04) (hash(x)=21674469)
974
+ 934 train 7.416131 (lr=1.7826e-04) (hash(x)=26213966)
975
+ 935 train 7.409687 (lr=1.7742e-04) (hash(x)=23770702)
976
+ 936 train 7.496741 (lr=1.7659e-04) (hash(x)=27829878)
977
+ 937 train 7.597492 (lr=1.7577e-04) (hash(x)=29177278)
978
+ 938 train 7.545094 (lr=1.7496e-04) (hash(x)=26020089)
979
+ 939 train 7.440755 (lr=1.7417e-04) (hash(x)=24676475)
980
+ 940 train 7.517210 (lr=1.7339e-04) (hash(x)=24056174)
981
+ 941 train 7.515801 (lr=1.7262e-04) (hash(x)=21887514)
982
+ 942 train 7.523308 (lr=1.7186e-04) (hash(x)=24196100)
983
+ 943 train 7.487014 (lr=1.7112e-04) (hash(x)=25700093)
984
+ 944 train 7.596696 (lr=1.7039e-04) (hash(x)=24365834)
985
+ 945 train 7.416509 (lr=1.6967e-04) (hash(x)=25631786)
986
+ 946 train 7.546913 (lr=1.6896e-04) (hash(x)=24452345)
987
+ 947 train 7.307942 (lr=1.6827e-04) (hash(x)=21465330)
988
+ 948 train 7.447302 (lr=1.6759e-04) (hash(x)=20427493)
989
+ 949 train 7.457048 (lr=1.6692e-04) (hash(x)=25439720)
990
+ 950 val loss 7.5220
991
+ 950 val perplexity 1848.2607
992
+ 950 train 7.540026 (lr=1.6627e-04) (hash(x)=21652988)
993
+ 951 train 7.457851 (lr=1.6563e-04) (hash(x)=22546779)
994
+ 952 train 7.340933 (lr=1.6500e-04) (hash(x)=20855736)
995
+ 953 train 7.514574 (lr=1.6438e-04) (hash(x)=27725873)
996
+ 954 train 7.561169 (lr=1.6378e-04) (hash(x)=24582989)
997
+ 955 train 7.437168 (lr=1.6319e-04) (hash(x)=24822372)
998
+ 956 train 7.498088 (lr=1.6261e-04) (hash(x)=25624588)
999
+ 957 train 7.454485 (lr=1.6205e-04) (hash(x)=23566489)
1000
+ 958 train 7.629837 (lr=1.6149e-04) (hash(x)=28117387)
1001
+ 959 train 7.437499 (lr=1.6095e-04) (hash(x)=26874283)
1002
+ 960 train 7.421173 (lr=1.6043e-04) (hash(x)=28570123)
1003
+ 961 train 7.541732 (lr=1.5991e-04) (hash(x)=24807017)
1004
+ 962 train 7.461828 (lr=1.5941e-04) (hash(x)=25189993)
1005
+ 963 train 7.571015 (lr=1.5893e-04) (hash(x)=29202226)
1006
+ 964 train 7.708316 (lr=1.5845e-04) (hash(x)=26060713)
1007
+ 965 train 7.615880 (lr=1.5799e-04) (hash(x)=23053285)
1008
+ 966 train 7.579921 (lr=1.5754e-04) (hash(x)=21968481)
1009
+ 967 train 7.521470 (lr=1.5710e-04) (hash(x)=24492698)
1010
+ 968 train 7.464500 (lr=1.5668e-04) (hash(x)=25760806)
1011
+ 969 train 7.494230 (lr=1.5627e-04) (hash(x)=27417392)
1012
+ 970 train 7.488170 (lr=1.5587e-04) (hash(x)=26767105)
1013
+ 971 train 7.549668 (lr=1.5549e-04) (hash(x)=25117678)
1014
+ 972 train 7.445091 (lr=1.5512e-04) (hash(x)=24233758)
1015
+ 973 train 7.502533 (lr=1.5476e-04) (hash(x)=29907634)
1016
+ 974 train 7.335837 (lr=1.5441e-04) (hash(x)=23552436)
1017
+ 975 train 7.613617 (lr=1.5408e-04) (hash(x)=28610406)
1018
+ 976 train 7.502740 (lr=1.5376e-04) (hash(x)=23858212)
1019
+ 977 train 7.391057 (lr=1.5345e-04) (hash(x)=20567764)
1020
+ 978 train 7.411984 (lr=1.5316e-04) (hash(x)=29631106)
1021
+ 979 train 7.390311 (lr=1.5288e-04) (hash(x)=19843564)
1022
+ 980 train 7.499676 (lr=1.5261e-04) (hash(x)=22589891)
1023
+ 981 train 7.283542 (lr=1.5236e-04) (hash(x)=21183943)
1024
+ 982 train 7.470164 (lr=1.5212e-04) (hash(x)=26152241)
1025
+ 983 train 7.463962 (lr=1.5189e-04) (hash(x)=22700785)
1026
+ 984 train 7.659617 (lr=1.5167e-04) (hash(x)=29611280)
1027
+ 985 train 7.790913 (lr=1.5147e-04) (hash(x)=25972823)
1028
+ 986 train 7.482958 (lr=1.5128e-04) (hash(x)=23447782)
1029
+ 987 train 7.462581 (lr=1.5110e-04) (hash(x)=22578453)
1030
+ 988 train 7.454294 (lr=1.5094e-04) (hash(x)=25474622)
1031
+ 989 train 7.522985 (lr=1.5079e-04) (hash(x)=22965762)
1032
+ 990 train 7.528740 (lr=1.5065e-04) (hash(x)=26116321)
1033
+ 991 train 7.455849 (lr=1.5053e-04) (hash(x)=25934549)
1034
+ 992 train 7.395969 (lr=1.5042e-04) (hash(x)=23937495)
1035
+ 993 train 7.522591 (lr=1.5032e-04) (hash(x)=26286442)
1036
+ 994 train 7.738519 (lr=1.5024e-04) (hash(x)=32737711)
1037
+ 995 train 7.412487 (lr=1.5016e-04) (hash(x)=20563613)
1038
+ 996 train 7.623408 (lr=1.5010e-04) (hash(x)=26234770)
1039
+ 997 train 7.549173 (lr=1.5006e-04) (hash(x)=26179422)
1040
+ 998 train 7.513829 (lr=1.5003e-04) (hash(x)=23289651)
1041
+ 999 val loss 7.5176
1042
+ 999 val perplexity 1840.0624
1043
+ 999 train 7.826050 (lr=1.5001e-04) (hash(x)=36436617)
lr1.5e-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:ad4041d8a9f1bb25cda78370497e93c5b6f061c32b9ad5c2257e205d065e1649
3
+ size 48156226
lr1.5e-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 sha256:5abc3f5ee477481bbe00af0ee838c03e2d1f8113414fbe5d6c4cf8c08fae9666
3
+ size 71157574