andrew-healey commited on
Commit
d8325ad
·
verified ·
1 Parent(s): 76bb476

Upload folder using huggingface_hub

Browse files
lr1e-4_total_batch_size20480_seq_len1024/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep/lr1e-4_total_batch_size20480_seq_len1024", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 2, "protect_bos_token": true, "prevent_from_masking_myself": true, "max_steps": null, "group": "proxy_model_sweep", "use_wandb": true, "kill_self_after_run": false, "random_seed": 1337, "memory_penalty_epsilon": 0.1, "selection_head_linear_combo": "none", "selection_head_linear_combo_scale": 1.0, "protection_kind": "none", "leaky_relu_alpha": null, "leaky_relu_bias": null, "use_compile": true, "use_mini_model": false, "upload_to_hf": true, "seq_len": 1024, "batch_size": null, "total_batch_size": 20480, "protection_head_scaling_factor": 1.0, "protection_head_bias": 0.0, "n_sliced_masks": null, "n_latent_masks": null, "mask_layernorm": false, "residual_attention_masks": false, "compute_base_shapes": false, "base_shapes_savefile": null, "mup": true, "disable_selection": true, "mup_enable_coord_check_logging": false, "max_lr": 0.0001, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr1e-4_total_batch_size20480_seq_len1024/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8513131d2bf732f8f6d97ff82d32c505b2a2f3f9f2247bac315bb09ce0e1f97
3
+ size 964
lr1e-4_total_batch_size20480_seq_len1024/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.7486
3
+ 0 val perplexity 126579.2500
4
+ 0 train 11.749327 (lr=3.4965e-07) (hash(x)=57833283)
5
+ 1 train 11.742214 (lr=6.9930e-07) (hash(x)=52769534)
6
+ 2 train 11.724283 (lr=1.0490e-06) (hash(x)=50373191)
7
+ 3 train 11.749997 (lr=1.3986e-06) (hash(x)=48487754)
8
+ 4 train 11.734939 (lr=1.7483e-06) (hash(x)=47781406)
9
+ 5 train 11.746944 (lr=2.0979e-06) (hash(x)=48658307)
10
+ 6 train 11.756604 (lr=2.4476e-06) (hash(x)=42956464)
11
+ 7 train 11.743142 (lr=2.7972e-06) (hash(x)=50106566)
12
+ 8 train 11.734677 (lr=3.1469e-06) (hash(x)=50260247)
13
+ 9 train 11.722575 (lr=3.4965e-06) (hash(x)=45839663)
14
+ 10 train 11.710037 (lr=3.8462e-06) (hash(x)=54489985)
15
+ 11 train 11.701223 (lr=4.1958e-06) (hash(x)=50770821)
16
+ 12 train 11.697615 (lr=4.5455e-06) (hash(x)=49416075)
17
+ 13 train 11.685614 (lr=4.8951e-06) (hash(x)=52323135)
18
+ 14 train 11.687724 (lr=5.2448e-06) (hash(x)=53922434)
19
+ 15 train 11.669305 (lr=5.5944e-06) (hash(x)=50183213)
20
+ 16 train 11.643473 (lr=5.9441e-06) (hash(x)=52003225)
21
+ 17 train 11.651104 (lr=6.2937e-06) (hash(x)=56845779)
22
+ 18 train 11.631777 (lr=6.6434e-06) (hash(x)=55034939)
23
+ 19 train 11.621422 (lr=6.9930e-06) (hash(x)=52671072)
24
+ 20 train 11.613849 (lr=7.3427e-06) (hash(x)=46979552)
25
+ 21 train 11.607208 (lr=7.6923e-06) (hash(x)=52068996)
26
+ 22 train 11.578172 (lr=8.0420e-06) (hash(x)=54071193)
27
+ 23 train 11.573407 (lr=8.3916e-06) (hash(x)=47640895)
28
+ 24 train 11.552900 (lr=8.7413e-06) (hash(x)=46150086)
29
+ 25 train 11.529486 (lr=9.0909e-06) (hash(x)=45438450)
30
+ 26 train 11.537867 (lr=9.4406e-06) (hash(x)=44573591)
31
+ 27 train 11.525861 (lr=9.7902e-06) (hash(x)=62300690)
32
+ 28 train 11.487083 (lr=1.0140e-05) (hash(x)=50319996)
33
+ 29 train 11.503645 (lr=1.0490e-05) (hash(x)=54520776)
34
+ 30 train 11.446773 (lr=1.0839e-05) (hash(x)=44855848)
35
+ 31 train 11.424117 (lr=1.1189e-05) (hash(x)=54982586)
36
+ 32 train 11.453508 (lr=1.1538e-05) (hash(x)=50784472)
37
+ 33 train 11.373411 (lr=1.1888e-05) (hash(x)=50188334)
38
+ 34 train 11.364918 (lr=1.2238e-05) (hash(x)=50861237)
39
+ 35 train 11.338379 (lr=1.2587e-05) (hash(x)=60945351)
40
+ 36 train 11.341898 (lr=1.2937e-05) (hash(x)=56694638)
41
+ 37 train 11.344574 (lr=1.3287e-05) (hash(x)=58412625)
42
+ 38 train 11.299179 (lr=1.3636e-05) (hash(x)=50750450)
43
+ 39 train 11.296921 (lr=1.3986e-05) (hash(x)=51819337)
44
+ 40 train 11.241918 (lr=1.4336e-05) (hash(x)=48267882)
45
+ 41 train 11.201263 (lr=1.4685e-05) (hash(x)=52889428)
46
+ 42 train 11.135584 (lr=1.5035e-05) (hash(x)=49536098)
47
+ 43 train 11.109211 (lr=1.5385e-05) (hash(x)=50686547)
48
+ 44 train 11.126301 (lr=1.5734e-05) (hash(x)=50818630)
49
+ 45 train 11.050308 (lr=1.6084e-05) (hash(x)=49697510)
50
+ 46 train 11.041220 (lr=1.6434e-05) (hash(x)=53436760)
51
+ 47 train 11.027643 (lr=1.6783e-05) (hash(x)=46807632)
52
+ 48 train 10.982990 (lr=1.7133e-05) (hash(x)=54016065)
53
+ 49 train 11.031576 (lr=1.7483e-05) (hash(x)=57723080)
54
+ 50 val loss 10.9265
55
+ 50 val perplexity 55633.3164
56
+ 50 train 10.913174 (lr=1.7832e-05) (hash(x)=53422909)
57
+ 51 train 10.941579 (lr=1.8182e-05) (hash(x)=48459663)
58
+ 52 train 10.856948 (lr=1.8531e-05) (hash(x)=48794753)
59
+ 53 train 10.843208 (lr=1.8881e-05) (hash(x)=47148322)
60
+ 54 train 10.838713 (lr=1.9231e-05) (hash(x)=52460055)
61
+ 55 train 10.800026 (lr=1.9580e-05) (hash(x)=45337189)
62
+ 56 train 10.742849 (lr=1.9930e-05) (hash(x)=50930055)
63
+ 57 train 10.713366 (lr=2.0280e-05) (hash(x)=42125704)
64
+ 58 train 10.585735 (lr=2.0629e-05) (hash(x)=49174011)
65
+ 59 train 10.658677 (lr=2.0979e-05) (hash(x)=50618057)
66
+ 60 train 10.685741 (lr=2.1329e-05) (hash(x)=52423338)
67
+ 61 train 10.633089 (lr=2.1678e-05) (hash(x)=51754551)
68
+ 62 train 10.689787 (lr=2.2028e-05) (hash(x)=47059587)
69
+ 63 train 10.549004 (lr=2.2378e-05) (hash(x)=52187435)
70
+ 64 train 10.587015 (lr=2.2727e-05) (hash(x)=50919108)
71
+ 65 train 10.560100 (lr=2.3077e-05) (hash(x)=51831972)
72
+ 66 train 10.523424 (lr=2.3427e-05) (hash(x)=49049513)
73
+ 67 train 10.384525 (lr=2.3776e-05) (hash(x)=56443459)
74
+ 68 train 10.536256 (lr=2.4126e-05) (hash(x)=54156964)
75
+ 69 train 10.578791 (lr=2.4476e-05) (hash(x)=54578790)
76
+ 70 train 10.453396 (lr=2.4825e-05) (hash(x)=52813407)
77
+ 71 train 10.457096 (lr=2.5175e-05) (hash(x)=45098704)
78
+ 72 train 10.486520 (lr=2.5524e-05) (hash(x)=42452493)
79
+ 73 train 10.481006 (lr=2.5874e-05) (hash(x)=45655812)
80
+ 74 train 10.434572 (lr=2.6224e-05) (hash(x)=45415908)
81
+ 75 train 10.559283 (lr=2.6573e-05) (hash(x)=49436419)
82
+ 76 train 10.490873 (lr=2.6923e-05) (hash(x)=48033921)
83
+ 77 train 10.475286 (lr=2.7273e-05) (hash(x)=56987876)
84
+ 78 train 10.400213 (lr=2.7622e-05) (hash(x)=48024575)
85
+ 79 train 10.380251 (lr=2.7972e-05) (hash(x)=48339908)
86
+ 80 train 10.353804 (lr=2.8322e-05) (hash(x)=53135323)
87
+ 81 train 10.359755 (lr=2.8671e-05) (hash(x)=46142231)
88
+ 82 train 10.416197 (lr=2.9021e-05) (hash(x)=41898441)
89
+ 83 train 10.440245 (lr=2.9371e-05) (hash(x)=46263050)
90
+ 84 train 10.393560 (lr=2.9720e-05) (hash(x)=37281177)
91
+ 85 train 10.401260 (lr=3.0070e-05) (hash(x)=48285373)
92
+ 86 train 10.374063 (lr=3.0420e-05) (hash(x)=51190878)
93
+ 87 train 10.373065 (lr=3.0769e-05) (hash(x)=56873965)
94
+ 88 train 10.279132 (lr=3.1119e-05) (hash(x)=52595463)
95
+ 89 train 10.407680 (lr=3.1469e-05) (hash(x)=50421608)
96
+ 90 train 10.298191 (lr=3.1818e-05) (hash(x)=45921624)
97
+ 91 train 10.391988 (lr=3.2168e-05) (hash(x)=52802882)
98
+ 92 train 10.355448 (lr=3.2517e-05) (hash(x)=45185272)
99
+ 93 train 10.367887 (lr=3.2867e-05) (hash(x)=49146692)
100
+ 94 train 10.310324 (lr=3.3217e-05) (hash(x)=52690444)
101
+ 95 train 10.337873 (lr=3.3566e-05) (hash(x)=45627699)
102
+ 96 train 10.379922 (lr=3.3916e-05) (hash(x)=50636707)
103
+ 97 train 10.193415 (lr=3.4266e-05) (hash(x)=44771118)
104
+ 98 train 10.222715 (lr=3.4615e-05) (hash(x)=48109117)
105
+ 99 train 10.345989 (lr=3.4965e-05) (hash(x)=55122837)
106
+ 100 val loss 10.3178
107
+ 100 val perplexity 30266.2969
108
+ 100 train 10.360267 (lr=3.5315e-05) (hash(x)=58304741)
109
+ 101 train 10.316904 (lr=3.5664e-05) (hash(x)=48891446)
110
+ 102 train 10.307606 (lr=3.6014e-05) (hash(x)=46085486)
111
+ 103 train 10.318693 (lr=3.6364e-05) (hash(x)=50609171)
112
+ 104 train 10.301366 (lr=3.6713e-05) (hash(x)=44693623)
113
+ 105 train 10.354119 (lr=3.7063e-05) (hash(x)=45080235)
114
+ 106 train 10.237966 (lr=3.7413e-05) (hash(x)=46731794)
115
+ 107 train 10.315882 (lr=3.7762e-05) (hash(x)=46674118)
116
+ 108 train 10.221298 (lr=3.8112e-05) (hash(x)=46647221)
117
+ 109 train 10.211369 (lr=3.8462e-05) (hash(x)=51868493)
118
+ 110 train 10.269386 (lr=3.8811e-05) (hash(x)=52926855)
119
+ 111 train 10.223500 (lr=3.9161e-05) (hash(x)=50306555)
120
+ 112 train 10.199757 (lr=3.9510e-05) (hash(x)=47017597)
121
+ 113 train 10.216875 (lr=3.9860e-05) (hash(x)=45930697)
122
+ 114 train 10.268049 (lr=4.0210e-05) (hash(x)=55896424)
123
+ 115 train 10.230568 (lr=4.0559e-05) (hash(x)=50806347)
124
+ 116 train 10.242318 (lr=4.0909e-05) (hash(x)=53269239)
125
+ 117 train 10.212035 (lr=4.1259e-05) (hash(x)=50115446)
126
+ 118 train 10.271646 (lr=4.1608e-05) (hash(x)=53956776)
127
+ 119 train 10.188864 (lr=4.1958e-05) (hash(x)=51243947)
128
+ 120 train 10.222794 (lr=4.2308e-05) (hash(x)=55362900)
129
+ 121 train 10.189230 (lr=4.2657e-05) (hash(x)=34965283)
130
+ 122 train 10.184373 (lr=4.3007e-05) (hash(x)=52583594)
131
+ 123 train 10.189457 (lr=4.3357e-05) (hash(x)=53806649)
132
+ 124 train 10.172391 (lr=4.3706e-05) (hash(x)=51974675)
133
+ 125 train 10.203533 (lr=4.4056e-05) (hash(x)=45810574)
134
+ 126 train 10.133426 (lr=4.4406e-05) (hash(x)=34273240)
135
+ 127 train 10.097095 (lr=4.4755e-05) (hash(x)=50536970)
136
+ 128 train 10.058943 (lr=4.5105e-05) (hash(x)=47843837)
137
+ 129 train 10.222813 (lr=4.5455e-05) (hash(x)=56940944)
138
+ 130 train 10.182350 (lr=4.5804e-05) (hash(x)=54162688)
139
+ 131 train 10.158394 (lr=4.6154e-05) (hash(x)=51669498)
140
+ 132 train 10.139881 (lr=4.6503e-05) (hash(x)=53020601)
141
+ 133 train 10.153988 (lr=4.6853e-05) (hash(x)=46315245)
142
+ 134 train 10.076962 (lr=4.7203e-05) (hash(x)=51422940)
143
+ 135 train 10.113983 (lr=4.7552e-05) (hash(x)=50610610)
144
+ 136 train 10.032269 (lr=4.7902e-05) (hash(x)=47220467)
145
+ 137 train 10.114471 (lr=4.8252e-05) (hash(x)=51478185)
146
+ 138 train 10.033234 (lr=4.8601e-05) (hash(x)=48930577)
147
+ 139 train 10.037033 (lr=4.8951e-05) (hash(x)=53701929)
148
+ 140 train 10.170954 (lr=4.9301e-05) (hash(x)=54410011)
149
+ 141 train 10.142626 (lr=4.9650e-05) (hash(x)=51021250)
150
+ 142 train 10.054307 (lr=5.0000e-05) (hash(x)=55817179)
151
+ 143 train 10.118052 (lr=5.0350e-05) (hash(x)=46930369)
152
+ 144 train 10.089009 (lr=5.0699e-05) (hash(x)=51147542)
153
+ 145 train 9.977171 (lr=5.1049e-05) (hash(x)=39405431)
154
+ 146 train 10.082464 (lr=5.1399e-05) (hash(x)=48536998)
155
+ 147 train 10.112307 (lr=5.1748e-05) (hash(x)=57476018)
156
+ 148 train 10.072135 (lr=5.2098e-05) (hash(x)=58950633)
157
+ 149 train 10.007186 (lr=5.2448e-05) (hash(x)=48315677)
158
+ 150 val loss 10.0307
159
+ 150 val perplexity 22713.8086
160
+ 150 train 10.053496 (lr=5.2797e-05) (hash(x)=53677175)
161
+ 151 train 10.022027 (lr=5.3147e-05) (hash(x)=50199187)
162
+ 152 train 10.012892 (lr=5.3497e-05) (hash(x)=56596899)
163
+ 153 train 9.978958 (lr=5.3846e-05) (hash(x)=49586111)
164
+ 154 train 10.056583 (lr=5.4196e-05) (hash(x)=46819915)
165
+ 155 train 10.067059 (lr=5.4545e-05) (hash(x)=45949699)
166
+ 156 train 10.036072 (lr=5.4895e-05) (hash(x)=45515416)
167
+ 157 train 10.146334 (lr=5.5245e-05) (hash(x)=47334388)
168
+ 158 train 10.116514 (lr=5.5594e-05) (hash(x)=49474562)
169
+ 159 train 9.913442 (lr=5.5944e-05) (hash(x)=53294055)
170
+ 160 train 9.871157 (lr=5.6294e-05) (hash(x)=54082362)
171
+ 161 train 9.891584 (lr=5.6643e-05) (hash(x)=46499362)
172
+ 162 train 9.980222 (lr=5.6993e-05) (hash(x)=53531448)
173
+ 163 train 9.945437 (lr=5.7343e-05) (hash(x)=53097392)
174
+ 164 train 9.714823 (lr=5.7692e-05) (hash(x)=41633073)
175
+ 165 train 9.917764 (lr=5.8042e-05) (hash(x)=46479397)
176
+ 166 train 9.892694 (lr=5.8392e-05) (hash(x)=38977328)
177
+ 167 train 9.902845 (lr=5.8741e-05) (hash(x)=54479595)
178
+ 168 train 9.981182 (lr=5.9091e-05) (hash(x)=52103949)
179
+ 169 train 9.889009 (lr=5.9441e-05) (hash(x)=47738093)
180
+ 170 train 9.939163 (lr=5.9790e-05) (hash(x)=51395731)
181
+ 171 train 9.919722 (lr=6.0140e-05) (hash(x)=59006693)
182
+ 172 train 9.846137 (lr=6.0490e-05) (hash(x)=52545488)
183
+ 173 train 9.860308 (lr=6.0839e-05) (hash(x)=46312947)
184
+ 174 train 9.830961 (lr=6.1189e-05) (hash(x)=51488924)
185
+ 175 train 9.842703 (lr=6.1538e-05) (hash(x)=51396394)
186
+ 176 train 9.812124 (lr=6.1888e-05) (hash(x)=43705721)
187
+ 177 train 9.885639 (lr=6.2238e-05) (hash(x)=55830040)
188
+ 178 train 9.895674 (lr=6.2587e-05) (hash(x)=49340705)
189
+ 179 train 9.881191 (lr=6.2937e-05) (hash(x)=52098425)
190
+ 180 train 9.845003 (lr=6.3287e-05) (hash(x)=48643399)
191
+ 181 train 9.774168 (lr=6.3636e-05) (hash(x)=44163258)
192
+ 182 train 9.794942 (lr=6.3986e-05) (hash(x)=48803235)
193
+ 183 train 9.794351 (lr=6.4336e-05) (hash(x)=58459317)
194
+ 184 train 9.795506 (lr=6.4685e-05) (hash(x)=55950871)
195
+ 185 train 9.773528 (lr=6.5035e-05) (hash(x)=50402996)
196
+ 186 train 9.751158 (lr=6.5385e-05) (hash(x)=51861896)
197
+ 187 train 9.740086 (lr=6.5734e-05) (hash(x)=47781307)
198
+ 188 train 9.765444 (lr=6.6084e-05) (hash(x)=58591982)
199
+ 189 train 9.833458 (lr=6.6434e-05) (hash(x)=55269973)
200
+ 190 train 9.831350 (lr=6.6783e-05) (hash(x)=54250578)
201
+ 191 train 9.719925 (lr=6.7133e-05) (hash(x)=55088766)
202
+ 192 train 9.682926 (lr=6.7483e-05) (hash(x)=50679914)
203
+ 193 train 9.734671 (lr=6.7832e-05) (hash(x)=48450358)
204
+ 194 train 9.725243 (lr=6.8182e-05) (hash(x)=49197883)
205
+ 195 train 9.729073 (lr=6.8531e-05) (hash(x)=50198680)
206
+ 196 train 9.704466 (lr=6.8881e-05) (hash(x)=51419665)
207
+ 197 train 9.576101 (lr=6.9231e-05) (hash(x)=48650332)
208
+ 198 train 9.696732 (lr=6.9580e-05) (hash(x)=48143827)
209
+ 199 train 9.687710 (lr=6.9930e-05) (hash(x)=52234688)
210
+ 200 val loss 9.6666
211
+ 200 val perplexity 15781.6191
212
+ 200 train 9.642776 (lr=7.0280e-05) (hash(x)=52032234)
213
+ 201 train 9.695349 (lr=7.0629e-05) (hash(x)=51346801)
214
+ 202 train 9.752648 (lr=7.0979e-05) (hash(x)=58253034)
215
+ 203 train 9.715641 (lr=7.1329e-05) (hash(x)=52929403)
216
+ 204 train 9.651012 (lr=7.1678e-05) (hash(x)=56975127)
217
+ 205 train 9.611279 (lr=7.2028e-05) (hash(x)=56070928)
218
+ 206 train 9.641582 (lr=7.2378e-05) (hash(x)=48921492)
219
+ 207 train 9.569183 (lr=7.2727e-05) (hash(x)=50526384)
220
+ 208 train 9.572097 (lr=7.3077e-05) (hash(x)=49178590)
221
+ 209 train 9.635328 (lr=7.3427e-05) (hash(x)=60209461)
222
+ 210 train 9.648325 (lr=7.3776e-05) (hash(x)=51970769)
223
+ 211 train 9.553243 (lr=7.4126e-05) (hash(x)=52322092)
224
+ 212 train 9.490023 (lr=7.4476e-05) (hash(x)=49806109)
225
+ 213 train 9.586144 (lr=7.4825e-05) (hash(x)=52916383)
226
+ 214 train 9.505272 (lr=7.5175e-05) (hash(x)=43459378)
227
+ 215 train 9.430286 (lr=7.5524e-05) (hash(x)=52529702)
228
+ 216 train 9.545857 (lr=7.5874e-05) (hash(x)=49688097)
229
+ 217 train 9.551417 (lr=7.6224e-05) (hash(x)=51980724)
230
+ 218 train 9.531169 (lr=7.6573e-05) (hash(x)=47435088)
231
+ 219 train 9.515650 (lr=7.6923e-05) (hash(x)=46826784)
232
+ 220 train 9.497662 (lr=7.7273e-05) (hash(x)=51120264)
233
+ 221 train 9.500837 (lr=7.7622e-05) (hash(x)=51321857)
234
+ 222 train 9.411946 (lr=7.7972e-05) (hash(x)=51476600)
235
+ 223 train 9.520540 (lr=7.8322e-05) (hash(x)=53475454)
236
+ 224 train 9.519842 (lr=7.8671e-05) (hash(x)=39878629)
237
+ 225 train 9.404202 (lr=7.9021e-05) (hash(x)=54787418)
238
+ 226 train 9.597739 (lr=7.9371e-05) (hash(x)=52506608)
239
+ 227 train 9.426920 (lr=7.9720e-05) (hash(x)=46810389)
240
+ 228 train 9.317448 (lr=8.0070e-05) (hash(x)=57144694)
241
+ 229 train 9.455252 (lr=8.0420e-05) (hash(x)=53498382)
242
+ 230 train 9.417372 (lr=8.0769e-05) (hash(x)=51280415)
243
+ 231 train 9.418135 (lr=8.1119e-05) (hash(x)=46323632)
244
+ 232 train 9.421375 (lr=8.1469e-05) (hash(x)=46779238)
245
+ 233 train 9.396482 (lr=8.1818e-05) (hash(x)=52398798)
246
+ 234 train 9.385621 (lr=8.2168e-05) (hash(x)=48186608)
247
+ 235 train 9.371988 (lr=8.2517e-05) (hash(x)=53724068)
248
+ 236 train 9.371989 (lr=8.2867e-05) (hash(x)=52960197)
249
+ 237 train 9.426411 (lr=8.3217e-05) (hash(x)=52768582)
250
+ 238 train 9.323154 (lr=8.3566e-05) (hash(x)=52810796)
251
+ 239 train 9.385962 (lr=8.3916e-05) (hash(x)=46747022)
252
+ 240 train 9.311642 (lr=8.4266e-05) (hash(x)=53760168)
253
+ 241 train 9.293126 (lr=8.4615e-05) (hash(x)=50353293)
254
+ 242 train 9.326193 (lr=8.4965e-05) (hash(x)=50573284)
255
+ 243 train 9.278626 (lr=8.5315e-05) (hash(x)=50822751)
256
+ 244 train 9.340521 (lr=8.5664e-05) (hash(x)=44593201)
257
+ 245 train 9.204533 (lr=8.6014e-05) (hash(x)=47616170)
258
+ 246 train 9.243162 (lr=8.6364e-05) (hash(x)=49711034)
259
+ 247 train 9.317259 (lr=8.6713e-05) (hash(x)=48013779)
260
+ 248 train 9.311863 (lr=8.7063e-05) (hash(x)=49696467)
261
+ 249 train 9.259468 (lr=8.7413e-05) (hash(x)=45598720)
262
+ 250 val loss 9.2352
263
+ 250 val perplexity 10251.5254
264
+ 250 train 9.217905 (lr=8.7762e-05) (hash(x)=49421968)
265
+ 251 train 9.223650 (lr=8.8112e-05) (hash(x)=52658193)
266
+ 252 train 9.298532 (lr=8.8462e-05) (hash(x)=51581514)
267
+ 253 train 9.250868 (lr=8.8811e-05) (hash(x)=48531900)
268
+ 254 train 9.173897 (lr=8.9161e-05) (hash(x)=54903424)
269
+ 255 train 9.228841 (lr=8.9510e-05) (hash(x)=46134720)
270
+ 256 train 9.323370 (lr=8.9860e-05) (hash(x)=53833396)
271
+ 257 train 9.173330 (lr=9.0210e-05) (hash(x)=51156418)
272
+ 258 train 9.044612 (lr=9.0559e-05) (hash(x)=48618191)
273
+ 259 train 9.113873 (lr=9.0909e-05) (hash(x)=52902481)
274
+ 260 train 9.031700 (lr=9.1259e-05) (hash(x)=54117145)
275
+ 261 train 9.132502 (lr=9.1608e-05) (hash(x)=51604647)
276
+ 262 train 9.089002 (lr=9.1958e-05) (hash(x)=50691387)
277
+ 263 train 9.096542 (lr=9.2308e-05) (hash(x)=49163526)
278
+ 264 train 9.064934 (lr=9.2657e-05) (hash(x)=47039068)
279
+ 265 train 9.094458 (lr=9.3007e-05) (hash(x)=53829392)
280
+ 266 train 9.017063 (lr=9.3357e-05) (hash(x)=49800706)
281
+ 267 train 9.046715 (lr=9.3706e-05) (hash(x)=49028078)
282
+ 268 train 9.130728 (lr=9.4056e-05) (hash(x)=53372658)
283
+ 269 train 8.990587 (lr=9.4406e-05) (hash(x)=52517672)
284
+ 270 train 8.981007 (lr=9.4755e-05) (hash(x)=49248218)
285
+ 271 train 9.081336 (lr=9.5105e-05) (hash(x)=59110263)
286
+ 272 train 8.930595 (lr=9.5455e-05) (hash(x)=54533121)
287
+ 273 train 8.981985 (lr=9.5804e-05) (hash(x)=51195255)
288
+ 274 train 8.992915 (lr=9.6154e-05) (hash(x)=47467588)
289
+ 275 train 8.994531 (lr=9.6503e-05) (hash(x)=50517780)
290
+ 276 train 8.974243 (lr=9.6853e-05) (hash(x)=50640606)
291
+ 277 train 9.167339 (lr=9.7203e-05) (hash(x)=62341175)
292
+ 278 train 9.029578 (lr=9.7552e-05) (hash(x)=51670137)
293
+ 279 train 8.866892 (lr=9.7902e-05) (hash(x)=45890874)
294
+ 280 train 8.898098 (lr=9.8252e-05) (hash(x)=50817384)
295
+ 281 train 8.896505 (lr=9.8601e-05) (hash(x)=49188067)
296
+ 282 train 8.974751 (lr=9.8951e-05) (hash(x)=53943162)
297
+ 283 train 8.860237 (lr=9.9301e-05) (hash(x)=47888783)
298
+ 284 train 8.928330 (lr=9.9650e-05) (hash(x)=47770108)
299
+ 285 train 9.031767 (lr=1.0000e-04) (hash(x)=52775090)
300
+ 286 train 8.904856 (lr=1.0000e-04) (hash(x)=49971730)
301
+ 287 train 8.931861 (lr=1.0000e-04) (hash(x)=56693767)
302
+ 288 train 8.811897 (lr=9.9998e-05) (hash(x)=53347077)
303
+ 289 train 8.849640 (lr=9.9996e-05) (hash(x)=45014400)
304
+ 290 train 8.769115 (lr=9.9993e-05) (hash(x)=45398141)
305
+ 291 train 8.828358 (lr=9.9989e-05) (hash(x)=53102856)
306
+ 292 train 8.756334 (lr=9.9984e-05) (hash(x)=51707711)
307
+ 293 train 8.837679 (lr=9.9979e-05) (hash(x)=52954310)
308
+ 294 train 8.792414 (lr=9.9972e-05) (hash(x)=45943270)
309
+ 295 train 8.743010 (lr=9.9965e-05) (hash(x)=47965042)
310
+ 296 train 8.782447 (lr=9.9956e-05) (hash(x)=56913043)
311
+ 297 train 8.817985 (lr=9.9947e-05) (hash(x)=56016526)
312
+ 298 train 8.714312 (lr=9.9937e-05) (hash(x)=50655822)
313
+ 299 train 8.693994 (lr=9.9926e-05) (hash(x)=45600303)
314
+ 300 val loss 8.6850
315
+ 300 val perplexity 5913.7178
316
+ 300 train 8.705973 (lr=9.9915e-05) (hash(x)=48710188)
317
+ 301 train 8.589563 (lr=9.9902e-05) (hash(x)=45764012)
318
+ 302 train 8.588981 (lr=9.9889e-05) (hash(x)=52034772)
319
+ 303 train 8.591536 (lr=9.9874e-05) (hash(x)=44542942)
320
+ 304 train 8.555359 (lr=9.9859e-05) (hash(x)=48042083)
321
+ 305 train 8.699409 (lr=9.9843e-05) (hash(x)=51672319)
322
+ 306 train 8.636736 (lr=9.9826e-05) (hash(x)=43542246)
323
+ 307 train 8.590668 (lr=9.9808e-05) (hash(x)=49315253)
324
+ 308 train 8.712494 (lr=9.9789e-05) (hash(x)=52341036)
325
+ 309 train 8.503772 (lr=9.9770e-05) (hash(x)=45571167)
326
+ 310 train 8.520258 (lr=9.9749e-05) (hash(x)=48382543)
327
+ 311 train 8.491424 (lr=9.9728e-05) (hash(x)=45320225)
328
+ 312 train 8.565531 (lr=9.9706e-05) (hash(x)=49116111)
329
+ 313 train 8.503986 (lr=9.9683e-05) (hash(x)=50184363)
330
+ 314 train 8.514565 (lr=9.9659e-05) (hash(x)=42655518)
331
+ 315 train 8.445046 (lr=9.9634e-05) (hash(x)=45775751)
332
+ 316 train 8.463261 (lr=9.9609e-05) (hash(x)=47776033)
333
+ 317 train 8.462808 (lr=9.9582e-05) (hash(x)=42462763)
334
+ 318 train 8.483853 (lr=9.9555e-05) (hash(x)=50472276)
335
+ 319 train 8.535341 (lr=9.9526e-05) (hash(x)=53985957)
336
+ 320 train 8.410070 (lr=9.9497e-05) (hash(x)=50215750)
337
+ 321 train 8.468092 (lr=9.9467e-05) (hash(x)=55567006)
338
+ 322 train 8.566929 (lr=9.9437e-05) (hash(x)=50830962)
339
+ 323 train 8.363476 (lr=9.9405e-05) (hash(x)=41963014)
340
+ 324 train 8.502115 (lr=9.9372e-05) (hash(x)=55253867)
341
+ 325 train 8.459347 (lr=9.9339e-05) (hash(x)=53862673)
342
+ 326 train 8.508167 (lr=9.9305e-05) (hash(x)=57095498)
343
+ 327 train 8.508019 (lr=9.9270e-05) (hash(x)=56042923)
344
+ 328 train 8.328711 (lr=9.9234e-05) (hash(x)=51651483)
345
+ 329 train 8.303233 (lr=9.9197e-05) (hash(x)=43078023)
346
+ 330 train 8.371385 (lr=9.9159e-05) (hash(x)=55335538)
347
+ 331 train 8.667809 (lr=9.9121e-05) (hash(x)=55023379)
348
+ 332 train 8.357313 (lr=9.9081e-05) (hash(x)=48246811)
349
+ 333 train 8.264225 (lr=9.9041e-05) (hash(x)=43011460)
350
+ 334 train 8.246189 (lr=9.9000e-05) (hash(x)=51322872)
351
+ 335 train 8.290827 (lr=9.8958e-05) (hash(x)=52536929)
352
+ 336 train 8.324900 (lr=9.8915e-05) (hash(x)=46105549)
353
+ 337 train 8.228960 (lr=9.8872e-05) (hash(x)=49761494)
354
+ 338 train 8.305653 (lr=9.8827e-05) (hash(x)=61065657)
355
+ 339 train 8.187710 (lr=9.8782e-05) (hash(x)=51631508)
356
+ 340 train 8.183403 (lr=9.8736e-05) (hash(x)=49689666)
357
+ 341 train 8.320385 (lr=9.8689e-05) (hash(x)=50400339)
358
+ 342 train 8.252396 (lr=9.8641e-05) (hash(x)=54240120)
359
+ 343 train 8.255602 (lr=9.8592e-05) (hash(x)=46560264)
360
+ 344 train 8.158655 (lr=9.8543e-05) (hash(x)=48013446)
361
+ 345 train 8.418500 (lr=9.8492e-05) (hash(x)=70534997)
362
+ 346 train 8.241776 (lr=9.8441e-05) (hash(x)=49508415)
363
+ 347 train 8.224426 (lr=9.8389e-05) (hash(x)=52420475)
364
+ 348 train 8.235367 (lr=9.8336e-05) (hash(x)=60279802)
365
+ 349 train 8.189192 (lr=9.8282e-05) (hash(x)=53194976)
366
+ 350 val loss 8.1691
367
+ 350 val perplexity 3530.2522
368
+ 350 train 8.164589 (lr=9.8228e-05) (hash(x)=49459146)
369
+ 351 train 8.269157 (lr=9.8172e-05) (hash(x)=66781899)
370
+ 352 train 8.487047 (lr=9.8116e-05) (hash(x)=43442176)
371
+ 353 train 8.188129 (lr=9.8059e-05) (hash(x)=51986059)
372
+ 354 train 8.158380 (lr=9.8001e-05) (hash(x)=45517898)
373
+ 355 train 8.444658 (lr=9.7942e-05) (hash(x)=60275167)
374
+ 356 train 8.273684 (lr=9.7882e-05) (hash(x)=61199882)
375
+ 357 train 8.090530 (lr=9.7822e-05) (hash(x)=45483557)
376
+ 358 train 8.295349 (lr=9.7761e-05) (hash(x)=49436293)
377
+ 359 train 8.101776 (lr=9.7699e-05) (hash(x)=49182582)
378
+ 360 train 8.214804 (lr=9.7636e-05) (hash(x)=54622543)
379
+ 361 train 8.085806 (lr=9.7572e-05) (hash(x)=51446091)
380
+ 362 train 7.979252 (lr=9.7507e-05) (hash(x)=45019935)
381
+ 363 train 8.004611 (lr=9.7442e-05) (hash(x)=47827767)
382
+ 364 train 8.058871 (lr=9.7376e-05) (hash(x)=47297652)
383
+ 365 train 8.061963 (lr=9.7309e-05) (hash(x)=46988810)
384
+ 366 train 7.950613 (lr=9.7241e-05) (hash(x)=44457482)
385
+ 367 train 7.928934 (lr=9.7172e-05) (hash(x)=44918130)
386
+ 368 train 8.068607 (lr=9.7103e-05) (hash(x)=47962618)
387
+ 369 train 8.027823 (lr=9.7032e-05) (hash(x)=49693485)
388
+ 370 train 7.986973 (lr=9.6961e-05) (hash(x)=46911489)
389
+ 371 train 7.927818 (lr=9.6889e-05) (hash(x)=40969008)
390
+ 372 train 7.900148 (lr=9.6817e-05) (hash(x)=47877745)
391
+ 373 train 8.064613 (lr=9.6743e-05) (hash(x)=50344401)
392
+ 374 train 8.075175 (lr=9.6669e-05) (hash(x)=53074032)
393
+ 375 train 7.990712 (lr=9.6593e-05) (hash(x)=49833636)
394
+ 376 train 8.127217 (lr=9.6518e-05) (hash(x)=61445802)
395
+ 377 train 7.942898 (lr=9.6441e-05) (hash(x)=41326666)
396
+ 378 train 7.854801 (lr=9.6363e-05) (hash(x)=43474195)
397
+ 379 train 7.922547 (lr=9.6285e-05) (hash(x)=52346553)
398
+ 380 train 7.625037 (lr=9.6206e-05) (hash(x)=35732133)
399
+ 381 train 7.931313 (lr=9.6126e-05) (hash(x)=47970930)
400
+ 382 train 7.814545 (lr=9.6045e-05) (hash(x)=47182916)
401
+ 383 train 7.897712 (lr=9.5963e-05) (hash(x)=47009621)
402
+ 384 train 7.974102 (lr=9.5881e-05) (hash(x)=52003882)
403
+ 385 train 7.945288 (lr=9.5798e-05) (hash(x)=48602644)
404
+ 386 train 7.972735 (lr=9.5714e-05) (hash(x)=49970789)
405
+ 387 train 7.942152 (lr=9.5629e-05) (hash(x)=52115732)
406
+ 388 train 7.871017 (lr=9.5544e-05) (hash(x)=50715961)
407
+ 389 train 8.057526 (lr=9.5457e-05) (hash(x)=56527845)
408
+ 390 train 7.844823 (lr=9.5370e-05) (hash(x)=49623946)
409
+ 391 train 7.873331 (lr=9.5282e-05) (hash(x)=49459868)
410
+ 392 train 7.903777 (lr=9.5194e-05) (hash(x)=52550306)
411
+ 393 train 7.958516 (lr=9.5104e-05) (hash(x)=48955151)
412
+ 394 train 7.936601 (lr=9.5014e-05) (hash(x)=49655015)
413
+ 395 train 7.848641 (lr=9.4923e-05) (hash(x)=48473450)
414
+ 396 train 8.027847 (lr=9.4831e-05) (hash(x)=45290633)
415
+ 397 train 7.844560 (lr=9.4739e-05) (hash(x)=50306722)
416
+ 398 train 7.820518 (lr=9.4646e-05) (hash(x)=48795732)
417
+ 399 train 7.945980 (lr=9.4551e-05) (hash(x)=50898037)
418
+ 400 val loss 7.8949
419
+ 400 val perplexity 2683.6042
420
+ 400 train 7.964312 (lr=9.4457e-05) (hash(x)=32890763)
421
+ 401 train 7.891204 (lr=9.4361e-05) (hash(x)=35015807)
422
+ 402 train 7.926899 (lr=9.4265e-05) (hash(x)=41169444)
423
+ 403 train 7.864088 (lr=9.4168e-05) (hash(x)=46327297)
424
+ 404 train 7.866665 (lr=9.4070e-05) (hash(x)=48536826)
425
+ 405 train 7.803713 (lr=9.3971e-05) (hash(x)=44297138)
426
+ 406 train 8.374780 (lr=9.3872e-05) (hash(x)=48533607)
427
+ 407 train 7.684158 (lr=9.3772e-05) (hash(x)=47776613)
428
+ 408 train 7.857242 (lr=9.3671e-05) (hash(x)=53854198)
429
+ 409 train 7.905153 (lr=9.3569e-05) (hash(x)=51262732)
430
+ 410 train 7.917740 (lr=9.3467e-05) (hash(x)=47159040)
431
+ 411 train 7.921654 (lr=9.3364e-05) (hash(x)=51480608)
432
+ 412 train 7.757312 (lr=9.3260e-05) (hash(x)=44193147)
433
+ 413 train 7.861832 (lr=9.3155e-05) (hash(x)=50929448)
434
+ 414 train 7.855194 (lr=9.3050e-05) (hash(x)=48420333)
435
+ 415 train 7.944426 (lr=9.2944e-05) (hash(x)=46494961)
436
+ 416 train 7.912766 (lr=9.2837e-05) (hash(x)=49410253)
437
+ 417 train 8.015892 (lr=9.2729e-05) (hash(x)=59634178)
438
+ 418 train 7.908131 (lr=9.2621e-05) (hash(x)=48922609)
439
+ 419 train 7.756649 (lr=9.2512e-05) (hash(x)=47517395)
440
+ 420 train 7.817853 (lr=9.2402e-05) (hash(x)=49531327)
441
+ 421 train 8.180258 (lr=9.2292e-05) (hash(x)=48371589)
442
+ 422 train 8.135199 (lr=9.2181e-05) (hash(x)=55313671)
443
+ 423 train 7.923292 (lr=9.2069e-05) (hash(x)=61097106)
444
+ 424 train 7.804233 (lr=9.1956e-05) (hash(x)=52305923)
445
+ 425 train 7.955711 (lr=9.1843e-05) (hash(x)=61405427)
446
+ 426 train 8.012609 (lr=9.1729e-05) (hash(x)=59898062)
447
+ 427 train 7.889417 (lr=9.1614e-05) (hash(x)=50139108)
448
+ 428 train 7.777630 (lr=9.1499e-05) (hash(x)=47955919)
449
+ 429 train 7.720078 (lr=9.1382e-05) (hash(x)=45467872)
450
+ 430 train 7.810029 (lr=9.1266e-05) (hash(x)=59686276)
451
+ 431 train 7.869569 (lr=9.1148e-05) (hash(x)=49686665)
452
+ 432 train 7.974269 (lr=9.1030e-05) (hash(x)=47341207)
453
+ 433 train 7.912833 (lr=9.0911e-05) (hash(x)=51019411)
454
+ 434 train 7.580331 (lr=9.0791e-05) (hash(x)=50050759)
455
+ 435 train 7.841652 (lr=9.0671e-05) (hash(x)=49380702)
456
+ 436 train 7.824608 (lr=9.0550e-05) (hash(x)=50963362)
457
+ 437 train 7.823874 (lr=9.0428e-05) (hash(x)=50156382)
458
+ 438 train 7.971721 (lr=9.0306e-05) (hash(x)=49995857)
459
+ 439 train 8.006757 (lr=9.0182e-05) (hash(x)=48775872)
460
+ 440 train 7.788049 (lr=9.0059e-05) (hash(x)=53183314)
461
+ 441 train 7.780898 (lr=8.9934e-05) (hash(x)=49497141)
462
+ 442 train 7.719938 (lr=8.9809e-05) (hash(x)=51207084)
463
+ 443 train 7.569196 (lr=8.9683e-05) (hash(x)=49456326)
464
+ 444 train 8.042864 (lr=8.9557e-05) (hash(x)=49811760)
465
+ 445 train 7.782897 (lr=8.9430e-05) (hash(x)=53343959)
466
+ 446 train 7.789208 (lr=8.9302e-05) (hash(x)=54858850)
467
+ 447 train 7.737599 (lr=8.9173e-05) (hash(x)=44654425)
468
+ 448 train 7.770115 (lr=8.9044e-05) (hash(x)=45260452)
469
+ 449 train 7.783391 (lr=8.8914e-05) (hash(x)=51367775)
470
+ 450 val loss 7.7725
471
+ 450 val perplexity 2374.2844
472
+ 450 train 7.733618 (lr=8.8784e-05) (hash(x)=46562023)
473
+ 451 train 7.801526 (lr=8.8653e-05) (hash(x)=54477325)
474
+ 452 train 7.788702 (lr=8.8521e-05) (hash(x)=48571772)
475
+ 453 train 7.787848 (lr=8.8388e-05) (hash(x)=45888568)
476
+ 454 train 7.807592 (lr=8.8255e-05) (hash(x)=49219751)
477
+ 455 train 7.694363 (lr=8.8122e-05) (hash(x)=49384161)
478
+ 456 train 7.682570 (lr=8.7987e-05) (hash(x)=50565032)
479
+ 457 train 7.881937 (lr=8.7852e-05) (hash(x)=52839038)
480
+ 458 train 7.743201 (lr=8.7717e-05) (hash(x)=51799409)
481
+ 459 train 7.730517 (lr=8.7580e-05) (hash(x)=49188683)
482
+ 460 train 7.744628 (lr=8.7444e-05) (hash(x)=51812773)
483
+ 461 train 7.922749 (lr=8.7306e-05) (hash(x)=62547001)
484
+ 462 train 7.762225 (lr=8.7168e-05) (hash(x)=49563491)
485
+ 463 train 7.847117 (lr=8.7029e-05) (hash(x)=51229664)
486
+ 464 train 7.640627 (lr=8.6890e-05) (hash(x)=46252668)
487
+ 465 train 7.652713 (lr=8.6750e-05) (hash(x)=48746317)
488
+ 466 train 7.879863 (lr=8.6609e-05) (hash(x)=44204094)
489
+ 467 train 7.781458 (lr=8.6468e-05) (hash(x)=51961195)
490
+ 468 train 7.991567 (lr=8.6326e-05) (hash(x)=47683918)
491
+ 469 train 8.026352 (lr=8.6184e-05) (hash(x)=49881442)
492
+ 470 train 7.900483 (lr=8.6041e-05) (hash(x)=45500608)
493
+ 471 train 7.856077 (lr=8.5897e-05) (hash(x)=48726061)
494
+ 472 train 7.726526 (lr=8.5753e-05) (hash(x)=53235305)
495
+ 473 train 7.806311 (lr=8.5608e-05) (hash(x)=50492385)
496
+ 474 train 7.629232 (lr=8.5462e-05) (hash(x)=46951052)
497
+ 475 train 7.733204 (lr=8.5316e-05) (hash(x)=53617083)
498
+ 476 train 7.740380 (lr=8.5170e-05) (hash(x)=50267127)
499
+ 477 train 7.600465 (lr=8.5022e-05) (hash(x)=45972461)
500
+ 478 train 7.766829 (lr=8.4875e-05) (hash(x)=55325922)
501
+ 479 train 8.426651 (lr=8.4726e-05) (hash(x)=60830807)
502
+ 480 train 8.370094 (lr=8.4577e-05) (hash(x)=59357720)
503
+ 481 train 8.348041 (lr=8.4428e-05) (hash(x)=63145389)
504
+ 482 train 8.340787 (lr=8.4278e-05) (hash(x)=60412496)
505
+ 483 train 8.385477 (lr=8.4127e-05) (hash(x)=64013010)
506
+ 484 train 8.413848 (lr=8.3976e-05) (hash(x)=62177701)
507
+ 485 train 8.385735 (lr=8.3824e-05) (hash(x)=64418102)
508
+ 486 train 8.075117 (lr=8.3672e-05) (hash(x)=52915348)
509
+ 487 train 7.841768 (lr=8.3519e-05) (hash(x)=54461658)
510
+ 488 train 7.994188 (lr=8.3366e-05) (hash(x)=47729553)
511
+ 489 train 7.677774 (lr=8.3212e-05) (hash(x)=49407427)
512
+ 490 train 7.743290 (lr=8.3057e-05) (hash(x)=51432179)
513
+ 491 train 7.695695 (lr=8.2902e-05) (hash(x)=48800368)
514
+ 492 train 7.774661 (lr=8.2746e-05) (hash(x)=45866136)
515
+ 493 train 7.719901 (lr=8.2590e-05) (hash(x)=54376909)
516
+ 494 train 7.817982 (lr=8.2434e-05) (hash(x)=53519805)
517
+ 495 train 7.804413 (lr=8.2276e-05) (hash(x)=49838333)
518
+ 496 train 7.614710 (lr=8.2119e-05) (hash(x)=48643653)
519
+ 497 train 9.203229 (lr=8.1960e-05) (hash(x)=111351979)
520
+ 498 train 11.584175 (lr=8.1801e-05) (hash(x)=132049261)
521
+ 499 train 11.518976 (lr=8.1642e-05) (hash(x)=119446440)
522
+ 500 val loss 7.7521
523
+ 500 val perplexity 2326.4551
524
+ 500 train 8.374486 (lr=8.1482e-05) (hash(x)=49421392)
525
+ 501 train 7.394449 (lr=8.1322e-05) (hash(x)=39507633)
526
+ 502 train 7.777297 (lr=8.1161e-05) (hash(x)=49902066)
527
+ 503 train 7.803941 (lr=8.1000e-05) (hash(x)=53078397)
528
+ 504 train 7.700747 (lr=8.0838e-05) (hash(x)=51304862)
529
+ 505 train 7.619796 (lr=8.0676e-05) (hash(x)=47937986)
530
+ 506 train 7.706973 (lr=8.0513e-05) (hash(x)=45497594)
531
+ 507 train 7.770051 (lr=8.0349e-05) (hash(x)=58893354)
532
+ 508 train 7.742538 (lr=8.0186e-05) (hash(x)=48729086)
533
+ 509 train 7.701914 (lr=8.0021e-05) (hash(x)=50312697)
534
+ 510 train 7.740653 (lr=7.9856e-05) (hash(x)=51615714)
535
+ 511 train 7.649053 (lr=7.9691e-05) (hash(x)=45343733)
536
+ 512 train 7.706976 (lr=7.9525e-05) (hash(x)=47745691)
537
+ 513 train 7.597313 (lr=7.9359e-05) (hash(x)=47270668)
538
+ 514 train 7.875018 (lr=7.9192e-05) (hash(x)=55874841)
539
+ 515 train 7.667527 (lr=7.9025e-05) (hash(x)=45564330)
540
+ 516 train 7.710185 (lr=7.8858e-05) (hash(x)=43717130)
541
+ 517 train 7.870839 (lr=7.8689e-05) (hash(x)=54465487)
542
+ 518 train 7.767524 (lr=7.8521e-05) (hash(x)=49943376)
543
+ 519 train 7.831169 (lr=7.8352e-05) (hash(x)=46808110)
544
+ 520 train 7.710778 (lr=7.8182e-05) (hash(x)=45689370)
545
+ 521 train 7.722620 (lr=7.8012e-05) (hash(x)=48250815)
546
+ 522 train 7.829618 (lr=7.7842e-05) (hash(x)=52916515)
547
+ 523 train 7.612786 (lr=7.7671e-05) (hash(x)=47328751)
548
+ 524 train 7.806985 (lr=7.7500e-05) (hash(x)=52705525)
549
+ 525 train 7.600501 (lr=7.7328e-05) (hash(x)=50483774)
550
+ 526 train 7.729258 (lr=7.7156e-05) (hash(x)=49144420)
551
+ 527 train 7.732896 (lr=7.6984e-05) (hash(x)=49778011)
552
+ 528 train 7.637959 (lr=7.6811e-05) (hash(x)=47038718)
553
+ 529 train 7.940543 (lr=7.6637e-05) (hash(x)=69410091)
554
+ 530 train 7.648120 (lr=7.6463e-05) (hash(x)=53344027)
555
+ 531 train 7.690960 (lr=7.6289e-05) (hash(x)=50448372)
556
+ 532 train 7.645270 (lr=7.6115e-05) (hash(x)=45353654)
557
+ 533 train 7.738573 (lr=7.5940e-05) (hash(x)=53769931)
558
+ 534 train 7.638944 (lr=7.5764e-05) (hash(x)=51846900)
559
+ 535 train 7.779233 (lr=7.5588e-05) (hash(x)=53670840)
560
+ 536 train 7.667195 (lr=7.5412e-05) (hash(x)=51732397)
561
+ 537 train 7.701200 (lr=7.5235e-05) (hash(x)=53844862)
562
+ 538 train 7.732162 (lr=7.5058e-05) (hash(x)=49927336)
563
+ 539 train 7.693257 (lr=7.4881e-05) (hash(x)=50707781)
564
+ 540 train 7.623203 (lr=7.4703e-05) (hash(x)=49961492)
565
+ 541 train 8.029165 (lr=7.4525e-05) (hash(x)=54542136)
566
+ 542 train 7.770865 (lr=7.4346e-05) (hash(x)=58216693)
567
+ 543 train 7.753073 (lr=7.4167e-05) (hash(x)=52796041)
568
+ 544 train 7.744573 (lr=7.3988e-05) (hash(x)=49102765)
569
+ 545 train 7.595520 (lr=7.3808e-05) (hash(x)=46224031)
570
+ 546 train 7.609255 (lr=7.3628e-05) (hash(x)=50461161)
571
+ 547 train 7.936080 (lr=7.3448e-05) (hash(x)=63541876)
572
+ 548 train 7.594507 (lr=7.3267e-05) (hash(x)=40786237)
573
+ 549 train 7.624818 (lr=7.3086e-05) (hash(x)=52401705)
574
+ 550 val loss 7.6664
575
+ 550 val perplexity 2135.3342
576
+ 550 train 7.735859 (lr=7.2904e-05) (hash(x)=45579136)
577
+ 551 train 7.631225 (lr=7.2723e-05) (hash(x)=47778722)
578
+ 552 train 7.596293 (lr=7.2540e-05) (hash(x)=44533957)
579
+ 553 train 7.586092 (lr=7.2358e-05) (hash(x)=53126468)
580
+ 554 train 7.841352 (lr=7.2175e-05) (hash(x)=59824165)
581
+ 555 train 7.615225 (lr=7.1992e-05) (hash(x)=48019231)
582
+ 556 train 7.733600 (lr=7.1808e-05) (hash(x)=56989020)
583
+ 557 train 7.677994 (lr=7.1624e-05) (hash(x)=50049909)
584
+ 558 train 7.718346 (lr=7.1440e-05) (hash(x)=54564285)
585
+ 559 train 7.645346 (lr=7.1256e-05) (hash(x)=39174017)
586
+ 560 train 7.477643 (lr=7.1071e-05) (hash(x)=50882180)
587
+ 561 train 7.882744 (lr=7.0886e-05) (hash(x)=53181098)
588
+ 562 train 7.948976 (lr=7.0701e-05) (hash(x)=57349606)
589
+ 563 train 8.055887 (lr=7.0515e-05) (hash(x)=55894374)
590
+ 564 train 7.694658 (lr=7.0329e-05) (hash(x)=49212896)
591
+ 565 train 7.682662 (lr=7.0143e-05) (hash(x)=49643242)
592
+ 566 train 7.646237 (lr=6.9956e-05) (hash(x)=47202647)
593
+ 567 train 7.705485 (lr=6.9769e-05) (hash(x)=52422794)
594
+ 568 train 7.710778 (lr=6.9582e-05) (hash(x)=50536114)
595
+ 569 train 7.663518 (lr=6.9394e-05) (hash(x)=53466863)
596
+ 570 train 7.757085 (lr=6.9207e-05) (hash(x)=46665324)
597
+ 571 train 7.675768 (lr=6.9019e-05) (hash(x)=48367015)
598
+ 572 train 7.569024 (lr=6.8830e-05) (hash(x)=49994353)
599
+ 573 train 7.529009 (lr=6.8642e-05) (hash(x)=45469072)
600
+ 574 train 7.647378 (lr=6.8453e-05) (hash(x)=50371481)
601
+ 575 train 7.623846 (lr=6.8264e-05) (hash(x)=50495673)
602
+ 576 train 7.642751 (lr=6.8075e-05) (hash(x)=49261234)
603
+ 577 train 7.403503 (lr=6.7885e-05) (hash(x)=42520893)
604
+ 578 train 7.410302 (lr=6.7695e-05) (hash(x)=40250490)
605
+ 579 train 7.474455 (lr=6.7505e-05) (hash(x)=47779452)
606
+ 580 train 7.660229 (lr=6.7315e-05) (hash(x)=48877084)
607
+ 581 train 7.992363 (lr=6.7124e-05) (hash(x)=51356106)
608
+ 582 train 8.569576 (lr=6.6933e-05) (hash(x)=53610925)
609
+ 583 train 7.778265 (lr=6.6742e-05) (hash(x)=46394994)
610
+ 584 train 7.538184 (lr=6.6551e-05) (hash(x)=51698609)
611
+ 585 train 7.586971 (lr=6.6360e-05) (hash(x)=41504436)
612
+ 586 train 7.660706 (lr=6.6168e-05) (hash(x)=47031162)
613
+ 587 train 7.685075 (lr=6.5976e-05) (hash(x)=49997685)
614
+ 588 train 7.528809 (lr=6.5784e-05) (hash(x)=45411930)
615
+ 589 train 7.650934 (lr=6.5592e-05) (hash(x)=53238687)
616
+ 590 train 7.943157 (lr=6.5399e-05) (hash(x)=51433959)
617
+ 591 train 7.476210 (lr=6.5206e-05) (hash(x)=46877586)
618
+ 592 train 7.759818 (lr=6.5013e-05) (hash(x)=54985258)
619
+ 593 train 7.625884 (lr=6.4820e-05) (hash(x)=49989026)
620
+ 594 train 7.538972 (lr=6.4627e-05) (hash(x)=47165921)
621
+ 595 train 7.602657 (lr=6.4433e-05) (hash(x)=46923187)
622
+ 596 train 7.489304 (lr=6.4240e-05) (hash(x)=46499138)
623
+ 597 train 7.762550 (lr=6.4046e-05) (hash(x)=54278769)
624
+ 598 train 7.451940 (lr=6.3852e-05) (hash(x)=51528476)
625
+ 599 train 7.734181 (lr=6.3658e-05) (hash(x)=51669015)
626
+ 600 val loss 7.6427
627
+ 600 val perplexity 2085.3840
628
+ 600 train 7.607240 (lr=6.3463e-05) (hash(x)=53555358)
629
+ 601 train 7.964956 (lr=6.3269e-05) (hash(x)=51463767)
630
+ 602 train 7.821107 (lr=6.3074e-05) (hash(x)=56715332)
631
+ 603 train 7.962706 (lr=6.2879e-05) (hash(x)=59893724)
632
+ 604 train 7.831098 (lr=6.2684e-05) (hash(x)=61290263)
633
+ 605 train 8.101337 (lr=6.2489e-05) (hash(x)=59853505)
634
+ 606 train 8.373316 (lr=6.2294e-05) (hash(x)=71013017)
635
+ 607 train 7.585391 (lr=6.2098e-05) (hash(x)=49259399)
636
+ 608 train 7.678213 (lr=6.1903e-05) (hash(x)=51845623)
637
+ 609 train 7.769015 (lr=6.1707e-05) (hash(x)=51402337)
638
+ 610 train 7.660353 (lr=6.1511e-05) (hash(x)=49246911)
639
+ 611 train 7.665056 (lr=6.1315e-05) (hash(x)=51048761)
640
+ 612 train 7.567147 (lr=6.1119e-05) (hash(x)=48666995)
641
+ 613 train 7.664025 (lr=6.0923e-05) (hash(x)=49666603)
642
+ 614 train 7.573703 (lr=6.0726e-05) (hash(x)=47571427)
643
+ 615 train 7.562501 (lr=6.0530e-05) (hash(x)=43281885)
644
+ 616 train 7.596685 (lr=6.0333e-05) (hash(x)=47245391)
645
+ 617 train 7.556493 (lr=6.0137e-05) (hash(x)=45554894)
646
+ 618 train 7.555155 (lr=5.9940e-05) (hash(x)=49355712)
647
+ 619 train 7.572587 (lr=5.9743e-05) (hash(x)=49755839)
648
+ 620 train 7.580906 (lr=5.9546e-05) (hash(x)=53058206)
649
+ 621 train 7.554852 (lr=5.9349e-05) (hash(x)=43099820)
650
+ 622 train 7.590403 (lr=5.9152e-05) (hash(x)=51875828)
651
+ 623 train 7.541515 (lr=5.8955e-05) (hash(x)=43941060)
652
+ 624 train 7.811266 (lr=5.8758e-05) (hash(x)=45886411)
653
+ 625 train 7.547644 (lr=5.8560e-05) (hash(x)=43560282)
654
+ 626 train 7.556549 (lr=5.8363e-05) (hash(x)=49976768)
655
+ 627 train 7.486022 (lr=5.8165e-05) (hash(x)=46745578)
656
+ 628 train 7.488120 (lr=5.7968e-05) (hash(x)=45475922)
657
+ 629 train 7.886341 (lr=5.7770e-05) (hash(x)=53817256)
658
+ 630 train 7.825129 (lr=5.7573e-05) (hash(x)=50919338)
659
+ 631 train 7.674350 (lr=5.7375e-05) (hash(x)=49733529)
660
+ 632 train 7.633029 (lr=5.7177e-05) (hash(x)=48532027)
661
+ 633 train 7.554757 (lr=5.6979e-05) (hash(x)=46736835)
662
+ 634 train 7.588017 (lr=5.6782e-05) (hash(x)=47939943)
663
+ 635 train 7.577759 (lr=5.6584e-05) (hash(x)=50611003)
664
+ 636 train 7.793478 (lr=5.6386e-05) (hash(x)=63044230)
665
+ 637 train 7.606507 (lr=5.6188e-05) (hash(x)=48670015)
666
+ 638 train 7.554696 (lr=5.5990e-05) (hash(x)=46344368)
667
+ 639 train 7.571234 (lr=5.5792e-05) (hash(x)=47655500)
668
+ 640 train 7.539433 (lr=5.5594e-05) (hash(x)=46687266)
669
+ 641 train 7.639654 (lr=5.5396e-05) (hash(x)=48973562)
670
+ 642 train 7.687500 (lr=5.5198e-05) (hash(x)=52720524)
671
+ 643 train 7.435553 (lr=5.5000e-05) (hash(x)=42548986)
672
+ 644 train 7.680899 (lr=5.4802e-05) (hash(x)=52320816)
673
+ 645 train 7.657489 (lr=5.4604e-05) (hash(x)=49272586)
674
+ 646 train 7.541936 (lr=5.4406e-05) (hash(x)=42633826)
675
+ 647 train 7.569506 (lr=5.4208e-05) (hash(x)=50679082)
676
+ 648 train 7.364428 (lr=5.4010e-05) (hash(x)=39001722)
677
+ 649 train 7.308228 (lr=5.3812e-05) (hash(x)=41312345)
678
+ 650 val loss 7.6274
679
+ 650 val perplexity 2053.6318
680
+ 650 train 7.791310 (lr=5.3614e-05) (hash(x)=51704770)
681
+ 651 train 7.477046 (lr=5.3416e-05) (hash(x)=50521445)
682
+ 652 train 7.493764 (lr=5.3218e-05) (hash(x)=47764695)
683
+ 653 train 7.542336 (lr=5.3021e-05) (hash(x)=48798201)
684
+ 654 train 7.534912 (lr=5.2823e-05) (hash(x)=48787936)
685
+ 655 train 7.551382 (lr=5.2625e-05) (hash(x)=48888163)
686
+ 656 train 7.569124 (lr=5.2427e-05) (hash(x)=53554722)
687
+ 657 train 7.651301 (lr=5.2230e-05) (hash(x)=45931709)
688
+ 658 train 7.605824 (lr=5.2032e-05) (hash(x)=49823792)
689
+ 659 train 7.477081 (lr=5.1835e-05) (hash(x)=44644363)
690
+ 660 train 7.497652 (lr=5.1637e-05) (hash(x)=50127860)
691
+ 661 train 7.779952 (lr=5.1440e-05) (hash(x)=55945599)
692
+ 662 train 7.602030 (lr=5.1242e-05) (hash(x)=47054039)
693
+ 663 train 7.554463 (lr=5.1045e-05) (hash(x)=48165003)
694
+ 664 train 7.498350 (lr=5.0848e-05) (hash(x)=46774917)
695
+ 665 train 7.655295 (lr=5.0651e-05) (hash(x)=49310124)
696
+ 666 train 7.981230 (lr=5.0454e-05) (hash(x)=60198659)
697
+ 667 train 7.675303 (lr=5.0257e-05) (hash(x)=44013693)
698
+ 668 train 7.816993 (lr=5.0060e-05) (hash(x)=56074356)
699
+ 669 train 7.557651 (lr=4.9863e-05) (hash(x)=49879848)
700
+ 670 train 7.578484 (lr=4.9667e-05) (hash(x)=51788596)
701
+ 671 train 7.441279 (lr=4.9470e-05) (hash(x)=45152283)
702
+ 672 train 7.590115 (lr=4.9274e-05) (hash(x)=48382985)
703
+ 673 train 7.716647 (lr=4.9077e-05) (hash(x)=60540503)
704
+ 674 train 7.759088 (lr=4.8881e-05) (hash(x)=57599871)
705
+ 675 train 7.467402 (lr=4.8685e-05) (hash(x)=48497343)
706
+ 676 train 7.599638 (lr=4.8489e-05) (hash(x)=49591496)
707
+ 677 train 7.708095 (lr=4.8293e-05) (hash(x)=54318341)
708
+ 678 train 7.557671 (lr=4.8097e-05) (hash(x)=50066235)
709
+ 679 train 7.688815 (lr=4.7902e-05) (hash(x)=52297893)
710
+ 680 train 7.589535 (lr=4.7706e-05) (hash(x)=50598841)
711
+ 681 train 7.656049 (lr=4.7511e-05) (hash(x)=51000121)
712
+ 682 train 7.629562 (lr=4.7316e-05) (hash(x)=44685936)
713
+ 683 train 7.503745 (lr=4.7121e-05) (hash(x)=45974569)
714
+ 684 train 7.475693 (lr=4.6926e-05) (hash(x)=47997917)
715
+ 685 train 7.566861 (lr=4.6731e-05) (hash(x)=52735968)
716
+ 686 train 7.723028 (lr=4.6537e-05) (hash(x)=52390749)
717
+ 687 train 7.572774 (lr=4.6342e-05) (hash(x)=51792713)
718
+ 688 train 7.536964 (lr=4.6148e-05) (hash(x)=52301005)
719
+ 689 train 7.429371 (lr=4.5954e-05) (hash(x)=45786866)
720
+ 690 train 7.606642 (lr=4.5760e-05) (hash(x)=46845157)
721
+ 691 train 7.490306 (lr=4.5567e-05) (hash(x)=44500212)
722
+ 692 train 7.517464 (lr=4.5373e-05) (hash(x)=45570299)
723
+ 693 train 7.902999 (lr=4.5180e-05) (hash(x)=66566677)
724
+ 694 train 7.617316 (lr=4.4987e-05) (hash(x)=50150833)
725
+ 695 train 7.495341 (lr=4.4794e-05) (hash(x)=48433902)
726
+ 696 train 7.372928 (lr=4.4601e-05) (hash(x)=46133410)
727
+ 697 train 7.608619 (lr=4.4408e-05) (hash(x)=53700753)
728
+ 698 train 7.471334 (lr=4.4216e-05) (hash(x)=46993379)
729
+ 699 train 7.340081 (lr=4.4024e-05) (hash(x)=42614965)
730
+ 700 val loss 7.6145
731
+ 700 val perplexity 2027.4814
732
+ 700 train 7.442565 (lr=4.3832e-05) (hash(x)=48395071)
733
+ 701 train 7.536832 (lr=4.3640e-05) (hash(x)=51631419)
734
+ 702 train 7.574078 (lr=4.3449e-05) (hash(x)=47372156)
735
+ 703 train 7.556821 (lr=4.3258e-05) (hash(x)=50899313)
736
+ 704 train 7.672125 (lr=4.3067e-05) (hash(x)=52908771)
737
+ 705 train 7.591564 (lr=4.2876e-05) (hash(x)=50553266)
738
+ 706 train 7.579815 (lr=4.2685e-05) (hash(x)=51956565)
739
+ 707 train 7.477600 (lr=4.2495e-05) (hash(x)=47548858)
740
+ 708 train 7.533485 (lr=4.2305e-05) (hash(x)=55751995)
741
+ 709 train 7.500002 (lr=4.2115e-05) (hash(x)=44976927)
742
+ 710 train 7.631439 (lr=4.1925e-05) (hash(x)=47933336)
743
+ 711 train 7.445666 (lr=4.1736e-05) (hash(x)=40088877)
744
+ 712 train 7.549294 (lr=4.1547e-05) (hash(x)=47607930)
745
+ 713 train 7.511546 (lr=4.1358e-05) (hash(x)=48104733)
746
+ 714 train 7.715056 (lr=4.1170e-05) (hash(x)=49392536)
747
+ 715 train 7.608405 (lr=4.0981e-05) (hash(x)=53658356)
748
+ 716 train 7.621395 (lr=4.0793e-05) (hash(x)=53935147)
749
+ 717 train 7.472118 (lr=4.0606e-05) (hash(x)=49964977)
750
+ 718 train 7.442056 (lr=4.0418e-05) (hash(x)=46137984)
751
+ 719 train 7.684283 (lr=4.0231e-05) (hash(x)=53925563)
752
+ 720 train 7.742408 (lr=4.0044e-05) (hash(x)=56437609)
753
+ 721 train 7.589161 (lr=3.9857e-05) (hash(x)=52403880)
754
+ 722 train 7.604401 (lr=3.9671e-05) (hash(x)=50394634)
755
+ 723 train 7.446402 (lr=3.9485e-05) (hash(x)=48385306)
756
+ 724 train 7.434714 (lr=3.9299e-05) (hash(x)=49338260)
757
+ 725 train 7.420393 (lr=3.9114e-05) (hash(x)=43581460)
758
+ 726 train 7.727831 (lr=3.8929e-05) (hash(x)=50667652)
759
+ 727 train 7.515707 (lr=3.8744e-05) (hash(x)=52568018)
760
+ 728 train 7.530282 (lr=3.8560e-05) (hash(x)=53364920)
761
+ 729 train 7.645272 (lr=3.8376e-05) (hash(x)=50012317)
762
+ 730 train 7.543212 (lr=3.8192e-05) (hash(x)=49642790)
763
+ 731 train 7.490849 (lr=3.8008e-05) (hash(x)=51336015)
764
+ 732 train 7.577429 (lr=3.7825e-05) (hash(x)=48650182)
765
+ 733 train 7.519669 (lr=3.7642e-05) (hash(x)=51308604)
766
+ 734 train 7.538778 (lr=3.7460e-05) (hash(x)=54041783)
767
+ 735 train 7.657524 (lr=3.7277e-05) (hash(x)=52205768)
768
+ 736 train 7.549491 (lr=3.7096e-05) (hash(x)=50031361)
769
+ 737 train 7.451278 (lr=3.6914e-05) (hash(x)=41696587)
770
+ 738 train 7.498239 (lr=3.6733e-05) (hash(x)=51419472)
771
+ 739 train 7.611216 (lr=3.6552e-05) (hash(x)=45595459)
772
+ 740 train 7.443361 (lr=3.6372e-05) (hash(x)=51384564)
773
+ 741 train 7.580682 (lr=3.6192e-05) (hash(x)=47769902)
774
+ 742 train 7.593332 (lr=3.6012e-05) (hash(x)=50083182)
775
+ 743 train 7.473451 (lr=3.5833e-05) (hash(x)=43562643)
776
+ 744 train 7.459412 (lr=3.5654e-05) (hash(x)=46413138)
777
+ 745 train 7.478660 (lr=3.5475e-05) (hash(x)=44913267)
778
+ 746 train 7.491737 (lr=3.5297e-05) (hash(x)=53383974)
779
+ 747 train 7.408987 (lr=3.5119e-05) (hash(x)=49247249)
780
+ 748 train 7.486147 (lr=3.4942e-05) (hash(x)=49795339)
781
+ 749 train 7.495726 (lr=3.4765e-05) (hash(x)=48475648)
782
+ 750 val loss 7.6038
783
+ 750 val perplexity 2005.8490
784
+ 750 train 7.392776 (lr=3.4588e-05) (hash(x)=43484650)
785
+ 751 train 7.565915 (lr=3.4412e-05) (hash(x)=51933643)
786
+ 752 train 7.555256 (lr=3.4236e-05) (hash(x)=53183469)
787
+ 753 train 7.459953 (lr=3.4060e-05) (hash(x)=46907449)
788
+ 754 train 7.562097 (lr=3.3885e-05) (hash(x)=50187673)
789
+ 755 train 7.512752 (lr=3.3711e-05) (hash(x)=49807502)
790
+ 756 train 7.667318 (lr=3.3537e-05) (hash(x)=51504109)
791
+ 757 train 7.545677 (lr=3.3363e-05) (hash(x)=55135176)
792
+ 758 train 7.467242 (lr=3.3189e-05) (hash(x)=47324424)
793
+ 759 train 7.532368 (lr=3.3016e-05) (hash(x)=51568772)
794
+ 760 train 7.443818 (lr=3.2844e-05) (hash(x)=45835934)
795
+ 761 train 7.442894 (lr=3.2672e-05) (hash(x)=47804614)
796
+ 762 train 7.429226 (lr=3.2500e-05) (hash(x)=41166564)
797
+ 763 train 7.491077 (lr=3.2329e-05) (hash(x)=40443372)
798
+ 764 train 7.313212 (lr=3.2158e-05) (hash(x)=34934770)
799
+ 765 train 7.387020 (lr=3.1988e-05) (hash(x)=45544157)
800
+ 766 train 7.627196 (lr=3.1818e-05) (hash(x)=45031680)
801
+ 767 train 7.663571 (lr=3.1648e-05) (hash(x)=52906728)
802
+ 768 train 7.617058 (lr=3.1479e-05) (hash(x)=48742316)
803
+ 769 train 7.754642 (lr=3.1311e-05) (hash(x)=49019570)
804
+ 770 train 7.428237 (lr=3.1142e-05) (hash(x)=43151555)
805
+ 771 train 7.504897 (lr=3.0975e-05) (hash(x)=50496495)
806
+ 772 train 7.953298 (lr=3.0808e-05) (hash(x)=55699971)
807
+ 773 train 7.568635 (lr=3.0641e-05) (hash(x)=53872114)
808
+ 774 train 8.111635 (lr=3.0475e-05) (hash(x)=64483412)
809
+ 775 train 7.564518 (lr=3.0309e-05) (hash(x)=52247792)
810
+ 776 train 7.471014 (lr=3.0144e-05) (hash(x)=47444496)
811
+ 777 train 7.491913 (lr=2.9979e-05) (hash(x)=40758659)
812
+ 778 train 7.797387 (lr=2.9814e-05) (hash(x)=48971234)
813
+ 779 train 8.275244 (lr=2.9651e-05) (hash(x)=55069044)
814
+ 780 train 7.516926 (lr=2.9487e-05) (hash(x)=51507491)
815
+ 781 train 7.712115 (lr=2.9324e-05) (hash(x)=51128293)
816
+ 782 train 7.618046 (lr=2.9162e-05) (hash(x)=55781237)
817
+ 783 train 7.609096 (lr=2.9000e-05) (hash(x)=50654953)
818
+ 784 train 7.493342 (lr=2.8839e-05) (hash(x)=47044897)
819
+ 785 train 7.742622 (lr=2.8678e-05) (hash(x)=61276676)
820
+ 786 train 7.845954 (lr=2.8518e-05) (hash(x)=59615784)
821
+ 787 train 7.592699 (lr=2.8358e-05) (hash(x)=47259878)
822
+ 788 train 7.635357 (lr=2.8199e-05) (hash(x)=43465735)
823
+ 789 train 7.448195 (lr=2.8040e-05) (hash(x)=50687278)
824
+ 790 train 7.589228 (lr=2.7881e-05) (hash(x)=52564412)
825
+ 791 train 7.562247 (lr=2.7724e-05) (hash(x)=52045445)
826
+ 792 train 7.572367 (lr=2.7566e-05) (hash(x)=50520733)
827
+ 793 train 7.573066 (lr=2.7410e-05) (hash(x)=49960644)
828
+ 794 train 7.646253 (lr=2.7254e-05) (hash(x)=52390420)
829
+ 795 train 7.624815 (lr=2.7098e-05) (hash(x)=49503934)
830
+ 796 train 7.495804 (lr=2.6943e-05) (hash(x)=46690037)
831
+ 797 train 7.635771 (lr=2.6788e-05) (hash(x)=46047135)
832
+ 798 train 7.560480 (lr=2.6634e-05) (hash(x)=50495769)
833
+ 799 train 7.481207 (lr=2.6481e-05) (hash(x)=48029368)
834
+ 800 val loss 7.5948
835
+ 800 val perplexity 1987.8000
836
+ 800 train 7.573584 (lr=2.6328e-05) (hash(x)=53949628)
837
+ 801 train 7.551225 (lr=2.6176e-05) (hash(x)=45374968)
838
+ 802 train 7.497496 (lr=2.6024e-05) (hash(x)=47324845)
839
+ 803 train 7.555898 (lr=2.5873e-05) (hash(x)=52734397)
840
+ 804 train 7.699437 (lr=2.5722e-05) (hash(x)=49557665)
841
+ 805 train 7.633223 (lr=2.5572e-05) (hash(x)=52389164)
842
+ 806 train 7.623111 (lr=2.5423e-05) (hash(x)=43032321)
843
+ 807 train 7.734851 (lr=2.5274e-05) (hash(x)=51521570)
844
+ 808 train 7.748096 (lr=2.5125e-05) (hash(x)=50950738)
845
+ 809 train 7.697177 (lr=2.4978e-05) (hash(x)=50463113)
846
+ 810 train 7.515639 (lr=2.4830e-05) (hash(x)=49118761)
847
+ 811 train 7.541613 (lr=2.4684e-05) (hash(x)=54958341)
848
+ 812 train 7.410232 (lr=2.4538e-05) (hash(x)=44761939)
849
+ 813 train 7.488259 (lr=2.4392e-05) (hash(x)=46153985)
850
+ 814 train 7.413387 (lr=2.4247e-05) (hash(x)=47721297)
851
+ 815 train 7.558563 (lr=2.4103e-05) (hash(x)=49432790)
852
+ 816 train 7.557635 (lr=2.3959e-05) (hash(x)=51785495)
853
+ 817 train 7.515110 (lr=2.3816e-05) (hash(x)=48385222)
854
+ 818 train 7.575776 (lr=2.3674e-05) (hash(x)=53389329)
855
+ 819 train 7.767172 (lr=2.3532e-05) (hash(x)=51630204)
856
+ 820 train 7.420727 (lr=2.3391e-05) (hash(x)=40253250)
857
+ 821 train 7.413262 (lr=2.3250e-05) (hash(x)=55768152)
858
+ 822 train 7.391830 (lr=2.3110e-05) (hash(x)=42811521)
859
+ 823 train 7.477667 (lr=2.2971e-05) (hash(x)=50243462)
860
+ 824 train 7.506099 (lr=2.2832e-05) (hash(x)=45680789)
861
+ 825 train 7.380934 (lr=2.2694e-05) (hash(x)=45951312)
862
+ 826 train 7.789862 (lr=2.2556e-05) (hash(x)=55666375)
863
+ 827 train 8.830496 (lr=2.2420e-05) (hash(x)=55551835)
864
+ 828 train 7.546107 (lr=2.2283e-05) (hash(x)=50871371)
865
+ 829 train 7.506018 (lr=2.2148e-05) (hash(x)=47924861)
866
+ 830 train 7.571692 (lr=2.2013e-05) (hash(x)=44149183)
867
+ 831 train 7.367566 (lr=2.1878e-05) (hash(x)=44943064)
868
+ 832 train 7.555159 (lr=2.1745e-05) (hash(x)=49469986)
869
+ 833 train 7.510525 (lr=2.1612e-05) (hash(x)=45952924)
870
+ 834 train 7.617723 (lr=2.1479e-05) (hash(x)=48879755)
871
+ 835 train 7.911084 (lr=2.1347e-05) (hash(x)=52015898)
872
+ 836 train 7.535168 (lr=2.1216e-05) (hash(x)=52250198)
873
+ 837 train 7.577283 (lr=2.1086e-05) (hash(x)=52449480)
874
+ 838 train 7.427073 (lr=2.0956e-05) (hash(x)=47909761)
875
+ 839 train 7.552354 (lr=2.0827e-05) (hash(x)=47971635)
876
+ 840 train 7.467387 (lr=2.0698e-05) (hash(x)=44937642)
877
+ 841 train 7.942339 (lr=2.0570e-05) (hash(x)=59172908)
878
+ 842 train 7.399529 (lr=2.0443e-05) (hash(x)=45480485)
879
+ 843 train 7.708764 (lr=2.0317e-05) (hash(x)=50097527)
880
+ 844 train 7.522643 (lr=2.0191e-05) (hash(x)=47814961)
881
+ 845 train 7.510270 (lr=2.0066e-05) (hash(x)=50748027)
882
+ 846 train 7.971689 (lr=1.9941e-05) (hash(x)=61446920)
883
+ 847 train 7.344824 (lr=1.9818e-05) (hash(x)=46113670)
884
+ 848 train 7.656268 (lr=1.9694e-05) (hash(x)=49992162)
885
+ 849 train 7.716410 (lr=1.9572e-05) (hash(x)=52184218)
886
+ 850 val loss 7.5894
887
+ 850 val perplexity 1977.0925
888
+ 850 train 7.600089 (lr=1.9450e-05) (hash(x)=47179637)
889
+ 851 train 7.535968 (lr=1.9329e-05) (hash(x)=50005550)
890
+ 852 train 7.545384 (lr=1.9209e-05) (hash(x)=50347610)
891
+ 853 train 7.548382 (lr=1.9089e-05) (hash(x)=47110180)
892
+ 854 train 7.740948 (lr=1.8970e-05) (hash(x)=48075421)
893
+ 855 train 7.551274 (lr=1.8852e-05) (hash(x)=50198511)
894
+ 856 train 7.594034 (lr=1.8734e-05) (hash(x)=51895435)
895
+ 857 train 7.263291 (lr=1.8618e-05) (hash(x)=41108331)
896
+ 858 train 7.415341 (lr=1.8501e-05) (hash(x)=47254356)
897
+ 859 train 7.613036 (lr=1.8386e-05) (hash(x)=51366503)
898
+ 860 train 7.793385 (lr=1.8271e-05) (hash(x)=52602310)
899
+ 861 train 7.769320 (lr=1.8157e-05) (hash(x)=59593423)
900
+ 862 train 7.670403 (lr=1.8044e-05) (hash(x)=51030093)
901
+ 863 train 7.552414 (lr=1.7931e-05) (hash(x)=54958512)
902
+ 864 train 7.390744 (lr=1.7819e-05) (hash(x)=46896166)
903
+ 865 train 7.500262 (lr=1.7708e-05) (hash(x)=50095280)
904
+ 866 train 7.666028 (lr=1.7598e-05) (hash(x)=54998859)
905
+ 867 train 7.557383 (lr=1.7488e-05) (hash(x)=46982568)
906
+ 868 train 7.470465 (lr=1.7379e-05) (hash(x)=49055711)
907
+ 869 train 7.494187 (lr=1.7271e-05) (hash(x)=45920340)
908
+ 870 train 7.392742 (lr=1.7163e-05) (hash(x)=41656939)
909
+ 871 train 7.489355 (lr=1.7056e-05) (hash(x)=48153200)
910
+ 872 train 7.463745 (lr=1.6950e-05) (hash(x)=46299958)
911
+ 873 train 7.580584 (lr=1.6845e-05) (hash(x)=51503557)
912
+ 874 train 7.936938 (lr=1.6740e-05) (hash(x)=58209066)
913
+ 875 train 7.446280 (lr=1.6636e-05) (hash(x)=45447171)
914
+ 876 train 7.454513 (lr=1.6533e-05) (hash(x)=52768123)
915
+ 877 train 7.392230 (lr=1.6431e-05) (hash(x)=43283692)
916
+ 878 train 7.685833 (lr=1.6329e-05) (hash(x)=56521429)
917
+ 879 train 7.475355 (lr=1.6228e-05) (hash(x)=52017911)
918
+ 880 train 7.494340 (lr=1.6128e-05) (hash(x)=49022458)
919
+ 881 train 7.382243 (lr=1.6029e-05) (hash(x)=49830063)
920
+ 882 train 7.391344 (lr=1.5930e-05) (hash(x)=45949076)
921
+ 883 train 7.369357 (lr=1.5832e-05) (hash(x)=56227516)
922
+ 884 train 7.443240 (lr=1.5735e-05) (hash(x)=47617073)
923
+ 885 train 7.457449 (lr=1.5639e-05) (hash(x)=48017404)
924
+ 886 train 7.462682 (lr=1.5543e-05) (hash(x)=49163346)
925
+ 887 train 7.484112 (lr=1.5449e-05) (hash(x)=51984978)
926
+ 888 train 7.487196 (lr=1.5354e-05) (hash(x)=48719449)
927
+ 889 train 7.215100 (lr=1.5261e-05) (hash(x)=47784187)
928
+ 890 train 7.140913 (lr=1.5169e-05) (hash(x)=46255615)
929
+ 891 train 7.363958 (lr=1.5077e-05) (hash(x)=47512728)
930
+ 892 train 7.400325 (lr=1.4986e-05) (hash(x)=48533555)
931
+ 893 train 7.591549 (lr=1.4896e-05) (hash(x)=47778087)
932
+ 894 train 7.458993 (lr=1.4806e-05) (hash(x)=45899594)
933
+ 895 train 7.672999 (lr=1.4718e-05) (hash(x)=53173755)
934
+ 896 train 7.420683 (lr=1.4630e-05) (hash(x)=55457819)
935
+ 897 train 7.493675 (lr=1.4543e-05) (hash(x)=48702762)
936
+ 898 train 7.480086 (lr=1.4456e-05) (hash(x)=44948099)
937
+ 899 train 7.654958 (lr=1.4371e-05) (hash(x)=50108146)
938
+ 900 val loss 7.5794
939
+ 900 val perplexity 1957.4412
940
+ 900 train 7.361840 (lr=1.4286e-05) (hash(x)=50289308)
941
+ 901 train 7.277493 (lr=1.4202e-05) (hash(x)=46030207)
942
+ 902 train 7.499839 (lr=1.4119e-05) (hash(x)=48657580)
943
+ 903 train 7.434409 (lr=1.4037e-05) (hash(x)=52115953)
944
+ 904 train 7.520154 (lr=1.3955e-05) (hash(x)=49847830)
945
+ 905 train 7.538352 (lr=1.3874e-05) (hash(x)=47643321)
946
+ 906 train 7.426982 (lr=1.3794e-05) (hash(x)=49693997)
947
+ 907 train 7.423518 (lr=1.3715e-05) (hash(x)=46586365)
948
+ 908 train 7.436792 (lr=1.3637e-05) (hash(x)=50545145)
949
+ 909 train 7.606075 (lr=1.3559e-05) (hash(x)=49204978)
950
+ 910 train 7.548124 (lr=1.3482e-05) (hash(x)=49510257)
951
+ 911 train 7.517291 (lr=1.3407e-05) (hash(x)=51872266)
952
+ 912 train 7.619929 (lr=1.3331e-05) (hash(x)=50198903)
953
+ 913 train 7.568089 (lr=1.3257e-05) (hash(x)=51204581)
954
+ 914 train 7.509589 (lr=1.3183e-05) (hash(x)=47660020)
955
+ 915 train 7.524913 (lr=1.3111e-05) (hash(x)=52821094)
956
+ 916 train 7.519816 (lr=1.3039e-05) (hash(x)=49449313)
957
+ 917 train 7.526318 (lr=1.2968e-05) (hash(x)=53242967)
958
+ 918 train 7.413209 (lr=1.2897e-05) (hash(x)=44188232)
959
+ 919 train 7.552077 (lr=1.2828e-05) (hash(x)=49251652)
960
+ 920 train 7.471602 (lr=1.2759e-05) (hash(x)=53473946)
961
+ 921 train 7.633470 (lr=1.2691e-05) (hash(x)=50825785)
962
+ 922 train 7.352749 (lr=1.2624e-05) (hash(x)=46790592)
963
+ 923 train 7.513597 (lr=1.2558e-05) (hash(x)=54773264)
964
+ 924 train 7.611215 (lr=1.2493e-05) (hash(x)=55892625)
965
+ 925 train 7.284847 (lr=1.2428e-05) (hash(x)=39155629)
966
+ 926 train 7.631807 (lr=1.2364e-05) (hash(x)=52691267)
967
+ 927 train 7.594419 (lr=1.2301e-05) (hash(x)=48621984)
968
+ 928 train 7.498408 (lr=1.2239e-05) (hash(x)=48349653)
969
+ 929 train 7.624229 (lr=1.2178e-05) (hash(x)=49567244)
970
+ 930 train 7.446261 (lr=1.2118e-05) (hash(x)=53438269)
971
+ 931 train 7.557856 (lr=1.2058e-05) (hash(x)=49633335)
972
+ 932 train 7.444903 (lr=1.1999e-05) (hash(x)=47721650)
973
+ 933 train 7.421088 (lr=1.1941e-05) (hash(x)=45377574)
974
+ 934 train 8.035638 (lr=1.1884e-05) (hash(x)=60171841)
975
+ 935 train 7.711227 (lr=1.1828e-05) (hash(x)=56838834)
976
+ 936 train 7.671886 (lr=1.1772e-05) (hash(x)=46618083)
977
+ 937 train 7.478709 (lr=1.1718e-05) (hash(x)=48343128)
978
+ 938 train 7.492517 (lr=1.1664e-05) (hash(x)=50155542)
979
+ 939 train 7.455033 (lr=1.1611e-05) (hash(x)=50527481)
980
+ 940 train 7.693537 (lr=1.1559e-05) (hash(x)=48347382)
981
+ 941 train 7.459398 (lr=1.1508e-05) (hash(x)=44035071)
982
+ 942 train 7.553293 (lr=1.1457e-05) (hash(x)=48298535)
983
+ 943 train 7.473688 (lr=1.1408e-05) (hash(x)=45518917)
984
+ 944 train 7.547445 (lr=1.1359e-05) (hash(x)=48536783)
985
+ 945 train 7.541576 (lr=1.1311e-05) (hash(x)=49027591)
986
+ 946 train 7.404552 (lr=1.1264e-05) (hash(x)=47080212)
987
+ 947 train 7.660092 (lr=1.1218e-05) (hash(x)=48086649)
988
+ 948 train 7.468519 (lr=1.1173e-05) (hash(x)=47142179)
989
+ 949 train 7.398285 (lr=1.1128e-05) (hash(x)=44730458)
990
+ 950 val loss 7.5766
991
+ 950 val perplexity 1951.9625
992
+ 950 train 7.503413 (lr=1.1085e-05) (hash(x)=50767530)
993
+ 951 train 7.517926 (lr=1.1042e-05) (hash(x)=48085406)
994
+ 952 train 7.504004 (lr=1.1000e-05) (hash(x)=45496925)
995
+ 953 train 7.404015 (lr=1.0959e-05) (hash(x)=49401521)
996
+ 954 train 7.515340 (lr=1.0919e-05) (hash(x)=50951491)
997
+ 955 train 7.462502 (lr=1.0879e-05) (hash(x)=52347328)
998
+ 956 train 7.271656 (lr=1.0841e-05) (hash(x)=46184146)
999
+ 957 train 7.311028 (lr=1.0803e-05) (hash(x)=50307408)
1000
+ 958 train 7.415719 (lr=1.0766e-05) (hash(x)=48451061)
1001
+ 959 train 7.289578 (lr=1.0730e-05) (hash(x)=43959848)
1002
+ 960 train 7.892675 (lr=1.0695e-05) (hash(x)=54902400)
1003
+ 961 train 7.475367 (lr=1.0661e-05) (hash(x)=45469715)
1004
+ 962 train 7.570274 (lr=1.0628e-05) (hash(x)=51664545)
1005
+ 963 train 7.365221 (lr=1.0595e-05) (hash(x)=53522067)
1006
+ 964 train 7.660219 (lr=1.0563e-05) (hash(x)=49003631)
1007
+ 965 train 7.733752 (lr=1.0533e-05) (hash(x)=58676829)
1008
+ 966 train 7.327644 (lr=1.0503e-05) (hash(x)=42612608)
1009
+ 967 train 7.372300 (lr=1.0474e-05) (hash(x)=49691960)
1010
+ 968 train 7.490693 (lr=1.0445e-05) (hash(x)=46491451)
1011
+ 969 train 7.293292 (lr=1.0418e-05) (hash(x)=47344591)
1012
+ 970 train 7.412208 (lr=1.0391e-05) (hash(x)=49485798)
1013
+ 971 train 7.543429 (lr=1.0366e-05) (hash(x)=50456125)
1014
+ 972 train 7.414816 (lr=1.0341e-05) (hash(x)=44633773)
1015
+ 973 train 7.429656 (lr=1.0317e-05) (hash(x)=51037651)
1016
+ 974 train 7.326130 (lr=1.0294e-05) (hash(x)=37768372)
1017
+ 975 train 7.240862 (lr=1.0272e-05) (hash(x)=43100867)
1018
+ 976 train 7.422596 (lr=1.0251e-05) (hash(x)=45427120)
1019
+ 977 train 7.376473 (lr=1.0230e-05) (hash(x)=43778357)
1020
+ 978 train 7.579514 (lr=1.0211e-05) (hash(x)=46241381)
1021
+ 979 train 7.316247 (lr=1.0192e-05) (hash(x)=54476483)
1022
+ 980 train 7.395556 (lr=1.0174e-05) (hash(x)=49769099)
1023
+ 981 train 7.327474 (lr=1.0157e-05) (hash(x)=47098615)
1024
+ 982 train 7.555210 (lr=1.0141e-05) (hash(x)=50789245)
1025
+ 983 train 7.521550 (lr=1.0126e-05) (hash(x)=49945665)
1026
+ 984 train 7.421632 (lr=1.0111e-05) (hash(x)=45312913)
1027
+ 985 train 7.344438 (lr=1.0098e-05) (hash(x)=45656407)
1028
+ 986 train 7.565191 (lr=1.0085e-05) (hash(x)=48551089)
1029
+ 987 train 7.673343 (lr=1.0074e-05) (hash(x)=52658605)
1030
+ 988 train 7.358370 (lr=1.0063e-05) (hash(x)=44445344)
1031
+ 989 train 7.077899 (lr=1.0053e-05) (hash(x)=47928268)
1032
+ 990 train 7.761909 (lr=1.0044e-05) (hash(x)=55796602)
1033
+ 991 train 7.557084 (lr=1.0035e-05) (hash(x)=51979631)
1034
+ 992 train 7.628481 (lr=1.0028e-05) (hash(x)=51095729)
1035
+ 993 train 7.559216 (lr=1.0021e-05) (hash(x)=51164625)
1036
+ 994 train 7.664023 (lr=1.0016e-05) (hash(x)=47953011)
1037
+ 995 train 7.730391 (lr=1.0011e-05) (hash(x)=57967556)
1038
+ 996 train 7.540661 (lr=1.0007e-05) (hash(x)=47898361)
1039
+ 997 train 7.671741 (lr=1.0004e-05) (hash(x)=57145664)
1040
+ 998 train 7.314530 (lr=1.0002e-05) (hash(x)=43250700)
1041
+ 999 val loss 7.5697
1042
+ 999 val perplexity 1938.5776
1043
+ 999 train 7.306107 (lr=1.0000e-05) (hash(x)=37203059)
lr1e-4_total_batch_size20480_seq_len1024/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a589a0f5312c117ff20cd24bf8549141da63d7fd944e109ff10fd76fe05b5293
3
+ size 86167106
lr1e-4_total_batch_size20480_seq_len1024/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b20a4b8c99eb1ac210d004ed047886556d7e473e966bc67dc9ba65bec1946b0
3
+ size 71681862