andrew-healey commited on
Commit
2e68a9c
·
verified ·
1 Parent(s): 8d9beb3

Upload folder using huggingface_hub

Browse files
lr1.5e-3_total_batch_size5120_seq_len128/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep_2/lr1.5e-3_total_batch_size5120_seq_len128", "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": 128, "batch_size": null, "total_batch_size": 5120, "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_size5120_seq_len128/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60582085b67f79a59d68bfd29cd9c94a65421fc83fc6750cd587ed93c2812fa4
3
+ size 964
lr1.5e-3_total_batch_size5120_seq_len128/log2.txt ADDED
@@ -0,0 +1,1043 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 1000
2
+ 0 val loss 11.6870
3
+ 0 val perplexity 119016.5000
4
+ 0 train 11.683437 (lr=5.2448e-06) (hash(x)=6154740)
5
+ 1 train 11.695753 (lr=1.0490e-05) (hash(x)=6605789)
6
+ 2 train 11.672916 (lr=1.5734e-05) (hash(x)=6696707)
7
+ 3 train 11.668686 (lr=2.0979e-05) (hash(x)=5113283)
8
+ 4 train 11.556931 (lr=2.6224e-05) (hash(x)=5607069)
9
+ 5 train 11.500885 (lr=3.1469e-05) (hash(x)=7007748)
10
+ 6 train 11.414856 (lr=3.6713e-05) (hash(x)=7643299)
11
+ 7 train 11.350816 (lr=4.1958e-05) (hash(x)=5534142)
12
+ 8 train 11.261035 (lr=4.7203e-05) (hash(x)=6539293)
13
+ 9 train 11.155344 (lr=5.2448e-05) (hash(x)=5974274)
14
+ 10 train 11.093600 (lr=5.7692e-05) (hash(x)=7199085)
15
+ 11 train 10.953268 (lr=6.2937e-05) (hash(x)=6763231)
16
+ 12 train 10.765748 (lr=6.8182e-05) (hash(x)=6720398)
17
+ 13 train 10.724706 (lr=7.3427e-05) (hash(x)=6706513)
18
+ 14 train 10.708819 (lr=7.8671e-05) (hash(x)=5727925)
19
+ 15 train 10.545939 (lr=8.3916e-05) (hash(x)=6328012)
20
+ 16 train 10.422302 (lr=8.9161e-05) (hash(x)=5812722)
21
+ 17 train 10.346864 (lr=9.4406e-05) (hash(x)=6183012)
22
+ 18 train 10.402117 (lr=9.9650e-05) (hash(x)=5952539)
23
+ 19 train 10.342976 (lr=1.0490e-04) (hash(x)=7052161)
24
+ 20 train 10.147366 (lr=1.1014e-04) (hash(x)=5810871)
25
+ 21 train 10.315709 (lr=1.1538e-04) (hash(x)=6586466)
26
+ 22 train 10.119836 (lr=1.2063e-04) (hash(x)=5061781)
27
+ 23 train 10.186750 (lr=1.2587e-04) (hash(x)=5556098)
28
+ 24 train 10.241531 (lr=1.3112e-04) (hash(x)=7268535)
29
+ 25 train 10.198915 (lr=1.3636e-04) (hash(x)=8155572)
30
+ 26 train 10.060469 (lr=1.4161e-04) (hash(x)=5019590)
31
+ 27 train 9.903213 (lr=1.4685e-04) (hash(x)=4696943)
32
+ 28 train 9.875051 (lr=1.5210e-04) (hash(x)=5984010)
33
+ 29 train 10.092470 (lr=1.5734e-04) (hash(x)=7649389)
34
+ 30 train 9.776911 (lr=1.6259e-04) (hash(x)=6332817)
35
+ 31 train 9.710447 (lr=1.6783e-04) (hash(x)=6219138)
36
+ 32 train 9.823084 (lr=1.7308e-04) (hash(x)=6895681)
37
+ 33 train 9.784887 (lr=1.7832e-04) (hash(x)=7075794)
38
+ 34 train 9.608407 (lr=1.8357e-04) (hash(x)=5024163)
39
+ 35 train 9.612877 (lr=1.8881e-04) (hash(x)=6049878)
40
+ 36 train 9.657012 (lr=1.9406e-04) (hash(x)=4774943)
41
+ 37 train 9.628016 (lr=1.9930e-04) (hash(x)=6385927)
42
+ 38 train 9.408731 (lr=2.0455e-04) (hash(x)=5070217)
43
+ 39 train 9.310562 (lr=2.0979e-04) (hash(x)=5491972)
44
+ 40 train 9.372475 (lr=2.1503e-04) (hash(x)=5695751)
45
+ 41 train 9.272992 (lr=2.2028e-04) (hash(x)=6061420)
46
+ 42 train 9.181724 (lr=2.2552e-04) (hash(x)=7836710)
47
+ 43 train 9.271749 (lr=2.3077e-04) (hash(x)=8075458)
48
+ 44 train 9.038565 (lr=2.3601e-04) (hash(x)=7190584)
49
+ 45 train 9.156437 (lr=2.4126e-04) (hash(x)=7238769)
50
+ 46 train 9.310590 (lr=2.4650e-04) (hash(x)=4931205)
51
+ 47 train 8.928012 (lr=2.5175e-04) (hash(x)=5917741)
52
+ 48 train 8.871819 (lr=2.5699e-04) (hash(x)=5530785)
53
+ 49 train 8.844534 (lr=2.6224e-04) (hash(x)=5430605)
54
+ 50 val loss 8.7951
55
+ 50 val perplexity 6601.6724
56
+ 50 train 8.813455 (lr=2.6748e-04) (hash(x)=6056774)
57
+ 51 train 8.612374 (lr=2.7273e-04) (hash(x)=5750403)
58
+ 52 train 8.624952 (lr=2.7797e-04) (hash(x)=6187497)
59
+ 53 train 8.555968 (lr=2.8322e-04) (hash(x)=4792640)
60
+ 54 train 8.715820 (lr=2.8846e-04) (hash(x)=6868430)
61
+ 55 train 8.483093 (lr=2.9371e-04) (hash(x)=6112458)
62
+ 56 train 8.429692 (lr=2.9895e-04) (hash(x)=5523601)
63
+ 57 train 8.225450 (lr=3.0420e-04) (hash(x)=5801122)
64
+ 58 train 8.384053 (lr=3.0944e-04) (hash(x)=6826295)
65
+ 59 train 8.600574 (lr=3.1469e-04) (hash(x)=5806490)
66
+ 60 train 8.304635 (lr=3.1993e-04) (hash(x)=6223100)
67
+ 61 train 8.105806 (lr=3.2517e-04) (hash(x)=5754140)
68
+ 62 train 8.194729 (lr=3.3042e-04) (hash(x)=6675367)
69
+ 63 train 8.069908 (lr=3.3566e-04) (hash(x)=6568379)
70
+ 64 train 7.952576 (lr=3.4091e-04) (hash(x)=5648233)
71
+ 65 train 8.376801 (lr=3.4615e-04) (hash(x)=7134932)
72
+ 66 train 8.014552 (lr=3.5140e-04) (hash(x)=5242536)
73
+ 67 train 8.192007 (lr=3.5664e-04) (hash(x)=7212403)
74
+ 68 train 7.965927 (lr=3.6189e-04) (hash(x)=5342967)
75
+ 69 train 8.158902 (lr=3.6713e-04) (hash(x)=7720703)
76
+ 70 train 8.323509 (lr=3.7238e-04) (hash(x)=7351337)
77
+ 71 train 8.179846 (lr=3.7762e-04) (hash(x)=7386082)
78
+ 72 train 7.977974 (lr=3.8287e-04) (hash(x)=6931318)
79
+ 73 train 8.094933 (lr=3.8811e-04) (hash(x)=6848513)
80
+ 74 train 8.042456 (lr=3.9336e-04) (hash(x)=7714274)
81
+ 75 train 8.209312 (lr=3.9860e-04) (hash(x)=6359743)
82
+ 76 train 8.040416 (lr=4.0385e-04) (hash(x)=6801650)
83
+ 77 train 8.095227 (lr=4.0909e-04) (hash(x)=7536538)
84
+ 78 train 8.062044 (lr=4.1434e-04) (hash(x)=6957410)
85
+ 79 train 8.116162 (lr=4.1958e-04) (hash(x)=6151674)
86
+ 80 train 7.902263 (lr=4.2483e-04) (hash(x)=6033844)
87
+ 81 train 7.730409 (lr=4.3007e-04) (hash(x)=6317099)
88
+ 82 train 7.673531 (lr=4.3531e-04) (hash(x)=5357844)
89
+ 83 train 7.910425 (lr=4.4056e-04) (hash(x)=7168704)
90
+ 84 train 8.281580 (lr=4.4580e-04) (hash(x)=7157938)
91
+ 85 train 7.839498 (lr=4.5105e-04) (hash(x)=6736092)
92
+ 86 train 7.749297 (lr=4.5629e-04) (hash(x)=5383012)
93
+ 87 train 8.039031 (lr=4.6154e-04) (hash(x)=7859060)
94
+ 88 train 7.929762 (lr=4.6678e-04) (hash(x)=5869288)
95
+ 89 train 7.951465 (lr=4.7203e-04) (hash(x)=6958223)
96
+ 90 train 7.820102 (lr=4.7727e-04) (hash(x)=6731344)
97
+ 91 train 8.002259 (lr=4.8252e-04) (hash(x)=6976239)
98
+ 92 train 7.735435 (lr=4.8776e-04) (hash(x)=7769441)
99
+ 93 train 7.898823 (lr=4.9301e-04) (hash(x)=6956845)
100
+ 94 train 7.995382 (lr=4.9825e-04) (hash(x)=6638037)
101
+ 95 train 7.588033 (lr=5.0350e-04) (hash(x)=6391506)
102
+ 96 train 8.105309 (lr=5.0874e-04) (hash(x)=7319928)
103
+ 97 train 7.978100 (lr=5.1399e-04) (hash(x)=6113701)
104
+ 98 train 7.726049 (lr=5.1923e-04) (hash(x)=6246453)
105
+ 99 train 7.846437 (lr=5.2448e-04) (hash(x)=5245999)
106
+ 100 val loss 7.9049
107
+ 100 val perplexity 2710.5264
108
+ 100 train 7.979496 (lr=5.2972e-04) (hash(x)=6227256)
109
+ 101 train 7.833317 (lr=5.3497e-04) (hash(x)=5774189)
110
+ 102 train 7.791671 (lr=5.4021e-04) (hash(x)=5588220)
111
+ 103 train 7.720082 (lr=5.4545e-04) (hash(x)=5985675)
112
+ 104 train 7.684405 (lr=5.5070e-04) (hash(x)=6259214)
113
+ 105 train 7.914079 (lr=5.5594e-04) (hash(x)=6451776)
114
+ 106 train 7.581518 (lr=5.6119e-04) (hash(x)=4940406)
115
+ 107 train 7.702998 (lr=5.6643e-04) (hash(x)=4355733)
116
+ 108 train 7.718850 (lr=5.7168e-04) (hash(x)=6186065)
117
+ 109 train 7.945103 (lr=5.7692e-04) (hash(x)=6669188)
118
+ 110 train 9.077463 (lr=5.8217e-04) (hash(x)=7118997)
119
+ 111 train 8.276721 (lr=5.8741e-04) (hash(x)=6984772)
120
+ 112 train 7.774105 (lr=5.9266e-04) (hash(x)=6472877)
121
+ 113 train 7.981002 (lr=5.9790e-04) (hash(x)=7440580)
122
+ 114 train 7.972934 (lr=6.0315e-04) (hash(x)=6378428)
123
+ 115 train 8.092747 (lr=6.0839e-04) (hash(x)=6244096)
124
+ 116 train 8.080446 (lr=6.1364e-04) (hash(x)=9100578)
125
+ 117 train 8.335641 (lr=6.1888e-04) (hash(x)=8381560)
126
+ 118 train 8.097921 (lr=6.2413e-04) (hash(x)=6066072)
127
+ 119 train 7.796756 (lr=6.2937e-04) (hash(x)=6603717)
128
+ 120 train 7.643592 (lr=6.3462e-04) (hash(x)=5138438)
129
+ 121 train 7.589627 (lr=6.3986e-04) (hash(x)=6437516)
130
+ 122 train 7.791254 (lr=6.4510e-04) (hash(x)=5743807)
131
+ 123 train 7.600759 (lr=6.5035e-04) (hash(x)=5669522)
132
+ 124 train 7.886339 (lr=6.5559e-04) (hash(x)=5963623)
133
+ 125 train 7.631421 (lr=6.6084e-04) (hash(x)=5505889)
134
+ 126 train 8.049172 (lr=6.6608e-04) (hash(x)=7607512)
135
+ 127 train 7.854572 (lr=6.7133e-04) (hash(x)=6190579)
136
+ 128 train 8.071034 (lr=6.7657e-04) (hash(x)=7444834)
137
+ 129 train 7.929633 (lr=6.8182e-04) (hash(x)=5153196)
138
+ 130 train 7.869814 (lr=6.8706e-04) (hash(x)=6495326)
139
+ 131 train 7.945278 (lr=6.9231e-04) (hash(x)=4912983)
140
+ 132 train 8.270130 (lr=6.9755e-04) (hash(x)=7094046)
141
+ 133 train 7.776339 (lr=7.0280e-04) (hash(x)=5109125)
142
+ 134 train 7.765362 (lr=7.0804e-04) (hash(x)=7119136)
143
+ 135 train 7.779494 (lr=7.1329e-04) (hash(x)=7276303)
144
+ 136 train 8.020192 (lr=7.1853e-04) (hash(x)=9147123)
145
+ 137 train 8.027353 (lr=7.2378e-04) (hash(x)=6645716)
146
+ 138 train 7.996442 (lr=7.2902e-04) (hash(x)=6347875)
147
+ 139 train 7.726008 (lr=7.3427e-04) (hash(x)=5479947)
148
+ 140 train 7.866443 (lr=7.3951e-04) (hash(x)=5832595)
149
+ 141 train 8.173807 (lr=7.4476e-04) (hash(x)=7850360)
150
+ 142 train 8.557974 (lr=7.5000e-04) (hash(x)=8499683)
151
+ 143 train 8.103818 (lr=7.5524e-04) (hash(x)=6630378)
152
+ 144 train 8.081128 (lr=7.6049e-04) (hash(x)=5340846)
153
+ 145 train 7.898849 (lr=7.6573e-04) (hash(x)=6295358)
154
+ 146 train 7.919582 (lr=7.7098e-04) (hash(x)=5974729)
155
+ 147 train 8.445787 (lr=7.7622e-04) (hash(x)=8097010)
156
+ 148 train 8.144188 (lr=7.8147e-04) (hash(x)=6681250)
157
+ 149 train 7.603414 (lr=7.8671e-04) (hash(x)=3958188)
158
+ 150 val loss 7.8867
159
+ 150 val perplexity 2661.7212
160
+ 150 train 7.849049 (lr=7.9196e-04) (hash(x)=5975539)
161
+ 151 train 9.719590 (lr=7.9720e-04) (hash(x)=9989744)
162
+ 152 train 8.809632 (lr=8.0245e-04) (hash(x)=6748487)
163
+ 153 train 7.759539 (lr=8.0769e-04) (hash(x)=6343832)
164
+ 154 train 7.854332 (lr=8.1294e-04) (hash(x)=5688949)
165
+ 155 train 8.214021 (lr=8.1818e-04) (hash(x)=6694932)
166
+ 156 train 7.849502 (lr=8.2343e-04) (hash(x)=5872835)
167
+ 157 train 7.881592 (lr=8.2867e-04) (hash(x)=6429137)
168
+ 158 train 8.068076 (lr=8.3392e-04) (hash(x)=6999390)
169
+ 159 train 7.914161 (lr=8.3916e-04) (hash(x)=6317894)
170
+ 160 train 7.632976 (lr=8.4441e-04) (hash(x)=5813657)
171
+ 161 train 7.715853 (lr=8.4965e-04) (hash(x)=5165595)
172
+ 162 train 7.725824 (lr=8.5490e-04) (hash(x)=5682633)
173
+ 163 train 8.034738 (lr=8.6014e-04) (hash(x)=5046327)
174
+ 164 train 8.073313 (lr=8.6538e-04) (hash(x)=6195808)
175
+ 165 train 7.944864 (lr=8.7063e-04) (hash(x)=7060363)
176
+ 166 train 7.942782 (lr=8.7587e-04) (hash(x)=5837126)
177
+ 167 train 7.893779 (lr=8.8112e-04) (hash(x)=7947703)
178
+ 168 train 7.543088 (lr=8.8636e-04) (hash(x)=5277865)
179
+ 169 train 7.815063 (lr=8.9161e-04) (hash(x)=5971240)
180
+ 170 train 7.758972 (lr=8.9685e-04) (hash(x)=6590408)
181
+ 171 train 7.918045 (lr=9.0210e-04) (hash(x)=5916068)
182
+ 172 train 7.831188 (lr=9.0734e-04) (hash(x)=6069727)
183
+ 173 train 7.889788 (lr=9.1259e-04) (hash(x)=7426277)
184
+ 174 train 8.020996 (lr=9.1783e-04) (hash(x)=5356513)
185
+ 175 train 7.805211 (lr=9.2308e-04) (hash(x)=5777498)
186
+ 176 train 8.062244 (lr=9.2832e-04) (hash(x)=5989756)
187
+ 177 train 7.785749 (lr=9.3357e-04) (hash(x)=6403331)
188
+ 178 train 7.831416 (lr=9.3881e-04) (hash(x)=7650667)
189
+ 179 train 7.941766 (lr=9.4406e-04) (hash(x)=5753092)
190
+ 180 train 7.798845 (lr=9.4930e-04) (hash(x)=5751457)
191
+ 181 train 7.669033 (lr=9.5455e-04) (hash(x)=5981486)
192
+ 182 train 7.664661 (lr=9.5979e-04) (hash(x)=5918229)
193
+ 183 train 7.875666 (lr=9.6503e-04) (hash(x)=7723226)
194
+ 184 train 7.986787 (lr=9.7028e-04) (hash(x)=6111721)
195
+ 185 train 7.654712 (lr=9.7552e-04) (hash(x)=6069558)
196
+ 186 train 7.961263 (lr=9.8077e-04) (hash(x)=5626336)
197
+ 187 train 8.246021 (lr=9.8601e-04) (hash(x)=7007016)
198
+ 188 train 7.725991 (lr=9.9126e-04) (hash(x)=5883013)
199
+ 189 train 7.570884 (lr=9.9650e-04) (hash(x)=5027958)
200
+ 190 train 7.890906 (lr=1.0017e-03) (hash(x)=5474948)
201
+ 191 train 7.731412 (lr=1.0070e-03) (hash(x)=6491229)
202
+ 192 train 7.865808 (lr=1.0122e-03) (hash(x)=5518341)
203
+ 193 train 7.850874 (lr=1.0175e-03) (hash(x)=7036653)
204
+ 194 train 7.872590 (lr=1.0227e-03) (hash(x)=5527616)
205
+ 195 train 8.227156 (lr=1.0280e-03) (hash(x)=6390752)
206
+ 196 train 7.871863 (lr=1.0332e-03) (hash(x)=6032557)
207
+ 197 train 8.638637 (lr=1.0385e-03) (hash(x)=7758167)
208
+ 198 train 8.419679 (lr=1.0437e-03) (hash(x)=7968070)
209
+ 199 train 7.947768 (lr=1.0490e-03) (hash(x)=6019389)
210
+ 200 val loss 7.8995
211
+ 200 val perplexity 2695.9016
212
+ 200 train 7.986177 (lr=1.0542e-03) (hash(x)=7276744)
213
+ 201 train 8.007886 (lr=1.0594e-03) (hash(x)=6503191)
214
+ 202 train 7.827729 (lr=1.0647e-03) (hash(x)=6006880)
215
+ 203 train 8.051325 (lr=1.0699e-03) (hash(x)=7662067)
216
+ 204 train 7.784589 (lr=1.0752e-03) (hash(x)=6297345)
217
+ 205 train 8.046803 (lr=1.0804e-03) (hash(x)=7901992)
218
+ 206 train 8.096746 (lr=1.0857e-03) (hash(x)=6579950)
219
+ 207 train 7.710021 (lr=1.0909e-03) (hash(x)=4648609)
220
+ 208 train 7.923376 (lr=1.0962e-03) (hash(x)=6903216)
221
+ 209 train 7.758868 (lr=1.1014e-03) (hash(x)=5897288)
222
+ 210 train 7.857474 (lr=1.1066e-03) (hash(x)=7300160)
223
+ 211 train 7.683037 (lr=1.1119e-03) (hash(x)=4725966)
224
+ 212 train 7.490445 (lr=1.1171e-03) (hash(x)=5060060)
225
+ 213 train 8.044586 (lr=1.1224e-03) (hash(x)=6243442)
226
+ 214 train 7.711121 (lr=1.1276e-03) (hash(x)=5893816)
227
+ 215 train 7.748185 (lr=1.1329e-03) (hash(x)=5558355)
228
+ 216 train 7.861792 (lr=1.1381e-03) (hash(x)=4378747)
229
+ 217 train 7.721724 (lr=1.1434e-03) (hash(x)=5400407)
230
+ 218 train 8.011228 (lr=1.1486e-03) (hash(x)=6900554)
231
+ 219 train 7.910191 (lr=1.1538e-03) (hash(x)=6524933)
232
+ 220 train 7.808038 (lr=1.1591e-03) (hash(x)=5409944)
233
+ 221 train 7.949763 (lr=1.1643e-03) (hash(x)=5889724)
234
+ 222 train 7.841991 (lr=1.1696e-03) (hash(x)=4970496)
235
+ 223 train 7.595563 (lr=1.1748e-03) (hash(x)=6369326)
236
+ 224 train 7.689951 (lr=1.1801e-03) (hash(x)=6563975)
237
+ 225 train 7.864306 (lr=1.1853e-03) (hash(x)=5911906)
238
+ 226 train 7.958080 (lr=1.1906e-03) (hash(x)=6111462)
239
+ 227 train 7.993994 (lr=1.1958e-03) (hash(x)=6022625)
240
+ 228 train 7.787601 (lr=1.2010e-03) (hash(x)=5884663)
241
+ 229 train 7.884144 (lr=1.2063e-03) (hash(x)=6584810)
242
+ 230 train 7.812581 (lr=1.2115e-03) (hash(x)=5289998)
243
+ 231 train 7.505483 (lr=1.2168e-03) (hash(x)=4906853)
244
+ 232 train 7.586112 (lr=1.2220e-03) (hash(x)=5862071)
245
+ 233 train 7.228203 (lr=1.2273e-03) (hash(x)=4564564)
246
+ 234 train 7.617900 (lr=1.2325e-03) (hash(x)=5723971)
247
+ 235 train 7.708233 (lr=1.2378e-03) (hash(x)=5589269)
248
+ 236 train 7.692608 (lr=1.2430e-03) (hash(x)=5683368)
249
+ 237 train 7.941964 (lr=1.2483e-03) (hash(x)=6969180)
250
+ 238 train 7.775698 (lr=1.2535e-03) (hash(x)=5341421)
251
+ 239 train 7.551417 (lr=1.2587e-03) (hash(x)=6437376)
252
+ 240 train 8.039429 (lr=1.2640e-03) (hash(x)=7527007)
253
+ 241 train 7.858900 (lr=1.2692e-03) (hash(x)=7116258)
254
+ 242 train 7.885498 (lr=1.2745e-03) (hash(x)=6463930)
255
+ 243 train 7.868840 (lr=1.2797e-03) (hash(x)=7304892)
256
+ 244 train 8.057841 (lr=1.2850e-03) (hash(x)=5881808)
257
+ 245 train 7.814553 (lr=1.2902e-03) (hash(x)=7366721)
258
+ 246 train 7.788866 (lr=1.2955e-03) (hash(x)=6457521)
259
+ 247 train 8.122196 (lr=1.3007e-03) (hash(x)=7084093)
260
+ 248 train 7.817363 (lr=1.3059e-03) (hash(x)=6015300)
261
+ 249 train 8.050730 (lr=1.3112e-03) (hash(x)=7197652)
262
+ 250 val loss 7.7776
263
+ 250 val perplexity 2386.5112
264
+ 250 train 8.082134 (lr=1.3164e-03) (hash(x)=4830828)
265
+ 251 train 7.826513 (lr=1.3217e-03) (hash(x)=5908178)
266
+ 252 train 7.797794 (lr=1.3269e-03) (hash(x)=6299092)
267
+ 253 train 7.782921 (lr=1.3322e-03) (hash(x)=6349974)
268
+ 254 train 7.796968 (lr=1.3374e-03) (hash(x)=5641494)
269
+ 255 train 7.761440 (lr=1.3427e-03) (hash(x)=7048804)
270
+ 256 train 8.005521 (lr=1.3479e-03) (hash(x)=7035244)
271
+ 257 train 7.669155 (lr=1.3531e-03) (hash(x)=7909039)
272
+ 258 train 7.950775 (lr=1.3584e-03) (hash(x)=5470939)
273
+ 259 train 7.795498 (lr=1.3636e-03) (hash(x)=6085549)
274
+ 260 train 8.059212 (lr=1.3689e-03) (hash(x)=5882649)
275
+ 261 train 7.846640 (lr=1.3741e-03) (hash(x)=7463181)
276
+ 262 train 8.173861 (lr=1.3794e-03) (hash(x)=5215574)
277
+ 263 train 7.629200 (lr=1.3846e-03) (hash(x)=5752594)
278
+ 264 train 7.784893 (lr=1.3899e-03) (hash(x)=6416619)
279
+ 265 train 7.802832 (lr=1.3951e-03) (hash(x)=6114601)
280
+ 266 train 7.744247 (lr=1.4003e-03) (hash(x)=5646171)
281
+ 267 train 7.652860 (lr=1.4056e-03) (hash(x)=7662769)
282
+ 268 train 7.936056 (lr=1.4108e-03) (hash(x)=6394376)
283
+ 269 train 7.528862 (lr=1.4161e-03) (hash(x)=7485666)
284
+ 270 train 7.630411 (lr=1.4213e-03) (hash(x)=6636823)
285
+ 271 train 7.527545 (lr=1.4266e-03) (hash(x)=6393520)
286
+ 272 train 7.777289 (lr=1.4318e-03) (hash(x)=5235659)
287
+ 273 train 7.820323 (lr=1.4371e-03) (hash(x)=5586291)
288
+ 274 train 8.047457 (lr=1.4423e-03) (hash(x)=7034674)
289
+ 275 train 8.186968 (lr=1.4476e-03) (hash(x)=5942867)
290
+ 276 train 7.858039 (lr=1.4528e-03) (hash(x)=7812344)
291
+ 277 train 8.095165 (lr=1.4580e-03) (hash(x)=7574027)
292
+ 278 train 8.074847 (lr=1.4633e-03) (hash(x)=5960216)
293
+ 279 train 7.821369 (lr=1.4685e-03) (hash(x)=6793550)
294
+ 280 train 7.803275 (lr=1.4738e-03) (hash(x)=6879965)
295
+ 281 train 7.485213 (lr=1.4790e-03) (hash(x)=6264789)
296
+ 282 train 7.682151 (lr=1.4843e-03) (hash(x)=5175356)
297
+ 283 train 7.927521 (lr=1.4895e-03) (hash(x)=7105976)
298
+ 284 train 7.868359 (lr=1.4948e-03) (hash(x)=7286386)
299
+ 285 train 8.047548 (lr=1.5000e-03) (hash(x)=7244279)
300
+ 286 train 7.592000 (lr=1.5000e-03) (hash(x)=4877247)
301
+ 287 train 7.560244 (lr=1.5000e-03) (hash(x)=6581348)
302
+ 288 train 7.567259 (lr=1.5000e-03) (hash(x)=6397208)
303
+ 289 train 7.649177 (lr=1.4999e-03) (hash(x)=6543825)
304
+ 290 train 7.452191 (lr=1.4999e-03) (hash(x)=6088993)
305
+ 291 train 7.761714 (lr=1.4998e-03) (hash(x)=5555598)
306
+ 292 train 7.861984 (lr=1.4998e-03) (hash(x)=6924240)
307
+ 293 train 7.987164 (lr=1.4997e-03) (hash(x)=7254232)
308
+ 294 train 7.574181 (lr=1.4996e-03) (hash(x)=7689743)
309
+ 295 train 7.566612 (lr=1.4995e-03) (hash(x)=6235837)
310
+ 296 train 7.824226 (lr=1.4993e-03) (hash(x)=5216883)
311
+ 297 train 7.427088 (lr=1.4992e-03) (hash(x)=4738550)
312
+ 298 train 7.620013 (lr=1.4991e-03) (hash(x)=5126918)
313
+ 299 train 7.851313 (lr=1.4989e-03) (hash(x)=5591770)
314
+ 300 val loss 7.7050
315
+ 300 val perplexity 2219.4363
316
+ 300 train 8.695097 (lr=1.4987e-03) (hash(x)=8001573)
317
+ 301 train 8.324284 (lr=1.4985e-03) (hash(x)=8012491)
318
+ 302 train 8.006080 (lr=1.4983e-03) (hash(x)=3997428)
319
+ 303 train 7.435925 (lr=1.4981e-03) (hash(x)=7083021)
320
+ 304 train 8.015249 (lr=1.4979e-03) (hash(x)=7177371)
321
+ 305 train 7.779846 (lr=1.4976e-03) (hash(x)=6456022)
322
+ 306 train 7.836329 (lr=1.4974e-03) (hash(x)=5238861)
323
+ 307 train 7.822866 (lr=1.4971e-03) (hash(x)=7281348)
324
+ 308 train 8.021527 (lr=1.4968e-03) (hash(x)=6848441)
325
+ 309 train 8.081388 (lr=1.4965e-03) (hash(x)=7824794)
326
+ 310 train 8.199903 (lr=1.4962e-03) (hash(x)=7440208)
327
+ 311 train 7.981733 (lr=1.4959e-03) (hash(x)=6155821)
328
+ 312 train 7.711666 (lr=1.4956e-03) (hash(x)=5350461)
329
+ 313 train 7.915307 (lr=1.4952e-03) (hash(x)=7807564)
330
+ 314 train 7.788118 (lr=1.4949e-03) (hash(x)=5992165)
331
+ 315 train 7.772419 (lr=1.4945e-03) (hash(x)=5736241)
332
+ 316 train 7.711909 (lr=1.4941e-03) (hash(x)=6536253)
333
+ 317 train 7.939167 (lr=1.4937e-03) (hash(x)=5688415)
334
+ 318 train 7.536775 (lr=1.4933e-03) (hash(x)=5839705)
335
+ 319 train 7.730500 (lr=1.4929e-03) (hash(x)=5657123)
336
+ 320 train 7.743964 (lr=1.4925e-03) (hash(x)=6354364)
337
+ 321 train 7.760127 (lr=1.4920e-03) (hash(x)=6102985)
338
+ 322 train 8.018381 (lr=1.4915e-03) (hash(x)=7161967)
339
+ 323 train 7.919029 (lr=1.4911e-03) (hash(x)=6452095)
340
+ 324 train 7.635701 (lr=1.4906e-03) (hash(x)=5547396)
341
+ 325 train 7.941135 (lr=1.4901e-03) (hash(x)=5848921)
342
+ 326 train 7.768565 (lr=1.4896e-03) (hash(x)=4890294)
343
+ 327 train 7.492349 (lr=1.4890e-03) (hash(x)=5312267)
344
+ 328 train 7.488919 (lr=1.4885e-03) (hash(x)=5348929)
345
+ 329 train 7.418282 (lr=1.4880e-03) (hash(x)=5025297)
346
+ 330 train 7.411344 (lr=1.4874e-03) (hash(x)=5670052)
347
+ 331 train 7.269845 (lr=1.4868e-03) (hash(x)=5434493)
348
+ 332 train 8.113320 (lr=1.4862e-03) (hash(x)=9122453)
349
+ 333 train 7.789946 (lr=1.4856e-03) (hash(x)=5197474)
350
+ 334 train 7.665634 (lr=1.4850e-03) (hash(x)=5550786)
351
+ 335 train 7.713607 (lr=1.4844e-03) (hash(x)=6830813)
352
+ 336 train 7.581280 (lr=1.4837e-03) (hash(x)=5258435)
353
+ 337 train 7.593606 (lr=1.4831e-03) (hash(x)=5242548)
354
+ 338 train 7.492717 (lr=1.4824e-03) (hash(x)=3754886)
355
+ 339 train 7.400321 (lr=1.4817e-03) (hash(x)=4752771)
356
+ 340 train 7.563559 (lr=1.4810e-03) (hash(x)=5926875)
357
+ 341 train 8.057698 (lr=1.4803e-03) (hash(x)=6691249)
358
+ 342 train 7.782262 (lr=1.4796e-03) (hash(x)=7514623)
359
+ 343 train 7.868989 (lr=1.4789e-03) (hash(x)=6424933)
360
+ 344 train 7.812491 (lr=1.4781e-03) (hash(x)=6347757)
361
+ 345 train 7.736119 (lr=1.4774e-03) (hash(x)=6960120)
362
+ 346 train 7.789748 (lr=1.4766e-03) (hash(x)=7597197)
363
+ 347 train 7.686011 (lr=1.4758e-03) (hash(x)=5786517)
364
+ 348 train 7.731782 (lr=1.4750e-03) (hash(x)=6424363)
365
+ 349 train 7.842951 (lr=1.4742e-03) (hash(x)=7229689)
366
+ 350 val loss 7.7209
367
+ 350 val perplexity 2255.0947
368
+ 350 train 7.907228 (lr=1.4734e-03) (hash(x)=6772284)
369
+ 351 train 7.971339 (lr=1.4726e-03) (hash(x)=6680023)
370
+ 352 train 7.843497 (lr=1.4717e-03) (hash(x)=6218520)
371
+ 353 train 7.903519 (lr=1.4709e-03) (hash(x)=7688393)
372
+ 354 train 7.776500 (lr=1.4700e-03) (hash(x)=6327977)
373
+ 355 train 7.686389 (lr=1.4691e-03) (hash(x)=6474729)
374
+ 356 train 7.733762 (lr=1.4682e-03) (hash(x)=6657266)
375
+ 357 train 7.793602 (lr=1.4673e-03) (hash(x)=5946650)
376
+ 358 train 7.698977 (lr=1.4664e-03) (hash(x)=6734878)
377
+ 359 train 7.802058 (lr=1.4655e-03) (hash(x)=7523279)
378
+ 360 train 7.617442 (lr=1.4645e-03) (hash(x)=6179478)
379
+ 361 train 7.735754 (lr=1.4636e-03) (hash(x)=5531605)
380
+ 362 train 7.729928 (lr=1.4626e-03) (hash(x)=5844980)
381
+ 363 train 7.425499 (lr=1.4616e-03) (hash(x)=5508768)
382
+ 364 train 7.280815 (lr=1.4606e-03) (hash(x)=6337190)
383
+ 365 train 8.794675 (lr=1.4596e-03) (hash(x)=6415982)
384
+ 366 train 7.905602 (lr=1.4586e-03) (hash(x)=6828959)
385
+ 367 train 7.923635 (lr=1.4576e-03) (hash(x)=6635925)
386
+ 368 train 7.636601 (lr=1.4565e-03) (hash(x)=6358540)
387
+ 369 train 7.992207 (lr=1.4555e-03) (hash(x)=5923706)
388
+ 370 train 7.647240 (lr=1.4544e-03) (hash(x)=4843320)
389
+ 371 train 7.922570 (lr=1.4533e-03) (hash(x)=6663801)
390
+ 372 train 7.712862 (lr=1.4522e-03) (hash(x)=5583287)
391
+ 373 train 7.724082 (lr=1.4511e-03) (hash(x)=6113540)
392
+ 374 train 7.838981 (lr=1.4500e-03) (hash(x)=6389936)
393
+ 375 train 7.749400 (lr=1.4489e-03) (hash(x)=5869441)
394
+ 376 train 7.835543 (lr=1.4478e-03) (hash(x)=5148617)
395
+ 377 train 8.075262 (lr=1.4466e-03) (hash(x)=6913454)
396
+ 378 train 7.812488 (lr=1.4454e-03) (hash(x)=5356233)
397
+ 379 train 7.774578 (lr=1.4443e-03) (hash(x)=6698878)
398
+ 380 train 7.699169 (lr=1.4431e-03) (hash(x)=6280665)
399
+ 381 train 7.962288 (lr=1.4419e-03) (hash(x)=6094899)
400
+ 382 train 7.488599 (lr=1.4407e-03) (hash(x)=6249746)
401
+ 383 train 7.814994 (lr=1.4394e-03) (hash(x)=6349004)
402
+ 384 train 7.763897 (lr=1.4382e-03) (hash(x)=7013468)
403
+ 385 train 7.743126 (lr=1.4370e-03) (hash(x)=6045933)
404
+ 386 train 8.058770 (lr=1.4357e-03) (hash(x)=6571735)
405
+ 387 train 7.618285 (lr=1.4344e-03) (hash(x)=4986137)
406
+ 388 train 7.301319 (lr=1.4332e-03) (hash(x)=5244912)
407
+ 389 train 7.676934 (lr=1.4319e-03) (hash(x)=4798229)
408
+ 390 train 7.732220 (lr=1.4306e-03) (hash(x)=5815783)
409
+ 391 train 7.568341 (lr=1.4292e-03) (hash(x)=6008454)
410
+ 392 train 7.795481 (lr=1.4279e-03) (hash(x)=6407333)
411
+ 393 train 7.634244 (lr=1.4266e-03) (hash(x)=5938362)
412
+ 394 train 7.521190 (lr=1.4252e-03) (hash(x)=6077124)
413
+ 395 train 7.766332 (lr=1.4238e-03) (hash(x)=6550770)
414
+ 396 train 7.647344 (lr=1.4225e-03) (hash(x)=6181528)
415
+ 397 train 7.859614 (lr=1.4211e-03) (hash(x)=7055344)
416
+ 398 train 8.009485 (lr=1.4197e-03) (hash(x)=7689348)
417
+ 399 train 7.909332 (lr=1.4183e-03) (hash(x)=7682741)
418
+ 400 val loss 7.8091
419
+ 400 val perplexity 2462.8333
420
+ 400 train 7.728385 (lr=1.4169e-03) (hash(x)=5189401)
421
+ 401 train 7.756392 (lr=1.4154e-03) (hash(x)=7372095)
422
+ 402 train 7.839929 (lr=1.4140e-03) (hash(x)=6838219)
423
+ 403 train 8.346878 (lr=1.4125e-03) (hash(x)=7892158)
424
+ 404 train 7.963587 (lr=1.4110e-03) (hash(x)=5677077)
425
+ 405 train 7.590663 (lr=1.4096e-03) (hash(x)=5929740)
426
+ 406 train 7.732174 (lr=1.4081e-03) (hash(x)=5285055)
427
+ 407 train 7.801322 (lr=1.4066e-03) (hash(x)=7794028)
428
+ 408 train 7.880821 (lr=1.4051e-03) (hash(x)=6640802)
429
+ 409 train 7.963985 (lr=1.4035e-03) (hash(x)=6839923)
430
+ 410 train 7.441358 (lr=1.4020e-03) (hash(x)=4020997)
431
+ 411 train 7.625512 (lr=1.4005e-03) (hash(x)=7093523)
432
+ 412 train 7.903796 (lr=1.3989e-03) (hash(x)=7370045)
433
+ 413 train 7.649959 (lr=1.3973e-03) (hash(x)=5341312)
434
+ 414 train 7.640549 (lr=1.3957e-03) (hash(x)=5589312)
435
+ 415 train 8.138444 (lr=1.3942e-03) (hash(x)=6166062)
436
+ 416 train 8.029768 (lr=1.3926e-03) (hash(x)=8354149)
437
+ 417 train 7.663107 (lr=1.3909e-03) (hash(x)=5992164)
438
+ 418 train 7.649403 (lr=1.3893e-03) (hash(x)=4748657)
439
+ 419 train 7.688432 (lr=1.3877e-03) (hash(x)=6645781)
440
+ 420 train 7.924760 (lr=1.3860e-03) (hash(x)=6689147)
441
+ 421 train 7.788351 (lr=1.3844e-03) (hash(x)=6207724)
442
+ 422 train 7.734213 (lr=1.3827e-03) (hash(x)=6232863)
443
+ 423 train 7.491404 (lr=1.3810e-03) (hash(x)=4811192)
444
+ 424 train 7.506578 (lr=1.3793e-03) (hash(x)=6211761)
445
+ 425 train 7.685887 (lr=1.3776e-03) (hash(x)=6906335)
446
+ 426 train 7.614520 (lr=1.3759e-03) (hash(x)=4922365)
447
+ 427 train 7.688290 (lr=1.3742e-03) (hash(x)=5970866)
448
+ 428 train 7.583428 (lr=1.3725e-03) (hash(x)=6417833)
449
+ 429 train 7.672919 (lr=1.3707e-03) (hash(x)=5238015)
450
+ 430 train 7.615414 (lr=1.3690e-03) (hash(x)=5375451)
451
+ 431 train 7.735305 (lr=1.3672e-03) (hash(x)=7822680)
452
+ 432 train 7.757323 (lr=1.3654e-03) (hash(x)=6173200)
453
+ 433 train 7.555243 (lr=1.3637e-03) (hash(x)=5478019)
454
+ 434 train 7.764196 (lr=1.3619e-03) (hash(x)=5597859)
455
+ 435 train 7.521608 (lr=1.3601e-03) (hash(x)=6317317)
456
+ 436 train 7.745163 (lr=1.3582e-03) (hash(x)=7120467)
457
+ 437 train 7.873893 (lr=1.3564e-03) (hash(x)=5832344)
458
+ 438 train 7.777862 (lr=1.3546e-03) (hash(x)=6813802)
459
+ 439 train 7.827317 (lr=1.3527e-03) (hash(x)=6431409)
460
+ 440 train 7.934685 (lr=1.3509e-03) (hash(x)=5886861)
461
+ 441 train 7.641377 (lr=1.3490e-03) (hash(x)=5900130)
462
+ 442 train 7.851336 (lr=1.3471e-03) (hash(x)=6061563)
463
+ 443 train 7.844938 (lr=1.3452e-03) (hash(x)=6653337)
464
+ 444 train 7.774863 (lr=1.3434e-03) (hash(x)=6575992)
465
+ 445 train 7.725399 (lr=1.3414e-03) (hash(x)=6696784)
466
+ 446 train 7.839738 (lr=1.3395e-03) (hash(x)=6242161)
467
+ 447 train 7.716653 (lr=1.3376e-03) (hash(x)=5323032)
468
+ 448 train 7.505628 (lr=1.3357e-03) (hash(x)=6439367)
469
+ 449 train 7.682821 (lr=1.3337e-03) (hash(x)=5372100)
470
+ 450 val loss 7.7203
471
+ 450 val perplexity 2253.6221
472
+ 450 train 7.775460 (lr=1.3318e-03) (hash(x)=6344849)
473
+ 451 train 7.449935 (lr=1.3298e-03) (hash(x)=5125339)
474
+ 452 train 7.461927 (lr=1.3278e-03) (hash(x)=4876156)
475
+ 453 train 7.578495 (lr=1.3258e-03) (hash(x)=5664463)
476
+ 454 train 7.457846 (lr=1.3238e-03) (hash(x)=6010042)
477
+ 455 train 7.656401 (lr=1.3218e-03) (hash(x)=7363286)
478
+ 456 train 8.196832 (lr=1.3198e-03) (hash(x)=4873268)
479
+ 457 train 7.947922 (lr=1.3178e-03) (hash(x)=7444468)
480
+ 458 train 7.963708 (lr=1.3158e-03) (hash(x)=6698134)
481
+ 459 train 7.798575 (lr=1.3137e-03) (hash(x)=7670050)
482
+ 460 train 7.792019 (lr=1.3117e-03) (hash(x)=6280866)
483
+ 461 train 7.674026 (lr=1.3096e-03) (hash(x)=8189420)
484
+ 462 train 7.778609 (lr=1.3075e-03) (hash(x)=6878292)
485
+ 463 train 7.527696 (lr=1.3054e-03) (hash(x)=5616075)
486
+ 464 train 7.732141 (lr=1.3033e-03) (hash(x)=5839939)
487
+ 465 train 7.583973 (lr=1.3012e-03) (hash(x)=6360924)
488
+ 466 train 8.053835 (lr=1.2991e-03) (hash(x)=6466475)
489
+ 467 train 7.816132 (lr=1.2970e-03) (hash(x)=6593764)
490
+ 468 train 8.316963 (lr=1.2949e-03) (hash(x)=7389698)
491
+ 469 train 7.953768 (lr=1.2928e-03) (hash(x)=6764377)
492
+ 470 train 7.645196 (lr=1.2906e-03) (hash(x)=7293570)
493
+ 471 train 7.685136 (lr=1.2885e-03) (hash(x)=5672608)
494
+ 472 train 7.513605 (lr=1.2863e-03) (hash(x)=5587975)
495
+ 473 train 7.418018 (lr=1.2841e-03) (hash(x)=6985256)
496
+ 474 train 8.032349 (lr=1.2819e-03) (hash(x)=6549231)
497
+ 475 train 7.740306 (lr=1.2797e-03) (hash(x)=6828653)
498
+ 476 train 7.575130 (lr=1.2775e-03) (hash(x)=5478970)
499
+ 477 train 7.614217 (lr=1.2753e-03) (hash(x)=5199845)
500
+ 478 train 7.677417 (lr=1.2731e-03) (hash(x)=6482073)
501
+ 479 train 7.755421 (lr=1.2709e-03) (hash(x)=6839867)
502
+ 480 train 7.753657 (lr=1.2687e-03) (hash(x)=7073454)
503
+ 481 train 7.979162 (lr=1.2664e-03) (hash(x)=6928712)
504
+ 482 train 7.738640 (lr=1.2642e-03) (hash(x)=5962982)
505
+ 483 train 7.831345 (lr=1.2619e-03) (hash(x)=8426992)
506
+ 484 train 7.762107 (lr=1.2596e-03) (hash(x)=6422394)
507
+ 485 train 7.715695 (lr=1.2574e-03) (hash(x)=5330355)
508
+ 486 train 7.210474 (lr=1.2551e-03) (hash(x)=4111491)
509
+ 487 train 7.239897 (lr=1.2528e-03) (hash(x)=4916928)
510
+ 488 train 7.387963 (lr=1.2505e-03) (hash(x)=6610244)
511
+ 489 train 7.822598 (lr=1.2482e-03) (hash(x)=7287928)
512
+ 490 train 7.769091 (lr=1.2459e-03) (hash(x)=6525083)
513
+ 491 train 7.709898 (lr=1.2435e-03) (hash(x)=6484050)
514
+ 492 train 7.826107 (lr=1.2412e-03) (hash(x)=6573424)
515
+ 493 train 7.715997 (lr=1.2389e-03) (hash(x)=5062850)
516
+ 494 train 7.684770 (lr=1.2365e-03) (hash(x)=6215995)
517
+ 495 train 7.825922 (lr=1.2341e-03) (hash(x)=8353379)
518
+ 496 train 7.688693 (lr=1.2318e-03) (hash(x)=6344823)
519
+ 497 train 7.659286 (lr=1.2294e-03) (hash(x)=6805358)
520
+ 498 train 7.565568 (lr=1.2270e-03) (hash(x)=6018070)
521
+ 499 train 7.849277 (lr=1.2246e-03) (hash(x)=6552510)
522
+ 500 val loss 7.7022
523
+ 500 val perplexity 2213.2854
524
+ 500 train 7.616573 (lr=1.2222e-03) (hash(x)=7218108)
525
+ 501 train 7.790018 (lr=1.2198e-03) (hash(x)=5105545)
526
+ 502 train 7.788584 (lr=1.2174e-03) (hash(x)=5997860)
527
+ 503 train 7.612062 (lr=1.2150e-03) (hash(x)=4838871)
528
+ 504 train 7.655845 (lr=1.2126e-03) (hash(x)=7105888)
529
+ 505 train 7.929840 (lr=1.2101e-03) (hash(x)=5945286)
530
+ 506 train 7.226278 (lr=1.2077e-03) (hash(x)=4381720)
531
+ 507 train 7.011541 (lr=1.2052e-03) (hash(x)=3021697)
532
+ 508 train 7.379128 (lr=1.2028e-03) (hash(x)=6826973)
533
+ 509 train 7.847807 (lr=1.2003e-03) (hash(x)=6443070)
534
+ 510 train 7.551422 (lr=1.1978e-03) (hash(x)=5520637)
535
+ 511 train 7.781796 (lr=1.1954e-03) (hash(x)=6795665)
536
+ 512 train 7.579126 (lr=1.1929e-03) (hash(x)=6063891)
537
+ 513 train 8.066650 (lr=1.1904e-03) (hash(x)=7683068)
538
+ 514 train 7.320961 (lr=1.1879e-03) (hash(x)=5526694)
539
+ 515 train 7.426521 (lr=1.1854e-03) (hash(x)=5486935)
540
+ 516 train 7.791752 (lr=1.1829e-03) (hash(x)=6002252)
541
+ 517 train 7.771213 (lr=1.1803e-03) (hash(x)=6115232)
542
+ 518 train 8.104203 (lr=1.1778e-03) (hash(x)=9039136)
543
+ 519 train 7.723840 (lr=1.1753e-03) (hash(x)=6678038)
544
+ 520 train 7.866488 (lr=1.1727e-03) (hash(x)=5383926)
545
+ 521 train 7.872234 (lr=1.1702e-03) (hash(x)=5747629)
546
+ 522 train 7.764795 (lr=1.1676e-03) (hash(x)=7090040)
547
+ 523 train 7.999930 (lr=1.1651e-03) (hash(x)=6657714)
548
+ 524 train 7.674353 (lr=1.1625e-03) (hash(x)=5414034)
549
+ 525 train 7.758694 (lr=1.1599e-03) (hash(x)=6190642)
550
+ 526 train 7.864784 (lr=1.1573e-03) (hash(x)=7431177)
551
+ 527 train 7.742682 (lr=1.1548e-03) (hash(x)=6112215)
552
+ 528 train 7.855534 (lr=1.1522e-03) (hash(x)=6785874)
553
+ 529 train 7.603073 (lr=1.1496e-03) (hash(x)=7314989)
554
+ 530 train 7.824694 (lr=1.1470e-03) (hash(x)=6222614)
555
+ 531 train 7.964776 (lr=1.1443e-03) (hash(x)=8353143)
556
+ 532 train 8.078378 (lr=1.1417e-03) (hash(x)=6752498)
557
+ 533 train 7.823925 (lr=1.1391e-03) (hash(x)=5912570)
558
+ 534 train 7.952271 (lr=1.1365e-03) (hash(x)=5621785)
559
+ 535 train 7.842968 (lr=1.1338e-03) (hash(x)=5915361)
560
+ 536 train 7.730354 (lr=1.1312e-03) (hash(x)=6853672)
561
+ 537 train 7.679391 (lr=1.1285e-03) (hash(x)=6369494)
562
+ 538 train 7.796976 (lr=1.1259e-03) (hash(x)=6039652)
563
+ 539 train 7.967696 (lr=1.1232e-03) (hash(x)=6254885)
564
+ 540 train 7.774942 (lr=1.1205e-03) (hash(x)=5829860)
565
+ 541 train 7.698615 (lr=1.1179e-03) (hash(x)=5425120)
566
+ 542 train 7.458246 (lr=1.1152e-03) (hash(x)=5209746)
567
+ 543 train 8.052234 (lr=1.1125e-03) (hash(x)=5771588)
568
+ 544 train 7.761955 (lr=1.1098e-03) (hash(x)=8298790)
569
+ 545 train 7.852404 (lr=1.1071e-03) (hash(x)=6763967)
570
+ 546 train 7.833679 (lr=1.1044e-03) (hash(x)=4882397)
571
+ 547 train 7.598743 (lr=1.1017e-03) (hash(x)=5561507)
572
+ 548 train 7.557584 (lr=1.0990e-03) (hash(x)=5762755)
573
+ 549 train 7.723754 (lr=1.0963e-03) (hash(x)=7081151)
574
+ 550 val loss 7.7346
575
+ 550 val perplexity 2286.0605
576
+ 550 train 7.588625 (lr=1.0936e-03) (hash(x)=5047520)
577
+ 551 train 7.753310 (lr=1.0908e-03) (hash(x)=5688829)
578
+ 552 train 7.636267 (lr=1.0881e-03) (hash(x)=7008925)
579
+ 553 train 7.752542 (lr=1.0854e-03) (hash(x)=7821554)
580
+ 554 train 7.837445 (lr=1.0826e-03) (hash(x)=6469035)
581
+ 555 train 7.587337 (lr=1.0799e-03) (hash(x)=5371951)
582
+ 556 train 7.860281 (lr=1.0771e-03) (hash(x)=6436050)
583
+ 557 train 7.813841 (lr=1.0744e-03) (hash(x)=8120149)
584
+ 558 train 8.061674 (lr=1.0716e-03) (hash(x)=6189935)
585
+ 559 train 7.733269 (lr=1.0688e-03) (hash(x)=5443305)
586
+ 560 train 7.827634 (lr=1.0661e-03) (hash(x)=6294312)
587
+ 561 train 7.453705 (lr=1.0633e-03) (hash(x)=7066030)
588
+ 562 train 7.860213 (lr=1.0605e-03) (hash(x)=7744989)
589
+ 563 train 8.055133 (lr=1.0577e-03) (hash(x)=5016757)
590
+ 564 train 7.822429 (lr=1.0549e-03) (hash(x)=8001284)
591
+ 565 train 7.770813 (lr=1.0521e-03) (hash(x)=6296163)
592
+ 566 train 7.856140 (lr=1.0493e-03) (hash(x)=6027688)
593
+ 567 train 7.774981 (lr=1.0465e-03) (hash(x)=6901933)
594
+ 568 train 7.639420 (lr=1.0437e-03) (hash(x)=5124305)
595
+ 569 train 7.789966 (lr=1.0409e-03) (hash(x)=8056633)
596
+ 570 train 7.861121 (lr=1.0381e-03) (hash(x)=6677566)
597
+ 571 train 7.885187 (lr=1.0353e-03) (hash(x)=6019085)
598
+ 572 train 7.571488 (lr=1.0325e-03) (hash(x)=5924495)
599
+ 573 train 7.697712 (lr=1.0296e-03) (hash(x)=7003893)
600
+ 574 train 7.680776 (lr=1.0268e-03) (hash(x)=5563075)
601
+ 575 train 7.478675 (lr=1.0240e-03) (hash(x)=4294425)
602
+ 576 train 7.650817 (lr=1.0211e-03) (hash(x)=5677870)
603
+ 577 train 7.569826 (lr=1.0183e-03) (hash(x)=6568540)
604
+ 578 train 7.831289 (lr=1.0154e-03) (hash(x)=7473394)
605
+ 579 train 7.851219 (lr=1.0126e-03) (hash(x)=6095229)
606
+ 580 train 7.844685 (lr=1.0097e-03) (hash(x)=6946168)
607
+ 581 train 7.691648 (lr=1.0069e-03) (hash(x)=5512150)
608
+ 582 train 7.308927 (lr=1.0040e-03) (hash(x)=4380268)
609
+ 583 train 7.373047 (lr=1.0011e-03) (hash(x)=4363941)
610
+ 584 train 7.546951 (lr=9.9827e-04) (hash(x)=4836009)
611
+ 585 train 7.796813 (lr=9.9540e-04) (hash(x)=6457783)
612
+ 586 train 7.678595 (lr=9.9252e-04) (hash(x)=6378784)
613
+ 587 train 7.556292 (lr=9.8964e-04) (hash(x)=4693798)
614
+ 588 train 7.870188 (lr=9.8676e-04) (hash(x)=9193310)
615
+ 589 train 8.534296 (lr=9.8387e-04) (hash(x)=8759806)
616
+ 590 train 8.807196 (lr=9.8099e-04) (hash(x)=9176140)
617
+ 591 train 7.692350 (lr=9.7810e-04) (hash(x)=5486335)
618
+ 592 train 7.946766 (lr=9.7520e-04) (hash(x)=7868436)
619
+ 593 train 7.808455 (lr=9.7230e-04) (hash(x)=6964993)
620
+ 594 train 8.057377 (lr=9.6940e-04) (hash(x)=6741270)
621
+ 595 train 8.288191 (lr=9.6650e-04) (hash(x)=7907450)
622
+ 596 train 8.068618 (lr=9.6360e-04) (hash(x)=6443452)
623
+ 597 train 7.729785 (lr=9.6069e-04) (hash(x)=5236642)
624
+ 598 train 7.880871 (lr=9.5778e-04) (hash(x)=5604795)
625
+ 599 train 7.803565 (lr=9.5486e-04) (hash(x)=7295165)
626
+ 600 val loss 7.7423
627
+ 600 val perplexity 2303.7397
628
+ 600 train 7.798344 (lr=9.5195e-04) (hash(x)=6034549)
629
+ 601 train 7.774276 (lr=9.4903e-04) (hash(x)=4913602)
630
+ 602 train 7.724575 (lr=9.4611e-04) (hash(x)=6271294)
631
+ 603 train 7.894084 (lr=9.4319e-04) (hash(x)=6814026)
632
+ 604 train 7.788796 (lr=9.4026e-04) (hash(x)=6848040)
633
+ 605 train 8.143744 (lr=9.3733e-04) (hash(x)=7042051)
634
+ 606 train 7.991267 (lr=9.3440e-04) (hash(x)=6481002)
635
+ 607 train 7.641241 (lr=9.3147e-04) (hash(x)=6267424)
636
+ 608 train 8.024668 (lr=9.2854e-04) (hash(x)=7665306)
637
+ 609 train 7.880799 (lr=9.2560e-04) (hash(x)=5614727)
638
+ 610 train 7.894226 (lr=9.2267e-04) (hash(x)=7039197)
639
+ 611 train 7.925148 (lr=9.1973e-04) (hash(x)=8086437)
640
+ 612 train 7.997160 (lr=9.1678e-04) (hash(x)=6993846)
641
+ 613 train 8.056112 (lr=9.1384e-04) (hash(x)=5900143)
642
+ 614 train 7.911551 (lr=9.1090e-04) (hash(x)=4841318)
643
+ 615 train 7.868651 (lr=9.0795e-04) (hash(x)=5270452)
644
+ 616 train 7.774522 (lr=9.0500e-04) (hash(x)=5955026)
645
+ 617 train 7.881798 (lr=9.0205e-04) (hash(x)=8617707)
646
+ 618 train 7.706151 (lr=8.9910e-04) (hash(x)=5159401)
647
+ 619 train 7.712880 (lr=8.9615e-04) (hash(x)=6420820)
648
+ 620 train 7.681422 (lr=8.9319e-04) (hash(x)=6628863)
649
+ 621 train 7.444349 (lr=8.9024e-04) (hash(x)=6215930)
650
+ 622 train 7.771464 (lr=8.8728e-04) (hash(x)=5248645)
651
+ 623 train 7.657475 (lr=8.8432e-04) (hash(x)=6305297)
652
+ 624 train 8.155793 (lr=8.8136e-04) (hash(x)=6989704)
653
+ 625 train 7.777881 (lr=8.7840e-04) (hash(x)=4026682)
654
+ 626 train 7.634096 (lr=8.7544e-04) (hash(x)=6367688)
655
+ 627 train 7.707991 (lr=8.7248e-04) (hash(x)=7889849)
656
+ 628 train 7.883907 (lr=8.6952e-04) (hash(x)=7334719)
657
+ 629 train 7.948597 (lr=8.6655e-04) (hash(x)=6141721)
658
+ 630 train 8.701369 (lr=8.6359e-04) (hash(x)=5056572)
659
+ 631 train 8.065869 (lr=8.6062e-04) (hash(x)=6040077)
660
+ 632 train 8.181516 (lr=8.5766e-04) (hash(x)=6866651)
661
+ 633 train 8.443906 (lr=8.5469e-04) (hash(x)=6155449)
662
+ 634 train 8.041350 (lr=8.5172e-04) (hash(x)=6690174)
663
+ 635 train 7.793467 (lr=8.4876e-04) (hash(x)=5652497)
664
+ 636 train 7.805273 (lr=8.4579e-04) (hash(x)=5931050)
665
+ 637 train 7.934156 (lr=8.4282e-04) (hash(x)=6961314)
666
+ 638 train 7.764378 (lr=8.3985e-04) (hash(x)=5973636)
667
+ 639 train 7.746509 (lr=8.3688e-04) (hash(x)=7130251)
668
+ 640 train 7.452876 (lr=8.3391e-04) (hash(x)=7503390)
669
+ 641 train 7.452590 (lr=8.3094e-04) (hash(x)=5927461)
670
+ 642 train 7.734743 (lr=8.2797e-04) (hash(x)=6196741)
671
+ 643 train 7.709318 (lr=8.2500e-04) (hash(x)=6610177)
672
+ 644 train 7.650116 (lr=8.2203e-04) (hash(x)=6635147)
673
+ 645 train 7.972590 (lr=8.1906e-04) (hash(x)=7277580)
674
+ 646 train 7.571792 (lr=8.1609e-04) (hash(x)=5050080)
675
+ 647 train 7.618207 (lr=8.1312e-04) (hash(x)=6508350)
676
+ 648 train 7.855141 (lr=8.1015e-04) (hash(x)=5276338)
677
+ 649 train 8.018814 (lr=8.0718e-04) (hash(x)=6536034)
678
+ 650 val loss 7.7466
679
+ 650 val perplexity 2313.6189
680
+ 650 train 7.802743 (lr=8.0421e-04) (hash(x)=6944772)
681
+ 651 train 7.860034 (lr=8.0124e-04) (hash(x)=6994983)
682
+ 652 train 7.758874 (lr=7.9828e-04) (hash(x)=7172017)
683
+ 653 train 7.866012 (lr=7.9531e-04) (hash(x)=8700721)
684
+ 654 train 7.870422 (lr=7.9234e-04) (hash(x)=6774360)
685
+ 655 train 7.712948 (lr=7.8938e-04) (hash(x)=5859576)
686
+ 656 train 7.322742 (lr=7.8641e-04) (hash(x)=5899275)
687
+ 657 train 7.261871 (lr=7.8345e-04) (hash(x)=5264962)
688
+ 658 train 7.066477 (lr=7.8048e-04) (hash(x)=5679861)
689
+ 659 train 7.560245 (lr=7.7752e-04) (hash(x)=5487065)
690
+ 660 train 7.374225 (lr=7.7456e-04) (hash(x)=4239476)
691
+ 661 train 7.652552 (lr=7.7160e-04) (hash(x)=5731624)
692
+ 662 train 7.759950 (lr=7.6864e-04) (hash(x)=5883465)
693
+ 663 train 7.577617 (lr=7.6568e-04) (hash(x)=4892065)
694
+ 664 train 7.400456 (lr=7.6272e-04) (hash(x)=5858782)
695
+ 665 train 7.672049 (lr=7.5976e-04) (hash(x)=5489496)
696
+ 666 train 7.357478 (lr=7.5681e-04) (hash(x)=4485195)
697
+ 667 train 7.425244 (lr=7.5385e-04) (hash(x)=4933674)
698
+ 668 train 7.688301 (lr=7.5090e-04) (hash(x)=5746292)
699
+ 669 train 7.681175 (lr=7.4795e-04) (hash(x)=7021003)
700
+ 670 train 7.856562 (lr=7.4500e-04) (hash(x)=5876710)
701
+ 671 train 7.884448 (lr=7.4205e-04) (hash(x)=7317289)
702
+ 672 train 7.997550 (lr=7.3910e-04) (hash(x)=5598226)
703
+ 673 train 7.801004 (lr=7.3616e-04) (hash(x)=7869305)
704
+ 674 train 7.885681 (lr=7.3322e-04) (hash(x)=6611408)
705
+ 675 train 7.829226 (lr=7.3027e-04) (hash(x)=6811522)
706
+ 676 train 7.820507 (lr=7.2733e-04) (hash(x)=6704714)
707
+ 677 train 7.686789 (lr=7.2440e-04) (hash(x)=6601423)
708
+ 678 train 7.905441 (lr=7.2146e-04) (hash(x)=6726071)
709
+ 679 train 7.313822 (lr=7.1853e-04) (hash(x)=5510218)
710
+ 680 train 8.137687 (lr=7.1560e-04) (hash(x)=7950952)
711
+ 681 train 7.829082 (lr=7.1267e-04) (hash(x)=7180298)
712
+ 682 train 7.667292 (lr=7.0974e-04) (hash(x)=6068813)
713
+ 683 train 7.769867 (lr=7.0681e-04) (hash(x)=7304235)
714
+ 684 train 7.857284 (lr=7.0389e-04) (hash(x)=7441806)
715
+ 685 train 7.918674 (lr=7.0097e-04) (hash(x)=8111920)
716
+ 686 train 8.539438 (lr=6.9805e-04) (hash(x)=6222783)
717
+ 687 train 7.774892 (lr=6.9514e-04) (hash(x)=6752265)
718
+ 688 train 7.585413 (lr=6.9222e-04) (hash(x)=6147634)
719
+ 689 train 7.629037 (lr=6.8931e-04) (hash(x)=6788720)
720
+ 690 train 7.815164 (lr=6.8640e-04) (hash(x)=6413518)
721
+ 691 train 7.671808 (lr=6.8350e-04) (hash(x)=5994476)
722
+ 692 train 7.595612 (lr=6.8060e-04) (hash(x)=5462082)
723
+ 693 train 7.601749 (lr=6.7770e-04) (hash(x)=5862533)
724
+ 694 train 7.753817 (lr=6.7480e-04) (hash(x)=7132796)
725
+ 695 train 7.651483 (lr=6.7190e-04) (hash(x)=6530867)
726
+ 696 train 7.744561 (lr=6.6901e-04) (hash(x)=7932207)
727
+ 697 train 7.673405 (lr=6.6613e-04) (hash(x)=5914734)
728
+ 698 train 7.838843 (lr=6.6324e-04) (hash(x)=6361594)
729
+ 699 train 7.790298 (lr=6.6036e-04) (hash(x)=5746260)
730
+ 700 val loss 7.7090
731
+ 700 val perplexity 2228.3621
732
+ 700 train 7.752214 (lr=6.5748e-04) (hash(x)=7065743)
733
+ 701 train 7.674419 (lr=6.5460e-04) (hash(x)=6856097)
734
+ 702 train 7.762239 (lr=6.5173e-04) (hash(x)=7588648)
735
+ 703 train 7.664467 (lr=6.4886e-04) (hash(x)=5790078)
736
+ 704 train 7.687239 (lr=6.4600e-04) (hash(x)=6069422)
737
+ 705 train 7.999065 (lr=6.4314e-04) (hash(x)=5805002)
738
+ 706 train 7.640356 (lr=6.4028e-04) (hash(x)=5344711)
739
+ 707 train 7.519300 (lr=6.3742e-04) (hash(x)=6430135)
740
+ 708 train 7.682941 (lr=6.3457e-04) (hash(x)=6317763)
741
+ 709 train 7.574615 (lr=6.3172e-04) (hash(x)=6156715)
742
+ 710 train 7.924071 (lr=6.2888e-04) (hash(x)=6321494)
743
+ 711 train 8.020352 (lr=6.2604e-04) (hash(x)=7614023)
744
+ 712 train 7.928492 (lr=6.2320e-04) (hash(x)=6740380)
745
+ 713 train 7.771074 (lr=6.2037e-04) (hash(x)=4861744)
746
+ 714 train 7.656484 (lr=6.1754e-04) (hash(x)=6542179)
747
+ 715 train 7.709598 (lr=6.1472e-04) (hash(x)=5244861)
748
+ 716 train 7.800132 (lr=6.1190e-04) (hash(x)=7306636)
749
+ 717 train 7.680935 (lr=6.0908e-04) (hash(x)=7163697)
750
+ 718 train 7.874273 (lr=6.0627e-04) (hash(x)=6421642)
751
+ 719 train 7.627335 (lr=6.0346e-04) (hash(x)=5245146)
752
+ 720 train 7.705989 (lr=6.0066e-04) (hash(x)=6046027)
753
+ 721 train 7.552551 (lr=5.9786e-04) (hash(x)=6153866)
754
+ 722 train 7.661225 (lr=5.9507e-04) (hash(x)=5827481)
755
+ 723 train 7.642850 (lr=5.9228e-04) (hash(x)=6415565)
756
+ 724 train 7.767719 (lr=5.8949e-04) (hash(x)=6409570)
757
+ 725 train 8.186916 (lr=5.8671e-04) (hash(x)=7835853)
758
+ 726 train 7.344842 (lr=5.8393e-04) (hash(x)=4827589)
759
+ 727 train 7.464017 (lr=5.8116e-04) (hash(x)=5786972)
760
+ 728 train 7.880276 (lr=5.7839e-04) (hash(x)=6736612)
761
+ 729 train 7.743939 (lr=5.7563e-04) (hash(x)=6733560)
762
+ 730 train 7.900611 (lr=5.7288e-04) (hash(x)=4877208)
763
+ 731 train 7.449635 (lr=5.7012e-04) (hash(x)=6131703)
764
+ 732 train 7.552754 (lr=5.6737e-04) (hash(x)=6533769)
765
+ 733 train 7.499313 (lr=5.6463e-04) (hash(x)=6001331)
766
+ 734 train 8.020337 (lr=5.6189e-04) (hash(x)=10602643)
767
+ 735 train 7.980137 (lr=5.5916e-04) (hash(x)=6346459)
768
+ 736 train 7.675183 (lr=5.5643e-04) (hash(x)=6728215)
769
+ 737 train 7.975532 (lr=5.5371e-04) (hash(x)=8943770)
770
+ 738 train 7.829362 (lr=5.5100e-04) (hash(x)=7141912)
771
+ 739 train 7.778615 (lr=5.4828e-04) (hash(x)=6504131)
772
+ 740 train 7.604162 (lr=5.4558e-04) (hash(x)=6461667)
773
+ 741 train 7.800959 (lr=5.4288e-04) (hash(x)=5869339)
774
+ 742 train 8.153522 (lr=5.4018e-04) (hash(x)=7948065)
775
+ 743 train 7.444650 (lr=5.3749e-04) (hash(x)=5209234)
776
+ 744 train 7.591177 (lr=5.3481e-04) (hash(x)=6372244)
777
+ 745 train 7.732340 (lr=5.3213e-04) (hash(x)=7678937)
778
+ 746 train 7.667024 (lr=5.2946e-04) (hash(x)=6519438)
779
+ 747 train 7.750811 (lr=5.2679e-04) (hash(x)=6163272)
780
+ 748 train 7.816018 (lr=5.2413e-04) (hash(x)=7025209)
781
+ 749 train 7.670505 (lr=5.2147e-04) (hash(x)=5774172)
782
+ 750 val loss 7.7031
783
+ 750 val perplexity 2215.1321
784
+ 750 train 7.765107 (lr=5.1882e-04) (hash(x)=5327301)
785
+ 751 train 7.564687 (lr=5.1618e-04) (hash(x)=5676768)
786
+ 752 train 7.245097 (lr=5.1354e-04) (hash(x)=4823304)
787
+ 753 train 7.378439 (lr=5.1091e-04) (hash(x)=6443895)
788
+ 754 train 8.171389 (lr=5.0828e-04) (hash(x)=8032890)
789
+ 755 train 8.196582 (lr=5.0566e-04) (hash(x)=6090561)
790
+ 756 train 7.585968 (lr=5.0305e-04) (hash(x)=4514262)
791
+ 757 train 8.070734 (lr=5.0044e-04) (hash(x)=7788628)
792
+ 758 train 7.975684 (lr=4.9784e-04) (hash(x)=7239456)
793
+ 759 train 7.995500 (lr=4.9525e-04) (hash(x)=5863092)
794
+ 760 train 7.698906 (lr=4.9266e-04) (hash(x)=5572628)
795
+ 761 train 7.790657 (lr=4.9008e-04) (hash(x)=7141354)
796
+ 762 train 8.013627 (lr=4.8750e-04) (hash(x)=7272638)
797
+ 763 train 7.833672 (lr=4.8493e-04) (hash(x)=7201312)
798
+ 764 train 7.884776 (lr=4.8237e-04) (hash(x)=6015388)
799
+ 765 train 7.937025 (lr=4.7981e-04) (hash(x)=6414278)
800
+ 766 train 7.676180 (lr=4.7726e-04) (hash(x)=7137466)
801
+ 767 train 8.276816 (lr=4.7472e-04) (hash(x)=7019489)
802
+ 768 train 7.827485 (lr=4.7219e-04) (hash(x)=7233453)
803
+ 769 train 7.574904 (lr=4.6966e-04) (hash(x)=7914626)
804
+ 770 train 7.648393 (lr=4.6714e-04) (hash(x)=5764080)
805
+ 771 train 7.706488 (lr=4.6462e-04) (hash(x)=6225608)
806
+ 772 train 7.757379 (lr=4.6211e-04) (hash(x)=8097255)
807
+ 773 train 7.751238 (lr=4.5961e-04) (hash(x)=5998078)
808
+ 774 train 7.891580 (lr=4.5712e-04) (hash(x)=5416254)
809
+ 775 train 7.535143 (lr=4.5463e-04) (hash(x)=5483019)
810
+ 776 train 7.400356 (lr=4.5215e-04) (hash(x)=4702208)
811
+ 777 train 7.713267 (lr=4.4968e-04) (hash(x)=5911642)
812
+ 778 train 7.760357 (lr=4.4722e-04) (hash(x)=6132487)
813
+ 779 train 7.916764 (lr=4.4476e-04) (hash(x)=5903258)
814
+ 780 train 7.787920 (lr=4.4231e-04) (hash(x)=7915382)
815
+ 781 train 7.809634 (lr=4.3987e-04) (hash(x)=5632006)
816
+ 782 train 7.619497 (lr=4.3743e-04) (hash(x)=6518211)
817
+ 783 train 7.704856 (lr=4.3500e-04) (hash(x)=5968716)
818
+ 784 train 7.783559 (lr=4.3258e-04) (hash(x)=7344525)
819
+ 785 train 7.707767 (lr=4.3017e-04) (hash(x)=6401968)
820
+ 786 train 7.626086 (lr=4.2776e-04) (hash(x)=6276127)
821
+ 787 train 7.776628 (lr=4.2537e-04) (hash(x)=5778017)
822
+ 788 train 7.252889 (lr=4.2298e-04) (hash(x)=5387306)
823
+ 789 train 7.209515 (lr=4.2060e-04) (hash(x)=5772567)
824
+ 790 train 7.492179 (lr=4.1822e-04) (hash(x)=6383748)
825
+ 791 train 7.791690 (lr=4.1586e-04) (hash(x)=7780194)
826
+ 792 train 7.684799 (lr=4.1350e-04) (hash(x)=7119030)
827
+ 793 train 7.796825 (lr=4.1115e-04) (hash(x)=6424771)
828
+ 794 train 7.596975 (lr=4.0880e-04) (hash(x)=6540151)
829
+ 795 train 7.594086 (lr=4.0647e-04) (hash(x)=6140998)
830
+ 796 train 7.667497 (lr=4.0414e-04) (hash(x)=6208271)
831
+ 797 train 7.498034 (lr=4.0183e-04) (hash(x)=7859566)
832
+ 798 train 8.068578 (lr=3.9952e-04) (hash(x)=7064477)
833
+ 799 train 7.357923 (lr=3.9722e-04) (hash(x)=3784321)
834
+ 800 val loss 7.6832
835
+ 800 val perplexity 2171.5957
836
+ 800 train 7.643449 (lr=3.9492e-04) (hash(x)=4472758)
837
+ 801 train 7.587022 (lr=3.9264e-04) (hash(x)=5557891)
838
+ 802 train 7.736720 (lr=3.9036e-04) (hash(x)=7969325)
839
+ 803 train 7.734771 (lr=3.8809e-04) (hash(x)=5860821)
840
+ 804 train 7.757414 (lr=3.8583e-04) (hash(x)=6750848)
841
+ 805 train 7.702592 (lr=3.8358e-04) (hash(x)=5674826)
842
+ 806 train 7.487648 (lr=3.8134e-04) (hash(x)=5529163)
843
+ 807 train 7.909180 (lr=3.7910e-04) (hash(x)=7774109)
844
+ 808 train 7.695702 (lr=3.7688e-04) (hash(x)=6762509)
845
+ 809 train 7.795451 (lr=3.7466e-04) (hash(x)=5466449)
846
+ 810 train 7.671218 (lr=3.7245e-04) (hash(x)=7046935)
847
+ 811 train 8.476582 (lr=3.7026e-04) (hash(x)=7997664)
848
+ 812 train 7.833663 (lr=3.6806e-04) (hash(x)=6882594)
849
+ 813 train 7.996621 (lr=3.6588e-04) (hash(x)=7006517)
850
+ 814 train 7.877766 (lr=3.6371e-04) (hash(x)=6832530)
851
+ 815 train 7.688039 (lr=3.6155e-04) (hash(x)=6576749)
852
+ 816 train 7.788406 (lr=3.5939e-04) (hash(x)=8276629)
853
+ 817 train 7.607869 (lr=3.5725e-04) (hash(x)=6896198)
854
+ 818 train 7.825890 (lr=3.5511e-04) (hash(x)=5829252)
855
+ 819 train 8.003388 (lr=3.5298e-04) (hash(x)=7266655)
856
+ 820 train 7.674016 (lr=3.5086e-04) (hash(x)=6015975)
857
+ 821 train 7.667737 (lr=3.4875e-04) (hash(x)=5696696)
858
+ 822 train 7.672213 (lr=3.4665e-04) (hash(x)=5411666)
859
+ 823 train 7.940804 (lr=3.4456e-04) (hash(x)=7072404)
860
+ 824 train 7.497922 (lr=3.4248e-04) (hash(x)=4910095)
861
+ 825 train 7.663493 (lr=3.4041e-04) (hash(x)=6590657)
862
+ 826 train 7.635783 (lr=3.3835e-04) (hash(x)=7665574)
863
+ 827 train 7.674464 (lr=3.3629e-04) (hash(x)=6626459)
864
+ 828 train 7.727785 (lr=3.3425e-04) (hash(x)=6731971)
865
+ 829 train 7.717244 (lr=3.3221e-04) (hash(x)=5644977)
866
+ 830 train 7.674600 (lr=3.3019e-04) (hash(x)=6238654)
867
+ 831 train 7.577949 (lr=3.2817e-04) (hash(x)=6556025)
868
+ 832 train 7.887453 (lr=3.2617e-04) (hash(x)=7175563)
869
+ 833 train 7.790633 (lr=3.2417e-04) (hash(x)=6407492)
870
+ 834 train 7.553496 (lr=3.2219e-04) (hash(x)=4934335)
871
+ 835 train 7.583318 (lr=3.2021e-04) (hash(x)=5942129)
872
+ 836 train 7.639629 (lr=3.1824e-04) (hash(x)=5995643)
873
+ 837 train 7.696274 (lr=3.1629e-04) (hash(x)=6012163)
874
+ 838 train 7.795301 (lr=3.1434e-04) (hash(x)=8464831)
875
+ 839 train 8.035112 (lr=3.1240e-04) (hash(x)=7325027)
876
+ 840 train 7.897722 (lr=3.1047e-04) (hash(x)=6785865)
877
+ 841 train 7.808799 (lr=3.0856e-04) (hash(x)=4425520)
878
+ 842 train 7.593394 (lr=3.0665e-04) (hash(x)=5388267)
879
+ 843 train 7.866083 (lr=3.0475e-04) (hash(x)=7322467)
880
+ 844 train 7.677780 (lr=3.0286e-04) (hash(x)=6681766)
881
+ 845 train 7.689533 (lr=3.0099e-04) (hash(x)=7482800)
882
+ 846 train 7.656269 (lr=2.9912e-04) (hash(x)=5554493)
883
+ 847 train 7.849956 (lr=2.9726e-04) (hash(x)=6373412)
884
+ 848 train 7.271985 (lr=2.9542e-04) (hash(x)=5117517)
885
+ 849 train 7.544342 (lr=2.9358e-04) (hash(x)=6981426)
886
+ 850 val loss 7.6707
887
+ 850 val perplexity 2144.5098
888
+ 850 train 7.494731 (lr=2.9175e-04) (hash(x)=6886188)
889
+ 851 train 7.700901 (lr=2.8994e-04) (hash(x)=7332255)
890
+ 852 train 7.724967 (lr=2.8813e-04) (hash(x)=6172042)
891
+ 853 train 7.865021 (lr=2.8634e-04) (hash(x)=5930894)
892
+ 854 train 7.939175 (lr=2.8455e-04) (hash(x)=7448958)
893
+ 855 train 7.697061 (lr=2.8278e-04) (hash(x)=5262868)
894
+ 856 train 7.593933 (lr=2.8102e-04) (hash(x)=5558427)
895
+ 857 train 7.472877 (lr=2.7926e-04) (hash(x)=5585769)
896
+ 858 train 7.354484 (lr=2.7752e-04) (hash(x)=5838081)
897
+ 859 train 7.521293 (lr=2.7579e-04) (hash(x)=5688247)
898
+ 860 train 7.678061 (lr=2.7407e-04) (hash(x)=5162020)
899
+ 861 train 7.860267 (lr=2.7236e-04) (hash(x)=7462079)
900
+ 862 train 7.645485 (lr=2.7066e-04) (hash(x)=6516108)
901
+ 863 train 7.469573 (lr=2.6897e-04) (hash(x)=8055563)
902
+ 864 train 7.645856 (lr=2.6729e-04) (hash(x)=6271901)
903
+ 865 train 7.686551 (lr=2.6562e-04) (hash(x)=6221701)
904
+ 866 train 7.715338 (lr=2.6396e-04) (hash(x)=5772861)
905
+ 867 train 7.699280 (lr=2.6232e-04) (hash(x)=5352405)
906
+ 868 train 7.710816 (lr=2.6068e-04) (hash(x)=6111630)
907
+ 869 train 7.780591 (lr=2.5906e-04) (hash(x)=6730666)
908
+ 870 train 7.904331 (lr=2.5745e-04) (hash(x)=8698315)
909
+ 871 train 7.596307 (lr=2.5584e-04) (hash(x)=5932790)
910
+ 872 train 8.021507 (lr=2.5425e-04) (hash(x)=8099892)
911
+ 873 train 7.602163 (lr=2.5267e-04) (hash(x)=6223114)
912
+ 874 train 7.695676 (lr=2.5110e-04) (hash(x)=5798363)
913
+ 875 train 7.588592 (lr=2.4955e-04) (hash(x)=6249312)
914
+ 876 train 7.739114 (lr=2.4800e-04) (hash(x)=6929692)
915
+ 877 train 7.832918 (lr=2.4646e-04) (hash(x)=7242827)
916
+ 878 train 7.602659 (lr=2.4494e-04) (hash(x)=6332123)
917
+ 879 train 7.473407 (lr=2.4343e-04) (hash(x)=5680154)
918
+ 880 train 7.865421 (lr=2.4192e-04) (hash(x)=6352331)
919
+ 881 train 7.685202 (lr=2.4043e-04) (hash(x)=6332332)
920
+ 882 train 7.758831 (lr=2.3895e-04) (hash(x)=6180674)
921
+ 883 train 7.696301 (lr=2.3749e-04) (hash(x)=8156280)
922
+ 884 train 7.721699 (lr=2.3603e-04) (hash(x)=5874247)
923
+ 885 train 7.665419 (lr=2.3458e-04) (hash(x)=6659781)
924
+ 886 train 7.776476 (lr=2.3315e-04) (hash(x)=5780147)
925
+ 887 train 7.595325 (lr=2.3173e-04) (hash(x)=5914217)
926
+ 888 train 7.466845 (lr=2.3032e-04) (hash(x)=5411762)
927
+ 889 train 7.725003 (lr=2.2892e-04) (hash(x)=6714110)
928
+ 890 train 7.411947 (lr=2.2753e-04) (hash(x)=5999685)
929
+ 891 train 7.599505 (lr=2.2615e-04) (hash(x)=7120215)
930
+ 892 train 7.857563 (lr=2.2479e-04) (hash(x)=6546587)
931
+ 893 train 7.658692 (lr=2.2344e-04) (hash(x)=6593413)
932
+ 894 train 7.736757 (lr=2.2209e-04) (hash(x)=6369927)
933
+ 895 train 7.644802 (lr=2.2076e-04) (hash(x)=6424128)
934
+ 896 train 7.611937 (lr=2.1945e-04) (hash(x)=5736158)
935
+ 897 train 7.422611 (lr=2.1814e-04) (hash(x)=5903043)
936
+ 898 train 7.499266 (lr=2.1685e-04) (hash(x)=4419128)
937
+ 899 train 7.306860 (lr=2.1556e-04) (hash(x)=4390027)
938
+ 900 val loss 7.6569
939
+ 900 val perplexity 2115.1245
940
+ 900 train 7.580153 (lr=2.1429e-04) (hash(x)=6728135)
941
+ 901 train 7.705926 (lr=2.1303e-04) (hash(x)=6945760)
942
+ 902 train 7.673790 (lr=2.1179e-04) (hash(x)=6081534)
943
+ 903 train 7.739121 (lr=2.1055e-04) (hash(x)=7804089)
944
+ 904 train 7.749308 (lr=2.0933e-04) (hash(x)=6225832)
945
+ 905 train 7.776077 (lr=2.0812e-04) (hash(x)=6273417)
946
+ 906 train 7.822191 (lr=2.0692e-04) (hash(x)=7775633)
947
+ 907 train 7.813146 (lr=2.0573e-04) (hash(x)=7130267)
948
+ 908 train 7.592674 (lr=2.0455e-04) (hash(x)=6554076)
949
+ 909 train 7.711257 (lr=2.0339e-04) (hash(x)=6140697)
950
+ 910 train 7.524534 (lr=2.0224e-04) (hash(x)=6128181)
951
+ 911 train 7.585888 (lr=2.0110e-04) (hash(x)=6490149)
952
+ 912 train 7.228277 (lr=1.9997e-04) (hash(x)=5422426)
953
+ 913 train 7.630261 (lr=1.9885e-04) (hash(x)=4847733)
954
+ 914 train 7.657506 (lr=1.9775e-04) (hash(x)=6349747)
955
+ 915 train 7.943142 (lr=1.9666e-04) (hash(x)=9276378)
956
+ 916 train 7.552117 (lr=1.9558e-04) (hash(x)=6869524)
957
+ 917 train 7.594892 (lr=1.9451e-04) (hash(x)=7082053)
958
+ 918 train 7.632938 (lr=1.9346e-04) (hash(x)=7799351)
959
+ 919 train 7.835704 (lr=1.9242e-04) (hash(x)=6283540)
960
+ 920 train 7.607046 (lr=1.9139e-04) (hash(x)=6660860)
961
+ 921 train 7.537795 (lr=1.9037e-04) (hash(x)=5200956)
962
+ 922 train 7.665108 (lr=1.8936e-04) (hash(x)=5653241)
963
+ 923 train 7.787826 (lr=1.8837e-04) (hash(x)=5781680)
964
+ 924 train 7.766808 (lr=1.8739e-04) (hash(x)=6146666)
965
+ 925 train 7.590612 (lr=1.8642e-04) (hash(x)=8652907)
966
+ 926 train 7.395837 (lr=1.8547e-04) (hash(x)=5311546)
967
+ 927 train 7.577305 (lr=1.8452e-04) (hash(x)=5728337)
968
+ 928 train 7.529030 (lr=1.8359e-04) (hash(x)=5413240)
969
+ 929 train 7.635933 (lr=1.8267e-04) (hash(x)=5832401)
970
+ 930 train 7.353016 (lr=1.8176e-04) (hash(x)=7420230)
971
+ 931 train 7.710123 (lr=1.8087e-04) (hash(x)=6278202)
972
+ 932 train 7.635953 (lr=1.7999e-04) (hash(x)=6189873)
973
+ 933 train 7.760139 (lr=1.7912e-04) (hash(x)=7403868)
974
+ 934 train 7.591611 (lr=1.7826e-04) (hash(x)=6867459)
975
+ 935 train 7.789165 (lr=1.7742e-04) (hash(x)=7400272)
976
+ 936 train 7.795142 (lr=1.7659e-04) (hash(x)=7936275)
977
+ 937 train 7.518725 (lr=1.7577e-04) (hash(x)=5943413)
978
+ 938 train 7.744372 (lr=1.7496e-04) (hash(x)=5442166)
979
+ 939 train 7.630917 (lr=1.7417e-04) (hash(x)=5795128)
980
+ 940 train 7.863900 (lr=1.7339e-04) (hash(x)=6831680)
981
+ 941 train 7.516699 (lr=1.7262e-04) (hash(x)=5519980)
982
+ 942 train 7.564527 (lr=1.7186e-04) (hash(x)=6765461)
983
+ 943 train 7.548002 (lr=1.7112e-04) (hash(x)=6473432)
984
+ 944 train 7.476457 (lr=1.7039e-04) (hash(x)=6490395)
985
+ 945 train 7.664333 (lr=1.6967e-04) (hash(x)=5484669)
986
+ 946 train 7.712130 (lr=1.6896e-04) (hash(x)=7278970)
987
+ 947 train 7.645688 (lr=1.6827e-04) (hash(x)=5201502)
988
+ 948 train 7.856702 (lr=1.6759e-04) (hash(x)=7277915)
989
+ 949 train 7.636239 (lr=1.6692e-04) (hash(x)=7227057)
990
+ 950 val loss 7.6439
991
+ 950 val perplexity 2087.7759
992
+ 950 train 7.814282 (lr=1.6627e-04) (hash(x)=8178996)
993
+ 951 train 7.957230 (lr=1.6563e-04) (hash(x)=6922646)
994
+ 952 train 7.483897 (lr=1.6500e-04) (hash(x)=5941772)
995
+ 953 train 7.477944 (lr=1.6438e-04) (hash(x)=6056673)
996
+ 954 train 7.890904 (lr=1.6378e-04) (hash(x)=6912178)
997
+ 955 train 7.378562 (lr=1.6319e-04) (hash(x)=5484641)
998
+ 956 train 7.723104 (lr=1.6261e-04) (hash(x)=5121117)
999
+ 957 train 7.754626 (lr=1.6205e-04) (hash(x)=7100865)
1000
+ 958 train 7.448507 (lr=1.6149e-04) (hash(x)=5334280)
1001
+ 959 train 7.559502 (lr=1.6095e-04) (hash(x)=7277141)
1002
+ 960 train 7.528415 (lr=1.6043e-04) (hash(x)=6401960)
1003
+ 961 train 7.594846 (lr=1.5991e-04) (hash(x)=6407603)
1004
+ 962 train 7.761965 (lr=1.5941e-04) (hash(x)=5786089)
1005
+ 963 train 7.634901 (lr=1.5893e-04) (hash(x)=5604614)
1006
+ 964 train 7.643066 (lr=1.5845e-04) (hash(x)=5454287)
1007
+ 965 train 7.757003 (lr=1.5799e-04) (hash(x)=6200612)
1008
+ 966 train 7.698393 (lr=1.5754e-04) (hash(x)=5691759)
1009
+ 967 train 7.683063 (lr=1.5710e-04) (hash(x)=7443124)
1010
+ 968 train 7.701757 (lr=1.5668e-04) (hash(x)=6711366)
1011
+ 969 train 7.617918 (lr=1.5627e-04) (hash(x)=6728764)
1012
+ 970 train 7.898620 (lr=1.5587e-04) (hash(x)=6870240)
1013
+ 971 train 7.598179 (lr=1.5549e-04) (hash(x)=5601968)
1014
+ 972 train 7.398835 (lr=1.5512e-04) (hash(x)=5311504)
1015
+ 973 train 7.659829 (lr=1.5476e-04) (hash(x)=6877761)
1016
+ 974 train 7.732410 (lr=1.5441e-04) (hash(x)=6457033)
1017
+ 975 train 7.651257 (lr=1.5408e-04) (hash(x)=5632735)
1018
+ 976 train 7.787086 (lr=1.5376e-04) (hash(x)=6024880)
1019
+ 977 train 7.542364 (lr=1.5345e-04) (hash(x)=5658845)
1020
+ 978 train 7.431050 (lr=1.5316e-04) (hash(x)=6089461)
1021
+ 979 train 7.500515 (lr=1.5288e-04) (hash(x)=5594810)
1022
+ 980 train 7.270208 (lr=1.5261e-04) (hash(x)=4667026)
1023
+ 981 train 7.138039 (lr=1.5236e-04) (hash(x)=4738523)
1024
+ 982 train 7.463640 (lr=1.5212e-04) (hash(x)=7451876)
1025
+ 983 train 7.730307 (lr=1.5189e-04) (hash(x)=5881015)
1026
+ 984 train 7.678827 (lr=1.5167e-04) (hash(x)=6255031)
1027
+ 985 train 7.476201 (lr=1.5147e-04) (hash(x)=5263325)
1028
+ 986 train 7.535008 (lr=1.5128e-04) (hash(x)=6654585)
1029
+ 987 train 7.627296 (lr=1.5110e-04) (hash(x)=6865361)
1030
+ 988 train 7.705475 (lr=1.5094e-04) (hash(x)=6978986)
1031
+ 989 train 7.759354 (lr=1.5079e-04) (hash(x)=5644619)
1032
+ 990 train 7.605429 (lr=1.5065e-04) (hash(x)=6551023)
1033
+ 991 train 7.590601 (lr=1.5053e-04) (hash(x)=5809411)
1034
+ 992 train 7.664506 (lr=1.5042e-04) (hash(x)=5501837)
1035
+ 993 train 7.671896 (lr=1.5032e-04) (hash(x)=7044374)
1036
+ 994 train 7.612270 (lr=1.5024e-04) (hash(x)=6526130)
1037
+ 995 train 7.754712 (lr=1.5016e-04) (hash(x)=6293025)
1038
+ 996 train 7.752643 (lr=1.5010e-04) (hash(x)=6295768)
1039
+ 997 train 7.521879 (lr=1.5006e-04) (hash(x)=5189023)
1040
+ 998 train 7.345703 (lr=1.5003e-04) (hash(x)=6255467)
1041
+ 999 val loss 7.6334
1042
+ 999 val perplexity 2065.9690
1043
+ 999 train 7.719183 (lr=1.5001e-04) (hash(x)=7232964)
lr1.5e-3_total_batch_size5120_seq_len128/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78b314fc4cc7361f75bdaeebd31605037a6a50c43e4ce0db644aa6e4a44ffc72
3
+ size 36163010
lr1.5e-3_total_batch_size5120_seq_len128/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6edb7fbda12553b8fabea56f7da425c5868914bd1c7b2afd4b6ba4c6da77ee02
3
+ size 70764358