andrew-healey commited on
Commit
4463fd7
·
verified ·
1 Parent(s): 35e024d

Upload folder using huggingface_hub

Browse files
lr2e-4_total_batch_size10240_seq_len1024/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "proxy_model_sweep/lr2e-4_total_batch_size10240_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": 10240, "protection_head_scaling_factor": 1.0, "protection_head_bias": 0.0, "n_sliced_masks": null, "n_latent_masks": null, "mask_layernorm": false, "residual_attention_masks": false, "compute_base_shapes": false, "base_shapes_savefile": null, "mup": true, "disable_selection": true, "mup_enable_coord_check_logging": false, "max_lr": 0.0002, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "n_embd": 128}
lr2e-4_total_batch_size10240_seq_len1024/dataloader_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b9fd0bfa27ab267155e5ae2bb7e7ccf3640df9cfec6890472ce93f6d886a52e
3
+ size 964
lr2e-4_total_batch_size10240_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 11.752059 (lr=6.9930e-07) (hash(x)=49261872)
5
+ 1 train 11.748100 (lr=1.3986e-06) (hash(x)=57833283)
6
+ 2 train 11.760985 (lr=2.0979e-06) (hash(x)=51469900)
7
+ 3 train 11.741347 (lr=2.7972e-06) (hash(x)=52769534)
8
+ 4 train 11.735526 (lr=3.4965e-06) (hash(x)=48296575)
9
+ 5 train 11.701181 (lr=4.1958e-06) (hash(x)=50373191)
10
+ 6 train 11.728847 (lr=4.8951e-06) (hash(x)=51583947)
11
+ 7 train 11.698127 (lr=5.5944e-06) (hash(x)=48487754)
12
+ 8 train 11.701320 (lr=6.2937e-06) (hash(x)=54502191)
13
+ 9 train 11.686286 (lr=6.9930e-06) (hash(x)=47781406)
14
+ 10 train 11.714013 (lr=7.6923e-06) (hash(x)=53817479)
15
+ 11 train 11.670273 (lr=8.3916e-06) (hash(x)=48658307)
16
+ 12 train 11.680598 (lr=9.0909e-06) (hash(x)=56467437)
17
+ 13 train 11.671237 (lr=9.7902e-06) (hash(x)=42956464)
18
+ 14 train 11.659665 (lr=1.0490e-05) (hash(x)=50465744)
19
+ 15 train 11.602141 (lr=1.1189e-05) (hash(x)=50106566)
20
+ 16 train 11.601291 (lr=1.1888e-05) (hash(x)=57533378)
21
+ 17 train 11.599414 (lr=1.2587e-05) (hash(x)=50260247)
22
+ 18 train 11.592255 (lr=1.3287e-05) (hash(x)=51106171)
23
+ 19 train 11.554175 (lr=1.3986e-05) (hash(x)=45839663)
24
+ 20 train 11.544660 (lr=1.4685e-05) (hash(x)=50479757)
25
+ 21 train 11.486102 (lr=1.5385e-05) (hash(x)=54489985)
26
+ 22 train 11.474479 (lr=1.6084e-05) (hash(x)=52601470)
27
+ 23 train 11.492828 (lr=1.6783e-05) (hash(x)=50770821)
28
+ 24 train 11.454160 (lr=1.7483e-05) (hash(x)=48459756)
29
+ 25 train 11.406890 (lr=1.8182e-05) (hash(x)=49416075)
30
+ 26 train 11.397143 (lr=1.8881e-05) (hash(x)=51191917)
31
+ 27 train 11.369567 (lr=1.9580e-05) (hash(x)=52323135)
32
+ 28 train 11.351745 (lr=2.0280e-05) (hash(x)=47323142)
33
+ 29 train 11.351085 (lr=2.0979e-05) (hash(x)=53922434)
34
+ 30 train 11.297923 (lr=2.1678e-05) (hash(x)=46277237)
35
+ 31 train 11.225369 (lr=2.2378e-05) (hash(x)=50183213)
36
+ 32 train 11.173755 (lr=2.3077e-05) (hash(x)=49300734)
37
+ 33 train 11.171687 (lr=2.3776e-05) (hash(x)=52003225)
38
+ 34 train 11.141995 (lr=2.4476e-05) (hash(x)=52058527)
39
+ 35 train 11.114552 (lr=2.5175e-05) (hash(x)=56845779)
40
+ 36 train 10.986182 (lr=2.5874e-05) (hash(x)=53027784)
41
+ 37 train 11.003158 (lr=2.6573e-05) (hash(x)=55034939)
42
+ 38 train 10.913643 (lr=2.7273e-05) (hash(x)=55634780)
43
+ 39 train 10.931262 (lr=2.7972e-05) (hash(x)=52671072)
44
+ 40 train 10.928313 (lr=2.8671e-05) (hash(x)=46158462)
45
+ 41 train 10.819260 (lr=2.9371e-05) (hash(x)=46979552)
46
+ 42 train 10.883597 (lr=3.0070e-05) (hash(x)=53668921)
47
+ 43 train 10.812106 (lr=3.0769e-05) (hash(x)=52068996)
48
+ 44 train 10.787017 (lr=3.1469e-05) (hash(x)=49328499)
49
+ 45 train 10.595138 (lr=3.2168e-05) (hash(x)=54071193)
50
+ 46 train 10.634130 (lr=3.2867e-05) (hash(x)=50573962)
51
+ 47 train 10.738607 (lr=3.3566e-05) (hash(x)=47640895)
52
+ 48 train 10.775195 (lr=3.4266e-05) (hash(x)=52381297)
53
+ 49 train 10.605321 (lr=3.4965e-05) (hash(x)=46150086)
54
+ 50 val loss 10.5834
55
+ 50 val perplexity 39475.5938
56
+ 50 train 10.597545 (lr=3.5664e-05) (hash(x)=47852060)
57
+ 51 train 10.603583 (lr=3.6364e-05) (hash(x)=45438450)
58
+ 52 train 10.561132 (lr=3.7063e-05) (hash(x)=47030948)
59
+ 53 train 10.425859 (lr=3.7762e-05) (hash(x)=44573591)
60
+ 54 train 10.519282 (lr=3.8462e-05) (hash(x)=49868350)
61
+ 55 train 10.694158 (lr=3.9161e-05) (hash(x)=62300690)
62
+ 56 train 10.502460 (lr=3.9860e-05) (hash(x)=52900359)
63
+ 57 train 10.483752 (lr=4.0559e-05) (hash(x)=50319996)
64
+ 58 train 10.566067 (lr=4.1259e-05) (hash(x)=63972756)
65
+ 59 train 10.442919 (lr=4.1958e-05) (hash(x)=54520776)
66
+ 60 train 10.394032 (lr=4.2657e-05) (hash(x)=41677283)
67
+ 61 train 10.414810 (lr=4.3357e-05) (hash(x)=44855848)
68
+ 62 train 10.355506 (lr=4.4056e-05) (hash(x)=48794108)
69
+ 63 train 10.413976 (lr=4.4755e-05) (hash(x)=54982586)
70
+ 64 train 10.396308 (lr=4.5455e-05) (hash(x)=51251428)
71
+ 65 train 10.437695 (lr=4.6154e-05) (hash(x)=50784472)
72
+ 66 train 10.400275 (lr=4.6853e-05) (hash(x)=52402568)
73
+ 67 train 10.236213 (lr=4.7552e-05) (hash(x)=50188334)
74
+ 68 train 10.411622 (lr=4.8252e-05) (hash(x)=56568050)
75
+ 69 train 10.296094 (lr=4.8951e-05) (hash(x)=50861237)
76
+ 70 train 10.339705 (lr=4.9650e-05) (hash(x)=52123048)
77
+ 71 train 10.382611 (lr=5.0350e-05) (hash(x)=60945351)
78
+ 72 train 10.317969 (lr=5.1049e-05) (hash(x)=52352539)
79
+ 73 train 10.404978 (lr=5.1748e-05) (hash(x)=56694638)
80
+ 74 train 10.296813 (lr=5.2448e-05) (hash(x)=51084620)
81
+ 75 train 10.409731 (lr=5.3147e-05) (hash(x)=58412625)
82
+ 76 train 10.326200 (lr=5.3846e-05) (hash(x)=56756916)
83
+ 77 train 10.319846 (lr=5.4545e-05) (hash(x)=50750450)
84
+ 78 train 10.275464 (lr=5.5245e-05) (hash(x)=49918759)
85
+ 79 train 10.358861 (lr=5.5944e-05) (hash(x)=51819337)
86
+ 80 train 10.186170 (lr=5.6643e-05) (hash(x)=46506057)
87
+ 81 train 10.202340 (lr=5.7343e-05) (hash(x)=48267882)
88
+ 82 train 10.215323 (lr=5.8042e-05) (hash(x)=52004910)
89
+ 83 train 10.159957 (lr=5.8741e-05) (hash(x)=52889428)
90
+ 84 train 10.084986 (lr=5.9441e-05) (hash(x)=44915533)
91
+ 85 train 10.081439 (lr=6.0140e-05) (hash(x)=49536098)
92
+ 86 train 10.165557 (lr=6.0839e-05) (hash(x)=50097275)
93
+ 87 train 10.123878 (lr=6.1538e-05) (hash(x)=50686547)
94
+ 88 train 10.150065 (lr=6.2238e-05) (hash(x)=49463715)
95
+ 89 train 10.107719 (lr=6.2937e-05) (hash(x)=50818630)
96
+ 90 train 10.056193 (lr=6.3636e-05) (hash(x)=47363971)
97
+ 91 train 10.086582 (lr=6.4336e-05) (hash(x)=49697510)
98
+ 92 train 10.030208 (lr=6.5035e-05) (hash(x)=51247734)
99
+ 93 train 10.095222 (lr=6.5734e-05) (hash(x)=53436760)
100
+ 94 train 10.040941 (lr=6.6434e-05) (hash(x)=42551739)
101
+ 95 train 10.060394 (lr=6.7133e-05) (hash(x)=46807632)
102
+ 96 train 10.041682 (lr=6.7832e-05) (hash(x)=52261514)
103
+ 97 train 10.108607 (lr=6.8531e-05) (hash(x)=54016065)
104
+ 98 train 10.148406 (lr=6.9231e-05) (hash(x)=54782473)
105
+ 99 train 10.172426 (lr=6.9930e-05) (hash(x)=57723080)
106
+ 100 val loss 10.0282
107
+ 100 val perplexity 22655.3984
108
+ 100 train 10.102162 (lr=7.0629e-05) (hash(x)=49355093)
109
+ 101 train 9.948083 (lr=7.1329e-05) (hash(x)=53422909)
110
+ 102 train 10.059508 (lr=7.2028e-05) (hash(x)=51486849)
111
+ 103 train 10.066257 (lr=7.2727e-05) (hash(x)=48459663)
112
+ 104 train 10.001030 (lr=7.3427e-05) (hash(x)=51698063)
113
+ 105 train 9.956862 (lr=7.4126e-05) (hash(x)=48794753)
114
+ 106 train 9.953750 (lr=7.4825e-05) (hash(x)=44825422)
115
+ 107 train 9.984085 (lr=7.5524e-05) (hash(x)=47148322)
116
+ 108 train 10.051394 (lr=7.6224e-05) (hash(x)=49279091)
117
+ 109 train 9.972125 (lr=7.6923e-05) (hash(x)=52460055)
118
+ 110 train 9.996131 (lr=7.7622e-05) (hash(x)=46749333)
119
+ 111 train 9.956304 (lr=7.8322e-05) (hash(x)=45337189)
120
+ 112 train 9.854750 (lr=7.9021e-05) (hash(x)=49845142)
121
+ 113 train 9.967847 (lr=7.9720e-05) (hash(x)=50930055)
122
+ 114 train 9.861843 (lr=8.0420e-05) (hash(x)=50684591)
123
+ 115 train 9.929499 (lr=8.1119e-05) (hash(x)=42125704)
124
+ 116 train 9.731269 (lr=8.1818e-05) (hash(x)=41912026)
125
+ 117 train 9.763512 (lr=8.2517e-05) (hash(x)=49174011)
126
+ 118 train 9.859032 (lr=8.3217e-05) (hash(x)=49497097)
127
+ 119 train 9.814177 (lr=8.3916e-05) (hash(x)=50618057)
128
+ 120 train 9.817591 (lr=8.4615e-05) (hash(x)=59453505)
129
+ 121 train 9.895617 (lr=8.5315e-05) (hash(x)=52423338)
130
+ 122 train 9.827598 (lr=8.6014e-05) (hash(x)=53802843)
131
+ 123 train 9.826434 (lr=8.6713e-05) (hash(x)=51754551)
132
+ 124 train 9.896172 (lr=8.7413e-05) (hash(x)=51856340)
133
+ 125 train 9.847727 (lr=8.8112e-05) (hash(x)=47059587)
134
+ 126 train 9.685721 (lr=8.8811e-05) (hash(x)=51793651)
135
+ 127 train 9.709285 (lr=8.9510e-05) (hash(x)=52187435)
136
+ 128 train 9.718776 (lr=9.0210e-05) (hash(x)=54885326)
137
+ 129 train 9.774138 (lr=9.0909e-05) (hash(x)=50919108)
138
+ 130 train 9.797560 (lr=9.1608e-05) (hash(x)=54468731)
139
+ 131 train 9.690082 (lr=9.2308e-05) (hash(x)=51831972)
140
+ 132 train 9.652719 (lr=9.3007e-05) (hash(x)=48609553)
141
+ 133 train 9.662732 (lr=9.3706e-05) (hash(x)=49049513)
142
+ 134 train 9.606483 (lr=9.4406e-05) (hash(x)=51951796)
143
+ 135 train 9.380771 (lr=9.5105e-05) (hash(x)=56443459)
144
+ 136 train 9.643146 (lr=9.5804e-05) (hash(x)=47261375)
145
+ 137 train 9.745523 (lr=9.6503e-05) (hash(x)=54156964)
146
+ 138 train 9.777709 (lr=9.7203e-05) (hash(x)=55621436)
147
+ 139 train 9.673388 (lr=9.7902e-05) (hash(x)=54578790)
148
+ 140 train 9.598248 (lr=9.8601e-05) (hash(x)=46178169)
149
+ 141 train 9.544604 (lr=9.9301e-05) (hash(x)=52813407)
150
+ 142 train 9.559250 (lr=1.0000e-04) (hash(x)=55046177)
151
+ 143 train 9.570002 (lr=1.0070e-04) (hash(x)=45098704)
152
+ 144 train 9.568776 (lr=1.0140e-04) (hash(x)=45895117)
153
+ 145 train 9.525518 (lr=1.0210e-04) (hash(x)=42452493)
154
+ 146 train 9.552066 (lr=1.0280e-04) (hash(x)=50633976)
155
+ 147 train 9.515182 (lr=1.0350e-04) (hash(x)=45655812)
156
+ 148 train 9.452623 (lr=1.0420e-04) (hash(x)=47272687)
157
+ 149 train 9.513901 (lr=1.0490e-04) (hash(x)=45415908)
158
+ 150 val loss 9.4788
159
+ 150 val perplexity 13079.1172
160
+ 150 train 9.833732 (lr=1.0559e-04) (hash(x)=63111643)
161
+ 151 train 9.367700 (lr=1.0629e-04) (hash(x)=49436419)
162
+ 152 train 9.495153 (lr=1.0699e-04) (hash(x)=52633828)
163
+ 153 train 9.523623 (lr=1.0769e-04) (hash(x)=48033921)
164
+ 154 train 9.504397 (lr=1.0839e-04) (hash(x)=54348904)
165
+ 155 train 9.510643 (lr=1.0909e-04) (hash(x)=56987876)
166
+ 156 train 9.425293 (lr=1.0979e-04) (hash(x)=50155518)
167
+ 157 train 9.362266 (lr=1.1049e-04) (hash(x)=48024575)
168
+ 158 train 9.366742 (lr=1.1119e-04) (hash(x)=53059532)
169
+ 159 train 9.311000 (lr=1.1189e-04) (hash(x)=48339908)
170
+ 160 train 9.329988 (lr=1.1259e-04) (hash(x)=48990896)
171
+ 161 train 9.328168 (lr=1.1329e-04) (hash(x)=53135323)
172
+ 162 train 9.255060 (lr=1.1399e-04) (hash(x)=52351650)
173
+ 163 train 9.264178 (lr=1.1469e-04) (hash(x)=46142231)
174
+ 164 train 9.263778 (lr=1.1538e-04) (hash(x)=40473173)
175
+ 165 train 9.197341 (lr=1.1608e-04) (hash(x)=41898441)
176
+ 166 train 9.382715 (lr=1.1678e-04) (hash(x)=53594433)
177
+ 167 train 9.290586 (lr=1.1748e-04) (hash(x)=46263050)
178
+ 168 train 9.279752 (lr=1.1818e-04) (hash(x)=43863543)
179
+ 169 train 9.182536 (lr=1.1888e-04) (hash(x)=37281177)
180
+ 170 train 9.297235 (lr=1.1958e-04) (hash(x)=47276630)
181
+ 171 train 9.206841 (lr=1.2028e-04) (hash(x)=48285373)
182
+ 172 train 9.180015 (lr=1.2098e-04) (hash(x)=53372103)
183
+ 173 train 9.182198 (lr=1.2168e-04) (hash(x)=51190878)
184
+ 174 train 9.148402 (lr=1.2238e-04) (hash(x)=54854560)
185
+ 175 train 9.215400 (lr=1.2308e-04) (hash(x)=56873965)
186
+ 176 train 9.084628 (lr=1.2378e-04) (hash(x)=53533306)
187
+ 177 train 9.118213 (lr=1.2448e-04) (hash(x)=52595463)
188
+ 178 train 9.233507 (lr=1.2517e-04) (hash(x)=49109869)
189
+ 179 train 9.126205 (lr=1.2587e-04) (hash(x)=50421608)
190
+ 180 train 9.041238 (lr=1.2657e-04) (hash(x)=47228528)
191
+ 181 train 8.961076 (lr=1.2727e-04) (hash(x)=45921624)
192
+ 182 train 9.164844 (lr=1.2797e-04) (hash(x)=54444263)
193
+ 183 train 9.068473 (lr=1.2867e-04) (hash(x)=52802882)
194
+ 184 train 8.996111 (lr=1.2937e-04) (hash(x)=49970100)
195
+ 185 train 9.071529 (lr=1.3007e-04) (hash(x)=45185272)
196
+ 186 train 8.963196 (lr=1.3077e-04) (hash(x)=48064569)
197
+ 187 train 9.062699 (lr=1.3147e-04) (hash(x)=49146692)
198
+ 188 train 9.052874 (lr=1.3217e-04) (hash(x)=52836857)
199
+ 189 train 8.878774 (lr=1.3287e-04) (hash(x)=52690444)
200
+ 190 train 8.987842 (lr=1.3357e-04) (hash(x)=53043440)
201
+ 191 train 8.924240 (lr=1.3427e-04) (hash(x)=45627699)
202
+ 192 train 8.980259 (lr=1.3497e-04) (hash(x)=50337632)
203
+ 193 train 8.971118 (lr=1.3566e-04) (hash(x)=50636707)
204
+ 194 train 8.743868 (lr=1.3636e-04) (hash(x)=43415450)
205
+ 195 train 8.784453 (lr=1.3706e-04) (hash(x)=44771118)
206
+ 196 train 8.800592 (lr=1.3776e-04) (hash(x)=47474403)
207
+ 197 train 8.733015 (lr=1.3846e-04) (hash(x)=48109117)
208
+ 198 train 8.827570 (lr=1.3916e-04) (hash(x)=50752974)
209
+ 199 train 8.938959 (lr=1.3986e-04) (hash(x)=55122837)
210
+ 200 val loss 8.8616
211
+ 200 val perplexity 7055.8833
212
+ 200 train 8.786185 (lr=1.4056e-04) (hash(x)=49692660)
213
+ 201 train 9.044040 (lr=1.4126e-04) (hash(x)=58304741)
214
+ 202 train 8.805367 (lr=1.4196e-04) (hash(x)=51071912)
215
+ 203 train 8.832573 (lr=1.4266e-04) (hash(x)=48891446)
216
+ 204 train 8.904689 (lr=1.4336e-04) (hash(x)=51980344)
217
+ 205 train 8.649818 (lr=1.4406e-04) (hash(x)=46085486)
218
+ 206 train 8.799945 (lr=1.4476e-04) (hash(x)=50429471)
219
+ 207 train 8.835604 (lr=1.4545e-04) (hash(x)=50609171)
220
+ 208 train 8.748601 (lr=1.4615e-04) (hash(x)=56366529)
221
+ 209 train 8.802855 (lr=1.4685e-04) (hash(x)=44693623)
222
+ 210 train 8.812421 (lr=1.4755e-04) (hash(x)=52480393)
223
+ 211 train 8.711429 (lr=1.4825e-04) (hash(x)=45080235)
224
+ 212 train 8.695422 (lr=1.4895e-04) (hash(x)=47772363)
225
+ 213 train 8.594543 (lr=1.4965e-04) (hash(x)=46731794)
226
+ 214 train 8.642806 (lr=1.5035e-04) (hash(x)=44181238)
227
+ 215 train 8.705047 (lr=1.5105e-04) (hash(x)=46674118)
228
+ 216 train 8.535064 (lr=1.5175e-04) (hash(x)=49734489)
229
+ 217 train 8.651253 (lr=1.5245e-04) (hash(x)=46647221)
230
+ 218 train 8.606984 (lr=1.5315e-04) (hash(x)=53699708)
231
+ 219 train 8.622198 (lr=1.5385e-04) (hash(x)=51868493)
232
+ 220 train 8.612872 (lr=1.5455e-04) (hash(x)=49234242)
233
+ 221 train 8.638827 (lr=1.5524e-04) (hash(x)=52926855)
234
+ 222 train 8.497591 (lr=1.5594e-04) (hash(x)=48168200)
235
+ 223 train 8.560508 (lr=1.5664e-04) (hash(x)=50306555)
236
+ 224 train 8.449250 (lr=1.5734e-04) (hash(x)=47450495)
237
+ 225 train 8.463203 (lr=1.5804e-04) (hash(x)=47017597)
238
+ 226 train 8.406131 (lr=1.5874e-04) (hash(x)=45879487)
239
+ 227 train 8.421384 (lr=1.5944e-04) (hash(x)=45930697)
240
+ 228 train 8.702616 (lr=1.6014e-04) (hash(x)=51866248)
241
+ 229 train 8.560703 (lr=1.6084e-04) (hash(x)=55896424)
242
+ 230 train 8.505924 (lr=1.6154e-04) (hash(x)=49856927)
243
+ 231 train 8.419337 (lr=1.6224e-04) (hash(x)=50806347)
244
+ 232 train 8.455842 (lr=1.6294e-04) (hash(x)=48907697)
245
+ 233 train 8.519343 (lr=1.6364e-04) (hash(x)=53269239)
246
+ 234 train 8.616884 (lr=1.6434e-04) (hash(x)=58562988)
247
+ 235 train 8.337857 (lr=1.6503e-04) (hash(x)=50115446)
248
+ 236 train 8.289886 (lr=1.6573e-04) (hash(x)=39354357)
249
+ 237 train 8.500144 (lr=1.6643e-04) (hash(x)=53956776)
250
+ 238 train 8.264709 (lr=1.6713e-04) (hash(x)=49150615)
251
+ 239 train 8.362792 (lr=1.6783e-04) (hash(x)=51243947)
252
+ 240 train 8.476169 (lr=1.6853e-04) (hash(x)=54091498)
253
+ 241 train 8.315440 (lr=1.6923e-04) (hash(x)=55362900)
254
+ 242 train 8.350728 (lr=1.6993e-04) (hash(x)=49537560)
255
+ 243 train 8.054591 (lr=1.7063e-04) (hash(x)=34965283)
256
+ 244 train 8.260314 (lr=1.7133e-04) (hash(x)=43293134)
257
+ 245 train 8.269095 (lr=1.7203e-04) (hash(x)=52583594)
258
+ 246 train 8.350337 (lr=1.7273e-04) (hash(x)=50277570)
259
+ 247 train 8.264630 (lr=1.7343e-04) (hash(x)=53806649)
260
+ 248 train 8.233866 (lr=1.7413e-04) (hash(x)=48819401)
261
+ 249 train 8.216193 (lr=1.7483e-04) (hash(x)=51974675)
262
+ 250 val loss 8.2413
263
+ 250 val perplexity 3794.4939
264
+ 250 train 8.272585 (lr=1.7552e-04) (hash(x)=46284281)
265
+ 251 train 8.224204 (lr=1.7622e-04) (hash(x)=45810574)
266
+ 252 train 8.305145 (lr=1.7692e-04) (hash(x)=49217593)
267
+ 253 train 7.798548 (lr=1.7762e-04) (hash(x)=34273240)
268
+ 254 train 8.064310 (lr=1.7832e-04) (hash(x)=48362439)
269
+ 255 train 8.125449 (lr=1.7902e-04) (hash(x)=50536970)
270
+ 256 train 8.221976 (lr=1.7972e-04) (hash(x)=52438285)
271
+ 257 train 7.859964 (lr=1.8042e-04) (hash(x)=47843837)
272
+ 258 train 8.202934 (lr=1.8112e-04) (hash(x)=50928031)
273
+ 259 train 8.250717 (lr=1.8182e-04) (hash(x)=56940944)
274
+ 260 train 8.244458 (lr=1.8252e-04) (hash(x)=54715437)
275
+ 261 train 8.217026 (lr=1.8322e-04) (hash(x)=54162688)
276
+ 262 train 8.147473 (lr=1.8392e-04) (hash(x)=45398098)
277
+ 263 train 8.181998 (lr=1.8462e-04) (hash(x)=51669498)
278
+ 264 train 8.089635 (lr=1.8531e-04) (hash(x)=52890008)
279
+ 265 train 8.212882 (lr=1.8601e-04) (hash(x)=53020601)
280
+ 266 train 8.193078 (lr=1.8671e-04) (hash(x)=50728575)
281
+ 267 train 8.134582 (lr=1.8741e-04) (hash(x)=46315245)
282
+ 268 train 7.980864 (lr=1.8811e-04) (hash(x)=51700327)
283
+ 269 train 8.158723 (lr=1.8881e-04) (hash(x)=51422940)
284
+ 270 train 8.027968 (lr=1.8951e-04) (hash(x)=50079658)
285
+ 271 train 8.035962 (lr=1.9021e-04) (hash(x)=50610610)
286
+ 272 train 8.052531 (lr=1.9091e-04) (hash(x)=53349554)
287
+ 273 train 7.979824 (lr=1.9161e-04) (hash(x)=47220467)
288
+ 274 train 7.955947 (lr=1.9231e-04) (hash(x)=47059896)
289
+ 275 train 7.994769 (lr=1.9301e-04) (hash(x)=51478185)
290
+ 276 train 7.974135 (lr=1.9371e-04) (hash(x)=49910644)
291
+ 277 train 7.890980 (lr=1.9441e-04) (hash(x)=48930577)
292
+ 278 train 8.030375 (lr=1.9510e-04) (hash(x)=49020396)
293
+ 279 train 8.134684 (lr=1.9580e-04) (hash(x)=53701929)
294
+ 280 train 7.946855 (lr=1.9650e-04) (hash(x)=47184278)
295
+ 281 train 8.185374 (lr=1.9720e-04) (hash(x)=54410011)
296
+ 282 train 8.052638 (lr=1.9790e-04) (hash(x)=54549283)
297
+ 283 train 8.043493 (lr=1.9860e-04) (hash(x)=51021250)
298
+ 284 train 7.933570 (lr=1.9930e-04) (hash(x)=50272010)
299
+ 285 train 8.029544 (lr=2.0000e-04) (hash(x)=55817179)
300
+ 286 train 7.906720 (lr=2.0000e-04) (hash(x)=47596588)
301
+ 287 train 7.840889 (lr=2.0000e-04) (hash(x)=46930369)
302
+ 288 train 7.852684 (lr=2.0000e-04) (hash(x)=48874951)
303
+ 289 train 8.007041 (lr=1.9999e-04) (hash(x)=51147542)
304
+ 290 train 7.927900 (lr=1.9999e-04) (hash(x)=52125021)
305
+ 291 train 7.597225 (lr=1.9998e-04) (hash(x)=39405431)
306
+ 292 train 7.849229 (lr=1.9997e-04) (hash(x)=47289107)
307
+ 293 train 7.831790 (lr=1.9996e-04) (hash(x)=48536998)
308
+ 294 train 8.294857 (lr=1.9994e-04) (hash(x)=62523011)
309
+ 295 train 8.322829 (lr=1.9993e-04) (hash(x)=57476018)
310
+ 296 train 7.976010 (lr=1.9991e-04) (hash(x)=54119125)
311
+ 297 train 8.247438 (lr=1.9989e-04) (hash(x)=58950633)
312
+ 298 train 7.990517 (lr=1.9987e-04) (hash(x)=53212705)
313
+ 299 train 7.916055 (lr=1.9985e-04) (hash(x)=48315677)
314
+ 300 val loss 7.8479
315
+ 300 val perplexity 2560.2395
316
+ 300 train 7.869667 (lr=1.9983e-04) (hash(x)=48622485)
317
+ 301 train 7.897988 (lr=1.9980e-04) (hash(x)=53677175)
318
+ 302 train 7.993814 (lr=1.9978e-04) (hash(x)=55167389)
319
+ 303 train 7.920248 (lr=1.9975e-04) (hash(x)=50199187)
320
+ 304 train 7.997447 (lr=1.9972e-04) (hash(x)=52896102)
321
+ 305 train 7.910948 (lr=1.9969e-04) (hash(x)=56596899)
322
+ 306 train 8.021743 (lr=1.9965e-04) (hash(x)=53062758)
323
+ 307 train 7.938333 (lr=1.9962e-04) (hash(x)=49586111)
324
+ 308 train 7.877156 (lr=1.9958e-04) (hash(x)=52369544)
325
+ 309 train 7.783241 (lr=1.9954e-04) (hash(x)=46819915)
326
+ 310 train 7.659936 (lr=1.9950e-04) (hash(x)=49293953)
327
+ 311 train 7.870523 (lr=1.9946e-04) (hash(x)=45949699)
328
+ 312 train 8.022288 (lr=1.9941e-04) (hash(x)=53515595)
329
+ 313 train 7.747560 (lr=1.9937e-04) (hash(x)=45515416)
330
+ 314 train 7.979358 (lr=1.9932e-04) (hash(x)=49455078)
331
+ 315 train 8.466532 (lr=1.9927e-04) (hash(x)=47334388)
332
+ 316 train 8.430546 (lr=1.9922e-04) (hash(x)=48654816)
333
+ 317 train 7.996665 (lr=1.9916e-04) (hash(x)=49474562)
334
+ 318 train 7.915223 (lr=1.9911e-04) (hash(x)=47371125)
335
+ 319 train 7.742874 (lr=1.9905e-04) (hash(x)=53294055)
336
+ 320 train 7.572171 (lr=1.9899e-04) (hash(x)=51679677)
337
+ 321 train 7.755049 (lr=1.9893e-04) (hash(x)=54082362)
338
+ 322 train 7.820014 (lr=1.9887e-04) (hash(x)=53670230)
339
+ 323 train 7.621237 (lr=1.9881e-04) (hash(x)=46499362)
340
+ 324 train 7.918240 (lr=1.9874e-04) (hash(x)=51843454)
341
+ 325 train 7.853727 (lr=1.9868e-04) (hash(x)=53531448)
342
+ 326 train 7.821175 (lr=1.9861e-04) (hash(x)=55684813)
343
+ 327 train 7.773271 (lr=1.9854e-04) (hash(x)=53097392)
344
+ 328 train 7.403336 (lr=1.9847e-04) (hash(x)=45277852)
345
+ 329 train 7.401851 (lr=1.9839e-04) (hash(x)=41633073)
346
+ 330 train 7.517650 (lr=1.9832e-04) (hash(x)=42914820)
347
+ 331 train 7.677701 (lr=1.9824e-04) (hash(x)=46479397)
348
+ 332 train 7.572614 (lr=1.9816e-04) (hash(x)=44391371)
349
+ 333 train 7.426587 (lr=1.9808e-04) (hash(x)=38977328)
350
+ 334 train 7.688114 (lr=1.9800e-04) (hash(x)=50758211)
351
+ 335 train 7.854351 (lr=1.9792e-04) (hash(x)=54479595)
352
+ 336 train 7.902742 (lr=1.9783e-04) (hash(x)=51411646)
353
+ 337 train 7.858977 (lr=1.9774e-04) (hash(x)=52103949)
354
+ 338 train 7.757939 (lr=1.9765e-04) (hash(x)=50550907)
355
+ 339 train 7.662548 (lr=1.9756e-04) (hash(x)=47738093)
356
+ 340 train 7.996507 (lr=1.9747e-04) (hash(x)=56998616)
357
+ 341 train 7.741690 (lr=1.9738e-04) (hash(x)=51395731)
358
+ 342 train 7.904665 (lr=1.9728e-04) (hash(x)=61530180)
359
+ 343 train 8.146672 (lr=1.9718e-04) (hash(x)=59006693)
360
+ 344 train 7.660684 (lr=1.9709e-04) (hash(x)=48594367)
361
+ 345 train 7.766876 (lr=1.9698e-04) (hash(x)=52545488)
362
+ 346 train 7.645439 (lr=1.9688e-04) (hash(x)=47376565)
363
+ 347 train 7.724994 (lr=1.9678e-04) (hash(x)=46312947)
364
+ 348 train 7.732822 (lr=1.9667e-04) (hash(x)=49325729)
365
+ 349 train 7.831522 (lr=1.9656e-04) (hash(x)=51488924)
366
+ 350 val loss 7.7156
367
+ 350 val perplexity 2243.0256
368
+ 350 train 7.726972 (lr=1.9646e-04) (hash(x)=52194507)
369
+ 351 train 7.708641 (lr=1.9634e-04) (hash(x)=51396394)
370
+ 352 train 7.842031 (lr=1.9623e-04) (hash(x)=47730521)
371
+ 353 train 7.637547 (lr=1.9612e-04) (hash(x)=43705721)
372
+ 354 train 7.650529 (lr=1.9600e-04) (hash(x)=47078328)
373
+ 355 train 7.965048 (lr=1.9588e-04) (hash(x)=55830040)
374
+ 356 train 7.901582 (lr=1.9576e-04) (hash(x)=48559738)
375
+ 357 train 7.713269 (lr=1.9564e-04) (hash(x)=49340705)
376
+ 358 train 7.790738 (lr=1.9552e-04) (hash(x)=49635452)
377
+ 359 train 7.806571 (lr=1.9540e-04) (hash(x)=52098425)
378
+ 360 train 7.656087 (lr=1.9527e-04) (hash(x)=48852030)
379
+ 361 train 7.719658 (lr=1.9514e-04) (hash(x)=48643399)
380
+ 362 train 7.997684 (lr=1.9501e-04) (hash(x)=58138953)
381
+ 363 train 7.454658 (lr=1.9488e-04) (hash(x)=44163258)
382
+ 364 train 7.817750 (lr=1.9475e-04) (hash(x)=55032638)
383
+ 365 train 7.683890 (lr=1.9462e-04) (hash(x)=48803235)
384
+ 366 train 7.560815 (lr=1.9448e-04) (hash(x)=48930309)
385
+ 367 train 7.992860 (lr=1.9434e-04) (hash(x)=58459317)
386
+ 368 train 7.840917 (lr=1.9421e-04) (hash(x)=52969670)
387
+ 369 train 7.787548 (lr=1.9406e-04) (hash(x)=55950871)
388
+ 370 train 7.742256 (lr=1.9392e-04) (hash(x)=50190558)
389
+ 371 train 7.824319 (lr=1.9378e-04) (hash(x)=50402996)
390
+ 372 train 7.669240 (lr=1.9363e-04) (hash(x)=48331790)
391
+ 373 train 7.696492 (lr=1.9349e-04) (hash(x)=51861896)
392
+ 374 train 7.722473 (lr=1.9334e-04) (hash(x)=52442780)
393
+ 375 train 7.676956 (lr=1.9319e-04) (hash(x)=47781307)
394
+ 376 train 7.351990 (lr=1.9304e-04) (hash(x)=45305570)
395
+ 377 train 8.219782 (lr=1.9288e-04) (hash(x)=58591982)
396
+ 378 train 7.834163 (lr=1.9273e-04) (hash(x)=53348894)
397
+ 379 train 7.998423 (lr=1.9257e-04) (hash(x)=55269973)
398
+ 380 train 7.748958 (lr=1.9241e-04) (hash(x)=52427717)
399
+ 381 train 7.931667 (lr=1.9225e-04) (hash(x)=54250578)
400
+ 382 train 7.914312 (lr=1.9209e-04) (hash(x)=55042333)
401
+ 383 train 7.920208 (lr=1.9193e-04) (hash(x)=55088766)
402
+ 384 train 7.689182 (lr=1.9176e-04) (hash(x)=51982665)
403
+ 385 train 7.662782 (lr=1.9160e-04) (hash(x)=50679914)
404
+ 386 train 7.760742 (lr=1.9143e-04) (hash(x)=55557228)
405
+ 387 train 7.739777 (lr=1.9126e-04) (hash(x)=48450358)
406
+ 388 train 7.578718 (lr=1.9109e-04) (hash(x)=49288522)
407
+ 389 train 7.820223 (lr=1.9091e-04) (hash(x)=49197883)
408
+ 390 train 7.781827 (lr=1.9074e-04) (hash(x)=54555602)
409
+ 391 train 7.651784 (lr=1.9056e-04) (hash(x)=50198680)
410
+ 392 train 7.737273 (lr=1.9039e-04) (hash(x)=54766944)
411
+ 393 train 7.700063 (lr=1.9021e-04) (hash(x)=51419665)
412
+ 394 train 7.276377 (lr=1.9003e-04) (hash(x)=43612177)
413
+ 395 train 7.636800 (lr=1.8985e-04) (hash(x)=48650332)
414
+ 396 train 7.729390 (lr=1.8966e-04) (hash(x)=53120879)
415
+ 397 train 7.593544 (lr=1.8948e-04) (hash(x)=48143827)
416
+ 398 train 7.595224 (lr=1.8929e-04) (hash(x)=46766862)
417
+ 399 train 7.705740 (lr=1.8910e-04) (hash(x)=52234688)
418
+ 400 val loss 7.6683
419
+ 400 val perplexity 2139.5386
420
+ 400 train 7.613262 (lr=1.8891e-04) (hash(x)=49099804)
421
+ 401 train 7.732048 (lr=1.8872e-04) (hash(x)=52032234)
422
+ 402 train 7.721689 (lr=1.8853e-04) (hash(x)=51775957)
423
+ 403 train 7.703857 (lr=1.8834e-04) (hash(x)=51346801)
424
+ 404 train 7.714598 (lr=1.8814e-04) (hash(x)=49392856)
425
+ 405 train 8.036746 (lr=1.8794e-04) (hash(x)=58253034)
426
+ 406 train 7.903621 (lr=1.8774e-04) (hash(x)=54043298)
427
+ 407 train 7.765503 (lr=1.8754e-04) (hash(x)=52929403)
428
+ 408 train 7.682576 (lr=1.8734e-04) (hash(x)=52901232)
429
+ 409 train 7.874216 (lr=1.8714e-04) (hash(x)=56975127)
430
+ 410 train 7.695967 (lr=1.8693e-04) (hash(x)=50603607)
431
+ 411 train 7.824713 (lr=1.8673e-04) (hash(x)=56070928)
432
+ 412 train 7.662697 (lr=1.8652e-04) (hash(x)=44443158)
433
+ 413 train 7.670433 (lr=1.8631e-04) (hash(x)=48921492)
434
+ 414 train 7.732751 (lr=1.8610e-04) (hash(x)=50946739)
435
+ 415 train 7.532600 (lr=1.8589e-04) (hash(x)=50526384)
436
+ 416 train 7.784912 (lr=1.8567e-04) (hash(x)=56037628)
437
+ 417 train 7.562004 (lr=1.8546e-04) (hash(x)=49178590)
438
+ 418 train 7.653552 (lr=1.8524e-04) (hash(x)=50164285)
439
+ 419 train 7.910399 (lr=1.8502e-04) (hash(x)=60209461)
440
+ 420 train 7.888630 (lr=1.8480e-04) (hash(x)=51355917)
441
+ 421 train 7.723494 (lr=1.8458e-04) (hash(x)=51970769)
442
+ 422 train 7.671051 (lr=1.8436e-04) (hash(x)=51896253)
443
+ 423 train 7.766355 (lr=1.8414e-04) (hash(x)=52322092)
444
+ 424 train 7.382028 (lr=1.8391e-04) (hash(x)=46560171)
445
+ 425 train 7.598588 (lr=1.8369e-04) (hash(x)=49806109)
446
+ 426 train 7.771512 (lr=1.8346e-04) (hash(x)=52507118)
447
+ 427 train 7.769502 (lr=1.8323e-04) (hash(x)=52916383)
448
+ 428 train 7.538571 (lr=1.8300e-04) (hash(x)=46519875)
449
+ 429 train 7.470142 (lr=1.8276e-04) (hash(x)=43459378)
450
+ 430 train 7.716650 (lr=1.8253e-04) (hash(x)=54438747)
451
+ 431 train 7.523711 (lr=1.8230e-04) (hash(x)=52529702)
452
+ 432 train 7.655316 (lr=1.8206e-04) (hash(x)=52506037)
453
+ 433 train 7.676141 (lr=1.8182e-04) (hash(x)=49688097)
454
+ 434 train 7.723791 (lr=1.8158e-04) (hash(x)=52187692)
455
+ 435 train 7.743150 (lr=1.8134e-04) (hash(x)=51980724)
456
+ 436 train 7.771242 (lr=1.8110e-04) (hash(x)=57522870)
457
+ 437 train 7.641463 (lr=1.8086e-04) (hash(x)=47435088)
458
+ 438 train 7.754393 (lr=1.8061e-04) (hash(x)=54021184)
459
+ 439 train 7.515082 (lr=1.8036e-04) (hash(x)=46826784)
460
+ 440 train 7.760738 (lr=1.8012e-04) (hash(x)=46764832)
461
+ 441 train 7.671886 (lr=1.7987e-04) (hash(x)=51120264)
462
+ 442 train 7.657025 (lr=1.7962e-04) (hash(x)=52654128)
463
+ 443 train 7.664714 (lr=1.7937e-04) (hash(x)=51321857)
464
+ 444 train 7.573438 (lr=1.7911e-04) (hash(x)=45311446)
465
+ 445 train 7.465247 (lr=1.7886e-04) (hash(x)=51476600)
466
+ 446 train 7.713422 (lr=1.7860e-04) (hash(x)=51269835)
467
+ 447 train 7.652721 (lr=1.7835e-04) (hash(x)=53475454)
468
+ 448 train 7.488464 (lr=1.7809e-04) (hash(x)=46470991)
469
+ 449 train 7.412716 (lr=1.7783e-04) (hash(x)=39878629)
470
+ 450 val loss 7.6031
471
+ 450 val perplexity 2004.4531
472
+ 450 train 7.635339 (lr=1.7757e-04) (hash(x)=50326173)
473
+ 451 train 7.651618 (lr=1.7731e-04) (hash(x)=54787418)
474
+ 452 train 7.728883 (lr=1.7704e-04) (hash(x)=48949948)
475
+ 453 train 7.788944 (lr=1.7678e-04) (hash(x)=52506608)
476
+ 454 train 7.568782 (lr=1.7651e-04) (hash(x)=54323120)
477
+ 455 train 7.542637 (lr=1.7624e-04) (hash(x)=46810389)
478
+ 456 train 7.435760 (lr=1.7597e-04) (hash(x)=45396873)
479
+ 457 train 7.776685 (lr=1.7570e-04) (hash(x)=57144694)
480
+ 458 train 7.521841 (lr=1.7543e-04) (hash(x)=49878429)
481
+ 459 train 7.666568 (lr=1.7516e-04) (hash(x)=53498382)
482
+ 460 train 7.540543 (lr=1.7489e-04) (hash(x)=47417814)
483
+ 461 train 7.680439 (lr=1.7461e-04) (hash(x)=51280415)
484
+ 462 train 7.665544 (lr=1.7434e-04) (hash(x)=54075295)
485
+ 463 train 7.482296 (lr=1.7406e-04) (hash(x)=46323632)
486
+ 464 train 7.588748 (lr=1.7378e-04) (hash(x)=50211782)
487
+ 465 train 7.537257 (lr=1.7350e-04) (hash(x)=46779238)
488
+ 466 train 7.626318 (lr=1.7322e-04) (hash(x)=48313734)
489
+ 467 train 7.625603 (lr=1.7294e-04) (hash(x)=52398798)
490
+ 468 train 7.594667 (lr=1.7265e-04) (hash(x)=50640616)
491
+ 469 train 7.588582 (lr=1.7237e-04) (hash(x)=48186608)
492
+ 470 train 7.615999 (lr=1.7208e-04) (hash(x)=51158107)
493
+ 471 train 7.490911 (lr=1.7179e-04) (hash(x)=53724068)
494
+ 472 train 7.546930 (lr=1.7151e-04) (hash(x)=52537855)
495
+ 473 train 7.636984 (lr=1.7122e-04) (hash(x)=52960197)
496
+ 474 train 7.710744 (lr=1.7092e-04) (hash(x)=54156942)
497
+ 475 train 7.888799 (lr=1.7063e-04) (hash(x)=52768582)
498
+ 476 train 7.448968 (lr=1.7034e-04) (hash(x)=50479600)
499
+ 477 train 7.595882 (lr=1.7004e-04) (hash(x)=52810796)
500
+ 478 train 7.725255 (lr=1.6975e-04) (hash(x)=49731354)
501
+ 479 train 7.500096 (lr=1.6945e-04) (hash(x)=46747022)
502
+ 480 train 7.551336 (lr=1.6915e-04) (hash(x)=49495779)
503
+ 481 train 7.638721 (lr=1.6886e-04) (hash(x)=53760168)
504
+ 482 train 7.659782 (lr=1.6856e-04) (hash(x)=51460514)
505
+ 483 train 7.618024 (lr=1.6825e-04) (hash(x)=50353293)
506
+ 484 train 7.600713 (lr=1.6795e-04) (hash(x)=53396658)
507
+ 485 train 7.710873 (lr=1.6765e-04) (hash(x)=50573284)
508
+ 486 train 7.445994 (lr=1.6734e-04) (hash(x)=49168443)
509
+ 487 train 7.671410 (lr=1.6704e-04) (hash(x)=50822751)
510
+ 488 train 7.696812 (lr=1.6673e-04) (hash(x)=45669857)
511
+ 489 train 7.462562 (lr=1.6642e-04) (hash(x)=44593201)
512
+ 490 train 7.261325 (lr=1.6611e-04) (hash(x)=39734742)
513
+ 491 train 7.585416 (lr=1.6580e-04) (hash(x)=47616170)
514
+ 492 train 7.549239 (lr=1.6549e-04) (hash(x)=46558148)
515
+ 493 train 7.512651 (lr=1.6518e-04) (hash(x)=49711034)
516
+ 494 train 7.748211 (lr=1.6487e-04) (hash(x)=55210283)
517
+ 495 train 7.575337 (lr=1.6455e-04) (hash(x)=48013779)
518
+ 496 train 7.616418 (lr=1.6424e-04) (hash(x)=50644422)
519
+ 497 train 7.621373 (lr=1.6392e-04) (hash(x)=49696467)
520
+ 498 train 7.617441 (lr=1.6360e-04) (hash(x)=43246287)
521
+ 499 train 7.511930 (lr=1.6328e-04) (hash(x)=45598720)
522
+ 500 val loss 7.5915
523
+ 500 val perplexity 1981.2271
524
+ 500 train 7.571321 (lr=1.6296e-04) (hash(x)=51062334)
525
+ 501 train 7.482541 (lr=1.6264e-04) (hash(x)=49421968)
526
+ 502 train 7.605157 (lr=1.6232e-04) (hash(x)=52123859)
527
+ 503 train 7.616413 (lr=1.6200e-04) (hash(x)=52658193)
528
+ 504 train 7.554155 (lr=1.6168e-04) (hash(x)=51023513)
529
+ 505 train 7.783034 (lr=1.6135e-04) (hash(x)=51581514)
530
+ 506 train 7.673482 (lr=1.6103e-04) (hash(x)=49089133)
531
+ 507 train 7.551337 (lr=1.6070e-04) (hash(x)=48531900)
532
+ 508 train 7.639862 (lr=1.6037e-04) (hash(x)=52693475)
533
+ 509 train 7.634651 (lr=1.6004e-04) (hash(x)=54903424)
534
+ 510 train 7.662857 (lr=1.5971e-04) (hash(x)=51036262)
535
+ 511 train 7.515724 (lr=1.5938e-04) (hash(x)=46134720)
536
+ 512 train 7.725934 (lr=1.5905e-04) (hash(x)=53031267)
537
+ 513 train 7.736398 (lr=1.5872e-04) (hash(x)=53833396)
538
+ 514 train 7.569791 (lr=1.5838e-04) (hash(x)=55247788)
539
+ 515 train 7.621184 (lr=1.5805e-04) (hash(x)=51156418)
540
+ 516 train 7.528346 (lr=1.5772e-04) (hash(x)=51468309)
541
+ 517 train 7.477683 (lr=1.5738e-04) (hash(x)=48618191)
542
+ 518 train 7.764391 (lr=1.5704e-04) (hash(x)=53297837)
543
+ 519 train 7.574213 (lr=1.5670e-04) (hash(x)=52902481)
544
+ 520 train 7.571499 (lr=1.5636e-04) (hash(x)=52699185)
545
+ 521 train 7.611329 (lr=1.5602e-04) (hash(x)=54117145)
546
+ 522 train 7.753488 (lr=1.5568e-04) (hash(x)=53596095)
547
+ 523 train 7.610703 (lr=1.5534e-04) (hash(x)=51604647)
548
+ 524 train 7.564107 (lr=1.5500e-04) (hash(x)=52145920)
549
+ 525 train 7.532642 (lr=1.5466e-04) (hash(x)=50691387)
550
+ 526 train 7.657948 (lr=1.5431e-04) (hash(x)=52429064)
551
+ 527 train 7.495456 (lr=1.5397e-04) (hash(x)=49163526)
552
+ 528 train 7.692514 (lr=1.5362e-04) (hash(x)=53184723)
553
+ 529 train 7.456529 (lr=1.5327e-04) (hash(x)=47039068)
554
+ 530 train 7.727259 (lr=1.5293e-04) (hash(x)=47552672)
555
+ 531 train 7.516154 (lr=1.5258e-04) (hash(x)=53829392)
556
+ 532 train 7.376569 (lr=1.5223e-04) (hash(x)=47800177)
557
+ 533 train 7.387517 (lr=1.5188e-04) (hash(x)=49800706)
558
+ 534 train 7.722309 (lr=1.5153e-04) (hash(x)=55575943)
559
+ 535 train 7.522624 (lr=1.5118e-04) (hash(x)=49028078)
560
+ 536 train 7.667557 (lr=1.5082e-04) (hash(x)=54755546)
561
+ 537 train 7.619453 (lr=1.5047e-04) (hash(x)=53372658)
562
+ 538 train 7.568677 (lr=1.5012e-04) (hash(x)=52268865)
563
+ 539 train 7.579017 (lr=1.4976e-04) (hash(x)=52517672)
564
+ 540 train 7.553670 (lr=1.4941e-04) (hash(x)=50244085)
565
+ 541 train 7.447942 (lr=1.4905e-04) (hash(x)=49248218)
566
+ 542 train 7.608560 (lr=1.4869e-04) (hash(x)=49762688)
567
+ 543 train 7.717953 (lr=1.4833e-04) (hash(x)=59110263)
568
+ 544 train 7.443224 (lr=1.4798e-04) (hash(x)=47478150)
569
+ 545 train 7.560536 (lr=1.4762e-04) (hash(x)=54533121)
570
+ 546 train 7.484357 (lr=1.4726e-04) (hash(x)=49901224)
571
+ 547 train 7.609699 (lr=1.4690e-04) (hash(x)=51195255)
572
+ 548 train 7.536994 (lr=1.4653e-04) (hash(x)=49109169)
573
+ 549 train 7.446237 (lr=1.4617e-04) (hash(x)=47467588)
574
+ 550 val loss 7.5069
575
+ 550 val perplexity 1820.5686
576
+ 550 train 7.469481 (lr=1.4581e-04) (hash(x)=46269857)
577
+ 551 train 7.443704 (lr=1.4545e-04) (hash(x)=50517780)
578
+ 552 train 7.543475 (lr=1.4508e-04) (hash(x)=50183795)
579
+ 553 train 7.498013 (lr=1.4472e-04) (hash(x)=50640606)
580
+ 554 train 7.539389 (lr=1.4435e-04) (hash(x)=51738364)
581
+ 555 train 8.239552 (lr=1.4398e-04) (hash(x)=62341175)
582
+ 556 train 7.653484 (lr=1.4362e-04) (hash(x)=45189486)
583
+ 557 train 7.615215 (lr=1.4325e-04) (hash(x)=51670137)
584
+ 558 train 7.443668 (lr=1.4288e-04) (hash(x)=46152565)
585
+ 559 train 7.347608 (lr=1.4251e-04) (hash(x)=45890874)
586
+ 560 train 7.554988 (lr=1.4214e-04) (hash(x)=50865200)
587
+ 561 train 7.392642 (lr=1.4177e-04) (hash(x)=50817384)
588
+ 562 train 7.362367 (lr=1.4140e-04) (hash(x)=45716461)
589
+ 563 train 7.507634 (lr=1.4103e-04) (hash(x)=49188067)
590
+ 564 train 7.604416 (lr=1.4066e-04) (hash(x)=51104117)
591
+ 565 train 7.604581 (lr=1.4029e-04) (hash(x)=53943162)
592
+ 566 train 7.228746 (lr=1.3991e-04) (hash(x)=42808589)
593
+ 567 train 7.484834 (lr=1.3954e-04) (hash(x)=47888783)
594
+ 568 train 7.360088 (lr=1.3916e-04) (hash(x)=46060882)
595
+ 569 train 7.580092 (lr=1.3879e-04) (hash(x)=47770108)
596
+ 570 train 7.750607 (lr=1.3841e-04) (hash(x)=53150465)
597
+ 571 train 7.917108 (lr=1.3804e-04) (hash(x)=52775090)
598
+ 572 train 7.651674 (lr=1.3766e-04) (hash(x)=49924177)
599
+ 573 train 7.485525 (lr=1.3728e-04) (hash(x)=49971730)
600
+ 574 train 7.506099 (lr=1.3691e-04) (hash(x)=47952698)
601
+ 575 train 7.821842 (lr=1.3653e-04) (hash(x)=56693767)
602
+ 576 train 7.283060 (lr=1.3615e-04) (hash(x)=43516921)
603
+ 577 train 7.648258 (lr=1.3577e-04) (hash(x)=53347077)
604
+ 578 train 7.427532 (lr=1.3539e-04) (hash(x)=47410931)
605
+ 579 train 7.411890 (lr=1.3501e-04) (hash(x)=45014400)
606
+ 580 train 7.525362 (lr=1.3463e-04) (hash(x)=54677976)
607
+ 581 train 7.301271 (lr=1.3425e-04) (hash(x)=45398141)
608
+ 582 train 7.334895 (lr=1.3387e-04) (hash(x)=47181969)
609
+ 583 train 7.775783 (lr=1.3348e-04) (hash(x)=53102856)
610
+ 584 train 7.364972 (lr=1.3310e-04) (hash(x)=48780879)
611
+ 585 train 7.496053 (lr=1.3272e-04) (hash(x)=51707711)
612
+ 586 train 7.585763 (lr=1.3234e-04) (hash(x)=48191112)
613
+ 587 train 7.428010 (lr=1.3195e-04) (hash(x)=52954310)
614
+ 588 train 7.764925 (lr=1.3157e-04) (hash(x)=58807144)
615
+ 589 train 7.323857 (lr=1.3118e-04) (hash(x)=45943270)
616
+ 590 train 7.514905 (lr=1.3080e-04) (hash(x)=49191483)
617
+ 591 train 7.451535 (lr=1.3041e-04) (hash(x)=47965042)
618
+ 592 train 7.534734 (lr=1.3003e-04) (hash(x)=51074739)
619
+ 593 train 7.562151 (lr=1.2964e-04) (hash(x)=56913043)
620
+ 594 train 7.782994 (lr=1.2925e-04) (hash(x)=62500105)
621
+ 595 train 7.591753 (lr=1.2887e-04) (hash(x)=56016526)
622
+ 596 train 7.417368 (lr=1.2848e-04) (hash(x)=46762756)
623
+ 597 train 7.524739 (lr=1.2809e-04) (hash(x)=50655822)
624
+ 598 train 7.462200 (lr=1.2770e-04) (hash(x)=50751670)
625
+ 599 train 7.385035 (lr=1.2732e-04) (hash(x)=45600303)
626
+ 600 val loss 7.4717
627
+ 600 val perplexity 1757.6083
628
+ 600 train 7.559194 (lr=1.2693e-04) (hash(x)=54607172)
629
+ 601 train 7.452538 (lr=1.2654e-04) (hash(x)=48710188)
630
+ 602 train 7.362789 (lr=1.2615e-04) (hash(x)=46206284)
631
+ 603 train 7.345494 (lr=1.2576e-04) (hash(x)=45764012)
632
+ 604 train 7.375309 (lr=1.2537e-04) (hash(x)=50393806)
633
+ 605 train 7.465965 (lr=1.2498e-04) (hash(x)=52034772)
634
+ 606 train 7.457087 (lr=1.2459e-04) (hash(x)=48705579)
635
+ 607 train 7.219659 (lr=1.2420e-04) (hash(x)=44542942)
636
+ 608 train 7.376381 (lr=1.2381e-04) (hash(x)=48781316)
637
+ 609 train 7.470450 (lr=1.2341e-04) (hash(x)=48042083)
638
+ 610 train 7.564235 (lr=1.2302e-04) (hash(x)=49658564)
639
+ 611 train 7.616368 (lr=1.2263e-04) (hash(x)=51672319)
640
+ 612 train 7.525478 (lr=1.2224e-04) (hash(x)=52431673)
641
+ 613 train 7.495017 (lr=1.2185e-04) (hash(x)=43542246)
642
+ 614 train 7.489853 (lr=1.2145e-04) (hash(x)=52666378)
643
+ 615 train 7.480514 (lr=1.2106e-04) (hash(x)=49315253)
644
+ 616 train 7.556072 (lr=1.2067e-04) (hash(x)=50722863)
645
+ 617 train 7.744073 (lr=1.2027e-04) (hash(x)=52341036)
646
+ 618 train 7.505013 (lr=1.1988e-04) (hash(x)=48814169)
647
+ 619 train 7.341057 (lr=1.1949e-04) (hash(x)=45571167)
648
+ 620 train 7.384340 (lr=1.1909e-04) (hash(x)=47776763)
649
+ 621 train 7.430861 (lr=1.1870e-04) (hash(x)=48382543)
650
+ 622 train 7.377116 (lr=1.1830e-04) (hash(x)=45514205)
651
+ 623 train 7.272841 (lr=1.1791e-04) (hash(x)=45320225)
652
+ 624 train 7.415464 (lr=1.1752e-04) (hash(x)=48452761)
653
+ 625 train 7.648940 (lr=1.1712e-04) (hash(x)=49116111)
654
+ 626 train 7.312500 (lr=1.1673e-04) (hash(x)=47302076)
655
+ 627 train 7.436479 (lr=1.1633e-04) (hash(x)=50184363)
656
+ 628 train 7.332208 (lr=1.1594e-04) (hash(x)=49385175)
657
+ 629 train 7.510123 (lr=1.1554e-04) (hash(x)=42655518)
658
+ 630 train 7.462863 (lr=1.1515e-04) (hash(x)=45181685)
659
+ 631 train 7.256723 (lr=1.1475e-04) (hash(x)=45775751)
660
+ 632 train 7.485651 (lr=1.1435e-04) (hash(x)=51697844)
661
+ 633 train 7.378335 (lr=1.1396e-04) (hash(x)=47776033)
662
+ 634 train 7.573701 (lr=1.1356e-04) (hash(x)=54185547)
663
+ 635 train 7.327558 (lr=1.1317e-04) (hash(x)=42462763)
664
+ 636 train 7.420086 (lr=1.1277e-04) (hash(x)=49909171)
665
+ 637 train 7.477614 (lr=1.1238e-04) (hash(x)=50472276)
666
+ 638 train 7.700451 (lr=1.1198e-04) (hash(x)=53342806)
667
+ 639 train 7.541902 (lr=1.1158e-04) (hash(x)=53985957)
668
+ 640 train 7.410445 (lr=1.1119e-04) (hash(x)=49621437)
669
+ 641 train 7.562529 (lr=1.1079e-04) (hash(x)=50215750)
670
+ 642 train 7.337263 (lr=1.1040e-04) (hash(x)=45707324)
671
+ 643 train 7.722333 (lr=1.1000e-04) (hash(x)=55567006)
672
+ 644 train 7.903564 (lr=1.0960e-04) (hash(x)=63587960)
673
+ 645 train 7.538648 (lr=1.0921e-04) (hash(x)=50830962)
674
+ 646 train 7.702681 (lr=1.0881e-04) (hash(x)=50228793)
675
+ 647 train 7.264350 (lr=1.0842e-04) (hash(x)=41963014)
676
+ 648 train 7.415267 (lr=1.0802e-04) (hash(x)=52083811)
677
+ 649 train 7.961313 (lr=1.0762e-04) (hash(x)=55253867)
678
+ 650 val loss 7.4662
679
+ 650 val perplexity 1747.9375
680
+ 650 train 7.548727 (lr=1.0723e-04) (hash(x)=56596815)
681
+ 651 train 7.673913 (lr=1.0683e-04) (hash(x)=53862673)
682
+ 652 train 7.547476 (lr=1.0644e-04) (hash(x)=51032349)
683
+ 653 train 7.688118 (lr=1.0604e-04) (hash(x)=57095498)
684
+ 654 train 7.845939 (lr=1.0565e-04) (hash(x)=57840576)
685
+ 655 train 7.673074 (lr=1.0525e-04) (hash(x)=56042923)
686
+ 656 train 7.480969 (lr=1.0485e-04) (hash(x)=53170061)
687
+ 657 train 7.493700 (lr=1.0446e-04) (hash(x)=51651483)
688
+ 658 train 7.611675 (lr=1.0406e-04) (hash(x)=52052987)
689
+ 659 train 7.172945 (lr=1.0367e-04) (hash(x)=43078023)
690
+ 660 train 7.487012 (lr=1.0327e-04) (hash(x)=53974411)
691
+ 661 train 7.565363 (lr=1.0288e-04) (hash(x)=55335538)
692
+ 662 train 8.142560 (lr=1.0248e-04) (hash(x)=60627374)
693
+ 663 train 7.793506 (lr=1.0209e-04) (hash(x)=55023379)
694
+ 664 train 7.495417 (lr=1.0170e-04) (hash(x)=53081545)
695
+ 665 train 7.571742 (lr=1.0130e-04) (hash(x)=48246811)
696
+ 666 train 7.612133 (lr=1.0091e-04) (hash(x)=48563603)
697
+ 667 train 7.277937 (lr=1.0051e-04) (hash(x)=43011460)
698
+ 668 train 7.455652 (lr=1.0012e-04) (hash(x)=48368934)
699
+ 669 train 7.413377 (lr=9.9726e-05) (hash(x)=51322872)
700
+ 670 train 7.382810 (lr=9.9333e-05) (hash(x)=49196654)
701
+ 671 train 7.522335 (lr=9.8940e-05) (hash(x)=52536929)
702
+ 672 train 7.823111 (lr=9.8547e-05) (hash(x)=57300798)
703
+ 673 train 7.261487 (lr=9.8154e-05) (hash(x)=46105549)
704
+ 674 train 7.263463 (lr=9.7762e-05) (hash(x)=51612608)
705
+ 675 train 7.559962 (lr=9.7370e-05) (hash(x)=49761494)
706
+ 676 train 7.324200 (lr=9.6978e-05) (hash(x)=45375531)
707
+ 677 train 7.737980 (lr=9.6586e-05) (hash(x)=61065657)
708
+ 678 train 7.306383 (lr=9.6195e-05) (hash(x)=44152741)
709
+ 679 train 7.355774 (lr=9.5804e-05) (hash(x)=51631508)
710
+ 680 train 7.287109 (lr=9.5413e-05) (hash(x)=43922234)
711
+ 681 train 7.430236 (lr=9.5022e-05) (hash(x)=49689666)
712
+ 682 train 7.593435 (lr=9.4632e-05) (hash(x)=53751279)
713
+ 683 train 7.477890 (lr=9.4242e-05) (hash(x)=50400339)
714
+ 684 train 7.444281 (lr=9.3852e-05) (hash(x)=51029698)
715
+ 685 train 7.486862 (lr=9.3463e-05) (hash(x)=54240120)
716
+ 686 train 7.461469 (lr=9.3073e-05) (hash(x)=45708714)
717
+ 687 train 7.312148 (lr=9.2685e-05) (hash(x)=46560264)
718
+ 688 train 7.306785 (lr=9.2296e-05) (hash(x)=46841822)
719
+ 689 train 7.369553 (lr=9.1908e-05) (hash(x)=48013446)
720
+ 690 train 7.502515 (lr=9.1520e-05) (hash(x)=54564252)
721
+ 691 train 8.051479 (lr=9.1133e-05) (hash(x)=70534997)
722
+ 692 train 7.515357 (lr=9.0746e-05) (hash(x)=53363496)
723
+ 693 train 7.405049 (lr=9.0359e-05) (hash(x)=49508415)
724
+ 694 train 7.434710 (lr=8.9973e-05) (hash(x)=53503261)
725
+ 695 train 7.484006 (lr=8.9587e-05) (hash(x)=52420475)
726
+ 696 train 7.368260 (lr=8.9202e-05) (hash(x)=50191708)
727
+ 697 train 7.757932 (lr=8.8817e-05) (hash(x)=60279802)
728
+ 698 train 7.447044 (lr=8.8432e-05) (hash(x)=52626271)
729
+ 699 train 7.512129 (lr=8.8048e-05) (hash(x)=53194976)
730
+ 700 val loss 7.4423
731
+ 700 val perplexity 1706.6106
732
+ 700 train 7.486727 (lr=8.7664e-05) (hash(x)=50906515)
733
+ 701 train 7.342045 (lr=8.7281e-05) (hash(x)=49459146)
734
+ 702 train 7.361005 (lr=8.6898e-05) (hash(x)=48320631)
735
+ 703 train 7.781622 (lr=8.6515e-05) (hash(x)=66781899)
736
+ 704 train 7.804730 (lr=8.6133e-05) (hash(x)=57109514)
737
+ 705 train 8.026830 (lr=8.5751e-05) (hash(x)=43442176)
738
+ 706 train 7.565680 (lr=8.5370e-05) (hash(x)=51431734)
739
+ 707 train 7.435273 (lr=8.4990e-05) (hash(x)=51986059)
740
+ 708 train 7.574686 (lr=8.4610e-05) (hash(x)=54279714)
741
+ 709 train 7.414528 (lr=8.4230e-05) (hash(x)=45517898)
742
+ 710 train 7.992764 (lr=8.3851e-05) (hash(x)=57689755)
743
+ 711 train 7.686505 (lr=8.3472e-05) (hash(x)=60275167)
744
+ 712 train 7.378433 (lr=8.3094e-05) (hash(x)=46698776)
745
+ 713 train 7.816518 (lr=8.2716e-05) (hash(x)=61199882)
746
+ 714 train 7.356793 (lr=8.2339e-05) (hash(x)=48256229)
747
+ 715 train 7.365051 (lr=8.1963e-05) (hash(x)=45483557)
748
+ 716 train 7.447577 (lr=8.1587e-05) (hash(x)=50062280)
749
+ 717 train 7.772910 (lr=8.1211e-05) (hash(x)=49436293)
750
+ 718 train 7.420883 (lr=8.0836e-05) (hash(x)=47916352)
751
+ 719 train 7.382729 (lr=8.0462e-05) (hash(x)=49182582)
752
+ 720 train 7.454296 (lr=8.0088e-05) (hash(x)=50054568)
753
+ 721 train 7.565198 (lr=7.9715e-05) (hash(x)=54622543)
754
+ 722 train 7.465895 (lr=7.9342e-05) (hash(x)=53761778)
755
+ 723 train 7.363367 (lr=7.8970e-05) (hash(x)=51446091)
756
+ 724 train 7.412581 (lr=7.8599e-05) (hash(x)=50809582)
757
+ 725 train 7.250308 (lr=7.8228e-05) (hash(x)=45019935)
758
+ 726 train 7.269919 (lr=7.7858e-05) (hash(x)=44061718)
759
+ 727 train 7.302129 (lr=7.7488e-05) (hash(x)=47827767)
760
+ 728 train 7.469113 (lr=7.7119e-05) (hash(x)=51054368)
761
+ 729 train 7.232515 (lr=7.6751e-05) (hash(x)=47297652)
762
+ 730 train 7.295374 (lr=7.6383e-05) (hash(x)=45486271)
763
+ 731 train 7.463426 (lr=7.6016e-05) (hash(x)=46988810)
764
+ 732 train 7.201756 (lr=7.5650e-05) (hash(x)=44578285)
765
+ 733 train 7.234397 (lr=7.5284e-05) (hash(x)=44457482)
766
+ 734 train 7.363039 (lr=7.4919e-05) (hash(x)=47273057)
767
+ 735 train 7.165349 (lr=7.4555e-05) (hash(x)=44918130)
768
+ 736 train 7.436283 (lr=7.4191e-05) (hash(x)=46825217)
769
+ 737 train 7.446614 (lr=7.3828e-05) (hash(x)=47962618)
770
+ 738 train 7.361891 (lr=7.3466e-05) (hash(x)=45539511)
771
+ 739 train 7.454641 (lr=7.3105e-05) (hash(x)=49693485)
772
+ 740 train 7.416045 (lr=7.2744e-05) (hash(x)=49483023)
773
+ 741 train 7.263144 (lr=7.2384e-05) (hash(x)=46911489)
774
+ 742 train 7.415665 (lr=7.2024e-05) (hash(x)=52016723)
775
+ 743 train 7.148834 (lr=7.1666e-05) (hash(x)=40969008)
776
+ 744 train 7.273324 (lr=7.1308e-05) (hash(x)=48694211)
777
+ 745 train 7.255490 (lr=7.0950e-05) (hash(x)=47877745)
778
+ 746 train 7.449329 (lr=7.0594e-05) (hash(x)=53156899)
779
+ 747 train 7.498144 (lr=7.0238e-05) (hash(x)=50344401)
780
+ 748 train 7.536567 (lr=6.9884e-05) (hash(x)=46614361)
781
+ 749 train 7.427803 (lr=6.9529e-05) (hash(x)=53074032)
782
+ 750 val loss 7.4117
783
+ 750 val perplexity 1655.2217
784
+ 750 train 7.403168 (lr=6.9176e-05) (hash(x)=46046498)
785
+ 751 train 7.357453 (lr=6.8824e-05) (hash(x)=49833636)
786
+ 752 train 7.335294 (lr=6.8472e-05) (hash(x)=49165393)
787
+ 753 train 7.745853 (lr=6.8121e-05) (hash(x)=61445802)
788
+ 754 train 7.545136 (lr=6.7771e-05) (hash(x)=51012661)
789
+ 755 train 7.141525 (lr=6.7422e-05) (hash(x)=41326666)
790
+ 756 train 7.250174 (lr=6.7073e-05) (hash(x)=47169271)
791
+ 757 train 7.245010 (lr=6.6725e-05) (hash(x)=43474195)
792
+ 758 train 7.279457 (lr=6.6379e-05) (hash(x)=41683297)
793
+ 759 train 7.359571 (lr=6.6033e-05) (hash(x)=52346553)
794
+ 760 train 7.159418 (lr=6.5688e-05) (hash(x)=44385560)
795
+ 761 train 6.836008 (lr=6.5343e-05) (hash(x)=35732133)
796
+ 762 train 7.179364 (lr=6.5000e-05) (hash(x)=41423670)
797
+ 763 train 7.392880 (lr=6.4657e-05) (hash(x)=47970930)
798
+ 764 train 7.299768 (lr=6.4316e-05) (hash(x)=47092804)
799
+ 765 train 7.162051 (lr=6.3975e-05) (hash(x)=47182916)
800
+ 766 train 7.176779 (lr=6.3635e-05) (hash(x)=41791614)
801
+ 767 train 7.368469 (lr=6.3296e-05) (hash(x)=47009621)
802
+ 768 train 7.539227 (lr=6.2958e-05) (hash(x)=51633434)
803
+ 769 train 7.319921 (lr=6.2621e-05) (hash(x)=52003882)
804
+ 770 train 7.492424 (lr=6.2285e-05) (hash(x)=51340089)
805
+ 771 train 7.326988 (lr=6.1950e-05) (hash(x)=48602644)
806
+ 772 train 7.449613 (lr=6.1615e-05) (hash(x)=50457646)
807
+ 773 train 7.344125 (lr=6.1282e-05) (hash(x)=49970789)
808
+ 774 train 7.411069 (lr=6.0949e-05) (hash(x)=51611455)
809
+ 775 train 7.380174 (lr=6.0618e-05) (hash(x)=52115732)
810
+ 776 train 7.242937 (lr=6.0287e-05) (hash(x)=49032269)
811
+ 777 train 7.442569 (lr=5.9958e-05) (hash(x)=50715961)
812
+ 778 train 7.342155 (lr=5.9629e-05) (hash(x)=49693375)
813
+ 779 train 7.728542 (lr=5.9301e-05) (hash(x)=56527845)
814
+ 780 train 7.241998 (lr=5.8974e-05) (hash(x)=48381855)
815
+ 781 train 7.291766 (lr=5.8649e-05) (hash(x)=49623946)
816
+ 782 train 7.156475 (lr=5.8324e-05) (hash(x)=47393528)
817
+ 783 train 7.435060 (lr=5.8000e-05) (hash(x)=49459868)
818
+ 784 train 7.325522 (lr=5.7678e-05) (hash(x)=48696657)
819
+ 785 train 7.386584 (lr=5.7356e-05) (hash(x)=52550306)
820
+ 786 train 7.426018 (lr=5.7035e-05) (hash(x)=49285282)
821
+ 787 train 7.445265 (lr=5.6716e-05) (hash(x)=48955151)
822
+ 788 train 7.299976 (lr=5.6397e-05) (hash(x)=43743600)
823
+ 789 train 7.458034 (lr=5.6079e-05) (hash(x)=49655015)
824
+ 790 train 7.344597 (lr=5.5763e-05) (hash(x)=51090440)
825
+ 791 train 7.284200 (lr=5.5447e-05) (hash(x)=48473450)
826
+ 792 train 7.494042 (lr=5.5133e-05) (hash(x)=47938317)
827
+ 793 train 7.493947 (lr=5.4820e-05) (hash(x)=45290633)
828
+ 794 train 7.188344 (lr=5.4507e-05) (hash(x)=48542673)
829
+ 795 train 7.396185 (lr=5.4196e-05) (hash(x)=50306722)
830
+ 796 train 7.320820 (lr=5.3886e-05) (hash(x)=50632201)
831
+ 797 train 7.279037 (lr=5.3577e-05) (hash(x)=48795732)
832
+ 798 train 7.282547 (lr=5.3269e-05) (hash(x)=50141432)
833
+ 799 train 7.610158 (lr=5.2962e-05) (hash(x)=50898037)
834
+ 800 val loss 7.3849
835
+ 800 val perplexity 1611.4460
836
+ 800 train 7.474757 (lr=5.2656e-05) (hash(x)=37427531)
837
+ 801 train 7.444345 (lr=5.2352e-05) (hash(x)=32890763)
838
+ 802 train 7.517336 (lr=5.2048e-05) (hash(x)=31307609)
839
+ 803 train 7.282603 (lr=5.1746e-05) (hash(x)=35015807)
840
+ 804 train 7.433179 (lr=5.1444e-05) (hash(x)=37542227)
841
+ 805 train 7.490714 (lr=5.1144e-05) (hash(x)=41169444)
842
+ 806 train 7.296855 (lr=5.0845e-05) (hash(x)=47354793)
843
+ 807 train 7.381520 (lr=5.0547e-05) (hash(x)=46327297)
844
+ 808 train 7.391984 (lr=5.0251e-05) (hash(x)=50124976)
845
+ 809 train 7.272499 (lr=4.9955e-05) (hash(x)=48536826)
846
+ 810 train 7.220053 (lr=4.9661e-05) (hash(x)=43048468)
847
+ 811 train 7.291720 (lr=4.9367e-05) (hash(x)=44297138)
848
+ 812 train 8.433998 (lr=4.9075e-05) (hash(x)=46976056)
849
+ 813 train 7.525517 (lr=4.8784e-05) (hash(x)=48533607)
850
+ 814 train 7.119399 (lr=4.8495e-05) (hash(x)=45581560)
851
+ 815 train 7.160219 (lr=4.8206e-05) (hash(x)=47776613)
852
+ 816 train 7.083488 (lr=4.7919e-05) (hash(x)=39451311)
853
+ 817 train 7.670686 (lr=4.7633e-05) (hash(x)=53854198)
854
+ 818 train 7.334175 (lr=4.7348e-05) (hash(x)=47110254)
855
+ 819 train 7.394891 (lr=4.7064e-05) (hash(x)=51262732)
856
+ 820 train 7.339790 (lr=4.6782e-05) (hash(x)=48675819)
857
+ 821 train 7.383459 (lr=4.6501e-05) (hash(x)=47159040)
858
+ 822 train 7.473578 (lr=4.6221e-05) (hash(x)=52808063)
859
+ 823 train 7.356369 (lr=4.5942e-05) (hash(x)=51480608)
860
+ 824 train 7.152187 (lr=4.5664e-05) (hash(x)=46075661)
861
+ 825 train 7.278708 (lr=4.5388e-05) (hash(x)=44193147)
862
+ 826 train 7.280488 (lr=4.5113e-05) (hash(x)=51387298)
863
+ 827 train 7.421686 (lr=4.4839e-05) (hash(x)=50929448)
864
+ 828 train 7.474228 (lr=4.4567e-05) (hash(x)=54393657)
865
+ 829 train 7.214499 (lr=4.4295e-05) (hash(x)=48420333)
866
+ 830 train 7.499422 (lr=4.4025e-05) (hash(x)=54428349)
867
+ 831 train 7.340552 (lr=4.3757e-05) (hash(x)=46494961)
868
+ 832 train 7.253775 (lr=4.3489e-05) (hash(x)=50441352)
869
+ 833 train 7.631042 (lr=4.3223e-05) (hash(x)=49410253)
870
+ 834 train 7.520402 (lr=4.2958e-05) (hash(x)=54977154)
871
+ 835 train 7.499618 (lr=4.2695e-05) (hash(x)=59634178)
872
+ 836 train 7.507488 (lr=4.2432e-05) (hash(x)=51875554)
873
+ 837 train 7.329817 (lr=4.2171e-05) (hash(x)=48922609)
874
+ 838 train 7.223741 (lr=4.1912e-05) (hash(x)=46512842)
875
+ 839 train 7.283721 (lr=4.1654e-05) (hash(x)=47517395)
876
+ 840 train 7.382298 (lr=4.1397e-05) (hash(x)=51480081)
877
+ 841 train 7.316738 (lr=4.1141e-05) (hash(x)=49531327)
878
+ 842 train 7.661731 (lr=4.0887e-05) (hash(x)=50654362)
879
+ 843 train 7.836868 (lr=4.0634e-05) (hash(x)=48371589)
880
+ 844 train 7.665013 (lr=4.0382e-05) (hash(x)=57169652)
881
+ 845 train 7.717676 (lr=4.0132e-05) (hash(x)=55313671)
882
+ 846 train 7.309360 (lr=3.9883e-05) (hash(x)=45244232)
883
+ 847 train 7.655797 (lr=3.9635e-05) (hash(x)=61097106)
884
+ 848 train 7.351036 (lr=3.9389e-05) (hash(x)=51926581)
885
+ 849 train 7.335475 (lr=3.9144e-05) (hash(x)=52305923)
886
+ 850 val loss 7.3852
887
+ 850 val perplexity 1612.0162
888
+ 850 train 7.331872 (lr=3.8901e-05) (hash(x)=48839174)
889
+ 851 train 7.676151 (lr=3.8659e-05) (hash(x)=61405427)
890
+ 852 train 7.680559 (lr=3.8418e-05) (hash(x)=61444306)
891
+ 853 train 7.554700 (lr=3.8178e-05) (hash(x)=59898062)
892
+ 854 train 7.531210 (lr=3.7941e-05) (hash(x)=54795546)
893
+ 855 train 7.355371 (lr=3.7704e-05) (hash(x)=50139108)
894
+ 856 train 7.274999 (lr=3.7469e-05) (hash(x)=48975659)
895
+ 857 train 7.359896 (lr=3.7235e-05) (hash(x)=47955919)
896
+ 858 train 7.305573 (lr=3.7003e-05) (hash(x)=48949357)
897
+ 859 train 7.184465 (lr=3.6772e-05) (hash(x)=45467872)
898
+ 860 train 7.142914 (lr=3.6542e-05) (hash(x)=42449398)
899
+ 861 train 7.597887 (lr=3.6314e-05) (hash(x)=59686276)
900
+ 862 train 7.415793 (lr=3.6088e-05) (hash(x)=51946811)
901
+ 863 train 7.426998 (lr=3.5862e-05) (hash(x)=49686665)
902
+ 864 train 7.871588 (lr=3.5639e-05) (hash(x)=65282589)
903
+ 865 train 7.240194 (lr=3.5416e-05) (hash(x)=47341207)
904
+ 866 train 7.731713 (lr=3.5195e-05) (hash(x)=63814513)
905
+ 867 train 7.217216 (lr=3.4976e-05) (hash(x)=51019411)
906
+ 868 train 7.092075 (lr=3.4758e-05) (hash(x)=46653745)
907
+ 869 train 7.227548 (lr=3.4541e-05) (hash(x)=50050759)
908
+ 870 train 7.538797 (lr=3.4326e-05) (hash(x)=53878448)
909
+ 871 train 7.315566 (lr=3.4113e-05) (hash(x)=49380702)
910
+ 872 train 7.457017 (lr=3.3900e-05) (hash(x)=50922302)
911
+ 873 train 7.345914 (lr=3.3690e-05) (hash(x)=50963362)
912
+ 874 train 7.507467 (lr=3.3480e-05) (hash(x)=52460631)
913
+ 875 train 7.298195 (lr=3.3273e-05) (hash(x)=50156382)
914
+ 876 train 7.481524 (lr=3.3066e-05) (hash(x)=44475078)
915
+ 877 train 7.599507 (lr=3.2862e-05) (hash(x)=49995857)
916
+ 878 train 7.790624 (lr=3.2658e-05) (hash(x)=57032860)
917
+ 879 train 7.435220 (lr=3.2457e-05) (hash(x)=48775872)
918
+ 880 train 7.299661 (lr=3.2256e-05) (hash(x)=47643742)
919
+ 881 train 7.423371 (lr=3.2058e-05) (hash(x)=53183314)
920
+ 882 train 7.370245 (lr=3.1860e-05) (hash(x)=50268180)
921
+ 883 train 7.323378 (lr=3.1665e-05) (hash(x)=49497141)
922
+ 884 train 7.143096 (lr=3.1471e-05) (hash(x)=43420753)
923
+ 885 train 7.425904 (lr=3.1278e-05) (hash(x)=51207084)
924
+ 886 train 6.946695 (lr=3.1087e-05) (hash(x)=34492530)
925
+ 887 train 7.303625 (lr=3.0897e-05) (hash(x)=49456326)
926
+ 888 train 7.748999 (lr=3.0709e-05) (hash(x)=50766672)
927
+ 889 train 7.563818 (lr=3.0522e-05) (hash(x)=49811760)
928
+ 890 train 7.302639 (lr=3.0337e-05) (hash(x)=48142033)
929
+ 891 train 7.418552 (lr=3.0154e-05) (hash(x)=53343959)
930
+ 892 train 7.177756 (lr=2.9972e-05) (hash(x)=46916209)
931
+ 893 train 7.489772 (lr=2.9791e-05) (hash(x)=54858850)
932
+ 894 train 7.426958 (lr=2.9613e-05) (hash(x)=51700707)
933
+ 895 train 7.233140 (lr=2.9435e-05) (hash(x)=44654425)
934
+ 896 train 7.413579 (lr=2.9260e-05) (hash(x)=52740006)
935
+ 897 train 7.231884 (lr=2.9085e-05) (hash(x)=45260452)
936
+ 898 train 7.246109 (lr=2.8913e-05) (hash(x)=47151929)
937
+ 899 train 7.476698 (lr=2.8742e-05) (hash(x)=51367775)
938
+ 900 val loss 7.3605
939
+ 900 val perplexity 1572.6396
940
+ 900 train 7.377492 (lr=2.8572e-05) (hash(x)=50325207)
941
+ 901 train 7.259298 (lr=2.8404e-05) (hash(x)=46562023)
942
+ 902 train 7.401593 (lr=2.8238e-05) (hash(x)=48332912)
943
+ 903 train 7.400344 (lr=2.8073e-05) (hash(x)=54477325)
944
+ 904 train 7.452058 (lr=2.7910e-05) (hash(x)=49389150)
945
+ 905 train 7.291712 (lr=2.7749e-05) (hash(x)=48571772)
946
+ 906 train 7.469693 (lr=2.7589e-05) (hash(x)=53218072)
947
+ 907 train 7.225865 (lr=2.7430e-05) (hash(x)=45888568)
948
+ 908 train 7.396409 (lr=2.7274e-05) (hash(x)=50792965)
949
+ 909 train 7.408258 (lr=2.7119e-05) (hash(x)=49219751)
950
+ 910 train 7.292919 (lr=2.6965e-05) (hash(x)=49516452)
951
+ 911 train 7.283922 (lr=2.6813e-05) (hash(x)=49384161)
952
+ 912 train 7.252263 (lr=2.6663e-05) (hash(x)=45403134)
953
+ 913 train 7.282631 (lr=2.6514e-05) (hash(x)=50565032)
954
+ 914 train 7.480428 (lr=2.6367e-05) (hash(x)=51152823)
955
+ 915 train 7.540170 (lr=2.6221e-05) (hash(x)=52839038)
956
+ 916 train 7.121208 (lr=2.6077e-05) (hash(x)=44409465)
957
+ 917 train 7.523247 (lr=2.5935e-05) (hash(x)=51799409)
958
+ 918 train 7.279533 (lr=2.5795e-05) (hash(x)=51152758)
959
+ 919 train 7.338554 (lr=2.5656e-05) (hash(x)=49188683)
960
+ 920 train 7.298937 (lr=2.5518e-05) (hash(x)=50284940)
961
+ 921 train 7.386201 (lr=2.5383e-05) (hash(x)=51812773)
962
+ 922 train 7.468254 (lr=2.5249e-05) (hash(x)=53642545)
963
+ 923 train 7.645918 (lr=2.5116e-05) (hash(x)=62547001)
964
+ 924 train 7.364452 (lr=2.4985e-05) (hash(x)=47242049)
965
+ 925 train 7.338970 (lr=2.4856e-05) (hash(x)=49563491)
966
+ 926 train 7.406125 (lr=2.4729e-05) (hash(x)=50947272)
967
+ 927 train 7.481740 (lr=2.4603e-05) (hash(x)=51229664)
968
+ 928 train 7.251749 (lr=2.4479e-05) (hash(x)=47237060)
969
+ 929 train 7.209367 (lr=2.4356e-05) (hash(x)=46252668)
970
+ 930 train 7.332854 (lr=2.4235e-05) (hash(x)=43781107)
971
+ 931 train 7.191736 (lr=2.4116e-05) (hash(x)=48746317)
972
+ 932 train 7.932748 (lr=2.3998e-05) (hash(x)=67646788)
973
+ 933 train 7.061550 (lr=2.3883e-05) (hash(x)=44204094)
974
+ 934 train 7.300662 (lr=2.3768e-05) (hash(x)=50776803)
975
+ 935 train 7.503706 (lr=2.3656e-05) (hash(x)=51961195)
976
+ 936 train 7.616617 (lr=2.3545e-05) (hash(x)=48943726)
977
+ 937 train 7.553282 (lr=2.3436e-05) (hash(x)=47683918)
978
+ 938 train 7.720181 (lr=2.3328e-05) (hash(x)=50300577)
979
+ 939 train 7.605381 (lr=2.3222e-05) (hash(x)=49881442)
980
+ 940 train 7.407542 (lr=2.3118e-05) (hash(x)=46953117)
981
+ 941 train 7.599292 (lr=2.3016e-05) (hash(x)=45500608)
982
+ 942 train 7.688606 (lr=2.2915e-05) (hash(x)=56801743)
983
+ 943 train 7.226922 (lr=2.2816e-05) (hash(x)=48726061)
984
+ 944 train 7.266284 (lr=2.2718e-05) (hash(x)=46878089)
985
+ 945 train 7.379684 (lr=2.2623e-05) (hash(x)=53235305)
986
+ 946 train 7.320511 (lr=2.2528e-05) (hash(x)=50160069)
987
+ 947 train 7.487619 (lr=2.2436e-05) (hash(x)=50492385)
988
+ 948 train 7.260494 (lr=2.2345e-05) (hash(x)=49291969)
989
+ 949 train 7.216304 (lr=2.2256e-05) (hash(x)=46951052)
990
+ 950 val loss 7.3575
991
+ 950 val perplexity 1567.9277
992
+ 950 train 7.194428 (lr=2.2169e-05) (hash(x)=46580257)
993
+ 951 train 7.458827 (lr=2.2084e-05) (hash(x)=53617083)
994
+ 952 train 7.335783 (lr=2.2000e-05) (hash(x)=50964186)
995
+ 953 train 7.355107 (lr=2.1918e-05) (hash(x)=50267127)
996
+ 954 train 7.159370 (lr=2.1837e-05) (hash(x)=43627333)
997
+ 955 train 7.178555 (lr=2.1758e-05) (hash(x)=45972461)
998
+ 956 train 7.155941 (lr=2.1681e-05) (hash(x)=45851643)
999
+ 957 train 7.569690 (lr=2.1606e-05) (hash(x)=55325922)
1000
+ 958 train 8.057076 (lr=2.1532e-05) (hash(x)=61621353)
1001
+ 959 train 8.062300 (lr=2.1461e-05) (hash(x)=60830807)
1002
+ 960 train 7.990614 (lr=2.1390e-05) (hash(x)=59780537)
1003
+ 961 train 8.035106 (lr=2.1322e-05) (hash(x)=59357720)
1004
+ 962 train 7.972476 (lr=2.1255e-05) (hash(x)=60305936)
1005
+ 963 train 7.999022 (lr=2.1190e-05) (hash(x)=63145389)
1006
+ 964 train 7.943399 (lr=2.1127e-05) (hash(x)=62011447)
1007
+ 965 train 7.949921 (lr=2.1065e-05) (hash(x)=60412496)
1008
+ 966 train 8.014411 (lr=2.1005e-05) (hash(x)=62740258)
1009
+ 967 train 8.054291 (lr=2.0947e-05) (hash(x)=64013010)
1010
+ 968 train 7.950397 (lr=2.0891e-05) (hash(x)=59176481)
1011
+ 969 train 8.154876 (lr=2.0836e-05) (hash(x)=62177701)
1012
+ 970 train 7.965341 (lr=2.0783e-05) (hash(x)=60908204)
1013
+ 971 train 8.127275 (lr=2.0732e-05) (hash(x)=64418102)
1014
+ 972 train 8.025210 (lr=2.0682e-05) (hash(x)=60282281)
1015
+ 973 train 7.466790 (lr=2.0634e-05) (hash(x)=52915348)
1016
+ 974 train 7.480574 (lr=2.0588e-05) (hash(x)=49995009)
1017
+ 975 train 7.574721 (lr=2.0544e-05) (hash(x)=54461658)
1018
+ 976 train 7.800498 (lr=2.0501e-05) (hash(x)=63990903)
1019
+ 977 train 7.494689 (lr=2.0460e-05) (hash(x)=47729553)
1020
+ 978 train 7.403316 (lr=2.0421e-05) (hash(x)=50140658)
1021
+ 979 train 7.260858 (lr=2.0384e-05) (hash(x)=49407427)
1022
+ 980 train 7.408459 (lr=2.0348e-05) (hash(x)=51411433)
1023
+ 981 train 7.324688 (lr=2.0314e-05) (hash(x)=51432179)
1024
+ 982 train 7.338339 (lr=2.0282e-05) (hash(x)=51394518)
1025
+ 983 train 7.257405 (lr=2.0252e-05) (hash(x)=48800368)
1026
+ 984 train 7.595236 (lr=2.0223e-05) (hash(x)=51807114)
1027
+ 985 train 7.262997 (lr=2.0196e-05) (hash(x)=45866136)
1028
+ 986 train 7.247237 (lr=2.0171e-05) (hash(x)=45069575)
1029
+ 987 train 7.373635 (lr=2.0147e-05) (hash(x)=54376909)
1030
+ 988 train 7.508296 (lr=2.0125e-05) (hash(x)=59343246)
1031
+ 989 train 7.422824 (lr=2.0105e-05) (hash(x)=53519805)
1032
+ 990 train 7.390272 (lr=2.0087e-05) (hash(x)=49346697)
1033
+ 991 train 7.474020 (lr=2.0071e-05) (hash(x)=49838333)
1034
+ 992 train 7.185272 (lr=2.0056e-05) (hash(x)=45831325)
1035
+ 993 train 7.272745 (lr=2.0043e-05) (hash(x)=48643653)
1036
+ 994 train 7.418303 (lr=2.0031e-05) (hash(x)=54516111)
1037
+ 995 train 10.597061 (lr=2.0022e-05) (hash(x)=111351979)
1038
+ 996 train 11.618749 (lr=2.0014e-05) (hash(x)=127687623)
1039
+ 997 train 11.606437 (lr=2.0008e-05) (hash(x)=132049261)
1040
+ 998 train 12.063709 (lr=2.0003e-05) (hash(x)=141228471)
1041
+ 999 val loss 7.3487
1042
+ 999 val perplexity 1554.1748
1043
+ 999 train 10.956384 (lr=2.0001e-05) (hash(x)=119446440)
lr2e-4_total_batch_size10240_seq_len1024/model_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03550f17af0a3f611bfd4a5da5da74511134925622050bac4ed4ac8f5a17fb85
3
+ size 86167106
lr2e-4_total_batch_size10240_seq_len1024/optimizer_00999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f24b019824c198e8f01eaf7848ac9e1331912d703d0a0952b46e87498363ddda
3
+ size 71681862