Upload folder using huggingface_hub
Browse files- lr11e-4_total_batch_size122880_baseline_seed1340/args.json +1 -0
- lr11e-4_total_batch_size122880_baseline_seed1340/dataloader_04374.pt +3 -0
- lr11e-4_total_batch_size122880_baseline_seed1340/log2.txt +529 -0
- lr11e-4_total_batch_size122880_baseline_seed1340/model_04374.pt +3 -0
- lr11e-4_total_batch_size122880_baseline_seed1340/optimizer_04374.pt +3 -0
lr11e-4_total_batch_size122880_baseline_seed1340/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/lr11e-4_total_batch_size122880_baseline_seed1340", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 4, "protect_bos_token": true, "prevent_from_masking_myself": true, "max_steps": 4375, "warmup_steps": 250, "group": "wider_is_better_11", "use_wandb": true, "kill_self_after_run": false, "random_seed": 1340, "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": 256, "batch_size": 120, "total_batch_size": 122880, "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": false, "mup_enable_coord_check_logging": false, "max_lr": 0.0011, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "11e-4_122880", "n_embd": 256}
|
lr11e-4_total_batch_size122880_baseline_seed1340/dataloader_04374.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea4dadf39b77e4add965f5ecc2364cea20935ceda972d0ad2e4a448c01101668
|
| 3 |
+
size 964
|
lr11e-4_total_batch_size122880_baseline_seed1340/log2.txt
ADDED
|
@@ -0,0 +1,529 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
max_steps: 4375
|
| 2 |
+
0 val loss 11.2471
|
| 3 |
+
0 val perplexity 76658.6719
|
| 4 |
+
0 train 11.263495 (lr=4.4000e-06) (hash(x)=164406924)
|
| 5 |
+
10 train 9.959173 (lr=4.8400e-05) (hash(x)=153494457)
|
| 6 |
+
20 train 9.442564 (lr=9.2400e-05) (hash(x)=159333245)
|
| 7 |
+
30 train 8.638273 (lr=1.3640e-04) (hash(x)=127877799)
|
| 8 |
+
40 train 8.002560 (lr=1.8040e-04) (hash(x)=160249377)
|
| 9 |
+
50 train 7.651460 (lr=2.2440e-04) (hash(x)=144903932)
|
| 10 |
+
60 train 7.604034 (lr=2.6840e-04) (hash(x)=152347864)
|
| 11 |
+
70 train 7.540380 (lr=3.1240e-04) (hash(x)=142121048)
|
| 12 |
+
80 train 7.366329 (lr=3.5640e-04) (hash(x)=158755931)
|
| 13 |
+
90 train 7.320047 (lr=4.0040e-04) (hash(x)=179696111)
|
| 14 |
+
100 val loss 7.1040
|
| 15 |
+
100 val perplexity 1216.7859
|
| 16 |
+
100 train 7.074910 (lr=4.4440e-04) (hash(x)=167734596)
|
| 17 |
+
110 train 7.102465 (lr=4.8840e-04) (hash(x)=152234670)
|
| 18 |
+
120 train 6.831584 (lr=5.3240e-04) (hash(x)=139367591)
|
| 19 |
+
130 train 6.757787 (lr=5.7640e-04) (hash(x)=156667529)
|
| 20 |
+
140 train 6.610012 (lr=6.2040e-04) (hash(x)=151939311)
|
| 21 |
+
150 train 6.585110 (lr=6.6440e-04) (hash(x)=149619098)
|
| 22 |
+
160 train 6.444356 (lr=7.0840e-04) (hash(x)=151878766)
|
| 23 |
+
170 train 6.498874 (lr=7.5240e-04) (hash(x)=160761498)
|
| 24 |
+
180 train 6.378943 (lr=7.9640e-04) (hash(x)=143063312)
|
| 25 |
+
190 train 6.228820 (lr=8.4040e-04) (hash(x)=149579175)
|
| 26 |
+
200 val loss 6.2129
|
| 27 |
+
200 val perplexity 499.1441
|
| 28 |
+
200 train 6.096167 (lr=8.8440e-04) (hash(x)=153710890)
|
| 29 |
+
210 train 6.022976 (lr=9.2840e-04) (hash(x)=141112989)
|
| 30 |
+
220 train 6.022707 (lr=9.7240e-04) (hash(x)=157683479)
|
| 31 |
+
230 train 6.042252 (lr=1.0164e-03) (hash(x)=152133659)
|
| 32 |
+
240 train 5.875635 (lr=1.0604e-03) (hash(x)=148422063)
|
| 33 |
+
250 train 5.895082 (lr=1.1000e-03) (hash(x)=145450636)
|
| 34 |
+
260 train 6.006030 (lr=1.1000e-03) (hash(x)=144256573)
|
| 35 |
+
270 train 5.880665 (lr=1.0999e-03) (hash(x)=149347480)
|
| 36 |
+
280 train 5.816213 (lr=1.0999e-03) (hash(x)=143045783)
|
| 37 |
+
290 train 5.733803 (lr=1.0998e-03) (hash(x)=158018166)
|
| 38 |
+
300 val loss 5.7201
|
| 39 |
+
300 val perplexity 304.9480
|
| 40 |
+
300 train 5.693263 (lr=1.0996e-03) (hash(x)=145249251)
|
| 41 |
+
310 train 5.723708 (lr=1.0995e-03) (hash(x)=134768519)
|
| 42 |
+
320 train 5.594657 (lr=1.0993e-03) (hash(x)=148444988)
|
| 43 |
+
330 train 5.526741 (lr=1.0991e-03) (hash(x)=145121305)
|
| 44 |
+
340 train 5.466708 (lr=1.0988e-03) (hash(x)=137272598)
|
| 45 |
+
350 train 5.608034 (lr=1.0986e-03) (hash(x)=150475545)
|
| 46 |
+
360 train 5.444489 (lr=1.0983e-03) (hash(x)=143328717)
|
| 47 |
+
370 train 5.382236 (lr=1.0979e-03) (hash(x)=149452392)
|
| 48 |
+
380 train 5.325267 (lr=1.0976e-03) (hash(x)=153305312)
|
| 49 |
+
390 train 5.352235 (lr=1.0972e-03) (hash(x)=151035519)
|
| 50 |
+
400 val loss 5.3859
|
| 51 |
+
400 val perplexity 218.3094
|
| 52 |
+
400 train 5.338007 (lr=1.0968e-03) (hash(x)=144483776)
|
| 53 |
+
410 train 5.345759 (lr=1.0963e-03) (hash(x)=157946886)
|
| 54 |
+
420 train 5.364297 (lr=1.0959e-03) (hash(x)=144514796)
|
| 55 |
+
430 train 5.320783 (lr=1.0954e-03) (hash(x)=151365298)
|
| 56 |
+
440 train 5.313928 (lr=1.0948e-03) (hash(x)=151935250)
|
| 57 |
+
450 train 5.284443 (lr=1.0943e-03) (hash(x)=157916369)
|
| 58 |
+
460 train 5.193489 (lr=1.0937e-03) (hash(x)=145383083)
|
| 59 |
+
470 train 5.250412 (lr=1.0931e-03) (hash(x)=151846396)
|
| 60 |
+
480 train 5.181935 (lr=1.0924e-03) (hash(x)=136782937)
|
| 61 |
+
490 train 5.372674 (lr=1.0918e-03) (hash(x)=171270168)
|
| 62 |
+
500 val loss 5.1750
|
| 63 |
+
500 val perplexity 176.8017
|
| 64 |
+
500 train 5.108829 (lr=1.0911e-03) (hash(x)=154856891)
|
| 65 |
+
510 train 5.041064 (lr=1.0903e-03) (hash(x)=144618667)
|
| 66 |
+
520 train 5.017812 (lr=1.0896e-03) (hash(x)=147217952)
|
| 67 |
+
530 train 5.055416 (lr=1.0888e-03) (hash(x)=147473652)
|
| 68 |
+
540 train 4.956928 (lr=1.0880e-03) (hash(x)=155595779)
|
| 69 |
+
550 train 4.912854 (lr=1.0871e-03) (hash(x)=136191502)
|
| 70 |
+
560 train 5.124794 (lr=1.0863e-03) (hash(x)=144828302)
|
| 71 |
+
570 train 5.130631 (lr=1.0854e-03) (hash(x)=166352243)
|
| 72 |
+
580 train 5.063017 (lr=1.0844e-03) (hash(x)=178082599)
|
| 73 |
+
590 train 5.096383 (lr=1.0835e-03) (hash(x)=151910947)
|
| 74 |
+
600 val loss 4.9926
|
| 75 |
+
600 val perplexity 147.3121
|
| 76 |
+
600 train 5.021742 (lr=1.0825e-03) (hash(x)=148021541)
|
| 77 |
+
610 train 4.981276 (lr=1.0815e-03) (hash(x)=156983220)
|
| 78 |
+
620 train 4.994115 (lr=1.0805e-03) (hash(x)=171125590)
|
| 79 |
+
630 train 4.953625 (lr=1.0794e-03) (hash(x)=150132098)
|
| 80 |
+
640 train 4.925951 (lr=1.0783e-03) (hash(x)=148900016)
|
| 81 |
+
650 train 4.852721 (lr=1.0772e-03) (hash(x)=146005217)
|
| 82 |
+
660 train 4.842213 (lr=1.0761e-03) (hash(x)=165903661)
|
| 83 |
+
670 train 4.919178 (lr=1.0749e-03) (hash(x)=176281294)
|
| 84 |
+
680 train 4.834496 (lr=1.0737e-03) (hash(x)=144362722)
|
| 85 |
+
690 train 4.808679 (lr=1.0725e-03) (hash(x)=193332654)
|
| 86 |
+
700 val loss 4.8368
|
| 87 |
+
700 val perplexity 126.0649
|
| 88 |
+
700 train 4.741785 (lr=1.0712e-03) (hash(x)=146019502)
|
| 89 |
+
710 train 4.752848 (lr=1.0699e-03) (hash(x)=149839636)
|
| 90 |
+
720 train 4.957731 (lr=1.0686e-03) (hash(x)=159787060)
|
| 91 |
+
730 train 4.740715 (lr=1.0673e-03) (hash(x)=164169521)
|
| 92 |
+
740 train 4.748223 (lr=1.0659e-03) (hash(x)=148916053)
|
| 93 |
+
750 train 4.854013 (lr=1.0645e-03) (hash(x)=150127281)
|
| 94 |
+
760 train 4.812323 (lr=1.0631e-03) (hash(x)=144237370)
|
| 95 |
+
770 train 4.656075 (lr=1.0617e-03) (hash(x)=150979737)
|
| 96 |
+
780 train 4.730696 (lr=1.0602e-03) (hash(x)=135769745)
|
| 97 |
+
790 train 4.630120 (lr=1.0587e-03) (hash(x)=166874637)
|
| 98 |
+
800 val loss 4.7108
|
| 99 |
+
800 val perplexity 111.1446
|
| 100 |
+
800 train 4.652043 (lr=1.0572e-03) (hash(x)=154120875)
|
| 101 |
+
810 train 4.581267 (lr=1.0557e-03) (hash(x)=144479755)
|
| 102 |
+
820 train 4.450714 (lr=1.0541e-03) (hash(x)=139779397)
|
| 103 |
+
830 train 4.449131 (lr=1.0525e-03) (hash(x)=150235814)
|
| 104 |
+
840 train 4.559576 (lr=1.0509e-03) (hash(x)=150039009)
|
| 105 |
+
850 train 4.466327 (lr=1.0492e-03) (hash(x)=155517086)
|
| 106 |
+
860 train 4.453397 (lr=1.0475e-03) (hash(x)=151196765)
|
| 107 |
+
870 train 4.663859 (lr=1.0458e-03) (hash(x)=150265348)
|
| 108 |
+
880 train 4.672001 (lr=1.0441e-03) (hash(x)=158062105)
|
| 109 |
+
890 train 4.665825 (lr=1.0424e-03) (hash(x)=142902137)
|
| 110 |
+
900 val loss 4.6138
|
| 111 |
+
900 val perplexity 100.8696
|
| 112 |
+
900 train 4.600214 (lr=1.0406e-03) (hash(x)=148079623)
|
| 113 |
+
910 train 4.557000 (lr=1.0388e-03) (hash(x)=152019195)
|
| 114 |
+
920 train 4.641799 (lr=1.0369e-03) (hash(x)=145809848)
|
| 115 |
+
930 train 4.600927 (lr=1.0351e-03) (hash(x)=140746613)
|
| 116 |
+
940 train 4.570846 (lr=1.0332e-03) (hash(x)=152188465)
|
| 117 |
+
950 train 4.484059 (lr=1.0313e-03) (hash(x)=152449118)
|
| 118 |
+
960 train 4.554606 (lr=1.0294e-03) (hash(x)=144851080)
|
| 119 |
+
970 train 4.477393 (lr=1.0274e-03) (hash(x)=150584184)
|
| 120 |
+
980 train 4.523669 (lr=1.0254e-03) (hash(x)=151779783)
|
| 121 |
+
990 train 4.494651 (lr=1.0234e-03) (hash(x)=142240442)
|
| 122 |
+
1000 val loss 4.5690
|
| 123 |
+
1000 val perplexity 96.4435
|
| 124 |
+
1000 train 5.071653 (lr=1.0214e-03) (hash(x)=130109171)
|
| 125 |
+
1010 train 4.426258 (lr=1.0194e-03) (hash(x)=148167283)
|
| 126 |
+
1020 train 4.253577 (lr=1.0173e-03) (hash(x)=145435376)
|
| 127 |
+
1030 train 4.412729 (lr=1.0152e-03) (hash(x)=153598736)
|
| 128 |
+
1040 train 4.586697 (lr=1.0131e-03) (hash(x)=194205227)
|
| 129 |
+
1050 train 4.470949 (lr=1.0109e-03) (hash(x)=177820556)
|
| 130 |
+
1060 train 4.272526 (lr=1.0088e-03) (hash(x)=151270506)
|
| 131 |
+
1070 train 4.511893 (lr=1.0066e-03) (hash(x)=148645541)
|
| 132 |
+
1080 train 4.555798 (lr=1.0044e-03) (hash(x)=149498205)
|
| 133 |
+
1090 train 4.499393 (lr=1.0021e-03) (hash(x)=158964519)
|
| 134 |
+
1100 val loss 4.5036
|
| 135 |
+
1100 val perplexity 90.3392
|
| 136 |
+
1100 train 4.478963 (lr=9.9985e-04) (hash(x)=148307533)
|
| 137 |
+
1110 train 4.526629 (lr=9.9757e-04) (hash(x)=174592633)
|
| 138 |
+
1120 train 4.468705 (lr=9.9526e-04) (hash(x)=152082559)
|
| 139 |
+
1130 train 4.419008 (lr=9.9293e-04) (hash(x)=146506423)
|
| 140 |
+
1140 train 4.471407 (lr=9.9058e-04) (hash(x)=169165940)
|
| 141 |
+
1150 train 4.468191 (lr=9.8820e-04) (hash(x)=135090644)
|
| 142 |
+
1160 train 4.481871 (lr=9.8580e-04) (hash(x)=149055060)
|
| 143 |
+
1170 train 4.446899 (lr=9.8338e-04) (hash(x)=157018311)
|
| 144 |
+
1180 train 4.344027 (lr=9.8094e-04) (hash(x)=151179765)
|
| 145 |
+
1190 train 4.378269 (lr=9.7848e-04) (hash(x)=145475470)
|
| 146 |
+
1200 val loss 4.4723
|
| 147 |
+
1200 val perplexity 87.5545
|
| 148 |
+
1200 train 4.419554 (lr=9.7599e-04) (hash(x)=149670497)
|
| 149 |
+
1210 train 4.412409 (lr=9.7349e-04) (hash(x)=140976189)
|
| 150 |
+
1220 train 4.194368 (lr=9.7096e-04) (hash(x)=161365540)
|
| 151 |
+
1230 train 4.315267 (lr=9.6841e-04) (hash(x)=143397542)
|
| 152 |
+
1240 train 4.198786 (lr=9.6584e-04) (hash(x)=155314125)
|
| 153 |
+
1250 train 4.341029 (lr=9.6325e-04) (hash(x)=161648789)
|
| 154 |
+
1260 train 4.223669 (lr=9.6064e-04) (hash(x)=150725605)
|
| 155 |
+
1270 train 4.425663 (lr=9.5800e-04) (hash(x)=156319617)
|
| 156 |
+
1280 train 4.408995 (lr=9.5535e-04) (hash(x)=142029552)
|
| 157 |
+
1290 train 4.347293 (lr=9.5268e-04) (hash(x)=150029405)
|
| 158 |
+
1300 val loss 4.4190
|
| 159 |
+
1300 val perplexity 83.0133
|
| 160 |
+
1300 train 4.442119 (lr=9.4998e-04) (hash(x)=149062648)
|
| 161 |
+
1310 train 4.431651 (lr=9.4727e-04) (hash(x)=159397363)
|
| 162 |
+
1320 train 4.423408 (lr=9.4454e-04) (hash(x)=133311045)
|
| 163 |
+
1330 train 4.368833 (lr=9.4178e-04) (hash(x)=147678176)
|
| 164 |
+
1340 train 4.422555 (lr=9.3901e-04) (hash(x)=135442534)
|
| 165 |
+
1350 train 4.381899 (lr=9.3622e-04) (hash(x)=142785200)
|
| 166 |
+
1360 train 4.286940 (lr=9.3341e-04) (hash(x)=141536832)
|
| 167 |
+
1370 train 4.346538 (lr=9.3058e-04) (hash(x)=142797516)
|
| 168 |
+
1380 train 4.380779 (lr=9.2773e-04) (hash(x)=146896151)
|
| 169 |
+
1390 train 4.284380 (lr=9.2486e-04) (hash(x)=147455544)
|
| 170 |
+
1400 val loss 4.3851
|
| 171 |
+
1400 val perplexity 80.2468
|
| 172 |
+
1400 train 4.318984 (lr=9.2197e-04) (hash(x)=145925829)
|
| 173 |
+
1410 train 4.492310 (lr=9.1907e-04) (hash(x)=147898061)
|
| 174 |
+
1420 train 4.270732 (lr=9.1615e-04) (hash(x)=144105804)
|
| 175 |
+
1430 train 4.215138 (lr=9.1321e-04) (hash(x)=151558022)
|
| 176 |
+
1440 train 4.299285 (lr=9.1025e-04) (hash(x)=155595753)
|
| 177 |
+
1450 train 4.149493 (lr=9.0727e-04) (hash(x)=152241381)
|
| 178 |
+
1460 train 4.122514 (lr=9.0428e-04) (hash(x)=142570404)
|
| 179 |
+
1470 train 4.357702 (lr=9.0127e-04) (hash(x)=148567811)
|
| 180 |
+
1480 train 4.405492 (lr=8.9824e-04) (hash(x)=175444499)
|
| 181 |
+
1490 train 4.328969 (lr=8.9519e-04) (hash(x)=144155303)
|
| 182 |
+
1500 val loss 4.3439
|
| 183 |
+
1500 val perplexity 77.0068
|
| 184 |
+
1500 train 4.338792 (lr=8.9213e-04) (hash(x)=154267584)
|
| 185 |
+
1510 train 4.356894 (lr=8.8905e-04) (hash(x)=145815808)
|
| 186 |
+
1520 train 4.343436 (lr=8.8595e-04) (hash(x)=154655799)
|
| 187 |
+
1530 train 4.293290 (lr=8.8284e-04) (hash(x)=162647390)
|
| 188 |
+
1540 train 4.357986 (lr=8.7971e-04) (hash(x)=145614679)
|
| 189 |
+
1550 train 4.325496 (lr=8.7657e-04) (hash(x)=147295831)
|
| 190 |
+
1560 train 4.352436 (lr=8.7341e-04) (hash(x)=144484366)
|
| 191 |
+
1570 train 4.221092 (lr=8.7023e-04) (hash(x)=147123612)
|
| 192 |
+
1580 train 4.236620 (lr=8.6704e-04) (hash(x)=155700054)
|
| 193 |
+
1590 train 4.233782 (lr=8.6384e-04) (hash(x)=151697986)
|
| 194 |
+
1600 val loss 4.3341
|
| 195 |
+
1600 val perplexity 76.2565
|
| 196 |
+
1600 train 4.212145 (lr=8.6062e-04) (hash(x)=152670423)
|
| 197 |
+
1610 train 4.304256 (lr=8.5738e-04) (hash(x)=165856653)
|
| 198 |
+
1620 train 4.162912 (lr=8.5413e-04) (hash(x)=144655963)
|
| 199 |
+
1630 train 4.386599 (lr=8.5087e-04) (hash(x)=153422816)
|
| 200 |
+
1640 train 4.780506 (lr=8.4759e-04) (hash(x)=142771608)
|
| 201 |
+
1650 train 4.381827 (lr=8.4429e-04) (hash(x)=166691112)
|
| 202 |
+
1660 train 4.207466 (lr=8.4099e-04) (hash(x)=147472491)
|
| 203 |
+
1670 train 4.166740 (lr=8.3767e-04) (hash(x)=137752245)
|
| 204 |
+
1680 train 4.256335 (lr=8.3433e-04) (hash(x)=141597224)
|
| 205 |
+
1690 train 4.242840 (lr=8.3098e-04) (hash(x)=149210215)
|
| 206 |
+
1700 val loss 4.3026
|
| 207 |
+
1700 val perplexity 73.8933
|
| 208 |
+
1700 train 4.263448 (lr=8.2762e-04) (hash(x)=138186236)
|
| 209 |
+
1710 train 4.132275 (lr=8.2425e-04) (hash(x)=152918191)
|
| 210 |
+
1720 train 4.239828 (lr=8.2086e-04) (hash(x)=150605863)
|
| 211 |
+
1730 train 4.202507 (lr=8.1746e-04) (hash(x)=147041401)
|
| 212 |
+
1740 train 4.125714 (lr=8.1405e-04) (hash(x)=141708584)
|
| 213 |
+
1750 train 4.213872 (lr=8.1063e-04) (hash(x)=153868644)
|
| 214 |
+
1760 train 4.281731 (lr=8.0720e-04) (hash(x)=149511279)
|
| 215 |
+
1770 train 4.290467 (lr=8.0375e-04) (hash(x)=147387110)
|
| 216 |
+
1780 train 4.313306 (lr=8.0029e-04) (hash(x)=142199822)
|
| 217 |
+
1790 train 4.313581 (lr=7.9682e-04) (hash(x)=139690565)
|
| 218 |
+
1800 val loss 4.2912
|
| 219 |
+
1800 val perplexity 73.0549
|
| 220 |
+
1800 train 4.315189 (lr=7.9334e-04) (hash(x)=151310379)
|
| 221 |
+
1810 train 4.187924 (lr=7.8985e-04) (hash(x)=170212583)
|
| 222 |
+
1820 train 4.157825 (lr=7.8635e-04) (hash(x)=150615613)
|
| 223 |
+
1830 train 4.243822 (lr=7.8283e-04) (hash(x)=172001687)
|
| 224 |
+
1840 train 4.208661 (lr=7.7931e-04) (hash(x)=146656471)
|
| 225 |
+
1850 train 4.149408 (lr=7.7578e-04) (hash(x)=140857897)
|
| 226 |
+
1860 train 4.150750 (lr=7.7223e-04) (hash(x)=147051882)
|
| 227 |
+
1870 train 4.195446 (lr=7.6868e-04) (hash(x)=136098619)
|
| 228 |
+
1880 train 4.201004 (lr=7.6512e-04) (hash(x)=132637728)
|
| 229 |
+
1890 train 4.246308 (lr=7.6154e-04) (hash(x)=158923722)
|
| 230 |
+
1900 val loss 4.2787
|
| 231 |
+
1900 val perplexity 72.1488
|
| 232 |
+
1900 train 4.251356 (lr=7.5796e-04) (hash(x)=146970345)
|
| 233 |
+
1910 train 4.269830 (lr=7.5437e-04) (hash(x)=149389070)
|
| 234 |
+
1920 train 4.401567 (lr=7.5078e-04) (hash(x)=150969708)
|
| 235 |
+
1930 train 4.214802 (lr=7.4717e-04) (hash(x)=166654385)
|
| 236 |
+
1940 train 4.295810 (lr=7.4355e-04) (hash(x)=169244273)
|
| 237 |
+
1950 train 4.055508 (lr=7.3993e-04) (hash(x)=139742292)
|
| 238 |
+
1960 train 4.124078 (lr=7.3630e-04) (hash(x)=142333795)
|
| 239 |
+
1970 train 4.153060 (lr=7.3266e-04) (hash(x)=146791447)
|
| 240 |
+
1980 train 4.119567 (lr=7.2901e-04) (hash(x)=154345047)
|
| 241 |
+
1990 train 4.095316 (lr=7.2536e-04) (hash(x)=136077430)
|
| 242 |
+
2000 val loss 4.2495
|
| 243 |
+
2000 val perplexity 70.0721
|
| 244 |
+
2000 train 4.130911 (lr=7.2170e-04) (hash(x)=147050448)
|
| 245 |
+
2010 train 4.123305 (lr=7.1803e-04) (hash(x)=113928828)
|
| 246 |
+
2020 train 4.145559 (lr=7.1436e-04) (hash(x)=139671346)
|
| 247 |
+
2030 train 4.164702 (lr=7.1068e-04) (hash(x)=153313957)
|
| 248 |
+
2040 train 4.268473 (lr=7.0699e-04) (hash(x)=155134200)
|
| 249 |
+
2050 train 4.137594 (lr=7.0330e-04) (hash(x)=132506281)
|
| 250 |
+
2060 train 4.262299 (lr=6.9960e-04) (hash(x)=159158331)
|
| 251 |
+
2070 train 4.244782 (lr=6.9590e-04) (hash(x)=147952838)
|
| 252 |
+
2080 train 4.196351 (lr=6.9219e-04) (hash(x)=150479791)
|
| 253 |
+
2090 train 4.050746 (lr=6.8848e-04) (hash(x)=151450859)
|
| 254 |
+
2100 val loss 4.2294
|
| 255 |
+
2100 val perplexity 68.6762
|
| 256 |
+
2100 train 4.161159 (lr=6.8476e-04) (hash(x)=169711226)
|
| 257 |
+
2110 train 4.176114 (lr=6.8104e-04) (hash(x)=137986803)
|
| 258 |
+
2120 train 4.128428 (lr=6.7731e-04) (hash(x)=141835006)
|
| 259 |
+
2130 train 4.126761 (lr=6.7358e-04) (hash(x)=144700487)
|
| 260 |
+
2140 train 4.102862 (lr=6.6984e-04) (hash(x)=145916546)
|
| 261 |
+
2150 train 4.081702 (lr=6.6610e-04) (hash(x)=147814678)
|
| 262 |
+
2160 train 4.084265 (lr=6.6236e-04) (hash(x)=157812488)
|
| 263 |
+
2170 train 4.133375 (lr=6.5862e-04) (hash(x)=149012593)
|
| 264 |
+
2180 train 3.969809 (lr=6.5487e-04) (hash(x)=156529822)
|
| 265 |
+
2190 train 4.090622 (lr=6.5111e-04) (hash(x)=145020725)
|
| 266 |
+
2200 val loss 4.2175
|
| 267 |
+
2200 val perplexity 67.8665
|
| 268 |
+
2200 train 4.210747 (lr=6.4736e-04) (hash(x)=169397902)
|
| 269 |
+
2210 train 4.218596 (lr=6.4360e-04) (hash(x)=153734629)
|
| 270 |
+
2220 train 4.132414 (lr=6.3984e-04) (hash(x)=147000975)
|
| 271 |
+
2230 train 4.137308 (lr=6.3608e-04) (hash(x)=146993657)
|
| 272 |
+
2240 train 4.284544 (lr=6.3232e-04) (hash(x)=144806819)
|
| 273 |
+
2250 train 4.043282 (lr=6.2855e-04) (hash(x)=152839663)
|
| 274 |
+
2260 train 4.107489 (lr=6.2479e-04) (hash(x)=137641448)
|
| 275 |
+
2270 train 4.085018 (lr=6.2102e-04) (hash(x)=162502026)
|
| 276 |
+
2280 train 4.154428 (lr=6.1725e-04) (hash(x)=147831529)
|
| 277 |
+
2290 train 4.123548 (lr=6.1348e-04) (hash(x)=156470392)
|
| 278 |
+
2300 val loss 4.2190
|
| 279 |
+
2300 val perplexity 67.9630
|
| 280 |
+
2300 train 4.107846 (lr=6.0971e-04) (hash(x)=142958493)
|
| 281 |
+
2310 train 4.133773 (lr=6.0594e-04) (hash(x)=138409756)
|
| 282 |
+
2320 train 4.089504 (lr=6.0217e-04) (hash(x)=136455908)
|
| 283 |
+
2330 train 3.979148 (lr=5.9840e-04) (hash(x)=143492694)
|
| 284 |
+
2340 train 4.102666 (lr=5.9463e-04) (hash(x)=142435917)
|
| 285 |
+
2350 train 4.171623 (lr=5.9086e-04) (hash(x)=147640974)
|
| 286 |
+
2360 train 4.198441 (lr=5.8710e-04) (hash(x)=147561724)
|
| 287 |
+
2370 train 4.169827 (lr=5.8333e-04) (hash(x)=141018440)
|
| 288 |
+
2380 train 4.178359 (lr=5.7956e-04) (hash(x)=151845653)
|
| 289 |
+
2390 train 4.123283 (lr=5.7580e-04) (hash(x)=158544872)
|
| 290 |
+
2400 val loss 4.1851
|
| 291 |
+
2400 val perplexity 65.7030
|
| 292 |
+
2400 train 4.146954 (lr=5.7204e-04) (hash(x)=142448868)
|
| 293 |
+
2410 train 4.184751 (lr=5.6828e-04) (hash(x)=144442479)
|
| 294 |
+
2420 train 4.093115 (lr=5.6452e-04) (hash(x)=146896536)
|
| 295 |
+
2430 train 4.162473 (lr=5.6076e-04) (hash(x)=153470546)
|
| 296 |
+
2440 train 4.051999 (lr=5.5701e-04) (hash(x)=142442410)
|
| 297 |
+
2450 train 4.020195 (lr=5.5326e-04) (hash(x)=148437502)
|
| 298 |
+
2460 train 3.941043 (lr=5.4951e-04) (hash(x)=142582804)
|
| 299 |
+
2470 train 4.167199 (lr=5.4577e-04) (hash(x)=153318139)
|
| 300 |
+
2480 train 4.354605 (lr=5.4203e-04) (hash(x)=152014464)
|
| 301 |
+
2490 train 4.162665 (lr=5.3829e-04) (hash(x)=150210054)
|
| 302 |
+
2500 val loss 4.1714
|
| 303 |
+
2500 val perplexity 64.8043
|
| 304 |
+
2500 train 4.141375 (lr=5.3455e-04) (hash(x)=158281813)
|
| 305 |
+
2510 train 4.135333 (lr=5.3082e-04) (hash(x)=137105682)
|
| 306 |
+
2520 train 4.251081 (lr=5.2710e-04) (hash(x)=123681591)
|
| 307 |
+
2530 train 4.264548 (lr=5.2338e-04) (hash(x)=184836557)
|
| 308 |
+
2540 train 4.176606 (lr=5.1966e-04) (hash(x)=141842972)
|
| 309 |
+
2550 train 4.108375 (lr=5.1595e-04) (hash(x)=143265724)
|
| 310 |
+
2560 train 4.081249 (lr=5.1225e-04) (hash(x)=159065874)
|
| 311 |
+
2570 train 4.105830 (lr=5.0855e-04) (hash(x)=144937230)
|
| 312 |
+
2580 train 3.931513 (lr=5.0485e-04) (hash(x)=146574710)
|
| 313 |
+
2590 train 4.032984 (lr=5.0116e-04) (hash(x)=145400708)
|
| 314 |
+
2600 val loss 4.1631
|
| 315 |
+
2600 val perplexity 64.2699
|
| 316 |
+
2600 train 3.970534 (lr=4.9748e-04) (hash(x)=141630135)
|
| 317 |
+
2610 train 4.044245 (lr=4.9380e-04) (hash(x)=152922497)
|
| 318 |
+
2620 train 3.998410 (lr=4.9013e-04) (hash(x)=145133153)
|
| 319 |
+
2630 train 4.035231 (lr=4.8647e-04) (hash(x)=156613033)
|
| 320 |
+
2640 train 4.224907 (lr=4.8281e-04) (hash(x)=154954967)
|
| 321 |
+
2650 train 4.192335 (lr=4.7916e-04) (hash(x)=132068617)
|
| 322 |
+
2660 train 4.243421 (lr=4.7552e-04) (hash(x)=148853275)
|
| 323 |
+
2670 train 4.193583 (lr=4.7188e-04) (hash(x)=134717627)
|
| 324 |
+
2680 train 4.080743 (lr=4.6826e-04) (hash(x)=148037282)
|
| 325 |
+
2690 train 4.001574 (lr=4.6464e-04) (hash(x)=156895238)
|
| 326 |
+
2700 val loss 4.1344
|
| 327 |
+
2700 val perplexity 62.4496
|
| 328 |
+
2700 train 4.038228 (lr=4.6103e-04) (hash(x)=189576996)
|
| 329 |
+
2710 train 4.179460 (lr=4.5742e-04) (hash(x)=144644131)
|
| 330 |
+
2720 train 3.995640 (lr=4.5383e-04) (hash(x)=142483444)
|
| 331 |
+
2730 train 4.081125 (lr=4.5025e-04) (hash(x)=150738859)
|
| 332 |
+
2740 train 4.053293 (lr=4.4667e-04) (hash(x)=153894981)
|
| 333 |
+
2750 train 3.975680 (lr=4.4310e-04) (hash(x)=148088547)
|
| 334 |
+
2760 train 3.911200 (lr=4.3954e-04) (hash(x)=160569014)
|
| 335 |
+
2770 train 4.055459 (lr=4.3600e-04) (hash(x)=145601198)
|
| 336 |
+
2780 train 3.979975 (lr=4.3246e-04) (hash(x)=151418110)
|
| 337 |
+
2790 train 4.028044 (lr=4.2893e-04) (hash(x)=131554526)
|
| 338 |
+
2800 val loss 4.1257
|
| 339 |
+
2800 val perplexity 61.9081
|
| 340 |
+
2800 train 4.098881 (lr=4.2541e-04) (hash(x)=158428852)
|
| 341 |
+
2810 train 4.033390 (lr=4.2190e-04) (hash(x)=181206292)
|
| 342 |
+
2820 train 4.076114 (lr=4.1841e-04) (hash(x)=154581441)
|
| 343 |
+
2830 train 4.158673 (lr=4.1492e-04) (hash(x)=138700842)
|
| 344 |
+
2840 train 4.168755 (lr=4.1144e-04) (hash(x)=151823167)
|
| 345 |
+
2850 train 4.045992 (lr=4.0798e-04) (hash(x)=138037431)
|
| 346 |
+
2860 train 4.018258 (lr=4.0453e-04) (hash(x)=160467469)
|
| 347 |
+
2870 train 4.083372 (lr=4.0109e-04) (hash(x)=148062966)
|
| 348 |
+
2880 train 3.981836 (lr=3.9766e-04) (hash(x)=140843852)
|
| 349 |
+
2890 train 3.931169 (lr=3.9424e-04) (hash(x)=200607340)
|
| 350 |
+
2900 val loss 4.1132
|
| 351 |
+
2900 val perplexity 61.1425
|
| 352 |
+
2900 train 3.975844 (lr=3.9083e-04) (hash(x)=157147920)
|
| 353 |
+
2910 train 3.972280 (lr=3.8744e-04) (hash(x)=154021984)
|
| 354 |
+
2920 train 3.937454 (lr=3.8406e-04) (hash(x)=148003772)
|
| 355 |
+
2930 train 3.963856 (lr=3.8069e-04) (hash(x)=145604355)
|
| 356 |
+
2940 train 4.025877 (lr=3.7734e-04) (hash(x)=137973699)
|
| 357 |
+
2950 train 4.114318 (lr=3.7400e-04) (hash(x)=153903647)
|
| 358 |
+
2960 train 4.165935 (lr=3.7067e-04) (hash(x)=155843151)
|
| 359 |
+
2970 train 4.153424 (lr=3.6736e-04) (hash(x)=153316011)
|
| 360 |
+
2980 train 4.106166 (lr=3.6406e-04) (hash(x)=149133725)
|
| 361 |
+
2990 train 4.071146 (lr=3.6077e-04) (hash(x)=156874628)
|
| 362 |
+
3000 val loss 4.0954
|
| 363 |
+
3000 val perplexity 60.0631
|
| 364 |
+
3000 train 3.951290 (lr=3.5750e-04) (hash(x)=147523460)
|
| 365 |
+
3010 train 4.407073 (lr=3.5424e-04) (hash(x)=166151966)
|
| 366 |
+
3020 train 4.017253 (lr=3.5100e-04) (hash(x)=150197656)
|
| 367 |
+
3030 train 3.921692 (lr=3.4777e-04) (hash(x)=159489805)
|
| 368 |
+
3040 train 3.927548 (lr=3.4456e-04) (hash(x)=149007094)
|
| 369 |
+
3050 train 3.931156 (lr=3.4136e-04) (hash(x)=161628710)
|
| 370 |
+
3060 train 3.905401 (lr=3.3818e-04) (hash(x)=151248942)
|
| 371 |
+
3070 train 3.962527 (lr=3.3501e-04) (hash(x)=142698123)
|
| 372 |
+
3080 train 3.857145 (lr=3.3186e-04) (hash(x)=140287682)
|
| 373 |
+
3090 train 4.102464 (lr=3.2872e-04) (hash(x)=140646346)
|
| 374 |
+
3100 val loss 4.0952
|
| 375 |
+
3100 val perplexity 60.0506
|
| 376 |
+
3100 train 4.096986 (lr=3.2560e-04) (hash(x)=147627509)
|
| 377 |
+
3110 train 4.037481 (lr=3.2250e-04) (hash(x)=156147704)
|
| 378 |
+
3120 train 4.014810 (lr=3.1941e-04) (hash(x)=147113228)
|
| 379 |
+
3130 train 4.017513 (lr=3.1634e-04) (hash(x)=156114554)
|
| 380 |
+
3140 train 3.911408 (lr=3.1328e-04) (hash(x)=150952909)
|
| 381 |
+
3150 train 3.919784 (lr=3.1025e-04) (hash(x)=142675915)
|
| 382 |
+
3160 train 3.938860 (lr=3.0723e-04) (hash(x)=149921944)
|
| 383 |
+
3170 train 4.043483 (lr=3.0422e-04) (hash(x)=140523501)
|
| 384 |
+
3180 train 3.930140 (lr=3.0124e-04) (hash(x)=157975563)
|
| 385 |
+
3190 train 3.910792 (lr=2.9827e-04) (hash(x)=152171280)
|
| 386 |
+
3200 val loss 4.0787
|
| 387 |
+
3200 val perplexity 59.0670
|
| 388 |
+
3200 train 3.950057 (lr=2.9532e-04) (hash(x)=146924313)
|
| 389 |
+
3210 train 3.872628 (lr=2.9239e-04) (hash(x)=139648262)
|
| 390 |
+
3220 train 3.878872 (lr=2.8948e-04) (hash(x)=147862079)
|
| 391 |
+
3230 train 3.885676 (lr=2.8658e-04) (hash(x)=145163796)
|
| 392 |
+
3240 train 4.165257 (lr=2.8370e-04) (hash(x)=146607485)
|
| 393 |
+
3250 train 4.176262 (lr=2.8084e-04) (hash(x)=148311826)
|
| 394 |
+
3260 train 3.998693 (lr=2.7800e-04) (hash(x)=174453341)
|
| 395 |
+
3270 train 4.144380 (lr=2.7518e-04) (hash(x)=153492012)
|
| 396 |
+
3280 train 4.110825 (lr=2.7238e-04) (hash(x)=147033673)
|
| 397 |
+
3290 train 4.125224 (lr=2.6960e-04) (hash(x)=146335117)
|
| 398 |
+
3300 val loss 4.0739
|
| 399 |
+
3300 val perplexity 58.7847
|
| 400 |
+
3300 train 4.107918 (lr=2.6684e-04) (hash(x)=144889712)
|
| 401 |
+
3310 train 3.898463 (lr=2.6409e-04) (hash(x)=158710683)
|
| 402 |
+
3320 train 3.978341 (lr=2.6137e-04) (hash(x)=155649513)
|
| 403 |
+
3330 train 4.052434 (lr=2.5867e-04) (hash(x)=152126488)
|
| 404 |
+
3340 train 4.174520 (lr=2.5598e-04) (hash(x)=153065158)
|
| 405 |
+
3350 train 4.088984 (lr=2.5332e-04) (hash(x)=156794512)
|
| 406 |
+
3360 train 3.982209 (lr=2.5068e-04) (hash(x)=150068441)
|
| 407 |
+
3370 train 4.072134 (lr=2.4806e-04) (hash(x)=157067126)
|
| 408 |
+
3380 train 4.083263 (lr=2.4545e-04) (hash(x)=148532206)
|
| 409 |
+
3390 train 4.055603 (lr=2.4287e-04) (hash(x)=161451523)
|
| 410 |
+
3400 val loss 4.0596
|
| 411 |
+
3400 val perplexity 57.9501
|
| 412 |
+
3400 train 4.150742 (lr=2.4031e-04) (hash(x)=154180141)
|
| 413 |
+
3410 train 4.082441 (lr=2.3777e-04) (hash(x)=144970962)
|
| 414 |
+
3420 train 4.091723 (lr=2.3526e-04) (hash(x)=152419016)
|
| 415 |
+
3430 train 4.001191 (lr=2.3276e-04) (hash(x)=155684986)
|
| 416 |
+
3440 train 4.005031 (lr=2.3029e-04) (hash(x)=139412683)
|
| 417 |
+
3450 train 4.063642 (lr=2.2784e-04) (hash(x)=148573702)
|
| 418 |
+
3460 train 4.012835 (lr=2.2540e-04) (hash(x)=170350706)
|
| 419 |
+
3470 train 3.968839 (lr=2.2300e-04) (hash(x)=170819774)
|
| 420 |
+
3480 train 3.936097 (lr=2.2061e-04) (hash(x)=153451957)
|
| 421 |
+
3490 train 4.026844 (lr=2.1825e-04) (hash(x)=150825808)
|
| 422 |
+
3500 val loss 4.0533
|
| 423 |
+
3500 val perplexity 57.5897
|
| 424 |
+
3500 train 3.957995 (lr=2.1590e-04) (hash(x)=146408995)
|
| 425 |
+
3510 train 3.924611 (lr=2.1358e-04) (hash(x)=137372175)
|
| 426 |
+
3520 train 4.011914 (lr=2.1129e-04) (hash(x)=145142963)
|
| 427 |
+
3530 train 4.016952 (lr=2.0901e-04) (hash(x)=157480818)
|
| 428 |
+
3540 train 4.025852 (lr=2.0676e-04) (hash(x)=156275606)
|
| 429 |
+
3550 train 4.067762 (lr=2.0454e-04) (hash(x)=141839187)
|
| 430 |
+
3560 train 4.015676 (lr=2.0233e-04) (hash(x)=150300593)
|
| 431 |
+
3570 train 3.953553 (lr=2.0015e-04) (hash(x)=164629404)
|
| 432 |
+
3580 train 4.028643 (lr=1.9799e-04) (hash(x)=155412140)
|
| 433 |
+
3590 train 4.017467 (lr=1.9586e-04) (hash(x)=142442693)
|
| 434 |
+
3600 val loss 4.0437
|
| 435 |
+
3600 val perplexity 57.0363
|
| 436 |
+
3600 train 3.987803 (lr=1.9375e-04) (hash(x)=155238730)
|
| 437 |
+
3610 train 3.967182 (lr=1.9166e-04) (hash(x)=151247336)
|
| 438 |
+
3620 train 3.910336 (lr=1.8960e-04) (hash(x)=161874831)
|
| 439 |
+
3630 train 3.881087 (lr=1.8756e-04) (hash(x)=151013760)
|
| 440 |
+
3640 train 3.882146 (lr=1.8555e-04) (hash(x)=134009599)
|
| 441 |
+
3650 train 3.935123 (lr=1.8356e-04) (hash(x)=148395531)
|
| 442 |
+
3660 train 3.935794 (lr=1.8159e-04) (hash(x)=148638990)
|
| 443 |
+
3670 train 3.932975 (lr=1.7965e-04) (hash(x)=147772685)
|
| 444 |
+
3680 train 4.035126 (lr=1.7774e-04) (hash(x)=145966246)
|
| 445 |
+
3690 train 4.091619 (lr=1.7585e-04) (hash(x)=153830518)
|
| 446 |
+
3700 val loss 4.0287
|
| 447 |
+
3700 val perplexity 56.1877
|
| 448 |
+
3700 train 3.989758 (lr=1.7398e-04) (hash(x)=155895558)
|
| 449 |
+
3710 train 3.998194 (lr=1.7214e-04) (hash(x)=138937251)
|
| 450 |
+
3720 train 4.054901 (lr=1.7032e-04) (hash(x)=169878300)
|
| 451 |
+
3730 train 4.027521 (lr=1.6853e-04) (hash(x)=146435658)
|
| 452 |
+
3740 train 4.013133 (lr=1.6677e-04) (hash(x)=147382834)
|
| 453 |
+
3750 train 3.953195 (lr=1.6503e-04) (hash(x)=151661820)
|
| 454 |
+
3760 train 3.933287 (lr=1.6331e-04) (hash(x)=171343176)
|
| 455 |
+
3770 train 3.950737 (lr=1.6162e-04) (hash(x)=147320174)
|
| 456 |
+
3780 train 3.977979 (lr=1.5996e-04) (hash(x)=150222525)
|
| 457 |
+
3790 train 4.043586 (lr=1.5832e-04) (hash(x)=139887785)
|
| 458 |
+
3800 val loss 4.0341
|
| 459 |
+
3800 val perplexity 56.4915
|
| 460 |
+
3800 train 3.844631 (lr=1.5671e-04) (hash(x)=142657108)
|
| 461 |
+
3810 train 3.906592 (lr=1.5512e-04) (hash(x)=157251919)
|
| 462 |
+
3820 train 3.867588 (lr=1.5356e-04) (hash(x)=153577487)
|
| 463 |
+
3830 train 4.184392 (lr=1.5203e-04) (hash(x)=148610416)
|
| 464 |
+
3840 train 3.926206 (lr=1.5052e-04) (hash(x)=146926779)
|
| 465 |
+
3850 train 4.074868 (lr=1.4904e-04) (hash(x)=141301606)
|
| 466 |
+
3860 train 3.920280 (lr=1.4759e-04) (hash(x)=158319996)
|
| 467 |
+
3870 train 4.079248 (lr=1.4616e-04) (hash(x)=148736434)
|
| 468 |
+
3880 train 4.147045 (lr=1.4476e-04) (hash(x)=159149914)
|
| 469 |
+
3890 train 3.979076 (lr=1.4339e-04) (hash(x)=148106908)
|
| 470 |
+
3900 val loss 4.0180
|
| 471 |
+
3900 val perplexity 55.5885
|
| 472 |
+
3900 train 4.018702 (lr=1.4204e-04) (hash(x)=146912279)
|
| 473 |
+
3910 train 4.018827 (lr=1.4072e-04) (hash(x)=147533241)
|
| 474 |
+
3920 train 3.978503 (lr=1.3942e-04) (hash(x)=150264472)
|
| 475 |
+
3930 train 3.950378 (lr=1.3816e-04) (hash(x)=142146717)
|
| 476 |
+
3940 train 3.905123 (lr=1.3692e-04) (hash(x)=141271841)
|
| 477 |
+
3950 train 3.965622 (lr=1.3570e-04) (hash(x)=150264103)
|
| 478 |
+
3960 train 3.922793 (lr=1.3452e-04) (hash(x)=148106240)
|
| 479 |
+
3970 train 4.014437 (lr=1.3336e-04) (hash(x)=150686305)
|
| 480 |
+
3980 train 4.042370 (lr=1.3223e-04) (hash(x)=157207544)
|
| 481 |
+
3990 train 4.012194 (lr=1.3113e-04) (hash(x)=161785965)
|
| 482 |
+
4000 val loss 4.0117
|
| 483 |
+
4000 val perplexity 55.2392
|
| 484 |
+
4000 train 4.020181 (lr=1.3005e-04) (hash(x)=154408435)
|
| 485 |
+
4010 train 4.123298 (lr=1.2900e-04) (hash(x)=150193109)
|
| 486 |
+
4020 train 3.981863 (lr=1.2798e-04) (hash(x)=156081999)
|
| 487 |
+
4030 train 3.989304 (lr=1.2699e-04) (hash(x)=147449250)
|
| 488 |
+
4040 train 4.004150 (lr=1.2602e-04) (hash(x)=151777625)
|
| 489 |
+
4050 train 3.957389 (lr=1.2509e-04) (hash(x)=150400620)
|
| 490 |
+
4060 train 3.959405 (lr=1.2418e-04) (hash(x)=143029773)
|
| 491 |
+
4070 train 3.959200 (lr=1.2329e-04) (hash(x)=158058157)
|
| 492 |
+
4080 train 3.972541 (lr=1.2244e-04) (hash(x)=146151623)
|
| 493 |
+
4090 train 3.914376 (lr=1.2161e-04) (hash(x)=159060195)
|
| 494 |
+
4100 val loss 4.0108
|
| 495 |
+
4100 val perplexity 55.1902
|
| 496 |
+
4100 train 3.912867 (lr=1.2082e-04) (hash(x)=146910399)
|
| 497 |
+
4110 train 3.662918 (lr=1.2005e-04) (hash(x)=142413707)
|
| 498 |
+
4120 train 3.784782 (lr=1.1931e-04) (hash(x)=151334076)
|
| 499 |
+
4130 train 3.712782 (lr=1.1859e-04) (hash(x)=148263988)
|
| 500 |
+
4140 train 3.820978 (lr=1.1791e-04) (hash(x)=135236846)
|
| 501 |
+
4150 train 3.683109 (lr=1.1725e-04) (hash(x)=154462745)
|
| 502 |
+
4160 train 4.039414 (lr=1.1662e-04) (hash(x)=150997914)
|
| 503 |
+
4170 train 4.023367 (lr=1.1602e-04) (hash(x)=155771516)
|
| 504 |
+
4180 train 4.006623 (lr=1.1545e-04) (hash(x)=152545900)
|
| 505 |
+
4190 train 4.027162 (lr=1.1491e-04) (hash(x)=171859542)
|
| 506 |
+
4200 val loss 4.0038
|
| 507 |
+
4200 val perplexity 54.8060
|
| 508 |
+
4200 train 4.009897 (lr=1.1439e-04) (hash(x)=158114837)
|
| 509 |
+
4210 train 3.926680 (lr=1.1390e-04) (hash(x)=158900351)
|
| 510 |
+
4220 train 4.079187 (lr=1.1344e-04) (hash(x)=158356339)
|
| 511 |
+
4230 train 3.916374 (lr=1.1302e-04) (hash(x)=170440865)
|
| 512 |
+
4240 train 3.992813 (lr=1.1261e-04) (hash(x)=144330316)
|
| 513 |
+
4250 train 4.009221 (lr=1.1224e-04) (hash(x)=147989541)
|
| 514 |
+
4260 train 3.949828 (lr=1.1190e-04) (hash(x)=148956428)
|
| 515 |
+
4270 train 3.855286 (lr=1.1158e-04) (hash(x)=162405324)
|
| 516 |
+
4280 train 3.896265 (lr=1.1130e-04) (hash(x)=142279992)
|
| 517 |
+
4290 train 4.020116 (lr=1.1104e-04) (hash(x)=145078631)
|
| 518 |
+
4300 val loss 4.0053
|
| 519 |
+
4300 val perplexity 54.8884
|
| 520 |
+
4300 train 4.076707 (lr=1.1081e-04) (hash(x)=153246964)
|
| 521 |
+
4310 train 3.762347 (lr=1.1061e-04) (hash(x)=154138756)
|
| 522 |
+
4320 train 3.826060 (lr=1.1043e-04) (hash(x)=144960180)
|
| 523 |
+
4330 train 3.789114 (lr=1.1029e-04) (hash(x)=157055563)
|
| 524 |
+
4340 train 3.643862 (lr=1.1018e-04) (hash(x)=148810108)
|
| 525 |
+
4350 train 3.845565 (lr=1.1009e-04) (hash(x)=138823271)
|
| 526 |
+
4360 train 3.935041 (lr=1.1003e-04) (hash(x)=161925954)
|
| 527 |
+
4370 train 4.046629 (lr=1.1000e-04) (hash(x)=140518916)
|
| 528 |
+
4374 val loss 4.0023
|
| 529 |
+
4374 val perplexity 54.7247
|
lr11e-4_total_batch_size122880_baseline_seed1340/model_04374.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddd61bb0a1a6a6f955d20d30f0695de7c1b5455b697fd8bf4faeb9a5fb9e5a26
|
| 3 |
+
size 92843394
|
lr11e-4_total_batch_size122880_baseline_seed1340/optimizer_04374.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8cd56a0fa9ea94221bebc727995da6770ca403eb63d5d27478e2390e66d14cbc
|
| 3 |
+
size 179406214
|