Upload folder using huggingface_hub
Browse files- lr11e-4_total_batch_size122880_baseline_seed1339/args.json +1 -0
- lr11e-4_total_batch_size122880_baseline_seed1339/dataloader_04374.pt +3 -0
- lr11e-4_total_batch_size122880_baseline_seed1339/log2.txt +529 -0
- lr11e-4_total_batch_size122880_baseline_seed1339/model_04374.pt +3 -0
- lr11e-4_total_batch_size122880_baseline_seed1339/optimizer_04374.pt +3 -0
lr11e-4_total_batch_size122880_baseline_seed1339/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/lr11e-4_total_batch_size122880_baseline_seed1339", "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": 1339, "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_seed1339/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_seed1339/log2.txt
ADDED
|
@@ -0,0 +1,529 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
max_steps: 4375
|
| 2 |
+
0 val loss 11.2576
|
| 3 |
+
0 val perplexity 77469.3047
|
| 4 |
+
0 train 11.261513 (lr=4.4000e-06) (hash(x)=150724848)
|
| 5 |
+
10 train 9.937681 (lr=4.8400e-05) (hash(x)=153483789)
|
| 6 |
+
20 train 9.378946 (lr=9.2400e-05) (hash(x)=145287265)
|
| 7 |
+
30 train 8.594829 (lr=1.3640e-04) (hash(x)=150157569)
|
| 8 |
+
40 train 7.956697 (lr=1.8040e-04) (hash(x)=152157598)
|
| 9 |
+
50 train 7.656182 (lr=2.2440e-04) (hash(x)=149910534)
|
| 10 |
+
60 train 7.648769 (lr=2.6840e-04) (hash(x)=150731273)
|
| 11 |
+
70 train 7.564219 (lr=3.1240e-04) (hash(x)=166473281)
|
| 12 |
+
80 train 7.295581 (lr=3.5640e-04) (hash(x)=151371145)
|
| 13 |
+
90 train 7.125069 (lr=4.0040e-04) (hash(x)=153368917)
|
| 14 |
+
100 val loss 7.0995
|
| 15 |
+
100 val perplexity 1211.3257
|
| 16 |
+
100 train 7.027227 (lr=4.4440e-04) (hash(x)=148123706)
|
| 17 |
+
110 train 6.902231 (lr=4.8840e-04) (hash(x)=150622583)
|
| 18 |
+
120 train 6.679286 (lr=5.3240e-04) (hash(x)=151535384)
|
| 19 |
+
130 train 6.644165 (lr=5.7640e-04) (hash(x)=148726088)
|
| 20 |
+
140 train 6.504584 (lr=6.2040e-04) (hash(x)=157260375)
|
| 21 |
+
150 train 6.374407 (lr=6.6440e-04) (hash(x)=146678221)
|
| 22 |
+
160 train 6.540164 (lr=7.0840e-04) (hash(x)=144947971)
|
| 23 |
+
170 train 6.489006 (lr=7.5240e-04) (hash(x)=162318756)
|
| 24 |
+
180 train 6.483446 (lr=7.9640e-04) (hash(x)=178373797)
|
| 25 |
+
190 train 6.193312 (lr=8.4040e-04) (hash(x)=150369080)
|
| 26 |
+
200 val loss 6.2287
|
| 27 |
+
200 val perplexity 507.1138
|
| 28 |
+
200 train 6.255595 (lr=8.8440e-04) (hash(x)=151700982)
|
| 29 |
+
210 train 6.097174 (lr=9.2840e-04) (hash(x)=148930522)
|
| 30 |
+
220 train 6.107656 (lr=9.7240e-04) (hash(x)=148079227)
|
| 31 |
+
230 train 6.002536 (lr=1.0164e-03) (hash(x)=145272617)
|
| 32 |
+
240 train 5.905256 (lr=1.0604e-03) (hash(x)=156287490)
|
| 33 |
+
250 train 5.880959 (lr=1.1000e-03) (hash(x)=156182087)
|
| 34 |
+
260 train 5.905228 (lr=1.1000e-03) (hash(x)=162758254)
|
| 35 |
+
270 train 5.735993 (lr=1.0999e-03) (hash(x)=157363914)
|
| 36 |
+
280 train 5.626208 (lr=1.0999e-03) (hash(x)=149449280)
|
| 37 |
+
290 train 5.644699 (lr=1.0998e-03) (hash(x)=149266312)
|
| 38 |
+
300 val loss 5.7079
|
| 39 |
+
300 val perplexity 301.2429
|
| 40 |
+
300 train 5.533155 (lr=1.0996e-03) (hash(x)=149318660)
|
| 41 |
+
310 train 5.402924 (lr=1.0995e-03) (hash(x)=137184623)
|
| 42 |
+
320 train 5.361253 (lr=1.0993e-03) (hash(x)=143999212)
|
| 43 |
+
330 train 5.224095 (lr=1.0991e-03) (hash(x)=138360108)
|
| 44 |
+
340 train 5.254739 (lr=1.0988e-03) (hash(x)=143752601)
|
| 45 |
+
350 train 5.530335 (lr=1.0986e-03) (hash(x)=150482428)
|
| 46 |
+
360 train 5.504911 (lr=1.0983e-03) (hash(x)=159461069)
|
| 47 |
+
370 train 5.462457 (lr=1.0979e-03) (hash(x)=149967984)
|
| 48 |
+
380 train 5.472826 (lr=1.0976e-03) (hash(x)=155106202)
|
| 49 |
+
390 train 5.435772 (lr=1.0972e-03) (hash(x)=155825158)
|
| 50 |
+
400 val loss 5.3881
|
| 51 |
+
400 val perplexity 218.7831
|
| 52 |
+
400 train 5.371276 (lr=1.0968e-03) (hash(x)=143268605)
|
| 53 |
+
410 train 5.353130 (lr=1.0963e-03) (hash(x)=179384792)
|
| 54 |
+
420 train 5.508554 (lr=1.0959e-03) (hash(x)=142711393)
|
| 55 |
+
430 train 5.171904 (lr=1.0954e-03) (hash(x)=138495801)
|
| 56 |
+
440 train 5.247691 (lr=1.0948e-03) (hash(x)=151951501)
|
| 57 |
+
450 train 5.192624 (lr=1.0943e-03) (hash(x)=152322423)
|
| 58 |
+
460 train 5.177573 (lr=1.0937e-03) (hash(x)=148720342)
|
| 59 |
+
470 train 5.134316 (lr=1.0931e-03) (hash(x)=150349503)
|
| 60 |
+
480 train 5.120070 (lr=1.0924e-03) (hash(x)=146292562)
|
| 61 |
+
490 train 4.974632 (lr=1.0918e-03) (hash(x)=137640259)
|
| 62 |
+
500 val loss 5.1639
|
| 63 |
+
500 val perplexity 174.8473
|
| 64 |
+
500 train 4.930937 (lr=1.0911e-03) (hash(x)=147904298)
|
| 65 |
+
510 train 4.916759 (lr=1.0903e-03) (hash(x)=140706537)
|
| 66 |
+
520 train 4.874089 (lr=1.0896e-03) (hash(x)=144746445)
|
| 67 |
+
530 train 4.797421 (lr=1.0888e-03) (hash(x)=153065665)
|
| 68 |
+
540 train 4.915986 (lr=1.0880e-03) (hash(x)=157674756)
|
| 69 |
+
550 train 5.270940 (lr=1.0871e-03) (hash(x)=154343147)
|
| 70 |
+
560 train 4.955053 (lr=1.0863e-03) (hash(x)=162479784)
|
| 71 |
+
570 train 5.030447 (lr=1.0854e-03) (hash(x)=146934299)
|
| 72 |
+
580 train 4.984862 (lr=1.0844e-03) (hash(x)=145536832)
|
| 73 |
+
590 train 5.017288 (lr=1.0835e-03) (hash(x)=158271614)
|
| 74 |
+
600 val loss 4.9808
|
| 75 |
+
600 val perplexity 145.5914
|
| 76 |
+
600 train 5.024428 (lr=1.0825e-03) (hash(x)=141843115)
|
| 77 |
+
610 train 4.895489 (lr=1.0815e-03) (hash(x)=144149947)
|
| 78 |
+
620 train 4.967009 (lr=1.0805e-03) (hash(x)=152217389)
|
| 79 |
+
630 train 4.892469 (lr=1.0794e-03) (hash(x)=144029489)
|
| 80 |
+
640 train 4.900438 (lr=1.0783e-03) (hash(x)=151625838)
|
| 81 |
+
650 train 4.849727 (lr=1.0772e-03) (hash(x)=145279030)
|
| 82 |
+
660 train 4.672498 (lr=1.0761e-03) (hash(x)=133642994)
|
| 83 |
+
670 train 4.840752 (lr=1.0749e-03) (hash(x)=155919973)
|
| 84 |
+
680 train 4.743078 (lr=1.0737e-03) (hash(x)=150613636)
|
| 85 |
+
690 train 4.940382 (lr=1.0725e-03) (hash(x)=152803906)
|
| 86 |
+
700 val loss 4.8578
|
| 87 |
+
700 val perplexity 128.7404
|
| 88 |
+
700 train 4.663158 (lr=1.0712e-03) (hash(x)=152507639)
|
| 89 |
+
710 train 4.657321 (lr=1.0699e-03) (hash(x)=167824828)
|
| 90 |
+
720 train 4.589524 (lr=1.0686e-03) (hash(x)=150085494)
|
| 91 |
+
730 train 4.524959 (lr=1.0673e-03) (hash(x)=142956546)
|
| 92 |
+
740 train 4.527936 (lr=1.0659e-03) (hash(x)=150893345)
|
| 93 |
+
750 train 4.794313 (lr=1.0645e-03) (hash(x)=148473774)
|
| 94 |
+
760 train 4.697918 (lr=1.0631e-03) (hash(x)=166217855)
|
| 95 |
+
770 train 4.715325 (lr=1.0617e-03) (hash(x)=155013278)
|
| 96 |
+
780 train 4.745700 (lr=1.0602e-03) (hash(x)=152138735)
|
| 97 |
+
790 train 4.772672 (lr=1.0587e-03) (hash(x)=157176546)
|
| 98 |
+
800 val loss 4.7035
|
| 99 |
+
800 val perplexity 110.3370
|
| 100 |
+
800 train 4.670205 (lr=1.0572e-03) (hash(x)=151117002)
|
| 101 |
+
810 train 4.678448 (lr=1.0557e-03) (hash(x)=145015354)
|
| 102 |
+
820 train 4.667449 (lr=1.0541e-03) (hash(x)=148776668)
|
| 103 |
+
830 train 4.475302 (lr=1.0525e-03) (hash(x)=170658972)
|
| 104 |
+
840 train 4.756358 (lr=1.0509e-03) (hash(x)=155609234)
|
| 105 |
+
850 train 4.725739 (lr=1.0492e-03) (hash(x)=147018482)
|
| 106 |
+
860 train 4.575591 (lr=1.0475e-03) (hash(x)=136898027)
|
| 107 |
+
870 train 4.563887 (lr=1.0458e-03) (hash(x)=141846516)
|
| 108 |
+
880 train 4.688482 (lr=1.0441e-03) (hash(x)=161873750)
|
| 109 |
+
890 train 4.519667 (lr=1.0424e-03) (hash(x)=157878866)
|
| 110 |
+
900 val loss 4.6529
|
| 111 |
+
900 val perplexity 104.8938
|
| 112 |
+
900 train 4.612679 (lr=1.0406e-03) (hash(x)=145757704)
|
| 113 |
+
910 train 4.729454 (lr=1.0388e-03) (hash(x)=145786621)
|
| 114 |
+
920 train 4.616345 (lr=1.0369e-03) (hash(x)=142045584)
|
| 115 |
+
930 train 4.690601 (lr=1.0351e-03) (hash(x)=164426390)
|
| 116 |
+
940 train 4.574326 (lr=1.0332e-03) (hash(x)=139987987)
|
| 117 |
+
950 train 4.769899 (lr=1.0313e-03) (hash(x)=148921543)
|
| 118 |
+
960 train 4.564944 (lr=1.0294e-03) (hash(x)=154138655)
|
| 119 |
+
970 train 4.550048 (lr=1.0274e-03) (hash(x)=138488697)
|
| 120 |
+
980 train 4.582568 (lr=1.0254e-03) (hash(x)=147004214)
|
| 121 |
+
990 train 4.446077 (lr=1.0234e-03) (hash(x)=156465269)
|
| 122 |
+
1000 val loss 4.5520
|
| 123 |
+
1000 val perplexity 94.8258
|
| 124 |
+
1000 train 4.397688 (lr=1.0214e-03) (hash(x)=161795711)
|
| 125 |
+
1010 train 4.481196 (lr=1.0194e-03) (hash(x)=141132073)
|
| 126 |
+
1020 train 4.598031 (lr=1.0173e-03) (hash(x)=141901746)
|
| 127 |
+
1030 train 4.444379 (lr=1.0152e-03) (hash(x)=137935785)
|
| 128 |
+
1040 train 4.646878 (lr=1.0131e-03) (hash(x)=144386181)
|
| 129 |
+
1050 train 4.475689 (lr=1.0109e-03) (hash(x)=145581641)
|
| 130 |
+
1060 train 4.499939 (lr=1.0088e-03) (hash(x)=155473484)
|
| 131 |
+
1070 train 4.601479 (lr=1.0066e-03) (hash(x)=144911208)
|
| 132 |
+
1080 train 4.716285 (lr=1.0044e-03) (hash(x)=156601997)
|
| 133 |
+
1090 train 4.497831 (lr=1.0021e-03) (hash(x)=151299535)
|
| 134 |
+
1100 val loss 4.5123
|
| 135 |
+
1100 val perplexity 91.1274
|
| 136 |
+
1100 train 4.515325 (lr=9.9985e-04) (hash(x)=157673597)
|
| 137 |
+
1110 train 4.495099 (lr=9.9757e-04) (hash(x)=149224552)
|
| 138 |
+
1120 train 4.525291 (lr=9.9526e-04) (hash(x)=155650894)
|
| 139 |
+
1130 train 4.506378 (lr=9.9293e-04) (hash(x)=143649236)
|
| 140 |
+
1140 train 4.491844 (lr=9.9058e-04) (hash(x)=146796338)
|
| 141 |
+
1150 train 4.416926 (lr=9.8820e-04) (hash(x)=147384772)
|
| 142 |
+
1160 train 4.471188 (lr=9.8580e-04) (hash(x)=148232447)
|
| 143 |
+
1170 train 4.373789 (lr=9.8338e-04) (hash(x)=149417135)
|
| 144 |
+
1180 train 4.379251 (lr=9.8094e-04) (hash(x)=146962492)
|
| 145 |
+
1190 train 4.217916 (lr=9.7848e-04) (hash(x)=147119717)
|
| 146 |
+
1200 val loss 4.4841
|
| 147 |
+
1200 val perplexity 88.6012
|
| 148 |
+
1200 train 4.415198 (lr=9.7599e-04) (hash(x)=144115998)
|
| 149 |
+
1210 train 4.601022 (lr=9.7349e-04) (hash(x)=148830634)
|
| 150 |
+
1220 train 4.473250 (lr=9.7096e-04) (hash(x)=155788808)
|
| 151 |
+
1230 train 4.582223 (lr=9.6841e-04) (hash(x)=155420927)
|
| 152 |
+
1240 train 4.413548 (lr=9.6584e-04) (hash(x)=176100202)
|
| 153 |
+
1250 train 4.367055 (lr=9.6325e-04) (hash(x)=163245801)
|
| 154 |
+
1260 train 4.368310 (lr=9.6064e-04) (hash(x)=147341874)
|
| 155 |
+
1270 train 4.379790 (lr=9.5800e-04) (hash(x)=140477438)
|
| 156 |
+
1280 train 4.246964 (lr=9.5535e-04) (hash(x)=137109128)
|
| 157 |
+
1290 train 4.390248 (lr=9.5268e-04) (hash(x)=158887908)
|
| 158 |
+
1300 val loss 4.4175
|
| 159 |
+
1300 val perplexity 82.8875
|
| 160 |
+
1300 train 4.368987 (lr=9.4998e-04) (hash(x)=146922182)
|
| 161 |
+
1310 train 4.374203 (lr=9.4727e-04) (hash(x)=143857902)
|
| 162 |
+
1320 train 4.339495 (lr=9.4454e-04) (hash(x)=150276684)
|
| 163 |
+
1330 train 4.368726 (lr=9.4178e-04) (hash(x)=153362319)
|
| 164 |
+
1340 train 4.423662 (lr=9.3901e-04) (hash(x)=153140891)
|
| 165 |
+
1350 train 4.448573 (lr=9.3622e-04) (hash(x)=166976502)
|
| 166 |
+
1360 train 4.467198 (lr=9.3341e-04) (hash(x)=147648241)
|
| 167 |
+
1370 train 4.379729 (lr=9.3058e-04) (hash(x)=152606892)
|
| 168 |
+
1380 train 4.361859 (lr=9.2773e-04) (hash(x)=151845292)
|
| 169 |
+
1390 train 4.434896 (lr=9.2486e-04) (hash(x)=142307378)
|
| 170 |
+
1400 val loss 4.3849
|
| 171 |
+
1400 val perplexity 80.2294
|
| 172 |
+
1400 train 4.443091 (lr=9.2197e-04) (hash(x)=172927707)
|
| 173 |
+
1410 train 4.466107 (lr=9.1907e-04) (hash(x)=154387219)
|
| 174 |
+
1420 train 4.285017 (lr=9.1615e-04) (hash(x)=156081484)
|
| 175 |
+
1430 train 4.356402 (lr=9.1321e-04) (hash(x)=156584703)
|
| 176 |
+
1440 train 4.253954 (lr=9.1025e-04) (hash(x)=141998503)
|
| 177 |
+
1450 train 4.373148 (lr=9.0727e-04) (hash(x)=148708706)
|
| 178 |
+
1460 train 4.237295 (lr=9.0428e-04) (hash(x)=148096724)
|
| 179 |
+
1470 train 4.202815 (lr=9.0127e-04) (hash(x)=143995394)
|
| 180 |
+
1480 train 4.445337 (lr=8.9824e-04) (hash(x)=160523370)
|
| 181 |
+
1490 train 4.266096 (lr=8.9519e-04) (hash(x)=142703815)
|
| 182 |
+
1500 val loss 4.3753
|
| 183 |
+
1500 val perplexity 79.4635
|
| 184 |
+
1500 train 4.261194 (lr=8.9213e-04) (hash(x)=150145360)
|
| 185 |
+
1510 train 4.415318 (lr=8.8905e-04) (hash(x)=140955409)
|
| 186 |
+
1520 train 4.323521 (lr=8.8595e-04) (hash(x)=121541202)
|
| 187 |
+
1530 train 4.309261 (lr=8.8284e-04) (hash(x)=155534733)
|
| 188 |
+
1540 train 4.352991 (lr=8.7971e-04) (hash(x)=141973489)
|
| 189 |
+
1550 train 4.232413 (lr=8.7657e-04) (hash(x)=157912759)
|
| 190 |
+
1560 train 4.284106 (lr=8.7341e-04) (hash(x)=143045873)
|
| 191 |
+
1570 train 4.297627 (lr=8.7023e-04) (hash(x)=148243369)
|
| 192 |
+
1580 train 4.312193 (lr=8.6704e-04) (hash(x)=149719396)
|
| 193 |
+
1590 train 4.352123 (lr=8.6384e-04) (hash(x)=142111349)
|
| 194 |
+
1600 val loss 4.3305
|
| 195 |
+
1600 val perplexity 75.9803
|
| 196 |
+
1600 train 4.360600 (lr=8.6062e-04) (hash(x)=145393936)
|
| 197 |
+
1610 train 4.257246 (lr=8.5738e-04) (hash(x)=146389859)
|
| 198 |
+
1620 train 4.312943 (lr=8.5413e-04) (hash(x)=152903298)
|
| 199 |
+
1630 train 4.214379 (lr=8.5087e-04) (hash(x)=149203171)
|
| 200 |
+
1640 train 4.178735 (lr=8.4759e-04) (hash(x)=143289561)
|
| 201 |
+
1650 train 4.212951 (lr=8.4429e-04) (hash(x)=158724067)
|
| 202 |
+
1660 train 4.143166 (lr=8.4099e-04) (hash(x)=149609240)
|
| 203 |
+
1670 train 4.353038 (lr=8.3767e-04) (hash(x)=155240838)
|
| 204 |
+
1680 train 4.340947 (lr=8.3433e-04) (hash(x)=149973508)
|
| 205 |
+
1690 train 4.198972 (lr=8.3098e-04) (hash(x)=153481145)
|
| 206 |
+
1700 val loss 4.3067
|
| 207 |
+
1700 val perplexity 74.1944
|
| 208 |
+
1700 train 4.441032 (lr=8.2762e-04) (hash(x)=147623002)
|
| 209 |
+
1710 train 4.303020 (lr=8.2425e-04) (hash(x)=163392549)
|
| 210 |
+
1720 train 4.306327 (lr=8.2086e-04) (hash(x)=154608207)
|
| 211 |
+
1730 train 4.252850 (lr=8.1746e-04) (hash(x)=152409179)
|
| 212 |
+
1740 train 4.220674 (lr=8.1405e-04) (hash(x)=148273153)
|
| 213 |
+
1750 train 4.320292 (lr=8.1063e-04) (hash(x)=143259465)
|
| 214 |
+
1760 train 4.490232 (lr=8.0720e-04) (hash(x)=138718468)
|
| 215 |
+
1770 train 4.236032 (lr=8.0375e-04) (hash(x)=159255954)
|
| 216 |
+
1780 train 4.231622 (lr=8.0029e-04) (hash(x)=135857355)
|
| 217 |
+
1790 train 4.144902 (lr=7.9682e-04) (hash(x)=145791671)
|
| 218 |
+
1800 val loss 4.2911
|
| 219 |
+
1800 val perplexity 73.0498
|
| 220 |
+
1800 train 4.242648 (lr=7.9334e-04) (hash(x)=151062384)
|
| 221 |
+
1810 train 4.145452 (lr=7.8985e-04) (hash(x)=144984716)
|
| 222 |
+
1820 train 4.206643 (lr=7.8635e-04) (hash(x)=144698775)
|
| 223 |
+
1830 train 4.345541 (lr=7.8283e-04) (hash(x)=150245397)
|
| 224 |
+
1840 train 4.380797 (lr=7.7931e-04) (hash(x)=170661468)
|
| 225 |
+
1850 train 4.303674 (lr=7.7578e-04) (hash(x)=151431836)
|
| 226 |
+
1860 train 4.181296 (lr=7.7223e-04) (hash(x)=148600110)
|
| 227 |
+
1870 train 4.271167 (lr=7.6868e-04) (hash(x)=148519036)
|
| 228 |
+
1880 train 4.190219 (lr=7.6512e-04) (hash(x)=141697198)
|
| 229 |
+
1890 train 4.285067 (lr=7.6154e-04) (hash(x)=156289714)
|
| 230 |
+
1900 val loss 4.2632
|
| 231 |
+
1900 val perplexity 71.0358
|
| 232 |
+
1900 train 4.224982 (lr=7.5796e-04) (hash(x)=155386073)
|
| 233 |
+
1910 train 4.418484 (lr=7.5437e-04) (hash(x)=147240084)
|
| 234 |
+
1920 train 4.267061 (lr=7.5078e-04) (hash(x)=148388842)
|
| 235 |
+
1930 train 4.178484 (lr=7.4717e-04) (hash(x)=146204229)
|
| 236 |
+
1940 train 4.055166 (lr=7.4355e-04) (hash(x)=143556545)
|
| 237 |
+
1950 train 4.126127 (lr=7.3993e-04) (hash(x)=147956690)
|
| 238 |
+
1960 train 4.234937 (lr=7.3630e-04) (hash(x)=137526984)
|
| 239 |
+
1970 train 4.180138 (lr=7.3266e-04) (hash(x)=140760871)
|
| 240 |
+
1980 train 4.478081 (lr=7.2901e-04) (hash(x)=152103697)
|
| 241 |
+
1990 train 4.217707 (lr=7.2536e-04) (hash(x)=164248169)
|
| 242 |
+
2000 val loss 4.2414
|
| 243 |
+
2000 val perplexity 69.5051
|
| 244 |
+
2000 train 4.225309 (lr=7.2170e-04) (hash(x)=151842998)
|
| 245 |
+
2010 train 4.271152 (lr=7.1803e-04) (hash(x)=153881605)
|
| 246 |
+
2020 train 4.291853 (lr=7.1436e-04) (hash(x)=145254556)
|
| 247 |
+
2030 train 4.188123 (lr=7.1068e-04) (hash(x)=143283003)
|
| 248 |
+
2040 train 4.178802 (lr=7.0699e-04) (hash(x)=149520645)
|
| 249 |
+
2050 train 4.127707 (lr=7.0330e-04) (hash(x)=143001992)
|
| 250 |
+
2060 train 4.175628 (lr=6.9960e-04) (hash(x)=147300288)
|
| 251 |
+
2070 train 4.143950 (lr=6.9590e-04) (hash(x)=148000315)
|
| 252 |
+
2080 train 4.240504 (lr=6.9219e-04) (hash(x)=146292733)
|
| 253 |
+
2090 train 4.092942 (lr=6.8848e-04) (hash(x)=152253548)
|
| 254 |
+
2100 val loss 4.2372
|
| 255 |
+
2100 val perplexity 69.2111
|
| 256 |
+
2100 train 4.299405 (lr=6.8476e-04) (hash(x)=163228537)
|
| 257 |
+
2110 train 4.116930 (lr=6.8104e-04) (hash(x)=156319517)
|
| 258 |
+
2120 train 4.396369 (lr=6.7731e-04) (hash(x)=140368798)
|
| 259 |
+
2130 train 4.185340 (lr=6.7358e-04) (hash(x)=162813132)
|
| 260 |
+
2140 train 4.338749 (lr=6.6984e-04) (hash(x)=140817744)
|
| 261 |
+
2150 train 4.185870 (lr=6.6610e-04) (hash(x)=165620945)
|
| 262 |
+
2160 train 4.133113 (lr=6.6236e-04) (hash(x)=139579339)
|
| 263 |
+
2170 train 4.309495 (lr=6.5862e-04) (hash(x)=145505114)
|
| 264 |
+
2180 train 4.225069 (lr=6.5487e-04) (hash(x)=151040053)
|
| 265 |
+
2190 train 4.146818 (lr=6.5111e-04) (hash(x)=138105191)
|
| 266 |
+
2200 val loss 4.2122
|
| 267 |
+
2200 val perplexity 67.5030
|
| 268 |
+
2200 train 4.111054 (lr=6.4736e-04) (hash(x)=149758589)
|
| 269 |
+
2210 train 4.141785 (lr=6.4360e-04) (hash(x)=149125273)
|
| 270 |
+
2220 train 4.145434 (lr=6.3984e-04) (hash(x)=152919204)
|
| 271 |
+
2230 train 4.207420 (lr=6.3608e-04) (hash(x)=152120880)
|
| 272 |
+
2240 train 3.975594 (lr=6.3232e-04) (hash(x)=148124218)
|
| 273 |
+
2250 train 4.089655 (lr=6.2855e-04) (hash(x)=137089797)
|
| 274 |
+
2260 train 4.073557 (lr=6.2479e-04) (hash(x)=155806109)
|
| 275 |
+
2270 train 4.135233 (lr=6.2102e-04) (hash(x)=161868486)
|
| 276 |
+
2280 train 4.274323 (lr=6.1725e-04) (hash(x)=156017019)
|
| 277 |
+
2290 train 4.102240 (lr=6.1348e-04) (hash(x)=144077561)
|
| 278 |
+
2300 val loss 4.1964
|
| 279 |
+
2300 val perplexity 66.4472
|
| 280 |
+
2300 train 4.242555 (lr=6.0971e-04) (hash(x)=170869105)
|
| 281 |
+
2310 train 4.222500 (lr=6.0594e-04) (hash(x)=147916300)
|
| 282 |
+
2320 train 4.179608 (lr=6.0217e-04) (hash(x)=152467472)
|
| 283 |
+
2330 train 4.187126 (lr=5.9840e-04) (hash(x)=144785419)
|
| 284 |
+
2340 train 4.100389 (lr=5.9463e-04) (hash(x)=150432854)
|
| 285 |
+
2350 train 4.148602 (lr=5.9086e-04) (hash(x)=141272857)
|
| 286 |
+
2360 train 4.197967 (lr=5.8710e-04) (hash(x)=160564556)
|
| 287 |
+
2370 train 4.186948 (lr=5.8333e-04) (hash(x)=151879225)
|
| 288 |
+
2380 train 4.046690 (lr=5.7956e-04) (hash(x)=151530029)
|
| 289 |
+
2390 train 3.999575 (lr=5.7580e-04) (hash(x)=152832807)
|
| 290 |
+
2400 val loss 4.1861
|
| 291 |
+
2400 val perplexity 65.7642
|
| 292 |
+
2400 train 4.060021 (lr=5.7204e-04) (hash(x)=147479799)
|
| 293 |
+
2410 train 4.084641 (lr=5.6828e-04) (hash(x)=145055665)
|
| 294 |
+
2420 train 4.069935 (lr=5.6452e-04) (hash(x)=154768391)
|
| 295 |
+
2430 train 4.140815 (lr=5.6076e-04) (hash(x)=150388315)
|
| 296 |
+
2440 train 4.137802 (lr=5.5701e-04) (hash(x)=161064525)
|
| 297 |
+
2450 train 4.131504 (lr=5.5326e-04) (hash(x)=154887504)
|
| 298 |
+
2460 train 4.109966 (lr=5.4951e-04) (hash(x)=138201850)
|
| 299 |
+
2470 train 4.240063 (lr=5.4577e-04) (hash(x)=151711687)
|
| 300 |
+
2480 train 4.325788 (lr=5.4203e-04) (hash(x)=147643099)
|
| 301 |
+
2490 train 4.197303 (lr=5.3829e-04) (hash(x)=143125010)
|
| 302 |
+
2500 val loss 4.1560
|
| 303 |
+
2500 val perplexity 63.8187
|
| 304 |
+
2500 train 4.215398 (lr=5.3455e-04) (hash(x)=142429041)
|
| 305 |
+
2510 train 4.170473 (lr=5.3082e-04) (hash(x)=148050199)
|
| 306 |
+
2520 train 3.996959 (lr=5.2710e-04) (hash(x)=90941145)
|
| 307 |
+
2530 train 4.119590 (lr=5.2338e-04) (hash(x)=150165963)
|
| 308 |
+
2540 train 4.039886 (lr=5.1966e-04) (hash(x)=150919495)
|
| 309 |
+
2550 train 4.034595 (lr=5.1595e-04) (hash(x)=160130013)
|
| 310 |
+
2560 train 4.053505 (lr=5.1225e-04) (hash(x)=148050104)
|
| 311 |
+
2570 train 4.133963 (lr=5.0855e-04) (hash(x)=145749915)
|
| 312 |
+
2580 train 4.059870 (lr=5.0485e-04) (hash(x)=149001490)
|
| 313 |
+
2590 train 3.972236 (lr=5.0116e-04) (hash(x)=158310837)
|
| 314 |
+
2600 val loss 4.1581
|
| 315 |
+
2600 val perplexity 63.9491
|
| 316 |
+
2600 train 3.946504 (lr=4.9748e-04) (hash(x)=146370729)
|
| 317 |
+
2610 train 3.934126 (lr=4.9380e-04) (hash(x)=154169203)
|
| 318 |
+
2620 train 4.075926 (lr=4.9013e-04) (hash(x)=137329685)
|
| 319 |
+
2630 train 3.993725 (lr=4.8647e-04) (hash(x)=160226256)
|
| 320 |
+
2640 train 4.079663 (lr=4.8281e-04) (hash(x)=153533222)
|
| 321 |
+
2650 train 4.138337 (lr=4.7916e-04) (hash(x)=157476060)
|
| 322 |
+
2660 train 4.218637 (lr=4.7552e-04) (hash(x)=164837154)
|
| 323 |
+
2670 train 4.042444 (lr=4.7188e-04) (hash(x)=155851799)
|
| 324 |
+
2680 train 4.139318 (lr=4.6826e-04) (hash(x)=162095841)
|
| 325 |
+
2690 train 4.177805 (lr=4.6464e-04) (hash(x)=140059129)
|
| 326 |
+
2700 val loss 4.1245
|
| 327 |
+
2700 val perplexity 61.8338
|
| 328 |
+
2700 train 4.128977 (lr=4.6103e-04) (hash(x)=149982495)
|
| 329 |
+
2710 train 4.128967 (lr=4.5742e-04) (hash(x)=154730619)
|
| 330 |
+
2720 train 4.093436 (lr=4.5383e-04) (hash(x)=153673087)
|
| 331 |
+
2730 train 4.188054 (lr=4.5025e-04) (hash(x)=141967987)
|
| 332 |
+
2740 train 4.031495 (lr=4.4667e-04) (hash(x)=148302521)
|
| 333 |
+
2750 train 4.056394 (lr=4.4310e-04) (hash(x)=169526708)
|
| 334 |
+
2760 train 4.070383 (lr=4.3954e-04) (hash(x)=147659745)
|
| 335 |
+
2770 train 4.046369 (lr=4.3600e-04) (hash(x)=165422791)
|
| 336 |
+
2780 train 4.022936 (lr=4.3246e-04) (hash(x)=139378759)
|
| 337 |
+
2790 train 3.837113 (lr=4.2893e-04) (hash(x)=149362982)
|
| 338 |
+
2800 val loss 4.1324
|
| 339 |
+
2800 val perplexity 62.3265
|
| 340 |
+
2800 train 3.906381 (lr=4.2541e-04) (hash(x)=166852181)
|
| 341 |
+
2810 train 3.868318 (lr=4.2190e-04) (hash(x)=162830529)
|
| 342 |
+
2820 train 3.819986 (lr=4.1841e-04) (hash(x)=150130923)
|
| 343 |
+
2830 train 4.122179 (lr=4.1492e-04) (hash(x)=132554804)
|
| 344 |
+
2840 train 4.112358 (lr=4.1144e-04) (hash(x)=156662719)
|
| 345 |
+
2850 train 4.154697 (lr=4.0798e-04) (hash(x)=157531009)
|
| 346 |
+
2860 train 4.113059 (lr=4.0453e-04) (hash(x)=153415901)
|
| 347 |
+
2870 train 4.103221 (lr=4.0109e-04) (hash(x)=157332351)
|
| 348 |
+
2880 train 4.052597 (lr=3.9766e-04) (hash(x)=151410115)
|
| 349 |
+
2890 train 4.192080 (lr=3.9424e-04) (hash(x)=151755498)
|
| 350 |
+
2900 val loss 4.1035
|
| 351 |
+
2900 val perplexity 60.5539
|
| 352 |
+
2900 train 4.016640 (lr=3.9083e-04) (hash(x)=157395013)
|
| 353 |
+
2910 train 3.994758 (lr=3.8744e-04) (hash(x)=144568819)
|
| 354 |
+
2920 train 3.943294 (lr=3.8406e-04) (hash(x)=140411793)
|
| 355 |
+
2930 train 4.067595 (lr=3.8069e-04) (hash(x)=153672754)
|
| 356 |
+
2940 train 4.072390 (lr=3.7734e-04) (hash(x)=144747136)
|
| 357 |
+
2950 train 4.038503 (lr=3.7400e-04) (hash(x)=139873259)
|
| 358 |
+
2960 train 4.045554 (lr=3.7067e-04) (hash(x)=164010581)
|
| 359 |
+
2970 train 4.057638 (lr=3.6736e-04) (hash(x)=145020447)
|
| 360 |
+
2980 train 3.952604 (lr=3.6406e-04) (hash(x)=146536623)
|
| 361 |
+
2990 train 3.917713 (lr=3.6077e-04) (hash(x)=143152437)
|
| 362 |
+
3000 val loss 4.1137
|
| 363 |
+
3000 val perplexity 61.1727
|
| 364 |
+
3000 train 3.885116 (lr=3.5750e-04) (hash(x)=151165786)
|
| 365 |
+
3010 train 3.866446 (lr=3.5424e-04) (hash(x)=154439823)
|
| 366 |
+
3020 train 3.923578 (lr=3.5100e-04) (hash(x)=144196226)
|
| 367 |
+
3030 train 3.824551 (lr=3.4777e-04) (hash(x)=148787067)
|
| 368 |
+
3040 train 3.973383 (lr=3.4456e-04) (hash(x)=161017582)
|
| 369 |
+
3050 train 4.292279 (lr=3.4136e-04) (hash(x)=158552651)
|
| 370 |
+
3060 train 4.197832 (lr=3.3818e-04) (hash(x)=149243161)
|
| 371 |
+
3070 train 4.146321 (lr=3.3501e-04) (hash(x)=146431684)
|
| 372 |
+
3080 train 4.067007 (lr=3.3186e-04) (hash(x)=149547069)
|
| 373 |
+
3090 train 4.097353 (lr=3.2872e-04) (hash(x)=146636428)
|
| 374 |
+
3100 val loss 4.0792
|
| 375 |
+
3100 val perplexity 59.0958
|
| 376 |
+
3100 train 4.080646 (lr=3.2560e-04) (hash(x)=156381859)
|
| 377 |
+
3110 train 4.024241 (lr=3.2250e-04) (hash(x)=141462203)
|
| 378 |
+
3120 train 4.069233 (lr=3.1941e-04) (hash(x)=145671245)
|
| 379 |
+
3130 train 4.061683 (lr=3.1634e-04) (hash(x)=167814117)
|
| 380 |
+
3140 train 3.918876 (lr=3.1328e-04) (hash(x)=145540357)
|
| 381 |
+
3150 train 3.981985 (lr=3.1025e-04) (hash(x)=152783577)
|
| 382 |
+
3160 train 4.030642 (lr=3.0723e-04) (hash(x)=165882035)
|
| 383 |
+
3170 train 4.032357 (lr=3.0422e-04) (hash(x)=145400469)
|
| 384 |
+
3180 train 3.834909 (lr=3.0124e-04) (hash(x)=142466281)
|
| 385 |
+
3190 train 3.820429 (lr=2.9827e-04) (hash(x)=146748038)
|
| 386 |
+
3200 val loss 4.0827
|
| 387 |
+
3200 val perplexity 59.3041
|
| 388 |
+
3200 train 3.910803 (lr=2.9532e-04) (hash(x)=137714353)
|
| 389 |
+
3210 train 3.763382 (lr=2.9239e-04) (hash(x)=138832871)
|
| 390 |
+
3220 train 3.856369 (lr=2.8948e-04) (hash(x)=176981093)
|
| 391 |
+
3230 train 4.148771 (lr=2.8658e-04) (hash(x)=142798954)
|
| 392 |
+
3240 train 4.118919 (lr=2.8370e-04) (hash(x)=159942548)
|
| 393 |
+
3250 train 4.068669 (lr=2.8084e-04) (hash(x)=160734682)
|
| 394 |
+
3260 train 4.101717 (lr=2.7800e-04) (hash(x)=140312993)
|
| 395 |
+
3270 train 4.081190 (lr=2.7518e-04) (hash(x)=149573286)
|
| 396 |
+
3280 train 4.047223 (lr=2.7238e-04) (hash(x)=143477630)
|
| 397 |
+
3290 train 4.030579 (lr=2.6960e-04) (hash(x)=153981909)
|
| 398 |
+
3300 val loss 4.0667
|
| 399 |
+
3300 val perplexity 58.3651
|
| 400 |
+
3300 train 4.121929 (lr=2.6684e-04) (hash(x)=159001813)
|
| 401 |
+
3310 train 3.955050 (lr=2.6409e-04) (hash(x)=149855896)
|
| 402 |
+
3320 train 4.022514 (lr=2.6137e-04) (hash(x)=141107417)
|
| 403 |
+
3330 train 3.915199 (lr=2.5867e-04) (hash(x)=149073289)
|
| 404 |
+
3340 train 3.899860 (lr=2.5598e-04) (hash(x)=151668579)
|
| 405 |
+
3350 train 3.999893 (lr=2.5332e-04) (hash(x)=173643651)
|
| 406 |
+
3360 train 3.936888 (lr=2.5068e-04) (hash(x)=152242363)
|
| 407 |
+
3370 train 4.043324 (lr=2.4806e-04) (hash(x)=151562166)
|
| 408 |
+
3380 train 4.088456 (lr=2.4545e-04) (hash(x)=151581565)
|
| 409 |
+
3390 train 4.111154 (lr=2.4287e-04) (hash(x)=141212401)
|
| 410 |
+
3400 val loss 4.0513
|
| 411 |
+
3400 val perplexity 57.4747
|
| 412 |
+
3400 train 4.098230 (lr=2.4031e-04) (hash(x)=135183673)
|
| 413 |
+
3410 train 4.039259 (lr=2.3777e-04) (hash(x)=165421497)
|
| 414 |
+
3420 train 4.045000 (lr=2.3526e-04) (hash(x)=135505826)
|
| 415 |
+
3430 train 4.064657 (lr=2.3276e-04) (hash(x)=145700270)
|
| 416 |
+
3440 train 3.941707 (lr=2.3029e-04) (hash(x)=152515836)
|
| 417 |
+
3450 train 4.070548 (lr=2.2784e-04) (hash(x)=149810888)
|
| 418 |
+
3460 train 3.989117 (lr=2.2540e-04) (hash(x)=152352159)
|
| 419 |
+
3470 train 4.096410 (lr=2.2300e-04) (hash(x)=153188532)
|
| 420 |
+
3480 train 3.910647 (lr=2.2061e-04) (hash(x)=143030925)
|
| 421 |
+
3490 train 3.972712 (lr=2.1825e-04) (hash(x)=146377314)
|
| 422 |
+
3500 val loss 4.0447
|
| 423 |
+
3500 val perplexity 57.0950
|
| 424 |
+
3500 train 4.001482 (lr=2.1590e-04) (hash(x)=145858334)
|
| 425 |
+
3510 train 4.020387 (lr=2.1358e-04) (hash(x)=140413285)
|
| 426 |
+
3520 train 3.824541 (lr=2.1129e-04) (hash(x)=152355806)
|
| 427 |
+
3530 train 3.921628 (lr=2.0901e-04) (hash(x)=161039622)
|
| 428 |
+
3540 train 3.795853 (lr=2.0676e-04) (hash(x)=145875530)
|
| 429 |
+
3550 train 3.714938 (lr=2.0454e-04) (hash(x)=152121088)
|
| 430 |
+
3560 train 3.882571 (lr=2.0233e-04) (hash(x)=157278781)
|
| 431 |
+
3570 train 4.000728 (lr=2.0015e-04) (hash(x)=174173723)
|
| 432 |
+
3580 train 4.019886 (lr=1.9799e-04) (hash(x)=153900568)
|
| 433 |
+
3590 train 4.082007 (lr=1.9586e-04) (hash(x)=149821849)
|
| 434 |
+
3600 val loss 4.0349
|
| 435 |
+
3600 val perplexity 56.5378
|
| 436 |
+
3600 train 4.136543 (lr=1.9375e-04) (hash(x)=139218384)
|
| 437 |
+
3610 train 3.995315 (lr=1.9166e-04) (hash(x)=140631585)
|
| 438 |
+
3620 train 4.074436 (lr=1.8960e-04) (hash(x)=149050718)
|
| 439 |
+
3630 train 4.056933 (lr=1.8756e-04) (hash(x)=156692726)
|
| 440 |
+
3640 train 4.006634 (lr=1.8555e-04) (hash(x)=139322335)
|
| 441 |
+
3650 train 4.092237 (lr=1.8356e-04) (hash(x)=149932950)
|
| 442 |
+
3660 train 3.982006 (lr=1.8159e-04) (hash(x)=150054869)
|
| 443 |
+
3670 train 3.952618 (lr=1.7965e-04) (hash(x)=159536042)
|
| 444 |
+
3680 train 4.020516 (lr=1.7774e-04) (hash(x)=150125974)
|
| 445 |
+
3690 train 3.964711 (lr=1.7585e-04) (hash(x)=147491471)
|
| 446 |
+
3700 val loss 4.0291
|
| 447 |
+
3700 val perplexity 56.2102
|
| 448 |
+
3700 train 3.910765 (lr=1.7398e-04) (hash(x)=164791918)
|
| 449 |
+
3710 train 4.001108 (lr=1.7214e-04) (hash(x)=154903363)
|
| 450 |
+
3720 train 3.822361 (lr=1.7032e-04) (hash(x)=144773414)
|
| 451 |
+
3730 train 3.810828 (lr=1.6853e-04) (hash(x)=150811300)
|
| 452 |
+
3740 train 3.887110 (lr=1.6677e-04) (hash(x)=139704691)
|
| 453 |
+
3750 train 3.809560 (lr=1.6503e-04) (hash(x)=149158040)
|
| 454 |
+
3760 train 4.066328 (lr=1.6331e-04) (hash(x)=144392699)
|
| 455 |
+
3770 train 4.041584 (lr=1.6162e-04) (hash(x)=145892447)
|
| 456 |
+
3780 train 4.005530 (lr=1.5996e-04) (hash(x)=152909322)
|
| 457 |
+
3790 train 4.054597 (lr=1.5832e-04) (hash(x)=142840107)
|
| 458 |
+
3800 val loss 4.0200
|
| 459 |
+
3800 val perplexity 55.7035
|
| 460 |
+
3800 train 4.038223 (lr=1.5671e-04) (hash(x)=153669007)
|
| 461 |
+
3810 train 3.970003 (lr=1.5512e-04) (hash(x)=163222094)
|
| 462 |
+
3820 train 4.019078 (lr=1.5356e-04) (hash(x)=146535831)
|
| 463 |
+
3830 train 3.945134 (lr=1.5203e-04) (hash(x)=146838204)
|
| 464 |
+
3840 train 4.038815 (lr=1.5052e-04) (hash(x)=143898682)
|
| 465 |
+
3850 train 3.991074 (lr=1.4904e-04) (hash(x)=160530792)
|
| 466 |
+
3860 train 3.994405 (lr=1.4759e-04) (hash(x)=150523987)
|
| 467 |
+
3870 train 3.955498 (lr=1.4616e-04) (hash(x)=150824699)
|
| 468 |
+
3880 train 3.967726 (lr=1.4476e-04) (hash(x)=133425975)
|
| 469 |
+
3890 train 4.093961 (lr=1.4339e-04) (hash(x)=140181774)
|
| 470 |
+
3900 val loss 4.0174
|
| 471 |
+
3900 val perplexity 55.5541
|
| 472 |
+
3900 train 3.996239 (lr=1.4204e-04) (hash(x)=148206418)
|
| 473 |
+
3910 train 3.925895 (lr=1.4072e-04) (hash(x)=145147878)
|
| 474 |
+
3920 train 3.883774 (lr=1.3942e-04) (hash(x)=148302875)
|
| 475 |
+
3930 train 3.889167 (lr=1.3816e-04) (hash(x)=151233666)
|
| 476 |
+
3940 train 3.872264 (lr=1.3692e-04) (hash(x)=150629020)
|
| 477 |
+
3950 train 3.870873 (lr=1.3570e-04) (hash(x)=158912397)
|
| 478 |
+
3960 train 4.058496 (lr=1.3452e-04) (hash(x)=151226523)
|
| 479 |
+
3970 train 4.105200 (lr=1.3336e-04) (hash(x)=158058808)
|
| 480 |
+
3980 train 4.101491 (lr=1.3223e-04) (hash(x)=156692706)
|
| 481 |
+
3990 train 4.024655 (lr=1.3113e-04) (hash(x)=168505851)
|
| 482 |
+
4000 val loss 4.0100
|
| 483 |
+
4000 val perplexity 55.1463
|
| 484 |
+
4000 train 4.073246 (lr=1.3005e-04) (hash(x)=149296573)
|
| 485 |
+
4010 train 3.918898 (lr=1.2900e-04) (hash(x)=180932838)
|
| 486 |
+
4020 train 4.063310 (lr=1.2798e-04) (hash(x)=159234400)
|
| 487 |
+
4030 train 3.941642 (lr=1.2699e-04) (hash(x)=141611733)
|
| 488 |
+
4040 train 4.077459 (lr=1.2602e-04) (hash(x)=154160884)
|
| 489 |
+
4050 train 4.016549 (lr=1.2509e-04) (hash(x)=151275031)
|
| 490 |
+
4060 train 3.885850 (lr=1.2418e-04) (hash(x)=139772847)
|
| 491 |
+
4070 train 3.868974 (lr=1.2329e-04) (hash(x)=147858808)
|
| 492 |
+
4080 train 3.955659 (lr=1.2244e-04) (hash(x)=151808464)
|
| 493 |
+
4090 train 4.084455 (lr=1.2161e-04) (hash(x)=136602254)
|
| 494 |
+
4100 val loss 4.0067
|
| 495 |
+
4100 val perplexity 54.9654
|
| 496 |
+
4100 train 3.949826 (lr=1.2082e-04) (hash(x)=147779302)
|
| 497 |
+
4110 train 3.891533 (lr=1.2005e-04) (hash(x)=155402005)
|
| 498 |
+
4120 train 3.822579 (lr=1.1931e-04) (hash(x)=153020044)
|
| 499 |
+
4130 train 3.921874 (lr=1.1859e-04) (hash(x)=132377576)
|
| 500 |
+
4140 train 3.934016 (lr=1.1791e-04) (hash(x)=146920817)
|
| 501 |
+
4150 train 3.999616 (lr=1.1725e-04) (hash(x)=163754570)
|
| 502 |
+
4160 train 4.000860 (lr=1.1662e-04) (hash(x)=153950650)
|
| 503 |
+
4170 train 3.900442 (lr=1.1602e-04) (hash(x)=126995123)
|
| 504 |
+
4180 train 4.087460 (lr=1.1545e-04) (hash(x)=143060933)
|
| 505 |
+
4190 train 4.091929 (lr=1.1491e-04) (hash(x)=156383960)
|
| 506 |
+
4200 val loss 4.0027
|
| 507 |
+
4200 val perplexity 54.7434
|
| 508 |
+
4200 train 3.977340 (lr=1.1439e-04) (hash(x)=146024494)
|
| 509 |
+
4210 train 3.968023 (lr=1.1390e-04) (hash(x)=145999484)
|
| 510 |
+
4220 train 3.987368 (lr=1.1344e-04) (hash(x)=142228099)
|
| 511 |
+
4230 train 4.088376 (lr=1.1302e-04) (hash(x)=145916043)
|
| 512 |
+
4240 train 4.070943 (lr=1.1261e-04) (hash(x)=142443731)
|
| 513 |
+
4250 train 4.044500 (lr=1.1224e-04) (hash(x)=167955144)
|
| 514 |
+
4260 train 3.873861 (lr=1.1190e-04) (hash(x)=146210105)
|
| 515 |
+
4270 train 3.977408 (lr=1.1158e-04) (hash(x)=152592455)
|
| 516 |
+
4280 train 3.972079 (lr=1.1130e-04) (hash(x)=147241106)
|
| 517 |
+
4290 train 3.874514 (lr=1.1104e-04) (hash(x)=143797512)
|
| 518 |
+
4300 val loss 3.9999
|
| 519 |
+
4300 val perplexity 54.5952
|
| 520 |
+
4300 train 4.025804 (lr=1.1081e-04) (hash(x)=151764773)
|
| 521 |
+
4310 train 4.020394 (lr=1.1061e-04) (hash(x)=147835320)
|
| 522 |
+
4320 train 4.185558 (lr=1.1043e-04) (hash(x)=158566371)
|
| 523 |
+
4330 train 3.949465 (lr=1.1029e-04) (hash(x)=139985797)
|
| 524 |
+
4340 train 3.936380 (lr=1.1018e-04) (hash(x)=153707587)
|
| 525 |
+
4350 train 3.991140 (lr=1.1009e-04) (hash(x)=156639133)
|
| 526 |
+
4360 train 4.051152 (lr=1.1003e-04) (hash(x)=145986727)
|
| 527 |
+
4370 train 4.066224 (lr=1.1000e-04) (hash(x)=142862019)
|
| 528 |
+
4374 val loss 3.9938
|
| 529 |
+
4374 val perplexity 54.2603
|
lr11e-4_total_batch_size122880_baseline_seed1339/model_04374.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97a15a8fdd610176b757df33b977181d8127c3ccab40ffa0335c748d38c13b0f
|
| 3 |
+
size 92843394
|
lr11e-4_total_batch_size122880_baseline_seed1339/optimizer_04374.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:475e18929d7440c28db18798c8eeca803f6a99e1739e35c806b7ddc6b878baef
|
| 3 |
+
size 179406214
|