Upload folder using huggingface_hub
Browse files- lr13e-4_total_batch_size122880_baseline_seed1339/args.json +1 -0
- lr13e-4_total_batch_size122880_baseline_seed1339/dataloader_04374.pt +3 -0
- lr13e-4_total_batch_size122880_baseline_seed1339/log2.txt +529 -0
- lr13e-4_total_batch_size122880_baseline_seed1339/model_04374.pt +3 -0
- lr13e-4_total_batch_size122880_baseline_seed1339/optimizer_04374.pt +3 -0
lr13e-4_total_batch_size122880_baseline_seed1339/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/lr13e-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": 0.0013, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "13e-4_122880", "n_embd": 256}
|
lr13e-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
|
lr13e-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=5.2000e-06) (hash(x)=150724848)
|
| 5 |
+
10 train 9.899696 (lr=5.7200e-05) (hash(x)=153483789)
|
| 6 |
+
20 train 9.255864 (lr=1.0920e-04) (hash(x)=145287265)
|
| 7 |
+
30 train 8.395938 (lr=1.6120e-04) (hash(x)=150157569)
|
| 8 |
+
40 train 7.827944 (lr=2.1320e-04) (hash(x)=152157598)
|
| 9 |
+
50 train 7.644742 (lr=2.6520e-04) (hash(x)=149910534)
|
| 10 |
+
60 train 7.606067 (lr=3.1720e-04) (hash(x)=150731273)
|
| 11 |
+
70 train 7.509567 (lr=3.6920e-04) (hash(x)=166473281)
|
| 12 |
+
80 train 7.229414 (lr=4.2120e-04) (hash(x)=151371145)
|
| 13 |
+
90 train 7.079880 (lr=4.7320e-04) (hash(x)=153368917)
|
| 14 |
+
100 val loss 7.0661
|
| 15 |
+
100 val perplexity 1171.5461
|
| 16 |
+
100 train 6.994465 (lr=5.2520e-04) (hash(x)=148123706)
|
| 17 |
+
110 train 6.842206 (lr=5.7720e-04) (hash(x)=150622583)
|
| 18 |
+
120 train 6.648126 (lr=6.2920e-04) (hash(x)=151535384)
|
| 19 |
+
130 train 6.599626 (lr=6.8120e-04) (hash(x)=148726088)
|
| 20 |
+
140 train 6.478449 (lr=7.3320e-04) (hash(x)=157260375)
|
| 21 |
+
150 train 6.346583 (lr=7.8520e-04) (hash(x)=146678221)
|
| 22 |
+
160 train 6.526664 (lr=8.3720e-04) (hash(x)=144947971)
|
| 23 |
+
170 train 6.440884 (lr=8.8920e-04) (hash(x)=162318756)
|
| 24 |
+
180 train 6.440250 (lr=9.4120e-04) (hash(x)=178373797)
|
| 25 |
+
190 train 6.165244 (lr=9.9320e-04) (hash(x)=150369080)
|
| 26 |
+
200 val loss 6.2108
|
| 27 |
+
200 val perplexity 498.0982
|
| 28 |
+
200 train 6.240228 (lr=1.0452e-03) (hash(x)=151700982)
|
| 29 |
+
210 train 6.050064 (lr=1.0972e-03) (hash(x)=148930522)
|
| 30 |
+
220 train 6.074565 (lr=1.1492e-03) (hash(x)=148079227)
|
| 31 |
+
230 train 5.976666 (lr=1.2012e-03) (hash(x)=145272617)
|
| 32 |
+
240 train 5.870818 (lr=1.2532e-03) (hash(x)=156287490)
|
| 33 |
+
250 train 5.847607 (lr=1.3000e-03) (hash(x)=156182087)
|
| 34 |
+
260 train 5.885349 (lr=1.3000e-03) (hash(x)=162758254)
|
| 35 |
+
270 train 5.703783 (lr=1.2999e-03) (hash(x)=157363914)
|
| 36 |
+
280 train 5.588426 (lr=1.2998e-03) (hash(x)=149449280)
|
| 37 |
+
290 train 5.619856 (lr=1.2997e-03) (hash(x)=149266312)
|
| 38 |
+
300 val loss 5.6849
|
| 39 |
+
300 val perplexity 294.3742
|
| 40 |
+
300 train 5.517653 (lr=1.2996e-03) (hash(x)=149318660)
|
| 41 |
+
310 train 5.372299 (lr=1.2994e-03) (hash(x)=137184623)
|
| 42 |
+
320 train 5.339736 (lr=1.2992e-03) (hash(x)=143999212)
|
| 43 |
+
330 train 5.204526 (lr=1.2989e-03) (hash(x)=138360108)
|
| 44 |
+
340 train 5.232503 (lr=1.2986e-03) (hash(x)=143752601)
|
| 45 |
+
350 train 5.509124 (lr=1.2983e-03) (hash(x)=150482428)
|
| 46 |
+
360 train 5.473442 (lr=1.2979e-03) (hash(x)=159461069)
|
| 47 |
+
370 train 5.433396 (lr=1.2976e-03) (hash(x)=149967984)
|
| 48 |
+
380 train 5.448001 (lr=1.2971e-03) (hash(x)=155106202)
|
| 49 |
+
390 train 5.414634 (lr=1.2967e-03) (hash(x)=155825158)
|
| 50 |
+
400 val loss 5.3542
|
| 51 |
+
400 val perplexity 211.4961
|
| 52 |
+
400 train 5.340728 (lr=1.2962e-03) (hash(x)=143268605)
|
| 53 |
+
410 train 5.325308 (lr=1.2957e-03) (hash(x)=179384792)
|
| 54 |
+
420 train 5.492976 (lr=1.2951e-03) (hash(x)=142711393)
|
| 55 |
+
430 train 5.143899 (lr=1.2945e-03) (hash(x)=138495801)
|
| 56 |
+
440 train 5.226528 (lr=1.2939e-03) (hash(x)=151951501)
|
| 57 |
+
450 train 5.168422 (lr=1.2932e-03) (hash(x)=152322423)
|
| 58 |
+
460 train 5.160437 (lr=1.2925e-03) (hash(x)=148720342)
|
| 59 |
+
470 train 5.115890 (lr=1.2918e-03) (hash(x)=150349503)
|
| 60 |
+
480 train 5.091730 (lr=1.2910e-03) (hash(x)=146292562)
|
| 61 |
+
490 train 4.944960 (lr=1.2903e-03) (hash(x)=137640259)
|
| 62 |
+
500 val loss 5.1459
|
| 63 |
+
500 val perplexity 171.7290
|
| 64 |
+
500 train 4.906366 (lr=1.2894e-03) (hash(x)=147904298)
|
| 65 |
+
510 train 4.895032 (lr=1.2886e-03) (hash(x)=140706537)
|
| 66 |
+
520 train 4.849339 (lr=1.2877e-03) (hash(x)=144746445)
|
| 67 |
+
530 train 4.774483 (lr=1.2867e-03) (hash(x)=153065665)
|
| 68 |
+
540 train 4.891927 (lr=1.2858e-03) (hash(x)=157674756)
|
| 69 |
+
550 train 5.245577 (lr=1.2848e-03) (hash(x)=154343147)
|
| 70 |
+
560 train 4.920190 (lr=1.2838e-03) (hash(x)=162479784)
|
| 71 |
+
570 train 4.991161 (lr=1.2827e-03) (hash(x)=146934299)
|
| 72 |
+
580 train 4.964786 (lr=1.2816e-03) (hash(x)=145536832)
|
| 73 |
+
590 train 4.988544 (lr=1.2805e-03) (hash(x)=158271614)
|
| 74 |
+
600 val loss 4.9562
|
| 75 |
+
600 val perplexity 142.0592
|
| 76 |
+
600 train 4.999316 (lr=1.2793e-03) (hash(x)=141843115)
|
| 77 |
+
610 train 4.871832 (lr=1.2781e-03) (hash(x)=144149947)
|
| 78 |
+
620 train 4.946622 (lr=1.2769e-03) (hash(x)=152217389)
|
| 79 |
+
630 train 4.866689 (lr=1.2757e-03) (hash(x)=144029489)
|
| 80 |
+
640 train 4.888097 (lr=1.2744e-03) (hash(x)=151625838)
|
| 81 |
+
650 train 4.830356 (lr=1.2731e-03) (hash(x)=145279030)
|
| 82 |
+
660 train 4.652700 (lr=1.2717e-03) (hash(x)=133642994)
|
| 83 |
+
670 train 4.810959 (lr=1.2703e-03) (hash(x)=155919973)
|
| 84 |
+
680 train 4.716548 (lr=1.2689e-03) (hash(x)=150613636)
|
| 85 |
+
690 train 4.921138 (lr=1.2675e-03) (hash(x)=152803906)
|
| 86 |
+
700 val loss 4.8303
|
| 87 |
+
700 val perplexity 125.2537
|
| 88 |
+
700 train 4.635255 (lr=1.2660e-03) (hash(x)=152507639)
|
| 89 |
+
710 train 4.624895 (lr=1.2645e-03) (hash(x)=167824828)
|
| 90 |
+
720 train 4.569371 (lr=1.2629e-03) (hash(x)=150085494)
|
| 91 |
+
730 train 4.502444 (lr=1.2613e-03) (hash(x)=142956546)
|
| 92 |
+
740 train 4.488472 (lr=1.2597e-03) (hash(x)=150893345)
|
| 93 |
+
750 train 4.773106 (lr=1.2581e-03) (hash(x)=148473774)
|
| 94 |
+
760 train 4.675390 (lr=1.2564e-03) (hash(x)=166217855)
|
| 95 |
+
770 train 4.702281 (lr=1.2547e-03) (hash(x)=155013278)
|
| 96 |
+
780 train 4.719789 (lr=1.2530e-03) (hash(x)=152138735)
|
| 97 |
+
790 train 4.757064 (lr=1.2512e-03) (hash(x)=157176546)
|
| 98 |
+
800 val loss 4.6873
|
| 99 |
+
800 val perplexity 108.5599
|
| 100 |
+
800 train 4.640759 (lr=1.2494e-03) (hash(x)=151117002)
|
| 101 |
+
810 train 4.662795 (lr=1.2476e-03) (hash(x)=145015354)
|
| 102 |
+
820 train 4.660993 (lr=1.2457e-03) (hash(x)=148776668)
|
| 103 |
+
830 train 4.466759 (lr=1.2438e-03) (hash(x)=170658972)
|
| 104 |
+
840 train 4.737228 (lr=1.2419e-03) (hash(x)=155609234)
|
| 105 |
+
850 train 4.724372 (lr=1.2400e-03) (hash(x)=147018482)
|
| 106 |
+
860 train 4.571147 (lr=1.2380e-03) (hash(x)=136898027)
|
| 107 |
+
870 train 4.548481 (lr=1.2360e-03) (hash(x)=141846516)
|
| 108 |
+
880 train 4.676339 (lr=1.2339e-03) (hash(x)=161873750)
|
| 109 |
+
890 train 4.503053 (lr=1.2319e-03) (hash(x)=157878866)
|
| 110 |
+
900 val loss 4.6417
|
| 111 |
+
900 val perplexity 103.7195
|
| 112 |
+
900 train 4.595750 (lr=1.2298e-03) (hash(x)=145757704)
|
| 113 |
+
910 train 4.719009 (lr=1.2276e-03) (hash(x)=145786621)
|
| 114 |
+
920 train 4.611430 (lr=1.2255e-03) (hash(x)=142045584)
|
| 115 |
+
930 train 4.682396 (lr=1.2233e-03) (hash(x)=164426390)
|
| 116 |
+
940 train 4.564182 (lr=1.2211e-03) (hash(x)=139987987)
|
| 117 |
+
950 train 4.765593 (lr=1.2188e-03) (hash(x)=148921543)
|
| 118 |
+
960 train 4.566187 (lr=1.2165e-03) (hash(x)=154138655)
|
| 119 |
+
970 train 4.544908 (lr=1.2142e-03) (hash(x)=138488697)
|
| 120 |
+
980 train 4.574082 (lr=1.2119e-03) (hash(x)=147004214)
|
| 121 |
+
990 train 4.437335 (lr=1.2095e-03) (hash(x)=156465269)
|
| 122 |
+
1000 val loss 4.5496
|
| 123 |
+
1000 val perplexity 94.5930
|
| 124 |
+
1000 train 4.394066 (lr=1.2071e-03) (hash(x)=161795711)
|
| 125 |
+
1010 train 4.476116 (lr=1.2047e-03) (hash(x)=141132073)
|
| 126 |
+
1020 train 4.587778 (lr=1.2023e-03) (hash(x)=141901746)
|
| 127 |
+
1030 train 4.436138 (lr=1.1998e-03) (hash(x)=137935785)
|
| 128 |
+
1040 train 4.642429 (lr=1.1973e-03) (hash(x)=144386181)
|
| 129 |
+
1050 train 4.477895 (lr=1.1947e-03) (hash(x)=145581641)
|
| 130 |
+
1060 train 4.506867 (lr=1.1922e-03) (hash(x)=155473484)
|
| 131 |
+
1070 train 4.599480 (lr=1.1896e-03) (hash(x)=144911208)
|
| 132 |
+
1080 train 4.716012 (lr=1.1870e-03) (hash(x)=156601997)
|
| 133 |
+
1090 train 4.493799 (lr=1.1843e-03) (hash(x)=151299535)
|
| 134 |
+
1100 val loss 4.5085
|
| 135 |
+
1100 val perplexity 90.7829
|
| 136 |
+
1100 train 4.512877 (lr=1.1816e-03) (hash(x)=157673597)
|
| 137 |
+
1110 train 4.493972 (lr=1.1789e-03) (hash(x)=149224552)
|
| 138 |
+
1120 train 4.527355 (lr=1.1762e-03) (hash(x)=155650894)
|
| 139 |
+
1130 train 4.505894 (lr=1.1735e-03) (hash(x)=143649236)
|
| 140 |
+
1140 train 4.489985 (lr=1.1707e-03) (hash(x)=146796338)
|
| 141 |
+
1150 train 4.418267 (lr=1.1679e-03) (hash(x)=147384772)
|
| 142 |
+
1160 train 4.475804 (lr=1.1650e-03) (hash(x)=148232447)
|
| 143 |
+
1170 train 4.375490 (lr=1.1622e-03) (hash(x)=149417135)
|
| 144 |
+
1180 train 4.379090 (lr=1.1593e-03) (hash(x)=146962492)
|
| 145 |
+
1190 train 4.220077 (lr=1.1564e-03) (hash(x)=147119717)
|
| 146 |
+
1200 val loss 4.4887
|
| 147 |
+
1200 val perplexity 89.0013
|
| 148 |
+
1200 train 4.416452 (lr=1.1534e-03) (hash(x)=144115998)
|
| 149 |
+
1210 train 4.607876 (lr=1.1505e-03) (hash(x)=148830634)
|
| 150 |
+
1220 train 4.471360 (lr=1.1475e-03) (hash(x)=155788808)
|
| 151 |
+
1230 train 4.578928 (lr=1.1445e-03) (hash(x)=155420927)
|
| 152 |
+
1240 train 4.413360 (lr=1.1414e-03) (hash(x)=176100202)
|
| 153 |
+
1250 train 4.364956 (lr=1.1384e-03) (hash(x)=163245801)
|
| 154 |
+
1260 train 4.369982 (lr=1.1353e-03) (hash(x)=147341874)
|
| 155 |
+
1270 train 4.381311 (lr=1.1322e-03) (hash(x)=140477438)
|
| 156 |
+
1280 train 4.248888 (lr=1.1291e-03) (hash(x)=137109128)
|
| 157 |
+
1290 train 4.390874 (lr=1.1259e-03) (hash(x)=158887908)
|
| 158 |
+
1300 val loss 4.4217
|
| 159 |
+
1300 val perplexity 83.2404
|
| 160 |
+
1300 train 4.372225 (lr=1.1227e-03) (hash(x)=146922182)
|
| 161 |
+
1310 train 4.381297 (lr=1.1195e-03) (hash(x)=143857902)
|
| 162 |
+
1320 train 4.341904 (lr=1.1163e-03) (hash(x)=150276684)
|
| 163 |
+
1330 train 4.371931 (lr=1.1130e-03) (hash(x)=153362319)
|
| 164 |
+
1340 train 4.421975 (lr=1.1097e-03) (hash(x)=153140891)
|
| 165 |
+
1350 train 4.457381 (lr=1.1064e-03) (hash(x)=166976502)
|
| 166 |
+
1360 train 4.467720 (lr=1.1031e-03) (hash(x)=147648241)
|
| 167 |
+
1370 train 4.385450 (lr=1.0998e-03) (hash(x)=152606892)
|
| 168 |
+
1380 train 4.372163 (lr=1.0964e-03) (hash(x)=151845292)
|
| 169 |
+
1390 train 4.438265 (lr=1.0930e-03) (hash(x)=142307378)
|
| 170 |
+
1400 val loss 4.3866
|
| 171 |
+
1400 val perplexity 80.3645
|
| 172 |
+
1400 train 4.448207 (lr=1.0896e-03) (hash(x)=172927707)
|
| 173 |
+
1410 train 4.472632 (lr=1.0862e-03) (hash(x)=154387219)
|
| 174 |
+
1420 train 4.291861 (lr=1.0827e-03) (hash(x)=156081484)
|
| 175 |
+
1430 train 4.356396 (lr=1.0792e-03) (hash(x)=156584703)
|
| 176 |
+
1440 train 4.258140 (lr=1.0757e-03) (hash(x)=141998503)
|
| 177 |
+
1450 train 4.373052 (lr=1.0722e-03) (hash(x)=148708706)
|
| 178 |
+
1460 train 4.237865 (lr=1.0687e-03) (hash(x)=148096724)
|
| 179 |
+
1470 train 4.209556 (lr=1.0651e-03) (hash(x)=143995394)
|
| 180 |
+
1480 train 4.446826 (lr=1.0616e-03) (hash(x)=160523370)
|
| 181 |
+
1490 train 4.272639 (lr=1.0580e-03) (hash(x)=142703815)
|
| 182 |
+
1500 val loss 4.3767
|
| 183 |
+
1500 val perplexity 79.5748
|
| 184 |
+
1500 train 4.264549 (lr=1.0543e-03) (hash(x)=150145360)
|
| 185 |
+
1510 train 4.418281 (lr=1.0507e-03) (hash(x)=140955409)
|
| 186 |
+
1520 train 4.323253 (lr=1.0470e-03) (hash(x)=121541202)
|
| 187 |
+
1530 train 4.327451 (lr=1.0434e-03) (hash(x)=155534733)
|
| 188 |
+
1540 train 4.360955 (lr=1.0397e-03) (hash(x)=141973489)
|
| 189 |
+
1550 train 4.233703 (lr=1.0359e-03) (hash(x)=157912759)
|
| 190 |
+
1560 train 4.291804 (lr=1.0322e-03) (hash(x)=143045873)
|
| 191 |
+
1570 train 4.302728 (lr=1.0285e-03) (hash(x)=148243369)
|
| 192 |
+
1580 train 4.321183 (lr=1.0247e-03) (hash(x)=149719396)
|
| 193 |
+
1590 train 4.353673 (lr=1.0209e-03) (hash(x)=142111349)
|
| 194 |
+
1600 val loss 4.3334
|
| 195 |
+
1600 val perplexity 76.2063
|
| 196 |
+
1600 train 4.363453 (lr=1.0171e-03) (hash(x)=145393936)
|
| 197 |
+
1610 train 4.264659 (lr=1.0133e-03) (hash(x)=146389859)
|
| 198 |
+
1620 train 4.316600 (lr=1.0094e-03) (hash(x)=152903298)
|
| 199 |
+
1630 train 4.220786 (lr=1.0056e-03) (hash(x)=149203171)
|
| 200 |
+
1640 train 4.183780 (lr=1.0017e-03) (hash(x)=143289561)
|
| 201 |
+
1650 train 4.218773 (lr=9.9780e-04) (hash(x)=158724067)
|
| 202 |
+
1660 train 4.147505 (lr=9.9389e-04) (hash(x)=149609240)
|
| 203 |
+
1670 train 4.360039 (lr=9.8997e-04) (hash(x)=155240838)
|
| 204 |
+
1680 train 4.348688 (lr=9.8603e-04) (hash(x)=149973508)
|
| 205 |
+
1690 train 4.200274 (lr=9.8207e-04) (hash(x)=153481145)
|
| 206 |
+
1700 val loss 4.3123
|
| 207 |
+
1700 val perplexity 74.6097
|
| 208 |
+
1700 train 4.441117 (lr=9.7810e-04) (hash(x)=147623002)
|
| 209 |
+
1710 train 4.312900 (lr=9.7411e-04) (hash(x)=163392549)
|
| 210 |
+
1720 train 4.317345 (lr=9.7011e-04) (hash(x)=154608207)
|
| 211 |
+
1730 train 4.255018 (lr=9.6609e-04) (hash(x)=152409179)
|
| 212 |
+
1740 train 4.228596 (lr=9.6206e-04) (hash(x)=148273153)
|
| 213 |
+
1750 train 4.324174 (lr=9.5802e-04) (hash(x)=143259465)
|
| 214 |
+
1760 train 4.496826 (lr=9.5396e-04) (hash(x)=138718468)
|
| 215 |
+
1770 train 4.240559 (lr=9.4988e-04) (hash(x)=159255954)
|
| 216 |
+
1780 train 4.243383 (lr=9.4580e-04) (hash(x)=135857355)
|
| 217 |
+
1790 train 4.157434 (lr=9.4170e-04) (hash(x)=145791671)
|
| 218 |
+
1800 val loss 4.3002
|
| 219 |
+
1800 val perplexity 73.7141
|
| 220 |
+
1800 train 4.250954 (lr=9.3758e-04) (hash(x)=151062384)
|
| 221 |
+
1810 train 4.149126 (lr=9.3346e-04) (hash(x)=144984716)
|
| 222 |
+
1820 train 4.216257 (lr=9.2932e-04) (hash(x)=144698775)
|
| 223 |
+
1830 train 4.352280 (lr=9.2517e-04) (hash(x)=150245397)
|
| 224 |
+
1840 train 4.390389 (lr=9.2100e-04) (hash(x)=170661468)
|
| 225 |
+
1850 train 4.311930 (lr=9.1683e-04) (hash(x)=151431836)
|
| 226 |
+
1860 train 4.183702 (lr=9.1264e-04) (hash(x)=148600110)
|
| 227 |
+
1870 train 4.278683 (lr=9.0844e-04) (hash(x)=148519036)
|
| 228 |
+
1880 train 4.190976 (lr=9.0423e-04) (hash(x)=141697198)
|
| 229 |
+
1890 train 4.289632 (lr=9.0001e-04) (hash(x)=156289714)
|
| 230 |
+
1900 val loss 4.2707
|
| 231 |
+
1900 val perplexity 71.5745
|
| 232 |
+
1900 train 4.231341 (lr=8.9577e-04) (hash(x)=155386073)
|
| 233 |
+
1910 train 4.431067 (lr=8.9153e-04) (hash(x)=147240084)
|
| 234 |
+
1920 train 4.275314 (lr=8.8728e-04) (hash(x)=148388842)
|
| 235 |
+
1930 train 4.184752 (lr=8.8302e-04) (hash(x)=146204229)
|
| 236 |
+
1940 train 4.054188 (lr=8.7874e-04) (hash(x)=143556545)
|
| 237 |
+
1950 train 4.133732 (lr=8.7446e-04) (hash(x)=147956690)
|
| 238 |
+
1960 train 4.247025 (lr=8.7017e-04) (hash(x)=137526984)
|
| 239 |
+
1970 train 4.190096 (lr=8.6587e-04) (hash(x)=140760871)
|
| 240 |
+
1980 train 4.492144 (lr=8.6156e-04) (hash(x)=152103697)
|
| 241 |
+
1990 train 4.222541 (lr=8.5724e-04) (hash(x)=164248169)
|
| 242 |
+
2000 val loss 4.2467
|
| 243 |
+
2000 val perplexity 69.8737
|
| 244 |
+
2000 train 4.231060 (lr=8.5292e-04) (hash(x)=151842998)
|
| 245 |
+
2010 train 4.283141 (lr=8.4859e-04) (hash(x)=153881605)
|
| 246 |
+
2020 train 4.304131 (lr=8.4424e-04) (hash(x)=145254556)
|
| 247 |
+
2030 train 4.198712 (lr=8.3989e-04) (hash(x)=143283003)
|
| 248 |
+
2040 train 4.188898 (lr=8.3554e-04) (hash(x)=149520645)
|
| 249 |
+
2050 train 4.134045 (lr=8.3118e-04) (hash(x)=143001992)
|
| 250 |
+
2060 train 4.185867 (lr=8.2681e-04) (hash(x)=147300288)
|
| 251 |
+
2070 train 4.149887 (lr=8.2243e-04) (hash(x)=148000315)
|
| 252 |
+
2080 train 4.250723 (lr=8.1805e-04) (hash(x)=146292733)
|
| 253 |
+
2090 train 4.100210 (lr=8.1366e-04) (hash(x)=152253548)
|
| 254 |
+
2100 val loss 4.2450
|
| 255 |
+
2100 val perplexity 69.7582
|
| 256 |
+
2100 train 4.304736 (lr=8.0926e-04) (hash(x)=163228537)
|
| 257 |
+
2110 train 4.125165 (lr=8.0486e-04) (hash(x)=156319517)
|
| 258 |
+
2120 train 4.402017 (lr=8.0046e-04) (hash(x)=140368798)
|
| 259 |
+
2130 train 4.190157 (lr=7.9605e-04) (hash(x)=162813132)
|
| 260 |
+
2140 train 4.344852 (lr=7.9163e-04) (hash(x)=140817744)
|
| 261 |
+
2150 train 4.202003 (lr=7.8721e-04) (hash(x)=165620945)
|
| 262 |
+
2160 train 4.142948 (lr=7.8279e-04) (hash(x)=139579339)
|
| 263 |
+
2170 train 4.317454 (lr=7.7836e-04) (hash(x)=145505114)
|
| 264 |
+
2180 train 4.233964 (lr=7.7393e-04) (hash(x)=151040053)
|
| 265 |
+
2190 train 4.158587 (lr=7.6950e-04) (hash(x)=138105191)
|
| 266 |
+
2200 val loss 4.2189
|
| 267 |
+
2200 val perplexity 67.9621
|
| 268 |
+
2200 train 4.116855 (lr=7.6506e-04) (hash(x)=149758589)
|
| 269 |
+
2210 train 4.144750 (lr=7.6062e-04) (hash(x)=149125273)
|
| 270 |
+
2220 train 4.149987 (lr=7.5618e-04) (hash(x)=152919204)
|
| 271 |
+
2230 train 4.213974 (lr=7.5173e-04) (hash(x)=152120880)
|
| 272 |
+
2240 train 3.977977 (lr=7.4728e-04) (hash(x)=148124218)
|
| 273 |
+
2250 train 4.097326 (lr=7.4284e-04) (hash(x)=137089797)
|
| 274 |
+
2260 train 4.082672 (lr=7.3838e-04) (hash(x)=155806109)
|
| 275 |
+
2270 train 4.144444 (lr=7.3393e-04) (hash(x)=161868486)
|
| 276 |
+
2280 train 4.280127 (lr=7.2948e-04) (hash(x)=156017019)
|
| 277 |
+
2290 train 4.109678 (lr=7.2502e-04) (hash(x)=144077561)
|
| 278 |
+
2300 val loss 4.2041
|
| 279 |
+
2300 val perplexity 66.9608
|
| 280 |
+
2300 train 4.248718 (lr=7.2057e-04) (hash(x)=170869105)
|
| 281 |
+
2310 train 4.228435 (lr=7.1611e-04) (hash(x)=147916300)
|
| 282 |
+
2320 train 4.186760 (lr=7.1166e-04) (hash(x)=152467472)
|
| 283 |
+
2330 train 4.194165 (lr=7.0720e-04) (hash(x)=144785419)
|
| 284 |
+
2340 train 4.107856 (lr=7.0275e-04) (hash(x)=150432854)
|
| 285 |
+
2350 train 4.157103 (lr=6.9829e-04) (hash(x)=141272857)
|
| 286 |
+
2360 train 4.206523 (lr=6.9384e-04) (hash(x)=160564556)
|
| 287 |
+
2370 train 4.196453 (lr=6.8939e-04) (hash(x)=151879225)
|
| 288 |
+
2380 train 4.056332 (lr=6.8494e-04) (hash(x)=151530029)
|
| 289 |
+
2390 train 4.006645 (lr=6.8049e-04) (hash(x)=152832807)
|
| 290 |
+
2400 val loss 4.1931
|
| 291 |
+
2400 val perplexity 66.2253
|
| 292 |
+
2400 train 4.066688 (lr=6.7604e-04) (hash(x)=147479799)
|
| 293 |
+
2410 train 4.090053 (lr=6.7160e-04) (hash(x)=145055665)
|
| 294 |
+
2420 train 4.077344 (lr=6.6716e-04) (hash(x)=154768391)
|
| 295 |
+
2430 train 4.149685 (lr=6.6272e-04) (hash(x)=150388315)
|
| 296 |
+
2440 train 4.143412 (lr=6.5828e-04) (hash(x)=161064525)
|
| 297 |
+
2450 train 4.135171 (lr=6.5385e-04) (hash(x)=154887504)
|
| 298 |
+
2460 train 4.116873 (lr=6.4942e-04) (hash(x)=138201850)
|
| 299 |
+
2470 train 4.247937 (lr=6.4500e-04) (hash(x)=151711687)
|
| 300 |
+
2480 train 4.333182 (lr=6.4058e-04) (hash(x)=147643099)
|
| 301 |
+
2490 train 4.202165 (lr=6.3616e-04) (hash(x)=143125010)
|
| 302 |
+
2500 val loss 4.1643
|
| 303 |
+
2500 val perplexity 64.3495
|
| 304 |
+
2500 train 4.222904 (lr=6.3175e-04) (hash(x)=142429041)
|
| 305 |
+
2510 train 4.173513 (lr=6.2734e-04) (hash(x)=148050199)
|
| 306 |
+
2520 train 4.007150 (lr=6.2294e-04) (hash(x)=90941145)
|
| 307 |
+
2530 train 4.125525 (lr=6.1854e-04) (hash(x)=150165963)
|
| 308 |
+
2540 train 4.047495 (lr=6.1415e-04) (hash(x)=150919495)
|
| 309 |
+
2550 train 4.039433 (lr=6.0976e-04) (hash(x)=160130013)
|
| 310 |
+
2560 train 4.059017 (lr=6.0538e-04) (hash(x)=148050104)
|
| 311 |
+
2570 train 4.140086 (lr=6.0101e-04) (hash(x)=145749915)
|
| 312 |
+
2580 train 4.067544 (lr=5.9664e-04) (hash(x)=149001490)
|
| 313 |
+
2590 train 3.978201 (lr=5.9228e-04) (hash(x)=158310837)
|
| 314 |
+
2600 val loss 4.1636
|
| 315 |
+
2600 val perplexity 64.3048
|
| 316 |
+
2600 train 3.946771 (lr=5.8793e-04) (hash(x)=146370729)
|
| 317 |
+
2610 train 3.939321 (lr=5.8358e-04) (hash(x)=154169203)
|
| 318 |
+
2620 train 4.084901 (lr=5.7925e-04) (hash(x)=137329685)
|
| 319 |
+
2630 train 4.007136 (lr=5.7492e-04) (hash(x)=160226256)
|
| 320 |
+
2640 train 4.086180 (lr=5.7060e-04) (hash(x)=153533222)
|
| 321 |
+
2650 train 4.144921 (lr=5.6628e-04) (hash(x)=157476060)
|
| 322 |
+
2660 train 4.221601 (lr=5.6198e-04) (hash(x)=164837154)
|
| 323 |
+
2670 train 4.045523 (lr=5.5768e-04) (hash(x)=155851799)
|
| 324 |
+
2680 train 4.148285 (lr=5.5340e-04) (hash(x)=162095841)
|
| 325 |
+
2690 train 4.183107 (lr=5.4912e-04) (hash(x)=140059129)
|
| 326 |
+
2700 val loss 4.1297
|
| 327 |
+
2700 val perplexity 62.1613
|
| 328 |
+
2700 train 4.130066 (lr=5.4485e-04) (hash(x)=149982495)
|
| 329 |
+
2710 train 4.139279 (lr=5.4059e-04) (hash(x)=154730619)
|
| 330 |
+
2720 train 4.096407 (lr=5.3635e-04) (hash(x)=153673087)
|
| 331 |
+
2730 train 4.196774 (lr=5.3211e-04) (hash(x)=141967987)
|
| 332 |
+
2740 train 4.042142 (lr=5.2788e-04) (hash(x)=148302521)
|
| 333 |
+
2750 train 4.070089 (lr=5.2367e-04) (hash(x)=169526708)
|
| 334 |
+
2760 train 4.077280 (lr=5.1946e-04) (hash(x)=147659745)
|
| 335 |
+
2770 train 4.051155 (lr=5.1527e-04) (hash(x)=165422791)
|
| 336 |
+
2780 train 4.031677 (lr=5.1109e-04) (hash(x)=139378759)
|
| 337 |
+
2790 train 3.842335 (lr=5.0692e-04) (hash(x)=149362982)
|
| 338 |
+
2800 val loss 4.1398
|
| 339 |
+
2800 val perplexity 62.7908
|
| 340 |
+
2800 train 3.910407 (lr=5.0276e-04) (hash(x)=166852181)
|
| 341 |
+
2810 train 3.874354 (lr=4.9861e-04) (hash(x)=162830529)
|
| 342 |
+
2820 train 3.820621 (lr=4.9448e-04) (hash(x)=150130923)
|
| 343 |
+
2830 train 4.124973 (lr=4.9036e-04) (hash(x)=132554804)
|
| 344 |
+
2840 train 4.121341 (lr=4.8625e-04) (hash(x)=156662719)
|
| 345 |
+
2850 train 4.159529 (lr=4.8216e-04) (hash(x)=157531009)
|
| 346 |
+
2860 train 4.122797 (lr=4.7808e-04) (hash(x)=153415901)
|
| 347 |
+
2870 train 4.112464 (lr=4.7401e-04) (hash(x)=157332351)
|
| 348 |
+
2880 train 4.060658 (lr=4.6996e-04) (hash(x)=151410115)
|
| 349 |
+
2890 train 4.196651 (lr=4.6592e-04) (hash(x)=151755498)
|
| 350 |
+
2900 val loss 4.1089
|
| 351 |
+
2900 val perplexity 60.8799
|
| 352 |
+
2900 train 4.023666 (lr=4.6190e-04) (hash(x)=157395013)
|
| 353 |
+
2910 train 4.002108 (lr=4.5789e-04) (hash(x)=144568819)
|
| 354 |
+
2920 train 3.947620 (lr=4.5389e-04) (hash(x)=140411793)
|
| 355 |
+
2930 train 4.071809 (lr=4.4991e-04) (hash(x)=153672754)
|
| 356 |
+
2940 train 4.073793 (lr=4.4595e-04) (hash(x)=144747136)
|
| 357 |
+
2950 train 4.049365 (lr=4.4200e-04) (hash(x)=139873259)
|
| 358 |
+
2960 train 4.048344 (lr=4.3807e-04) (hash(x)=164010581)
|
| 359 |
+
2970 train 4.059557 (lr=4.3415e-04) (hash(x)=145020447)
|
| 360 |
+
2980 train 3.960986 (lr=4.3025e-04) (hash(x)=146536623)
|
| 361 |
+
2990 train 3.923598 (lr=4.2637e-04) (hash(x)=143152437)
|
| 362 |
+
3000 val loss 4.1206
|
| 363 |
+
3000 val perplexity 61.5961
|
| 364 |
+
3000 train 3.889336 (lr=4.2250e-04) (hash(x)=151165786)
|
| 365 |
+
3010 train 3.873349 (lr=4.1865e-04) (hash(x)=154439823)
|
| 366 |
+
3020 train 3.928835 (lr=4.1482e-04) (hash(x)=144196226)
|
| 367 |
+
3030 train 3.827914 (lr=4.1100e-04) (hash(x)=148787067)
|
| 368 |
+
3040 train 3.979329 (lr=4.0720e-04) (hash(x)=161017582)
|
| 369 |
+
3050 train 4.296149 (lr=4.0342e-04) (hash(x)=158552651)
|
| 370 |
+
3060 train 4.204640 (lr=3.9966e-04) (hash(x)=149243161)
|
| 371 |
+
3070 train 4.151746 (lr=3.9592e-04) (hash(x)=146431684)
|
| 372 |
+
3080 train 4.069300 (lr=3.9219e-04) (hash(x)=149547069)
|
| 373 |
+
3090 train 4.101101 (lr=3.8849e-04) (hash(x)=146636428)
|
| 374 |
+
3100 val loss 4.0840
|
| 375 |
+
3100 val perplexity 59.3815
|
| 376 |
+
3100 train 4.086329 (lr=3.8480e-04) (hash(x)=156381859)
|
| 377 |
+
3110 train 4.032917 (lr=3.8113e-04) (hash(x)=141462203)
|
| 378 |
+
3120 train 4.075357 (lr=3.7748e-04) (hash(x)=145671245)
|
| 379 |
+
3130 train 4.067023 (lr=3.7385e-04) (hash(x)=167814117)
|
| 380 |
+
3140 train 3.925965 (lr=3.7025e-04) (hash(x)=145540357)
|
| 381 |
+
3150 train 3.985930 (lr=3.6666e-04) (hash(x)=152783577)
|
| 382 |
+
3160 train 4.036351 (lr=3.6309e-04) (hash(x)=165882035)
|
| 383 |
+
3170 train 4.035054 (lr=3.5954e-04) (hash(x)=145400469)
|
| 384 |
+
3180 train 3.842237 (lr=3.5601e-04) (hash(x)=142466281)
|
| 385 |
+
3190 train 3.821066 (lr=3.5250e-04) (hash(x)=146748038)
|
| 386 |
+
3200 val loss 4.0865
|
| 387 |
+
3200 val perplexity 59.5323
|
| 388 |
+
3200 train 3.917698 (lr=3.4902e-04) (hash(x)=137714353)
|
| 389 |
+
3210 train 3.769162 (lr=3.4555e-04) (hash(x)=138832871)
|
| 390 |
+
3220 train 3.859823 (lr=3.4211e-04) (hash(x)=176981093)
|
| 391 |
+
3230 train 4.158112 (lr=3.3868e-04) (hash(x)=142798954)
|
| 392 |
+
3240 train 4.124155 (lr=3.3528e-04) (hash(x)=159942548)
|
| 393 |
+
3250 train 4.070099 (lr=3.3191e-04) (hash(x)=160734682)
|
| 394 |
+
3260 train 4.106826 (lr=3.2855e-04) (hash(x)=140312993)
|
| 395 |
+
3270 train 4.090263 (lr=3.2522e-04) (hash(x)=149573286)
|
| 396 |
+
3280 train 4.047934 (lr=3.2191e-04) (hash(x)=143477630)
|
| 397 |
+
3290 train 4.036378 (lr=3.1862e-04) (hash(x)=153981909)
|
| 398 |
+
3300 val loss 4.0706
|
| 399 |
+
3300 val perplexity 58.5904
|
| 400 |
+
3300 train 4.129128 (lr=3.1535e-04) (hash(x)=159001813)
|
| 401 |
+
3310 train 3.957122 (lr=3.1211e-04) (hash(x)=149855896)
|
| 402 |
+
3320 train 4.024250 (lr=3.0889e-04) (hash(x)=141107417)
|
| 403 |
+
3330 train 3.916763 (lr=3.0570e-04) (hash(x)=149073289)
|
| 404 |
+
3340 train 3.904666 (lr=3.0253e-04) (hash(x)=151668579)
|
| 405 |
+
3350 train 4.000971 (lr=2.9938e-04) (hash(x)=173643651)
|
| 406 |
+
3360 train 3.942657 (lr=2.9625e-04) (hash(x)=152242363)
|
| 407 |
+
3370 train 4.052671 (lr=2.9316e-04) (hash(x)=151562166)
|
| 408 |
+
3380 train 4.093901 (lr=2.9008e-04) (hash(x)=151581565)
|
| 409 |
+
3390 train 4.119246 (lr=2.8703e-04) (hash(x)=141212401)
|
| 410 |
+
3400 val loss 4.0557
|
| 411 |
+
3400 val perplexity 57.7232
|
| 412 |
+
3400 train 4.106876 (lr=2.8401e-04) (hash(x)=135183673)
|
| 413 |
+
3410 train 4.051873 (lr=2.8101e-04) (hash(x)=165421497)
|
| 414 |
+
3420 train 4.047119 (lr=2.7803e-04) (hash(x)=135505826)
|
| 415 |
+
3430 train 4.070667 (lr=2.7508e-04) (hash(x)=145700270)
|
| 416 |
+
3440 train 3.946532 (lr=2.7216e-04) (hash(x)=152515836)
|
| 417 |
+
3450 train 4.077172 (lr=2.6926e-04) (hash(x)=149810888)
|
| 418 |
+
3460 train 3.991472 (lr=2.6639e-04) (hash(x)=152352159)
|
| 419 |
+
3470 train 4.102365 (lr=2.6354e-04) (hash(x)=153188532)
|
| 420 |
+
3480 train 3.911688 (lr=2.6072e-04) (hash(x)=143030925)
|
| 421 |
+
3490 train 3.974265 (lr=2.5793e-04) (hash(x)=146377314)
|
| 422 |
+
3500 val loss 4.0483
|
| 423 |
+
3500 val perplexity 57.3001
|
| 424 |
+
3500 train 4.011786 (lr=2.5516e-04) (hash(x)=145858334)
|
| 425 |
+
3510 train 4.021500 (lr=2.5242e-04) (hash(x)=140413285)
|
| 426 |
+
3520 train 3.824435 (lr=2.4970e-04) (hash(x)=152355806)
|
| 427 |
+
3530 train 3.927970 (lr=2.4702e-04) (hash(x)=161039622)
|
| 428 |
+
3540 train 3.794086 (lr=2.4436e-04) (hash(x)=145875530)
|
| 429 |
+
3550 train 3.714986 (lr=2.4173e-04) (hash(x)=152121088)
|
| 430 |
+
3560 train 3.888797 (lr=2.3912e-04) (hash(x)=157278781)
|
| 431 |
+
3570 train 4.001958 (lr=2.3654e-04) (hash(x)=174173723)
|
| 432 |
+
3580 train 4.025462 (lr=2.3399e-04) (hash(x)=153900568)
|
| 433 |
+
3590 train 4.086500 (lr=2.3147e-04) (hash(x)=149821849)
|
| 434 |
+
3600 val loss 4.0388
|
| 435 |
+
3600 val perplexity 56.7594
|
| 436 |
+
3600 train 4.139464 (lr=2.2898e-04) (hash(x)=139218384)
|
| 437 |
+
3610 train 3.998794 (lr=2.2651e-04) (hash(x)=140631585)
|
| 438 |
+
3620 train 4.073269 (lr=2.2407e-04) (hash(x)=149050718)
|
| 439 |
+
3630 train 4.052765 (lr=2.2167e-04) (hash(x)=156692726)
|
| 440 |
+
3640 train 4.010030 (lr=2.1929e-04) (hash(x)=139322335)
|
| 441 |
+
3650 train 4.099500 (lr=2.1693e-04) (hash(x)=149932950)
|
| 442 |
+
3660 train 3.986239 (lr=2.1461e-04) (hash(x)=150054869)
|
| 443 |
+
3670 train 3.955571 (lr=2.1232e-04) (hash(x)=159536042)
|
| 444 |
+
3680 train 4.023818 (lr=2.1005e-04) (hash(x)=150125974)
|
| 445 |
+
3690 train 3.967466 (lr=2.0782e-04) (hash(x)=147491471)
|
| 446 |
+
3700 val loss 4.0319
|
| 447 |
+
3700 val perplexity 56.3673
|
| 448 |
+
3700 train 3.914078 (lr=2.0561e-04) (hash(x)=164791918)
|
| 449 |
+
3710 train 4.003021 (lr=2.0344e-04) (hash(x)=154903363)
|
| 450 |
+
3720 train 3.824463 (lr=2.0129e-04) (hash(x)=144773414)
|
| 451 |
+
3730 train 3.808792 (lr=1.9917e-04) (hash(x)=150811300)
|
| 452 |
+
3740 train 3.894309 (lr=1.9709e-04) (hash(x)=139704691)
|
| 453 |
+
3750 train 3.809958 (lr=1.9503e-04) (hash(x)=149158040)
|
| 454 |
+
3760 train 4.067762 (lr=1.9300e-04) (hash(x)=144392699)
|
| 455 |
+
3770 train 4.045252 (lr=1.9101e-04) (hash(x)=145892447)
|
| 456 |
+
3780 train 4.008160 (lr=1.8904e-04) (hash(x)=152909322)
|
| 457 |
+
3790 train 4.058767 (lr=1.8711e-04) (hash(x)=142840107)
|
| 458 |
+
3800 val loss 4.0218
|
| 459 |
+
3800 val perplexity 55.8012
|
| 460 |
+
3800 train 4.043006 (lr=1.8520e-04) (hash(x)=153669007)
|
| 461 |
+
3810 train 3.973555 (lr=1.8333e-04) (hash(x)=163222094)
|
| 462 |
+
3820 train 4.026102 (lr=1.8149e-04) (hash(x)=146535831)
|
| 463 |
+
3830 train 3.950185 (lr=1.7967e-04) (hash(x)=146838204)
|
| 464 |
+
3840 train 4.040135 (lr=1.7789e-04) (hash(x)=143898682)
|
| 465 |
+
3850 train 3.988523 (lr=1.7614e-04) (hash(x)=160530792)
|
| 466 |
+
3860 train 3.999667 (lr=1.7442e-04) (hash(x)=150523987)
|
| 467 |
+
3870 train 3.956961 (lr=1.7274e-04) (hash(x)=150824699)
|
| 468 |
+
3880 train 3.968313 (lr=1.7108e-04) (hash(x)=133425975)
|
| 469 |
+
3890 train 4.097871 (lr=1.6946e-04) (hash(x)=140181774)
|
| 470 |
+
3900 val loss 4.0191
|
| 471 |
+
3900 val perplexity 55.6492
|
| 472 |
+
3900 train 3.995764 (lr=1.6786e-04) (hash(x)=148206418)
|
| 473 |
+
3910 train 3.926682 (lr=1.6630e-04) (hash(x)=145147878)
|
| 474 |
+
3920 train 3.884958 (lr=1.6477e-04) (hash(x)=148302875)
|
| 475 |
+
3930 train 3.892116 (lr=1.6328e-04) (hash(x)=151233666)
|
| 476 |
+
3940 train 3.878435 (lr=1.6181e-04) (hash(x)=150629020)
|
| 477 |
+
3950 train 3.873196 (lr=1.6038e-04) (hash(x)=158912397)
|
| 478 |
+
3960 train 4.064729 (lr=1.5898e-04) (hash(x)=151226523)
|
| 479 |
+
3970 train 4.104826 (lr=1.5761e-04) (hash(x)=158058808)
|
| 480 |
+
3980 train 4.105172 (lr=1.5627e-04) (hash(x)=156692706)
|
| 481 |
+
3990 train 4.028969 (lr=1.5497e-04) (hash(x)=168505851)
|
| 482 |
+
4000 val loss 4.0123
|
| 483 |
+
4000 val perplexity 55.2753
|
| 484 |
+
4000 train 4.077477 (lr=1.5370e-04) (hash(x)=149296573)
|
| 485 |
+
4010 train 3.924878 (lr=1.5246e-04) (hash(x)=180932838)
|
| 486 |
+
4020 train 4.066677 (lr=1.5125e-04) (hash(x)=159234400)
|
| 487 |
+
4030 train 3.945107 (lr=1.5008e-04) (hash(x)=141611733)
|
| 488 |
+
4040 train 4.082750 (lr=1.4894e-04) (hash(x)=154160884)
|
| 489 |
+
4050 train 4.017225 (lr=1.4783e-04) (hash(x)=151275031)
|
| 490 |
+
4060 train 3.883554 (lr=1.4675e-04) (hash(x)=139772847)
|
| 491 |
+
4070 train 3.873773 (lr=1.4571e-04) (hash(x)=147858808)
|
| 492 |
+
4080 train 3.959400 (lr=1.4470e-04) (hash(x)=151808464)
|
| 493 |
+
4090 train 4.087124 (lr=1.4373e-04) (hash(x)=136602254)
|
| 494 |
+
4100 val loss 4.0086
|
| 495 |
+
4100 val perplexity 55.0676
|
| 496 |
+
4100 train 3.950099 (lr=1.4278e-04) (hash(x)=147779302)
|
| 497 |
+
4110 train 3.893926 (lr=1.4187e-04) (hash(x)=155402005)
|
| 498 |
+
4120 train 3.820303 (lr=1.4100e-04) (hash(x)=153020044)
|
| 499 |
+
4130 train 3.922626 (lr=1.4015e-04) (hash(x)=132377576)
|
| 500 |
+
4140 train 3.937971 (lr=1.3934e-04) (hash(x)=146920817)
|
| 501 |
+
4150 train 4.000101 (lr=1.3857e-04) (hash(x)=163754570)
|
| 502 |
+
4160 train 4.005728 (lr=1.3782e-04) (hash(x)=153950650)
|
| 503 |
+
4170 train 3.898536 (lr=1.3712e-04) (hash(x)=126995123)
|
| 504 |
+
4180 train 4.095392 (lr=1.3644e-04) (hash(x)=143060933)
|
| 505 |
+
4190 train 4.097093 (lr=1.3580e-04) (hash(x)=156383960)
|
| 506 |
+
4200 val loss 4.0039
|
| 507 |
+
4200 val perplexity 54.8113
|
| 508 |
+
4200 train 3.981920 (lr=1.3519e-04) (hash(x)=146024494)
|
| 509 |
+
4210 train 3.969098 (lr=1.3461e-04) (hash(x)=145999484)
|
| 510 |
+
4220 train 3.989855 (lr=1.3407e-04) (hash(x)=142228099)
|
| 511 |
+
4230 train 4.091044 (lr=1.3356e-04) (hash(x)=145916043)
|
| 512 |
+
4240 train 4.073423 (lr=1.3309e-04) (hash(x)=142443731)
|
| 513 |
+
4250 train 4.044354 (lr=1.3265e-04) (hash(x)=167955144)
|
| 514 |
+
4260 train 3.879480 (lr=1.3224e-04) (hash(x)=146210105)
|
| 515 |
+
4270 train 3.980434 (lr=1.3187e-04) (hash(x)=152592455)
|
| 516 |
+
4280 train 3.975655 (lr=1.3153e-04) (hash(x)=147241106)
|
| 517 |
+
4290 train 3.879006 (lr=1.3123e-04) (hash(x)=143797512)
|
| 518 |
+
4300 val loss 4.0017
|
| 519 |
+
4300 val perplexity 54.6904
|
| 520 |
+
4300 train 4.020705 (lr=1.3095e-04) (hash(x)=151764773)
|
| 521 |
+
4310 train 4.024444 (lr=1.3072e-04) (hash(x)=147835320)
|
| 522 |
+
4320 train 4.196402 (lr=1.3051e-04) (hash(x)=158566371)
|
| 523 |
+
4330 train 3.949323 (lr=1.3034e-04) (hash(x)=139985797)
|
| 524 |
+
4340 train 3.937892 (lr=1.3021e-04) (hash(x)=153707587)
|
| 525 |
+
4350 train 3.989393 (lr=1.3011e-04) (hash(x)=156639133)
|
| 526 |
+
4360 train 4.051330 (lr=1.3004e-04) (hash(x)=145986727)
|
| 527 |
+
4370 train 4.066138 (lr=1.3000e-04) (hash(x)=142862019)
|
| 528 |
+
4374 val loss 3.9954
|
| 529 |
+
4374 val perplexity 54.3490
|
lr13e-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:a129b050e927fe3966182890a13f527a79308ef7c9db2bd019261208b17a6e9e
|
| 3 |
+
size 92843394
|
lr13e-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 sha256:195dcc960d52f65c793dcbc429e84e7b53032d88bd3cf8ddb6c9daef0ac9d0b9
|
| 3 |
+
size 179406214
|