Upload folder using huggingface_hub
Browse files- lr6e-4_total_batch_size61440_baseline_seed1340/args.json +1 -0
- lr6e-4_total_batch_size61440_baseline_seed1340/dataloader_08749.pt +3 -0
- lr6e-4_total_batch_size61440_baseline_seed1340/log2.txt +1054 -0
- lr6e-4_total_batch_size61440_baseline_seed1340/model_08749.pt +3 -0
- lr6e-4_total_batch_size61440_baseline_seed1340/optimizer_08749.pt +3 -0
lr6e-4_total_batch_size61440_baseline_seed1340/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/lr6e-4_total_batch_size61440_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": 8750, "warmup_steps": 500, "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": 61440, "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": 0.0006, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "6e-4_61440", "n_embd": 256}
|
lr6e-4_total_batch_size61440_baseline_seed1340/dataloader_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:953385078aa3787b69fc6857dfd48b0a2cd2f4d27c6f8892e01211aca53d07f5
|
| 3 |
+
size 964
|
lr6e-4_total_batch_size61440_baseline_seed1340/log2.txt
ADDED
|
@@ -0,0 +1,1054 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
max_steps: 8750
|
| 2 |
+
0 val loss 11.2465
|
| 3 |
+
0 val perplexity 76607.9531
|
| 4 |
+
0 train 11.254622 (lr=1.2000e-06) (hash(x)=164406924)
|
| 5 |
+
10 train 10.486423 (lr=1.3200e-05) (hash(x)=152008797)
|
| 6 |
+
20 train 9.935457 (lr=2.5200e-05) (hash(x)=153494457)
|
| 7 |
+
30 train 9.582285 (lr=3.7200e-05) (hash(x)=137328499)
|
| 8 |
+
40 train 9.382899 (lr=4.9200e-05) (hash(x)=159333245)
|
| 9 |
+
50 train 8.991772 (lr=6.1200e-05) (hash(x)=177407419)
|
| 10 |
+
60 train 8.484509 (lr=7.3200e-05) (hash(x)=127877799)
|
| 11 |
+
70 train 8.067822 (lr=8.5200e-05) (hash(x)=140873918)
|
| 12 |
+
80 train 7.910814 (lr=9.7200e-05) (hash(x)=160249377)
|
| 13 |
+
90 train 7.667356 (lr=1.0920e-04) (hash(x)=154282910)
|
| 14 |
+
100 val loss 7.5881
|
| 15 |
+
100 val perplexity 1974.5780
|
| 16 |
+
100 train 7.494137 (lr=1.2120e-04) (hash(x)=144903932)
|
| 17 |
+
110 train 7.504959 (lr=1.3320e-04) (hash(x)=151685723)
|
| 18 |
+
120 train 7.299349 (lr=1.4520e-04) (hash(x)=152347864)
|
| 19 |
+
130 train 7.293032 (lr=1.5720e-04) (hash(x)=152230926)
|
| 20 |
+
140 train 7.148556 (lr=1.6920e-04) (hash(x)=142121048)
|
| 21 |
+
150 train 7.265087 (lr=1.8120e-04) (hash(x)=173839165)
|
| 22 |
+
160 train 7.050098 (lr=1.9320e-04) (hash(x)=158755931)
|
| 23 |
+
170 train 6.881308 (lr=2.0520e-04) (hash(x)=151645697)
|
| 24 |
+
180 train 7.167406 (lr=2.1720e-04) (hash(x)=179696111)
|
| 25 |
+
190 train 6.812750 (lr=2.2920e-04) (hash(x)=150511423)
|
| 26 |
+
200 val loss 6.8505
|
| 27 |
+
200 val perplexity 944.3774
|
| 28 |
+
200 train 6.975245 (lr=2.4120e-04) (hash(x)=167734596)
|
| 29 |
+
210 train 6.547418 (lr=2.5320e-04) (hash(x)=133157560)
|
| 30 |
+
220 train 6.813692 (lr=2.6520e-04) (hash(x)=152234670)
|
| 31 |
+
230 train 6.621839 (lr=2.7720e-04) (hash(x)=160995039)
|
| 32 |
+
240 train 6.495830 (lr=2.8920e-04) (hash(x)=139367591)
|
| 33 |
+
250 train 6.664530 (lr=3.0120e-04) (hash(x)=153224076)
|
| 34 |
+
260 train 6.560595 (lr=3.1320e-04) (hash(x)=156667529)
|
| 35 |
+
270 train 6.467413 (lr=3.2520e-04) (hash(x)=133883725)
|
| 36 |
+
280 train 6.291535 (lr=3.3720e-04) (hash(x)=151939311)
|
| 37 |
+
290 train 6.328707 (lr=3.4920e-04) (hash(x)=150290478)
|
| 38 |
+
300 val loss 6.3322
|
| 39 |
+
300 val perplexity 562.4131
|
| 40 |
+
300 train 6.290477 (lr=3.6120e-04) (hash(x)=149619098)
|
| 41 |
+
310 train 6.208458 (lr=3.7320e-04) (hash(x)=142344072)
|
| 42 |
+
320 train 6.271106 (lr=3.8520e-04) (hash(x)=151878766)
|
| 43 |
+
330 train 6.121476 (lr=3.9720e-04) (hash(x)=136419763)
|
| 44 |
+
340 train 6.240640 (lr=4.0920e-04) (hash(x)=160761498)
|
| 45 |
+
350 train 6.089416 (lr=4.2120e-04) (hash(x)=146539909)
|
| 46 |
+
360 train 6.200458 (lr=4.3320e-04) (hash(x)=143063312)
|
| 47 |
+
370 train 6.056172 (lr=4.4520e-04) (hash(x)=153705818)
|
| 48 |
+
380 train 5.972964 (lr=4.5720e-04) (hash(x)=149579175)
|
| 49 |
+
390 train 5.918429 (lr=4.6920e-04) (hash(x)=150904406)
|
| 50 |
+
400 val loss 6.0112
|
| 51 |
+
400 val perplexity 407.9679
|
| 52 |
+
400 train 5.876863 (lr=4.8120e-04) (hash(x)=153710890)
|
| 53 |
+
410 train 5.771727 (lr=4.9320e-04) (hash(x)=138302968)
|
| 54 |
+
420 train 5.777105 (lr=5.0520e-04) (hash(x)=141112989)
|
| 55 |
+
430 train 5.830154 (lr=5.1720e-04) (hash(x)=149846944)
|
| 56 |
+
440 train 5.828116 (lr=5.2920e-04) (hash(x)=157683479)
|
| 57 |
+
450 train 5.824431 (lr=5.4120e-04) (hash(x)=155873620)
|
| 58 |
+
460 train 6.028512 (lr=5.5320e-04) (hash(x)=152133659)
|
| 59 |
+
470 train 5.799159 (lr=5.6520e-04) (hash(x)=158095308)
|
| 60 |
+
480 train 5.649451 (lr=5.7720e-04) (hash(x)=148422063)
|
| 61 |
+
490 train 5.826170 (lr=5.8920e-04) (hash(x)=145665071)
|
| 62 |
+
500 val loss 5.7841
|
| 63 |
+
500 val perplexity 325.1038
|
| 64 |
+
500 train 5.776611 (lr=6.0000e-04) (hash(x)=145450636)
|
| 65 |
+
510 train 5.759080 (lr=6.0000e-04) (hash(x)=148375186)
|
| 66 |
+
520 train 5.775891 (lr=5.9999e-04) (hash(x)=144256573)
|
| 67 |
+
530 train 5.816446 (lr=5.9998e-04) (hash(x)=141443048)
|
| 68 |
+
540 train 5.650346 (lr=5.9997e-04) (hash(x)=149347480)
|
| 69 |
+
550 train 5.734946 (lr=5.9995e-04) (hash(x)=154123388)
|
| 70 |
+
560 train 5.695611 (lr=5.9993e-04) (hash(x)=143045783)
|
| 71 |
+
570 train 5.565760 (lr=5.9990e-04) (hash(x)=157244060)
|
| 72 |
+
580 train 5.561405 (lr=5.9987e-04) (hash(x)=158018166)
|
| 73 |
+
590 train 5.635768 (lr=5.9984e-04) (hash(x)=153794251)
|
| 74 |
+
600 val loss 5.6070
|
| 75 |
+
600 val perplexity 272.3222
|
| 76 |
+
600 train 5.557906 (lr=5.9980e-04) (hash(x)=145249251)
|
| 77 |
+
610 train 5.607287 (lr=5.9976e-04) (hash(x)=150377688)
|
| 78 |
+
620 train 5.564731 (lr=5.9972e-04) (hash(x)=134768519)
|
| 79 |
+
630 train 5.479506 (lr=5.9967e-04) (hash(x)=141559208)
|
| 80 |
+
640 train 5.457463 (lr=5.9962e-04) (hash(x)=148444988)
|
| 81 |
+
650 train 5.460835 (lr=5.9956e-04) (hash(x)=148937127)
|
| 82 |
+
660 train 5.367904 (lr=5.9950e-04) (hash(x)=145121305)
|
| 83 |
+
670 train 5.500012 (lr=5.9943e-04) (hash(x)=156860740)
|
| 84 |
+
680 train 5.281329 (lr=5.9937e-04) (hash(x)=137272598)
|
| 85 |
+
690 train 5.285691 (lr=5.9929e-04) (hash(x)=139021179)
|
| 86 |
+
700 val loss 5.4152
|
| 87 |
+
700 val perplexity 224.7959
|
| 88 |
+
700 train 5.586763 (lr=5.9922e-04) (hash(x)=150475545)
|
| 89 |
+
710 train 5.376775 (lr=5.9914e-04) (hash(x)=148476077)
|
| 90 |
+
720 train 5.262850 (lr=5.9905e-04) (hash(x)=143328717)
|
| 91 |
+
730 train 5.183437 (lr=5.9897e-04) (hash(x)=137342488)
|
| 92 |
+
740 train 5.278402 (lr=5.9887e-04) (hash(x)=149452392)
|
| 93 |
+
750 train 5.261243 (lr=5.9878e-04) (hash(x)=154653428)
|
| 94 |
+
760 train 5.379678 (lr=5.9868e-04) (hash(x)=153305312)
|
| 95 |
+
770 train 5.227743 (lr=5.9857e-04) (hash(x)=153862299)
|
| 96 |
+
780 train 5.243790 (lr=5.9847e-04) (hash(x)=151035519)
|
| 97 |
+
790 train 5.340232 (lr=5.9836e-04) (hash(x)=145990773)
|
| 98 |
+
800 val loss 5.2937
|
| 99 |
+
800 val perplexity 199.0835
|
| 100 |
+
800 train 5.304738 (lr=5.9824e-04) (hash(x)=144483776)
|
| 101 |
+
810 train 5.202996 (lr=5.9812e-04) (hash(x)=142045737)
|
| 102 |
+
820 train 5.332297 (lr=5.9800e-04) (hash(x)=157946886)
|
| 103 |
+
830 train 5.202717 (lr=5.9787e-04) (hash(x)=150425385)
|
| 104 |
+
840 train 5.245149 (lr=5.9774e-04) (hash(x)=144514796)
|
| 105 |
+
850 train 5.278734 (lr=5.9761e-04) (hash(x)=157395496)
|
| 106 |
+
860 train 5.250891 (lr=5.9747e-04) (hash(x)=151365298)
|
| 107 |
+
870 train 5.275786 (lr=5.9732e-04) (hash(x)=151437005)
|
| 108 |
+
880 train 5.180932 (lr=5.9718e-04) (hash(x)=151935250)
|
| 109 |
+
890 train 5.223048 (lr=5.9703e-04) (hash(x)=159838534)
|
| 110 |
+
900 val loss 5.1791
|
| 111 |
+
900 val perplexity 177.5242
|
| 112 |
+
900 train 5.195571 (lr=5.9687e-04) (hash(x)=157916369)
|
| 113 |
+
910 train 5.251864 (lr=5.9672e-04) (hash(x)=165272484)
|
| 114 |
+
920 train 5.105950 (lr=5.9655e-04) (hash(x)=145383083)
|
| 115 |
+
930 train 5.092994 (lr=5.9639e-04) (hash(x)=152078393)
|
| 116 |
+
940 train 5.209867 (lr=5.9622e-04) (hash(x)=151846396)
|
| 117 |
+
950 train 5.135397 (lr=5.9605e-04) (hash(x)=166073923)
|
| 118 |
+
960 train 5.104422 (lr=5.9587e-04) (hash(x)=136782937)
|
| 119 |
+
970 train 5.076737 (lr=5.9569e-04) (hash(x)=146747407)
|
| 120 |
+
980 train 5.097136 (lr=5.9550e-04) (hash(x)=171270168)
|
| 121 |
+
990 train 5.102144 (lr=5.9531e-04) (hash(x)=141668491)
|
| 122 |
+
1000 val loss 5.1032
|
| 123 |
+
1000 val perplexity 164.5528
|
| 124 |
+
1000 train 5.003941 (lr=5.9512e-04) (hash(x)=154856891)
|
| 125 |
+
1010 train 5.093828 (lr=5.9492e-04) (hash(x)=145288527)
|
| 126 |
+
1020 train 4.962240 (lr=5.9472e-04) (hash(x)=144618667)
|
| 127 |
+
1030 train 5.199950 (lr=5.9452e-04) (hash(x)=139276042)
|
| 128 |
+
1040 train 4.887685 (lr=5.9431e-04) (hash(x)=147217952)
|
| 129 |
+
1050 train 4.988291 (lr=5.9410e-04) (hash(x)=151925203)
|
| 130 |
+
1060 train 4.949405 (lr=5.9388e-04) (hash(x)=147473652)
|
| 131 |
+
1070 train 4.911823 (lr=5.9366e-04) (hash(x)=145345524)
|
| 132 |
+
1080 train 4.801612 (lr=5.9344e-04) (hash(x)=155595779)
|
| 133 |
+
1090 train 4.851241 (lr=5.9321e-04) (hash(x)=147405225)
|
| 134 |
+
1100 val loss 5.0080
|
| 135 |
+
1100 val perplexity 149.6043
|
| 136 |
+
1100 train 4.702862 (lr=5.9298e-04) (hash(x)=136191502)
|
| 137 |
+
1110 train 4.935129 (lr=5.9275e-04) (hash(x)=151847694)
|
| 138 |
+
1120 train 5.024825 (lr=5.9251e-04) (hash(x)=144828302)
|
| 139 |
+
1130 train 5.058961 (lr=5.9227e-04) (hash(x)=153817096)
|
| 140 |
+
1140 train 5.107350 (lr=5.9202e-04) (hash(x)=166352243)
|
| 141 |
+
1150 train 5.144423 (lr=5.9177e-04) (hash(x)=153273362)
|
| 142 |
+
1160 train 4.983146 (lr=5.9152e-04) (hash(x)=178082599)
|
| 143 |
+
1170 train 4.779559 (lr=5.9126e-04) (hash(x)=149460553)
|
| 144 |
+
1180 train 4.984323 (lr=5.9100e-04) (hash(x)=151910947)
|
| 145 |
+
1190 train 4.881153 (lr=5.9073e-04) (hash(x)=147115336)
|
| 146 |
+
1200 val loss 4.9122
|
| 147 |
+
1200 val perplexity 135.9390
|
| 148 |
+
1200 train 4.933949 (lr=5.9046e-04) (hash(x)=148021541)
|
| 149 |
+
1210 train 4.826488 (lr=5.9019e-04) (hash(x)=151495989)
|
| 150 |
+
1220 train 4.946360 (lr=5.8992e-04) (hash(x)=156983220)
|
| 151 |
+
1230 train 5.015411 (lr=5.8963e-04) (hash(x)=147788187)
|
| 152 |
+
1240 train 4.985041 (lr=5.8935e-04) (hash(x)=171125590)
|
| 153 |
+
1250 train 4.886112 (lr=5.8906e-04) (hash(x)=141356608)
|
| 154 |
+
1260 train 4.929478 (lr=5.8877e-04) (hash(x)=150132098)
|
| 155 |
+
1270 train 4.807566 (lr=5.8848e-04) (hash(x)=147917245)
|
| 156 |
+
1280 train 4.900241 (lr=5.8818e-04) (hash(x)=148900016)
|
| 157 |
+
1290 train 4.870604 (lr=5.8787e-04) (hash(x)=144978605)
|
| 158 |
+
1300 val loss 4.8334
|
| 159 |
+
1300 val perplexity 125.6381
|
| 160 |
+
1300 train 4.781497 (lr=5.8757e-04) (hash(x)=146005217)
|
| 161 |
+
1310 train 4.797524 (lr=5.8726e-04) (hash(x)=144892309)
|
| 162 |
+
1320 train 4.859780 (lr=5.8694e-04) (hash(x)=165903661)
|
| 163 |
+
1330 train 4.807590 (lr=5.8663e-04) (hash(x)=168489989)
|
| 164 |
+
1340 train 4.975265 (lr=5.8630e-04) (hash(x)=176281294)
|
| 165 |
+
1350 train 4.759254 (lr=5.8598e-04) (hash(x)=144511718)
|
| 166 |
+
1360 train 4.823272 (lr=5.8565e-04) (hash(x)=144362722)
|
| 167 |
+
1370 train 4.693036 (lr=5.8532e-04) (hash(x)=139964999)
|
| 168 |
+
1380 train 4.716559 (lr=5.8498e-04) (hash(x)=193332654)
|
| 169 |
+
1390 train 4.761081 (lr=5.8464e-04) (hash(x)=138180866)
|
| 170 |
+
1400 val loss 4.7848
|
| 171 |
+
1400 val perplexity 119.6811
|
| 172 |
+
1400 train 4.762902 (lr=5.8430e-04) (hash(x)=146019502)
|
| 173 |
+
1410 train 4.809028 (lr=5.8395e-04) (hash(x)=153245388)
|
| 174 |
+
1420 train 4.693506 (lr=5.8360e-04) (hash(x)=149839636)
|
| 175 |
+
1430 train 4.802482 (lr=5.8324e-04) (hash(x)=142844796)
|
| 176 |
+
1440 train 5.023189 (lr=5.8289e-04) (hash(x)=159787060)
|
| 177 |
+
1450 train 4.732884 (lr=5.8252e-04) (hash(x)=146496200)
|
| 178 |
+
1460 train 4.655418 (lr=5.8216e-04) (hash(x)=164169521)
|
| 179 |
+
1470 train 4.702697 (lr=5.8179e-04) (hash(x)=150906386)
|
| 180 |
+
1480 train 4.720006 (lr=5.8142e-04) (hash(x)=148916053)
|
| 181 |
+
1490 train 4.651597 (lr=5.8104e-04) (hash(x)=148746554)
|
| 182 |
+
1500 val loss 4.7859
|
| 183 |
+
1500 val perplexity 119.8145
|
| 184 |
+
1500 train 4.742775 (lr=5.8066e-04) (hash(x)=150127281)
|
| 185 |
+
1510 train 4.912547 (lr=5.8028e-04) (hash(x)=149760530)
|
| 186 |
+
1520 train 4.759301 (lr=5.7989e-04) (hash(x)=144237370)
|
| 187 |
+
1530 train 4.567895 (lr=5.7950e-04) (hash(x)=132692898)
|
| 188 |
+
1540 train 4.553084 (lr=5.7910e-04) (hash(x)=150979737)
|
| 189 |
+
1550 train 4.742966 (lr=5.7870e-04) (hash(x)=142022255)
|
| 190 |
+
1560 train 4.673608 (lr=5.7830e-04) (hash(x)=135769745)
|
| 191 |
+
1570 train 4.536185 (lr=5.7790e-04) (hash(x)=162241131)
|
| 192 |
+
1580 train 4.635792 (lr=5.7749e-04) (hash(x)=166874637)
|
| 193 |
+
1590 train 4.600491 (lr=5.7707e-04) (hash(x)=157401041)
|
| 194 |
+
1600 val loss 4.6843
|
| 195 |
+
1600 val perplexity 108.2363
|
| 196 |
+
1600 train 4.623621 (lr=5.7666e-04) (hash(x)=154120875)
|
| 197 |
+
1610 train 4.613076 (lr=5.7624e-04) (hash(x)=149407490)
|
| 198 |
+
1620 train 4.560949 (lr=5.7581e-04) (hash(x)=144479755)
|
| 199 |
+
1630 train 4.640452 (lr=5.7539e-04) (hash(x)=170907716)
|
| 200 |
+
1640 train 4.289717 (lr=5.7496e-04) (hash(x)=151914010)
|
| 201 |
+
1650 train 4.579615 (lr=5.7452e-04) (hash(x)=149843610)
|
| 202 |
+
1660 train 4.474859 (lr=5.7408e-04) (hash(x)=133217001)
|
| 203 |
+
1670 train 4.449507 (lr=5.7364e-04) (hash(x)=168227774)
|
| 204 |
+
1680 train 4.403447 (lr=5.7320e-04) (hash(x)=157093189)
|
| 205 |
+
1690 train 4.241521 (lr=5.7275e-04) (hash(x)=154989819)
|
| 206 |
+
1700 val loss 4.6601
|
| 207 |
+
1700 val perplexity 105.6499
|
| 208 |
+
1700 train 4.764375 (lr=5.7230e-04) (hash(x)=155797680)
|
| 209 |
+
1710 train 4.438869 (lr=5.7184e-04) (hash(x)=166050772)
|
| 210 |
+
1720 train 4.415730 (lr=5.7138e-04) (hash(x)=151663443)
|
| 211 |
+
1730 train 4.637233 (lr=5.7092e-04) (hash(x)=152112619)
|
| 212 |
+
1740 train 4.559391 (lr=5.7045e-04) (hash(x)=156132679)
|
| 213 |
+
1750 train 4.410774 (lr=5.6999e-04) (hash(x)=149044477)
|
| 214 |
+
1760 train 4.588952 (lr=5.6951e-04) (hash(x)=142924719)
|
| 215 |
+
1770 train 4.492014 (lr=5.6904e-04) (hash(x)=153467406)
|
| 216 |
+
1780 train 4.576634 (lr=5.6856e-04) (hash(x)=148889581)
|
| 217 |
+
1790 train 4.516285 (lr=5.6807e-04) (hash(x)=164831182)
|
| 218 |
+
1800 val loss 4.5968
|
| 219 |
+
1800 val perplexity 99.1682
|
| 220 |
+
1800 train 4.524964 (lr=5.6759e-04) (hash(x)=156809396)
|
| 221 |
+
1810 train 4.623642 (lr=5.6710e-04) (hash(x)=149245741)
|
| 222 |
+
1820 train 4.570191 (lr=5.6660e-04) (hash(x)=146865466)
|
| 223 |
+
1830 train 4.640428 (lr=5.6611e-04) (hash(x)=148068618)
|
| 224 |
+
1840 train 4.473223 (lr=5.6561e-04) (hash(x)=150088183)
|
| 225 |
+
1850 train 4.549740 (lr=5.6510e-04) (hash(x)=146695082)
|
| 226 |
+
1860 train 4.488262 (lr=5.6459e-04) (hash(x)=169008624)
|
| 227 |
+
1870 train 4.424570 (lr=5.6408e-04) (hash(x)=148645124)
|
| 228 |
+
1880 train 4.590400 (lr=5.6357e-04) (hash(x)=139701422)
|
| 229 |
+
1890 train 4.548831 (lr=5.6305e-04) (hash(x)=156374257)
|
| 230 |
+
1900 val loss 4.5746
|
| 231 |
+
1900 val perplexity 96.9929
|
| 232 |
+
1900 train 4.496499 (lr=5.6253e-04) (hash(x)=144640294)
|
| 233 |
+
1910 train 4.637846 (lr=5.6201e-04) (hash(x)=147484985)
|
| 234 |
+
1920 train 4.494401 (lr=5.6148e-04) (hash(x)=142317889)
|
| 235 |
+
1930 train 4.482625 (lr=5.6095e-04) (hash(x)=144906216)
|
| 236 |
+
1940 train 4.534515 (lr=5.6041e-04) (hash(x)=165332621)
|
| 237 |
+
1950 train 4.510262 (lr=5.5988e-04) (hash(x)=145701919)
|
| 238 |
+
1960 train 4.410273 (lr=5.5934e-04) (hash(x)=146414118)
|
| 239 |
+
1970 train 4.572169 (lr=5.5879e-04) (hash(x)=146766958)
|
| 240 |
+
1980 train 4.492216 (lr=5.5824e-04) (hash(x)=148146847)
|
| 241 |
+
1990 train 4.256873 (lr=5.5769e-04) (hash(x)=159684604)
|
| 242 |
+
2000 val loss 4.5661
|
| 243 |
+
2000 val perplexity 96.1705
|
| 244 |
+
2000 train 4.404471 (lr=5.5714e-04) (hash(x)=162831106)
|
| 245 |
+
2010 train 4.424958 (lr=5.5658e-04) (hash(x)=150583346)
|
| 246 |
+
2020 train 4.373349 (lr=5.5602e-04) (hash(x)=142779458)
|
| 247 |
+
2030 train 4.354062 (lr=5.5546e-04) (hash(x)=143755114)
|
| 248 |
+
2040 train 4.268564 (lr=5.5489e-04) (hash(x)=147324095)
|
| 249 |
+
2050 train 4.371172 (lr=5.5432e-04) (hash(x)=141727373)
|
| 250 |
+
2060 train 4.443020 (lr=5.5374e-04) (hash(x)=162053052)
|
| 251 |
+
2070 train 4.490764 (lr=5.5317e-04) (hash(x)=162596975)
|
| 252 |
+
2080 train 4.409820 (lr=5.5259e-04) (hash(x)=161242340)
|
| 253 |
+
2090 train 4.390230 (lr=5.5200e-04) (hash(x)=148583522)
|
| 254 |
+
2100 val loss 4.5780
|
| 255 |
+
2100 val perplexity 97.3198
|
| 256 |
+
2100 train 4.465339 (lr=5.5142e-04) (hash(x)=158239484)
|
| 257 |
+
2110 train 4.378835 (lr=5.5083e-04) (hash(x)=152610058)
|
| 258 |
+
2120 train 4.409603 (lr=5.5023e-04) (hash(x)=146582203)
|
| 259 |
+
2130 train 4.148681 (lr=5.4964e-04) (hash(x)=146686436)
|
| 260 |
+
2140 train 4.555778 (lr=5.4904e-04) (hash(x)=142827295)
|
| 261 |
+
2150 train 4.478709 (lr=5.4843e-04) (hash(x)=150747398)
|
| 262 |
+
2160 train 4.763395 (lr=5.4783e-04) (hash(x)=150777134)
|
| 263 |
+
2170 train 4.433492 (lr=5.4722e-04) (hash(x)=157002856)
|
| 264 |
+
2180 train 4.492192 (lr=5.4661e-04) (hash(x)=150442337)
|
| 265 |
+
2190 train 4.460946 (lr=5.4599e-04) (hash(x)=150799081)
|
| 266 |
+
2200 val loss 4.4986
|
| 267 |
+
2200 val perplexity 89.8879
|
| 268 |
+
2200 train 4.381059 (lr=5.4537e-04) (hash(x)=140504180)
|
| 269 |
+
2210 train 4.517085 (lr=5.4475e-04) (hash(x)=149088475)
|
| 270 |
+
2220 train 4.481184 (lr=5.4413e-04) (hash(x)=155180847)
|
| 271 |
+
2230 train 4.402691 (lr=5.4350e-04) (hash(x)=159347164)
|
| 272 |
+
2240 train 4.563639 (lr=5.4287e-04) (hash(x)=153574288)
|
| 273 |
+
2250 train 4.508776 (lr=5.4223e-04) (hash(x)=142603512)
|
| 274 |
+
2260 train 4.519448 (lr=5.4160e-04) (hash(x)=142953831)
|
| 275 |
+
2270 train 4.413083 (lr=5.4096e-04) (hash(x)=137642681)
|
| 276 |
+
2280 train 4.433607 (lr=5.4031e-04) (hash(x)=158106614)
|
| 277 |
+
2290 train 4.436308 (lr=5.3967e-04) (hash(x)=160787891)
|
| 278 |
+
2300 val loss 4.4679
|
| 279 |
+
2300 val perplexity 87.1697
|
| 280 |
+
2300 train 4.533278 (lr=5.3902e-04) (hash(x)=142234024)
|
| 281 |
+
2310 train 4.434681 (lr=5.3837e-04) (hash(x)=138192210)
|
| 282 |
+
2320 train 4.458347 (lr=5.3771e-04) (hash(x)=147727662)
|
| 283 |
+
2330 train 4.403520 (lr=5.3705e-04) (hash(x)=169324653)
|
| 284 |
+
2340 train 4.514364 (lr=5.3639e-04) (hash(x)=145350355)
|
| 285 |
+
2350 train 4.508618 (lr=5.3573e-04) (hash(x)=156909016)
|
| 286 |
+
2360 train 4.430801 (lr=5.3506e-04) (hash(x)=144559543)
|
| 287 |
+
2370 train 4.417887 (lr=5.3439e-04) (hash(x)=153212312)
|
| 288 |
+
2380 train 4.389775 (lr=5.3372e-04) (hash(x)=131816284)
|
| 289 |
+
2390 train 4.450120 (lr=5.3304e-04) (hash(x)=165818309)
|
| 290 |
+
2400 val loss 4.4651
|
| 291 |
+
2400 val perplexity 86.9287
|
| 292 |
+
2400 train 4.476176 (lr=5.3236e-04) (hash(x)=143091562)
|
| 293 |
+
2410 train 4.372552 (lr=5.3168e-04) (hash(x)=154010697)
|
| 294 |
+
2420 train 4.343225 (lr=5.3099e-04) (hash(x)=144844001)
|
| 295 |
+
2430 train 4.501669 (lr=5.3030e-04) (hash(x)=156389834)
|
| 296 |
+
2440 train 4.198299 (lr=5.2961e-04) (hash(x)=162425919)
|
| 297 |
+
2450 train 4.446963 (lr=5.2892e-04) (hash(x)=154200479)
|
| 298 |
+
2460 train 4.370580 (lr=5.2822e-04) (hash(x)=159066617)
|
| 299 |
+
2470 train 4.185802 (lr=5.2752e-04) (hash(x)=148278058)
|
| 300 |
+
2480 train 4.150192 (lr=5.2682e-04) (hash(x)=165598522)
|
| 301 |
+
2490 train 4.454783 (lr=5.2612e-04) (hash(x)=141306086)
|
| 302 |
+
2500 val loss 4.4934
|
| 303 |
+
2500 val perplexity 89.4246
|
| 304 |
+
2500 train 4.308168 (lr=5.2541e-04) (hash(x)=149857456)
|
| 305 |
+
2510 train 4.301208 (lr=5.2470e-04) (hash(x)=150368907)
|
| 306 |
+
2520 train 4.256428 (lr=5.2398e-04) (hash(x)=153394920)
|
| 307 |
+
2530 train 4.233460 (lr=5.2327e-04) (hash(x)=159740116)
|
| 308 |
+
2540 train 4.392202 (lr=5.2255e-04) (hash(x)=161620367)
|
| 309 |
+
2550 train 4.467092 (lr=5.2183e-04) (hash(x)=147223658)
|
| 310 |
+
2560 train 4.554312 (lr=5.2110e-04) (hash(x)=158146613)
|
| 311 |
+
2570 train 4.460134 (lr=5.2037e-04) (hash(x)=136375336)
|
| 312 |
+
2580 train 4.500267 (lr=5.1964e-04) (hash(x)=149298016)
|
| 313 |
+
2590 train 4.415389 (lr=5.1891e-04) (hash(x)=150720933)
|
| 314 |
+
2600 val loss 4.4306
|
| 315 |
+
2600 val perplexity 83.9809
|
| 316 |
+
2600 train 4.525354 (lr=5.1817e-04) (hash(x)=146191551)
|
| 317 |
+
2610 train 4.342191 (lr=5.1743e-04) (hash(x)=151413395)
|
| 318 |
+
2620 train 4.543557 (lr=5.1669e-04) (hash(x)=162094106)
|
| 319 |
+
2630 train 4.428928 (lr=5.1595e-04) (hash(x)=150608302)
|
| 320 |
+
2640 train 4.333382 (lr=5.1520e-04) (hash(x)=146115160)
|
| 321 |
+
2650 train 4.430802 (lr=5.1445e-04) (hash(x)=138242788)
|
| 322 |
+
2660 train 4.641551 (lr=5.1370e-04) (hash(x)=159921837)
|
| 323 |
+
2670 train 4.250050 (lr=5.1295e-04) (hash(x)=137205989)
|
| 324 |
+
2680 train 4.269504 (lr=5.1219e-04) (hash(x)=158608524)
|
| 325 |
+
2690 train 4.314833 (lr=5.1143e-04) (hash(x)=153322085)
|
| 326 |
+
2700 val loss 4.4073
|
| 327 |
+
2700 val perplexity 82.0512
|
| 328 |
+
2700 train 4.354226 (lr=5.1067e-04) (hash(x)=145375752)
|
| 329 |
+
2710 train 4.365409 (lr=5.0990e-04) (hash(x)=145036398)
|
| 330 |
+
2720 train 4.384626 (lr=5.0913e-04) (hash(x)=140064355)
|
| 331 |
+
2730 train 4.415036 (lr=5.0836e-04) (hash(x)=148983355)
|
| 332 |
+
2740 train 4.311646 (lr=5.0759e-04) (hash(x)=147636026)
|
| 333 |
+
2750 train 4.367444 (lr=5.0681e-04) (hash(x)=151763585)
|
| 334 |
+
2760 train 4.299557 (lr=5.0603e-04) (hash(x)=178033416)
|
| 335 |
+
2770 train 4.262702 (lr=5.0525e-04) (hash(x)=147097669)
|
| 336 |
+
2780 train 4.223789 (lr=5.0447e-04) (hash(x)=140475447)
|
| 337 |
+
2790 train 4.337634 (lr=5.0368e-04) (hash(x)=141135962)
|
| 338 |
+
2800 val loss 4.3988
|
| 339 |
+
2800 val perplexity 81.3571
|
| 340 |
+
2800 train 4.240672 (lr=5.0290e-04) (hash(x)=151568014)
|
| 341 |
+
2810 train 4.953083 (lr=5.0210e-04) (hash(x)=170629615)
|
| 342 |
+
2820 train 4.317946 (lr=5.0131e-04) (hash(x)=151858146)
|
| 343 |
+
2830 train 4.232313 (lr=5.0052e-04) (hash(x)=152170305)
|
| 344 |
+
2840 train 4.149346 (lr=4.9972e-04) (hash(x)=150705881)
|
| 345 |
+
2850 train 4.182802 (lr=4.9892e-04) (hash(x)=146997394)
|
| 346 |
+
2860 train 4.164981 (lr=4.9811e-04) (hash(x)=159749180)
|
| 347 |
+
2870 train 4.046839 (lr=4.9731e-04) (hash(x)=147618423)
|
| 348 |
+
2880 train 4.240561 (lr=4.9650e-04) (hash(x)=140778993)
|
| 349 |
+
2890 train 4.124408 (lr=4.9569e-04) (hash(x)=150735837)
|
| 350 |
+
2900 val loss 4.3982
|
| 351 |
+
2900 val perplexity 81.3030
|
| 352 |
+
2900 train 4.178474 (lr=4.9487e-04) (hash(x)=149366597)
|
| 353 |
+
2910 train 4.299051 (lr=4.9406e-04) (hash(x)=181602500)
|
| 354 |
+
2920 train 4.265178 (lr=4.9324e-04) (hash(x)=148080200)
|
| 355 |
+
2930 train 4.104599 (lr=4.9242e-04) (hash(x)=150629961)
|
| 356 |
+
2940 train 4.279860 (lr=4.9160e-04) (hash(x)=155786888)
|
| 357 |
+
2950 train 4.278832 (lr=4.9077e-04) (hash(x)=148553059)
|
| 358 |
+
2960 train 4.396752 (lr=4.8995e-04) (hash(x)=158494862)
|
| 359 |
+
2970 train 4.517593 (lr=4.8912e-04) (hash(x)=147684099)
|
| 360 |
+
2980 train 4.273145 (lr=4.8829e-04) (hash(x)=159008790)
|
| 361 |
+
2990 train 4.419888 (lr=4.8745e-04) (hash(x)=148288782)
|
| 362 |
+
3000 val loss 4.3568
|
| 363 |
+
3000 val perplexity 78.0109
|
| 364 |
+
3000 train 4.292861 (lr=4.8662e-04) (hash(x)=150464442)
|
| 365 |
+
3010 train 4.371593 (lr=4.8578e-04) (hash(x)=148665916)
|
| 366 |
+
3020 train 4.316908 (lr=4.8494e-04) (hash(x)=159788454)
|
| 367 |
+
3030 train 4.389796 (lr=4.8409e-04) (hash(x)=170224765)
|
| 368 |
+
3040 train 4.277595 (lr=4.8325e-04) (hash(x)=139746299)
|
| 369 |
+
3050 train 4.338845 (lr=4.8240e-04) (hash(x)=173491884)
|
| 370 |
+
3060 train 4.332809 (lr=4.8155e-04) (hash(x)=146098725)
|
| 371 |
+
3070 train 4.309294 (lr=4.8070e-04) (hash(x)=154713525)
|
| 372 |
+
3080 train 4.338870 (lr=4.7984e-04) (hash(x)=151206978)
|
| 373 |
+
3090 train 4.370872 (lr=4.7899e-04) (hash(x)=144250687)
|
| 374 |
+
3100 val loss 4.3782
|
| 375 |
+
3100 val perplexity 79.6952
|
| 376 |
+
3100 train 4.409786 (lr=4.7813e-04) (hash(x)=182449036)
|
| 377 |
+
3110 train 4.221416 (lr=4.7727e-04) (hash(x)=145504538)
|
| 378 |
+
3120 train 4.258609 (lr=4.7641e-04) (hash(x)=147948751)
|
| 379 |
+
3130 train 4.212210 (lr=4.7554e-04) (hash(x)=144547711)
|
| 380 |
+
3140 train 4.326009 (lr=4.7467e-04) (hash(x)=158622029)
|
| 381 |
+
3150 train 4.281703 (lr=4.7380e-04) (hash(x)=155949476)
|
| 382 |
+
3160 train 4.287187 (lr=4.7293e-04) (hash(x)=129245664)
|
| 383 |
+
3170 train 4.309855 (lr=4.7206e-04) (hash(x)=157162959)
|
| 384 |
+
3180 train 4.374291 (lr=4.7118e-04) (hash(x)=163102434)
|
| 385 |
+
3190 train 4.401355 (lr=4.7031e-04) (hash(x)=155952009)
|
| 386 |
+
3200 val loss 4.3503
|
| 387 |
+
3200 val perplexity 77.5037
|
| 388 |
+
3200 train 4.289626 (lr=4.6943e-04) (hash(x)=140141286)
|
| 389 |
+
3210 train 4.280479 (lr=4.6855e-04) (hash(x)=148811694)
|
| 390 |
+
3220 train 4.223735 (lr=4.6766e-04) (hash(x)=143433404)
|
| 391 |
+
3230 train 4.279935 (lr=4.6678e-04) (hash(x)=150525065)
|
| 392 |
+
3240 train 4.176414 (lr=4.6589e-04) (hash(x)=152014854)
|
| 393 |
+
3250 train 4.048805 (lr=4.6500e-04) (hash(x)=144716153)
|
| 394 |
+
3260 train 4.385694 (lr=4.6411e-04) (hash(x)=145324818)
|
| 395 |
+
3270 train 4.317697 (lr=4.6322e-04) (hash(x)=154930419)
|
| 396 |
+
3280 train 4.319047 (lr=4.6232e-04) (hash(x)=144649958)
|
| 397 |
+
3290 train 4.301317 (lr=4.6142e-04) (hash(x)=156496147)
|
| 398 |
+
3300 val loss 4.3369
|
| 399 |
+
3300 val perplexity 76.4678
|
| 400 |
+
3300 train 4.258340 (lr=4.6052e-04) (hash(x)=148099414)
|
| 401 |
+
3310 train 4.315618 (lr=4.5962e-04) (hash(x)=141081470)
|
| 402 |
+
3320 train 4.287163 (lr=4.5872e-04) (hash(x)=151572529)
|
| 403 |
+
3330 train 4.253131 (lr=4.5782e-04) (hash(x)=146887343)
|
| 404 |
+
3340 train 4.288600 (lr=4.5691e-04) (hash(x)=148612634)
|
| 405 |
+
3350 train 4.246165 (lr=4.5600e-04) (hash(x)=145218304)
|
| 406 |
+
3360 train 4.167013 (lr=4.5509e-04) (hash(x)=146893345)
|
| 407 |
+
3370 train 4.254632 (lr=4.5418e-04) (hash(x)=159302018)
|
| 408 |
+
3380 train 4.323816 (lr=4.5326e-04) (hash(x)=164117611)
|
| 409 |
+
3390 train 4.194248 (lr=4.5235e-04) (hash(x)=142401925)
|
| 410 |
+
3400 val loss 4.3225
|
| 411 |
+
3400 val perplexity 75.3787
|
| 412 |
+
3400 train 4.173793 (lr=4.5143e-04) (hash(x)=142633951)
|
| 413 |
+
3410 train 4.300762 (lr=4.5051e-04) (hash(x)=148491904)
|
| 414 |
+
3420 train 4.351557 (lr=4.4959e-04) (hash(x)=144013244)
|
| 415 |
+
3430 train 4.161072 (lr=4.4867e-04) (hash(x)=146939843)
|
| 416 |
+
3440 train 4.182618 (lr=4.4774e-04) (hash(x)=161055964)
|
| 417 |
+
3450 train 4.087534 (lr=4.4682e-04) (hash(x)=134825681)
|
| 418 |
+
3460 train 4.276597 (lr=4.4589e-04) (hash(x)=145087511)
|
| 419 |
+
3470 train 4.100667 (lr=4.4496e-04) (hash(x)=144200286)
|
| 420 |
+
3480 train 4.148541 (lr=4.4403e-04) (hash(x)=157605428)
|
| 421 |
+
3490 train 4.325905 (lr=4.4310e-04) (hash(x)=153636990)
|
| 422 |
+
3500 val loss 4.3239
|
| 423 |
+
3500 val perplexity 75.4790
|
| 424 |
+
3500 train 4.178516 (lr=4.4216e-04) (hash(x)=148368965)
|
| 425 |
+
3510 train 4.088175 (lr=4.4123e-04) (hash(x)=144775557)
|
| 426 |
+
3520 train 4.261998 (lr=4.4029e-04) (hash(x)=151539855)
|
| 427 |
+
3530 train 4.280823 (lr=4.3935e-04) (hash(x)=168384321)
|
| 428 |
+
3540 train 4.422899 (lr=4.3841e-04) (hash(x)=155671447)
|
| 429 |
+
3550 train 4.444036 (lr=4.3747e-04) (hash(x)=157403334)
|
| 430 |
+
3560 train 4.223127 (lr=4.3652e-04) (hash(x)=139495714)
|
| 431 |
+
3570 train 4.357045 (lr=4.3558e-04) (hash(x)=147685555)
|
| 432 |
+
3580 train 4.476269 (lr=4.3463e-04) (hash(x)=154209753)
|
| 433 |
+
3590 train 4.097392 (lr=4.3368e-04) (hash(x)=166864372)
|
| 434 |
+
3600 val loss 4.3065
|
| 435 |
+
3600 val perplexity 74.1818
|
| 436 |
+
3600 train 4.208117 (lr=4.3273e-04) (hash(x)=152372067)
|
| 437 |
+
3610 train 4.488429 (lr=4.3178e-04) (hash(x)=156579291)
|
| 438 |
+
3620 train 4.076468 (lr=4.3083e-04) (hash(x)=152969451)
|
| 439 |
+
3630 train 4.051098 (lr=4.2987e-04) (hash(x)=164428105)
|
| 440 |
+
3640 train 4.226479 (lr=4.2892e-04) (hash(x)=153325907)
|
| 441 |
+
3650 train 4.200468 (lr=4.2796e-04) (hash(x)=159197101)
|
| 442 |
+
3660 train 4.191760 (lr=4.2700e-04) (hash(x)=157503290)
|
| 443 |
+
3670 train 4.112825 (lr=4.2604e-04) (hash(x)=149036650)
|
| 444 |
+
3680 train 4.122258 (lr=4.2508e-04) (hash(x)=144525088)
|
| 445 |
+
3690 train 4.232194 (lr=4.2411e-04) (hash(x)=143154211)
|
| 446 |
+
3700 val loss 4.3016
|
| 447 |
+
3700 val perplexity 73.8149
|
| 448 |
+
3700 train 4.222100 (lr=4.2315e-04) (hash(x)=168885609)
|
| 449 |
+
3710 train 4.189565 (lr=4.2218e-04) (hash(x)=148815644)
|
| 450 |
+
3720 train 4.120987 (lr=4.2122e-04) (hash(x)=153279629)
|
| 451 |
+
3730 train 4.140110 (lr=4.2025e-04) (hash(x)=148181200)
|
| 452 |
+
3740 train 4.113040 (lr=4.1928e-04) (hash(x)=151357364)
|
| 453 |
+
3750 train 4.080936 (lr=4.1831e-04) (hash(x)=145269246)
|
| 454 |
+
3760 train 4.149215 (lr=4.1734e-04) (hash(x)=148923398)
|
| 455 |
+
3770 train 4.154061 (lr=4.1636e-04) (hash(x)=141205226)
|
| 456 |
+
3780 train 4.041055 (lr=4.1539e-04) (hash(x)=146180296)
|
| 457 |
+
3790 train 4.117515 (lr=4.1441e-04) (hash(x)=169790000)
|
| 458 |
+
3800 val loss 4.3008
|
| 459 |
+
3800 val perplexity 73.7604
|
| 460 |
+
3800 train 4.153253 (lr=4.1343e-04) (hash(x)=141633734)
|
| 461 |
+
3810 train 4.277663 (lr=4.1246e-04) (hash(x)=156306070)
|
| 462 |
+
3820 train 4.264000 (lr=4.1148e-04) (hash(x)=158769870)
|
| 463 |
+
3830 train 4.237772 (lr=4.1050e-04) (hash(x)=165301927)
|
| 464 |
+
3840 train 4.267307 (lr=4.0951e-04) (hash(x)=147025475)
|
| 465 |
+
3850 train 4.208849 (lr=4.0853e-04) (hash(x)=141223580)
|
| 466 |
+
3860 train 4.236732 (lr=4.0755e-04) (hash(x)=139668795)
|
| 467 |
+
3870 train 4.295900 (lr=4.0656e-04) (hash(x)=155839599)
|
| 468 |
+
3880 train 4.365773 (lr=4.0557e-04) (hash(x)=150635541)
|
| 469 |
+
3890 train 4.353963 (lr=4.0459e-04) (hash(x)=153702524)
|
| 470 |
+
3900 val loss 4.2767
|
| 471 |
+
3900 val perplexity 72.0022
|
| 472 |
+
3900 train 4.179128 (lr=4.0360e-04) (hash(x)=153141007)
|
| 473 |
+
3910 train 4.262108 (lr=4.0261e-04) (hash(x)=145483115)
|
| 474 |
+
3920 train 4.056787 (lr=4.0162e-04) (hash(x)=147480523)
|
| 475 |
+
3930 train 4.230914 (lr=4.0063e-04) (hash(x)=146229467)
|
| 476 |
+
3940 train 4.207361 (lr=3.9963e-04) (hash(x)=151376187)
|
| 477 |
+
3950 train 4.179211 (lr=3.9864e-04) (hash(x)=153745186)
|
| 478 |
+
3960 train 4.150270 (lr=3.9764e-04) (hash(x)=147595615)
|
| 479 |
+
3970 train 4.275993 (lr=3.9665e-04) (hash(x)=157979848)
|
| 480 |
+
3980 train 4.144835 (lr=3.9565e-04) (hash(x)=153714091)
|
| 481 |
+
3990 train 4.175827 (lr=3.9465e-04) (hash(x)=155637629)
|
| 482 |
+
4000 val loss 4.2712
|
| 483 |
+
4000 val perplexity 71.6050
|
| 484 |
+
4000 train 4.201859 (lr=3.9365e-04) (hash(x)=160577202)
|
| 485 |
+
4010 train 4.116205 (lr=3.9266e-04) (hash(x)=147432640)
|
| 486 |
+
4020 train 4.075618 (lr=3.9165e-04) (hash(x)=135542902)
|
| 487 |
+
4030 train 4.209609 (lr=3.9065e-04) (hash(x)=143137909)
|
| 488 |
+
4040 train 4.202774 (lr=3.8965e-04) (hash(x)=148269908)
|
| 489 |
+
4050 train 4.087222 (lr=3.8865e-04) (hash(x)=141954182)
|
| 490 |
+
4060 train 4.169354 (lr=3.8764e-04) (hash(x)=157035179)
|
| 491 |
+
4070 train 3.901249 (lr=3.8664e-04) (hash(x)=151361423)
|
| 492 |
+
4080 train 4.023126 (lr=3.8563e-04) (hash(x)=144373988)
|
| 493 |
+
4090 train 4.241362 (lr=3.8463e-04) (hash(x)=156461428)
|
| 494 |
+
4100 val loss 4.2644
|
| 495 |
+
4100 val perplexity 71.1221
|
| 496 |
+
4100 train 4.343929 (lr=3.8362e-04) (hash(x)=153858169)
|
| 497 |
+
4110 train 4.315270 (lr=3.8261e-04) (hash(x)=165506959)
|
| 498 |
+
4120 train 4.319119 (lr=3.8160e-04) (hash(x)=158709009)
|
| 499 |
+
4130 train 4.398439 (lr=3.8059e-04) (hash(x)=151836522)
|
| 500 |
+
4140 train 4.161856 (lr=3.7958e-04) (hash(x)=152917389)
|
| 501 |
+
4150 train 4.210320 (lr=3.7857e-04) (hash(x)=146973868)
|
| 502 |
+
4160 train 4.311877 (lr=3.7756e-04) (hash(x)=157875887)
|
| 503 |
+
4170 train 4.578331 (lr=3.7655e-04) (hash(x)=167933111)
|
| 504 |
+
4180 train 4.178298 (lr=3.7553e-04) (hash(x)=161938168)
|
| 505 |
+
4190 train 4.148326 (lr=3.7452e-04) (hash(x)=158102630)
|
| 506 |
+
4200 val loss 4.2498
|
| 507 |
+
4200 val perplexity 70.0920
|
| 508 |
+
4200 train 4.269673 (lr=3.7351e-04) (hash(x)=155889149)
|
| 509 |
+
4210 train 4.349800 (lr=3.7249e-04) (hash(x)=131046288)
|
| 510 |
+
4220 train 4.115254 (lr=3.7148e-04) (hash(x)=149423408)
|
| 511 |
+
4230 train 4.205424 (lr=3.7046e-04) (hash(x)=150033580)
|
| 512 |
+
4240 train 4.096198 (lr=3.6944e-04) (hash(x)=137509644)
|
| 513 |
+
4250 train 4.267327 (lr=3.6843e-04) (hash(x)=161159362)
|
| 514 |
+
4260 train 4.173113 (lr=3.6741e-04) (hash(x)=148117355)
|
| 515 |
+
4270 train 4.100213 (lr=3.6639e-04) (hash(x)=147315384)
|
| 516 |
+
4280 train 4.166073 (lr=3.6537e-04) (hash(x)=156577316)
|
| 517 |
+
4290 train 4.342910 (lr=3.6435e-04) (hash(x)=161553761)
|
| 518 |
+
4300 val loss 4.2421
|
| 519 |
+
4300 val perplexity 69.5559
|
| 520 |
+
4300 train 4.006040 (lr=3.6333e-04) (hash(x)=152294662)
|
| 521 |
+
4310 train 4.035946 (lr=3.6231e-04) (hash(x)=144616611)
|
| 522 |
+
4320 train 4.183397 (lr=3.6129e-04) (hash(x)=154134591)
|
| 523 |
+
4330 train 4.045680 (lr=3.6027e-04) (hash(x)=159947834)
|
| 524 |
+
4340 train 4.027687 (lr=3.5925e-04) (hash(x)=156261313)
|
| 525 |
+
4350 train 4.147388 (lr=3.5822e-04) (hash(x)=141245643)
|
| 526 |
+
4360 train 4.113534 (lr=3.5720e-04) (hash(x)=157291204)
|
| 527 |
+
4370 train 4.145621 (lr=3.5618e-04) (hash(x)=142877676)
|
| 528 |
+
4380 train 4.186522 (lr=3.5515e-04) (hash(x)=155174402)
|
| 529 |
+
4390 train 4.036535 (lr=3.5413e-04) (hash(x)=154675451)
|
| 530 |
+
4400 val loss 4.2396
|
| 531 |
+
4400 val perplexity 69.3832
|
| 532 |
+
4400 train 4.203861 (lr=3.5311e-04) (hash(x)=141804386)
|
| 533 |
+
4410 train 4.426002 (lr=3.5208e-04) (hash(x)=163930619)
|
| 534 |
+
4420 train 4.298167 (lr=3.5106e-04) (hash(x)=139437472)
|
| 535 |
+
4430 train 4.234465 (lr=3.5003e-04) (hash(x)=142474831)
|
| 536 |
+
4440 train 4.258693 (lr=3.4901e-04) (hash(x)=150102428)
|
| 537 |
+
4450 train 4.227980 (lr=3.4798e-04) (hash(x)=135805460)
|
| 538 |
+
4460 train 4.189848 (lr=3.4695e-04) (hash(x)=154732100)
|
| 539 |
+
4470 train 4.195867 (lr=3.4593e-04) (hash(x)=148554435)
|
| 540 |
+
4480 train 4.133001 (lr=3.4490e-04) (hash(x)=142410065)
|
| 541 |
+
4490 train 4.155505 (lr=3.4387e-04) (hash(x)=148108338)
|
| 542 |
+
4500 val loss 4.2217
|
| 543 |
+
4500 val perplexity 68.1509
|
| 544 |
+
4500 train 4.071621 (lr=3.4285e-04) (hash(x)=151095242)
|
| 545 |
+
4510 train 4.075661 (lr=3.4182e-04) (hash(x)=154911617)
|
| 546 |
+
4520 train 4.115407 (lr=3.4079e-04) (hash(x)=150858662)
|
| 547 |
+
4530 train 4.061187 (lr=3.3977e-04) (hash(x)=146850830)
|
| 548 |
+
4540 train 4.118964 (lr=3.3874e-04) (hash(x)=153506103)
|
| 549 |
+
4550 train 4.134123 (lr=3.3771e-04) (hash(x)=157068400)
|
| 550 |
+
4560 train 4.206189 (lr=3.3668e-04) (hash(x)=163640327)
|
| 551 |
+
4570 train 4.263094 (lr=3.3565e-04) (hash(x)=158962962)
|
| 552 |
+
4580 train 4.125972 (lr=3.3463e-04) (hash(x)=143454481)
|
| 553 |
+
4590 train 4.134850 (lr=3.3360e-04) (hash(x)=155380269)
|
| 554 |
+
4600 val loss 4.2556
|
| 555 |
+
4600 val perplexity 70.4971
|
| 556 |
+
4600 train 4.288472 (lr=3.3257e-04) (hash(x)=156414699)
|
| 557 |
+
4610 train 4.108760 (lr=3.3154e-04) (hash(x)=153520595)
|
| 558 |
+
4620 train 4.129999 (lr=3.3051e-04) (hash(x)=151287061)
|
| 559 |
+
4630 train 3.961761 (lr=3.2949e-04) (hash(x)=144972877)
|
| 560 |
+
4640 train 4.081348 (lr=3.2846e-04) (hash(x)=152031134)
|
| 561 |
+
4650 train 4.220479 (lr=3.2743e-04) (hash(x)=155348609)
|
| 562 |
+
4660 train 4.006318 (lr=3.2640e-04) (hash(x)=144855343)
|
| 563 |
+
4670 train 4.161096 (lr=3.2537e-04) (hash(x)=159247995)
|
| 564 |
+
4680 train 4.053130 (lr=3.2435e-04) (hash(x)=133146878)
|
| 565 |
+
4690 train 4.289427 (lr=3.2332e-04) (hash(x)=148797338)
|
| 566 |
+
4700 val loss 4.2147
|
| 567 |
+
4700 val perplexity 67.6768
|
| 568 |
+
4700 train 4.169440 (lr=3.2229e-04) (hash(x)=161556686)
|
| 569 |
+
4710 train 4.353711 (lr=3.2126e-04) (hash(x)=142528636)
|
| 570 |
+
4720 train 4.279312 (lr=3.2023e-04) (hash(x)=148322603)
|
| 571 |
+
4730 train 4.196731 (lr=3.1921e-04) (hash(x)=145051555)
|
| 572 |
+
4740 train 4.113234 (lr=3.1818e-04) (hash(x)=146193153)
|
| 573 |
+
4750 train 4.310147 (lr=3.1715e-04) (hash(x)=165802167)
|
| 574 |
+
4760 train 4.260219 (lr=3.1613e-04) (hash(x)=158948628)
|
| 575 |
+
4770 train 4.301055 (lr=3.1510e-04) (hash(x)=156177788)
|
| 576 |
+
4780 train 4.156542 (lr=3.1407e-04) (hash(x)=175446069)
|
| 577 |
+
4790 train 4.179591 (lr=3.1305e-04) (hash(x)=141100706)
|
| 578 |
+
4800 val loss 4.2115
|
| 579 |
+
4800 val perplexity 67.4604
|
| 580 |
+
4800 train 4.103304 (lr=3.1202e-04) (hash(x)=149000293)
|
| 581 |
+
4810 train 4.028175 (lr=3.1099e-04) (hash(x)=135891778)
|
| 582 |
+
4820 train 4.081861 (lr=3.0997e-04) (hash(x)=145474733)
|
| 583 |
+
4830 train 4.141610 (lr=3.0894e-04) (hash(x)=145187742)
|
| 584 |
+
4840 train 4.190108 (lr=3.0792e-04) (hash(x)=157685237)
|
| 585 |
+
4850 train 4.026490 (lr=3.0689e-04) (hash(x)=128599506)
|
| 586 |
+
4860 train 4.117213 (lr=3.0587e-04) (hash(x)=142067051)
|
| 587 |
+
4870 train 4.065267 (lr=3.0485e-04) (hash(x)=156215711)
|
| 588 |
+
4880 train 4.259776 (lr=3.0382e-04) (hash(x)=150456895)
|
| 589 |
+
4890 train 3.884215 (lr=3.0280e-04) (hash(x)=141647202)
|
| 590 |
+
4900 val loss 4.2000
|
| 591 |
+
4900 val perplexity 66.6849
|
| 592 |
+
4900 train 3.945271 (lr=3.0178e-04) (hash(x)=154349989)
|
| 593 |
+
4910 train 3.981321 (lr=3.0075e-04) (hash(x)=151563396)
|
| 594 |
+
4920 train 3.994511 (lr=2.9973e-04) (hash(x)=138868314)
|
| 595 |
+
4930 train 4.139812 (lr=2.9871e-04) (hash(x)=148533460)
|
| 596 |
+
4940 train 4.129040 (lr=2.9769e-04) (hash(x)=144892493)
|
| 597 |
+
4950 train 4.150645 (lr=2.9667e-04) (hash(x)=170326431)
|
| 598 |
+
4960 train 4.149920 (lr=2.9565e-04) (hash(x)=163312680)
|
| 599 |
+
4970 train 4.111387 (lr=2.9463e-04) (hash(x)=184352734)
|
| 600 |
+
4980 train 4.187467 (lr=2.9361e-04) (hash(x)=140363733)
|
| 601 |
+
4990 train 4.085557 (lr=2.9259e-04) (hash(x)=147794873)
|
| 602 |
+
5000 val loss 4.1943
|
| 603 |
+
5000 val perplexity 66.3055
|
| 604 |
+
5000 train 4.177243 (lr=2.9157e-04) (hash(x)=131475967)
|
| 605 |
+
5010 train 4.281310 (lr=2.9056e-04) (hash(x)=139560000)
|
| 606 |
+
5020 train 4.186268 (lr=2.8954e-04) (hash(x)=160128701)
|
| 607 |
+
5030 train 4.298525 (lr=2.8852e-04) (hash(x)=154698531)
|
| 608 |
+
5040 train 4.261716 (lr=2.8751e-04) (hash(x)=153833791)
|
| 609 |
+
5050 train 4.180938 (lr=2.8649e-04) (hash(x)=145953388)
|
| 610 |
+
5060 train 4.522370 (lr=2.8548e-04) (hash(x)=134297881)
|
| 611 |
+
5070 train 4.186088 (lr=2.8447e-04) (hash(x)=159987550)
|
| 612 |
+
5080 train 4.177457 (lr=2.8345e-04) (hash(x)=147699302)
|
| 613 |
+
5090 train 3.901203 (lr=2.8244e-04) (hash(x)=140534876)
|
| 614 |
+
5100 val loss 4.1898
|
| 615 |
+
5100 val perplexity 66.0123
|
| 616 |
+
5100 train 4.105705 (lr=2.8143e-04) (hash(x)=149717902)
|
| 617 |
+
5110 train 4.023517 (lr=2.8042e-04) (hash(x)=147648148)
|
| 618 |
+
5120 train 4.042538 (lr=2.7941e-04) (hash(x)=170962791)
|
| 619 |
+
5130 train 4.088052 (lr=2.7840e-04) (hash(x)=139633288)
|
| 620 |
+
5140 train 4.256811 (lr=2.7739e-04) (hash(x)=160163221)
|
| 621 |
+
5150 train 4.087533 (lr=2.7638e-04) (hash(x)=146625393)
|
| 622 |
+
5160 train 4.150156 (lr=2.7537e-04) (hash(x)=147399092)
|
| 623 |
+
5170 train 3.991426 (lr=2.7437e-04) (hash(x)=147172792)
|
| 624 |
+
5180 train 4.097240 (lr=2.7336e-04) (hash(x)=140600568)
|
| 625 |
+
5190 train 4.018775 (lr=2.7236e-04) (hash(x)=152423962)
|
| 626 |
+
5200 val loss 4.1866
|
| 627 |
+
5200 val perplexity 65.8001
|
| 628 |
+
5200 train 4.127081 (lr=2.7135e-04) (hash(x)=151407999)
|
| 629 |
+
5210 train 4.075913 (lr=2.7035e-04) (hash(x)=154874903)
|
| 630 |
+
5220 train 3.890605 (lr=2.6935e-04) (hash(x)=154133697)
|
| 631 |
+
5230 train 4.077150 (lr=2.6835e-04) (hash(x)=148148721)
|
| 632 |
+
5240 train 4.055752 (lr=2.6734e-04) (hash(x)=149681665)
|
| 633 |
+
5250 train 4.022038 (lr=2.6635e-04) (hash(x)=150136904)
|
| 634 |
+
5260 train 4.137352 (lr=2.6535e-04) (hash(x)=149261170)
|
| 635 |
+
5270 train 4.107561 (lr=2.6435e-04) (hash(x)=155540595)
|
| 636 |
+
5280 train 4.124571 (lr=2.6335e-04) (hash(x)=141973714)
|
| 637 |
+
5290 train 4.570252 (lr=2.6236e-04) (hash(x)=146827439)
|
| 638 |
+
5300 val loss 4.1745
|
| 639 |
+
5300 val perplexity 65.0058
|
| 640 |
+
5300 train 4.140273 (lr=2.6136e-04) (hash(x)=168602728)
|
| 641 |
+
5310 train 4.119944 (lr=2.6037e-04) (hash(x)=144035757)
|
| 642 |
+
5320 train 4.179716 (lr=2.5937e-04) (hash(x)=143287560)
|
| 643 |
+
5330 train 4.110137 (lr=2.5838e-04) (hash(x)=142461814)
|
| 644 |
+
5340 train 4.154832 (lr=2.5739e-04) (hash(x)=149905536)
|
| 645 |
+
5350 train 4.251827 (lr=2.5640e-04) (hash(x)=134597061)
|
| 646 |
+
5360 train 4.276795 (lr=2.5541e-04) (hash(x)=156520228)
|
| 647 |
+
5370 train 4.185418 (lr=2.5443e-04) (hash(x)=152690323)
|
| 648 |
+
5380 train 4.125360 (lr=2.5344e-04) (hash(x)=140092622)
|
| 649 |
+
5390 train 4.088529 (lr=2.5245e-04) (hash(x)=153504017)
|
| 650 |
+
5400 val loss 4.1616
|
| 651 |
+
5400 val perplexity 64.1746
|
| 652 |
+
5400 train 4.112967 (lr=2.5147e-04) (hash(x)=158344511)
|
| 653 |
+
5410 train 4.042314 (lr=2.5049e-04) (hash(x)=149681960)
|
| 654 |
+
5420 train 4.093867 (lr=2.4950e-04) (hash(x)=143459968)
|
| 655 |
+
5430 train 4.008551 (lr=2.4852e-04) (hash(x)=134759020)
|
| 656 |
+
5440 train 3.986306 (lr=2.4754e-04) (hash(x)=164760471)
|
| 657 |
+
5450 train 4.039769 (lr=2.4657e-04) (hash(x)=148551310)
|
| 658 |
+
5460 train 4.175451 (lr=2.4559e-04) (hash(x)=155464239)
|
| 659 |
+
5470 train 3.987691 (lr=2.4461e-04) (hash(x)=147309485)
|
| 660 |
+
5480 train 3.940682 (lr=2.4364e-04) (hash(x)=163937590)
|
| 661 |
+
5490 train 4.226748 (lr=2.4266e-04) (hash(x)=142830147)
|
| 662 |
+
5500 val loss 4.1583
|
| 663 |
+
5500 val perplexity 63.9600
|
| 664 |
+
5500 train 4.054812 (lr=2.4169e-04) (hash(x)=148350057)
|
| 665 |
+
5510 train 4.028842 (lr=2.4072e-04) (hash(x)=149007838)
|
| 666 |
+
5520 train 3.997600 (lr=2.3975e-04) (hash(x)=158176239)
|
| 667 |
+
5530 train 4.104340 (lr=2.3878e-04) (hash(x)=156395740)
|
| 668 |
+
5540 train 4.270691 (lr=2.3782e-04) (hash(x)=152453211)
|
| 669 |
+
5550 train 3.986491 (lr=2.3685e-04) (hash(x)=144447218)
|
| 670 |
+
5560 train 4.157467 (lr=2.3589e-04) (hash(x)=153858804)
|
| 671 |
+
5570 train 3.988536 (lr=2.3492e-04) (hash(x)=133929681)
|
| 672 |
+
5580 train 4.237970 (lr=2.3396e-04) (hash(x)=159170988)
|
| 673 |
+
5590 train 4.175734 (lr=2.3300e-04) (hash(x)=144978886)
|
| 674 |
+
5600 val loss 4.1527
|
| 675 |
+
5600 val perplexity 63.6048
|
| 676 |
+
5600 train 4.194169 (lr=2.3204e-04) (hash(x)=153847323)
|
| 677 |
+
5610 train 4.143672 (lr=2.3108e-04) (hash(x)=145553636)
|
| 678 |
+
5620 train 5.290322 (lr=2.3013e-04) (hash(x)=153712417)
|
| 679 |
+
5630 train 4.174591 (lr=2.2917e-04) (hash(x)=148397520)
|
| 680 |
+
5640 train 4.117166 (lr=2.2822e-04) (hash(x)=149424351)
|
| 681 |
+
5650 train 4.210706 (lr=2.2727e-04) (hash(x)=140531069)
|
| 682 |
+
5660 train 4.148504 (lr=2.2632e-04) (hash(x)=144904009)
|
| 683 |
+
5670 train 3.999589 (lr=2.2537e-04) (hash(x)=162469666)
|
| 684 |
+
5680 train 4.124962 (lr=2.2442e-04) (hash(x)=149254569)
|
| 685 |
+
5690 train 4.041796 (lr=2.2348e-04) (hash(x)=146949494)
|
| 686 |
+
5700 val loss 4.1440
|
| 687 |
+
5700 val perplexity 63.0532
|
| 688 |
+
5700 train 4.104294 (lr=2.2253e-04) (hash(x)=156607405)
|
| 689 |
+
5710 train 3.951504 (lr=2.2159e-04) (hash(x)=146349425)
|
| 690 |
+
5720 train 4.117228 (lr=2.2065e-04) (hash(x)=160347239)
|
| 691 |
+
5730 train 4.053485 (lr=2.1971e-04) (hash(x)=153785045)
|
| 692 |
+
5740 train 4.068789 (lr=2.1877e-04) (hash(x)=135967367)
|
| 693 |
+
5750 train 4.032428 (lr=2.1784e-04) (hash(x)=140181987)
|
| 694 |
+
5760 train 4.184039 (lr=2.1690e-04) (hash(x)=153217075)
|
| 695 |
+
5770 train 4.046869 (lr=2.1597e-04) (hash(x)=146360100)
|
| 696 |
+
5780 train 4.070055 (lr=2.1504e-04) (hash(x)=161111351)
|
| 697 |
+
5790 train 4.006124 (lr=2.1411e-04) (hash(x)=147891055)
|
| 698 |
+
5800 val loss 4.1410
|
| 699 |
+
5800 val perplexity 62.8682
|
| 700 |
+
5800 train 3.884526 (lr=2.1318e-04) (hash(x)=145115031)
|
| 701 |
+
5810 train 3.921891 (lr=2.1226e-04) (hash(x)=153661465)
|
| 702 |
+
5820 train 4.016182 (lr=2.1133e-04) (hash(x)=148264581)
|
| 703 |
+
5830 train 4.116851 (lr=2.1041e-04) (hash(x)=171144748)
|
| 704 |
+
5840 train 4.046432 (lr=2.0949e-04) (hash(x)=157863238)
|
| 705 |
+
5850 train 3.985251 (lr=2.0857e-04) (hash(x)=144669655)
|
| 706 |
+
5860 train 3.898940 (lr=2.0765e-04) (hash(x)=154948909)
|
| 707 |
+
5870 train 4.066360 (lr=2.0674e-04) (hash(x)=158981837)
|
| 708 |
+
5880 train 4.108955 (lr=2.0582e-04) (hash(x)=151102851)
|
| 709 |
+
5890 train 4.306065 (lr=2.0491e-04) (hash(x)=153490628)
|
| 710 |
+
5900 val loss 4.1310
|
| 711 |
+
5900 val perplexity 62.2424
|
| 712 |
+
5900 train 4.096318 (lr=2.0400e-04) (hash(x)=141584622)
|
| 713 |
+
5910 train 3.982910 (lr=2.0309e-04) (hash(x)=142896324)
|
| 714 |
+
5920 train 4.269152 (lr=2.0218e-04) (hash(x)=148264764)
|
| 715 |
+
5930 train 4.129268 (lr=2.0128e-04) (hash(x)=153157848)
|
| 716 |
+
5940 train 4.105019 (lr=2.0038e-04) (hash(x)=151086429)
|
| 717 |
+
5950 train 4.079765 (lr=1.9948e-04) (hash(x)=146009598)
|
| 718 |
+
5960 train 4.190752 (lr=1.9858e-04) (hash(x)=149488374)
|
| 719 |
+
5970 train 4.024580 (lr=1.9768e-04) (hash(x)=149307478)
|
| 720 |
+
5980 train 3.795582 (lr=1.9678e-04) (hash(x)=156742339)
|
| 721 |
+
5990 train 4.059119 (lr=1.9589e-04) (hash(x)=164296391)
|
| 722 |
+
6000 val loss 4.1255
|
| 723 |
+
6000 val perplexity 61.8984
|
| 724 |
+
6000 train 4.154809 (lr=1.9500e-04) (hash(x)=146613857)
|
| 725 |
+
6010 train 4.030589 (lr=1.9411e-04) (hash(x)=149742104)
|
| 726 |
+
6020 train 3.977369 (lr=1.9322e-04) (hash(x)=145645994)
|
| 727 |
+
6030 train 4.059315 (lr=1.9234e-04) (hash(x)=156324150)
|
| 728 |
+
6040 train 4.231585 (lr=1.9145e-04) (hash(x)=154655300)
|
| 729 |
+
6050 train 4.094217 (lr=1.9057e-04) (hash(x)=153714860)
|
| 730 |
+
6060 train 4.046013 (lr=1.8969e-04) (hash(x)=139981556)
|
| 731 |
+
6070 train 3.881154 (lr=1.8882e-04) (hash(x)=158753458)
|
| 732 |
+
6080 train 4.043500 (lr=1.8794e-04) (hash(x)=152559930)
|
| 733 |
+
6090 train 4.016532 (lr=1.8707e-04) (hash(x)=137128715)
|
| 734 |
+
6100 val loss 4.1262
|
| 735 |
+
6100 val perplexity 61.9437
|
| 736 |
+
6100 train 4.030574 (lr=1.8620e-04) (hash(x)=144621768)
|
| 737 |
+
6110 train 4.032393 (lr=1.8533e-04) (hash(x)=155614333)
|
| 738 |
+
6120 train 4.017185 (lr=1.8446e-04) (hash(x)=145682343)
|
| 739 |
+
6130 train 3.926471 (lr=1.8359e-04) (hash(x)=142909600)
|
| 740 |
+
6140 train 3.959304 (lr=1.8273e-04) (hash(x)=152993494)
|
| 741 |
+
6150 train 3.844943 (lr=1.8187e-04) (hash(x)=157151527)
|
| 742 |
+
6160 train 3.993061 (lr=1.8101e-04) (hash(x)=150653611)
|
| 743 |
+
6170 train 4.244534 (lr=1.8016e-04) (hash(x)=165237934)
|
| 744 |
+
6180 train 4.094344 (lr=1.7930e-04) (hash(x)=155730197)
|
| 745 |
+
6190 train 4.112364 (lr=1.7845e-04) (hash(x)=149687169)
|
| 746 |
+
6200 val loss 4.1250
|
| 747 |
+
6200 val perplexity 61.8648
|
| 748 |
+
6200 train 4.106633 (lr=1.7760e-04) (hash(x)=146521760)
|
| 749 |
+
6210 train 4.114485 (lr=1.7675e-04) (hash(x)=141008090)
|
| 750 |
+
6220 train 4.076965 (lr=1.7591e-04) (hash(x)=143407095)
|
| 751 |
+
6230 train 4.213218 (lr=1.7506e-04) (hash(x)=152947604)
|
| 752 |
+
6240 train 4.522073 (lr=1.7422e-04) (hash(x)=153876740)
|
| 753 |
+
6250 train 4.159676 (lr=1.7338e-04) (hash(x)=144250615)
|
| 754 |
+
6260 train 4.153634 (lr=1.7255e-04) (hash(x)=155236959)
|
| 755 |
+
6270 train 4.127711 (lr=1.7171e-04) (hash(x)=139785369)
|
| 756 |
+
6280 train 4.057211 (lr=1.7088e-04) (hash(x)=165036565)
|
| 757 |
+
6290 train 4.078882 (lr=1.7005e-04) (hash(x)=144585028)
|
| 758 |
+
6300 val loss 4.1120
|
| 759 |
+
6300 val perplexity 61.0669
|
| 760 |
+
6300 train 3.931736 (lr=1.6923e-04) (hash(x)=161378136)
|
| 761 |
+
6310 train 3.981984 (lr=1.6840e-04) (hash(x)=148483421)
|
| 762 |
+
6320 train 4.031734 (lr=1.6758e-04) (hash(x)=149835040)
|
| 763 |
+
6330 train 3.944311 (lr=1.6676e-04) (hash(x)=149459414)
|
| 764 |
+
6340 train 3.975904 (lr=1.6594e-04) (hash(x)=154290067)
|
| 765 |
+
6350 train 4.075600 (lr=1.6513e-04) (hash(x)=147407391)
|
| 766 |
+
6360 train 4.010707 (lr=1.6431e-04) (hash(x)=146317149)
|
| 767 |
+
6370 train 3.914214 (lr=1.6350e-04) (hash(x)=141970628)
|
| 768 |
+
6380 train 3.949024 (lr=1.6269e-04) (hash(x)=139170535)
|
| 769 |
+
6390 train 4.145844 (lr=1.6189e-04) (hash(x)=148905963)
|
| 770 |
+
6400 val loss 4.1111
|
| 771 |
+
6400 val perplexity 61.0108
|
| 772 |
+
6400 train 3.937335 (lr=1.6108e-04) (hash(x)=141624235)
|
| 773 |
+
6410 train 3.942767 (lr=1.6028e-04) (hash(x)=150520968)
|
| 774 |
+
6420 train 3.830738 (lr=1.5948e-04) (hash(x)=155019129)
|
| 775 |
+
6430 train 3.888057 (lr=1.5869e-04) (hash(x)=150031836)
|
| 776 |
+
6440 train 3.857171 (lr=1.5790e-04) (hash(x)=112835661)
|
| 777 |
+
6450 train 3.855033 (lr=1.5710e-04) (hash(x)=141072709)
|
| 778 |
+
6460 train 3.998727 (lr=1.5632e-04) (hash(x)=153933796)
|
| 779 |
+
6470 train 4.050470 (lr=1.5553e-04) (hash(x)=153315715)
|
| 780 |
+
6480 train 4.126259 (lr=1.5475e-04) (hash(x)=158089228)
|
| 781 |
+
6490 train 3.978586 (lr=1.5397e-04) (hash(x)=149471788)
|
| 782 |
+
6500 val loss 4.1078
|
| 783 |
+
6500 val perplexity 60.8111
|
| 784 |
+
6500 train 4.108046 (lr=1.5319e-04) (hash(x)=151197095)
|
| 785 |
+
6510 train 4.070200 (lr=1.5241e-04) (hash(x)=165554266)
|
| 786 |
+
6520 train 4.232692 (lr=1.5164e-04) (hash(x)=157822242)
|
| 787 |
+
6530 train 4.166207 (lr=1.5087e-04) (hash(x)=156267861)
|
| 788 |
+
6540 train 4.076926 (lr=1.5010e-04) (hash(x)=145825803)
|
| 789 |
+
6550 train 4.155082 (lr=1.4933e-04) (hash(x)=151303683)
|
| 790 |
+
6560 train 4.217830 (lr=1.4857e-04) (hash(x)=150786942)
|
| 791 |
+
6570 train 4.034340 (lr=1.4781e-04) (hash(x)=146237093)
|
| 792 |
+
6580 train 3.999360 (lr=1.4705e-04) (hash(x)=147063866)
|
| 793 |
+
6590 train 4.233628 (lr=1.4630e-04) (hash(x)=130513396)
|
| 794 |
+
6600 val loss 4.1058
|
| 795 |
+
6600 val perplexity 60.6909
|
| 796 |
+
6600 train 4.125781 (lr=1.4555e-04) (hash(x)=153269571)
|
| 797 |
+
6610 train 4.198629 (lr=1.4480e-04) (hash(x)=150340530)
|
| 798 |
+
6620 train 4.071691 (lr=1.4405e-04) (hash(x)=153843616)
|
| 799 |
+
6630 train 4.143201 (lr=1.4331e-04) (hash(x)=144540858)
|
| 800 |
+
6640 train 3.941280 (lr=1.4257e-04) (hash(x)=153330434)
|
| 801 |
+
6650 train 3.955559 (lr=1.4183e-04) (hash(x)=146851492)
|
| 802 |
+
6660 train 3.974411 (lr=1.4109e-04) (hash(x)=153601788)
|
| 803 |
+
6670 train 4.007741 (lr=1.4036e-04) (hash(x)=140005742)
|
| 804 |
+
6680 train 3.877739 (lr=1.3963e-04) (hash(x)=154102392)
|
| 805 |
+
6690 train 3.874567 (lr=1.3890e-04) (hash(x)=152774975)
|
| 806 |
+
6700 val loss 4.1016
|
| 807 |
+
6700 val perplexity 60.4390
|
| 808 |
+
6700 train 4.074510 (lr=1.3817e-04) (hash(x)=146111181)
|
| 809 |
+
6710 train 4.026546 (lr=1.3745e-04) (hash(x)=143988017)
|
| 810 |
+
6720 train 4.016078 (lr=1.3673e-04) (hash(x)=146073959)
|
| 811 |
+
6730 train 4.267421 (lr=1.3602e-04) (hash(x)=152642956)
|
| 812 |
+
6740 train 4.157714 (lr=1.3530e-04) (hash(x)=146964363)
|
| 813 |
+
6750 train 3.954356 (lr=1.3459e-04) (hash(x)=162251871)
|
| 814 |
+
6760 train 4.301650 (lr=1.3388e-04) (hash(x)=171122166)
|
| 815 |
+
6770 train 4.080363 (lr=1.3318e-04) (hash(x)=135199617)
|
| 816 |
+
6780 train 4.048437 (lr=1.3248e-04) (hash(x)=160480410)
|
| 817 |
+
6790 train 4.193917 (lr=1.3178e-04) (hash(x)=151933948)
|
| 818 |
+
6800 val loss 4.0881
|
| 819 |
+
6800 val perplexity 59.6239
|
| 820 |
+
6800 train 4.094066 (lr=1.3108e-04) (hash(x)=147269760)
|
| 821 |
+
6810 train 4.161630 (lr=1.3039e-04) (hash(x)=156710316)
|
| 822 |
+
6820 train 4.175751 (lr=1.2970e-04) (hash(x)=157300754)
|
| 823 |
+
6830 train 3.989247 (lr=1.2901e-04) (hash(x)=133838057)
|
| 824 |
+
6840 train 4.171911 (lr=1.2832e-04) (hash(x)=143518263)
|
| 825 |
+
6850 train 4.247186 (lr=1.2764e-04) (hash(x)=160328446)
|
| 826 |
+
6860 train 4.085854 (lr=1.2696e-04) (hash(x)=142143427)
|
| 827 |
+
6870 train 4.144669 (lr=1.2628e-04) (hash(x)=149045075)
|
| 828 |
+
6880 train 4.058600 (lr=1.2561e-04) (hash(x)=150708270)
|
| 829 |
+
6890 train 4.058171 (lr=1.2494e-04) (hash(x)=153066841)
|
| 830 |
+
6900 val loss 4.0867
|
| 831 |
+
6900 val perplexity 59.5415
|
| 832 |
+
6900 train 4.168222 (lr=1.2427e-04) (hash(x)=152912762)
|
| 833 |
+
6910 train 4.011430 (lr=1.2361e-04) (hash(x)=148346240)
|
| 834 |
+
6920 train 4.135716 (lr=1.2295e-04) (hash(x)=152806752)
|
| 835 |
+
6930 train 4.104208 (lr=1.2229e-04) (hash(x)=161831829)
|
| 836 |
+
6940 train 4.017661 (lr=1.2163e-04) (hash(x)=149483673)
|
| 837 |
+
6950 train 4.001894 (lr=1.2098e-04) (hash(x)=119634555)
|
| 838 |
+
6960 train 3.924197 (lr=1.2033e-04) (hash(x)=144709540)
|
| 839 |
+
6970 train 3.955101 (lr=1.1969e-04) (hash(x)=141277017)
|
| 840 |
+
6980 train 4.011095 (lr=1.1904e-04) (hash(x)=140618792)
|
| 841 |
+
6990 train 3.988875 (lr=1.1840e-04) (hash(x)=157443505)
|
| 842 |
+
7000 val loss 4.0861
|
| 843 |
+
7000 val perplexity 59.5067
|
| 844 |
+
7000 train 4.222891 (lr=1.1777e-04) (hash(x)=165412343)
|
| 845 |
+
7010 train 3.872253 (lr=1.1713e-04) (hash(x)=140746035)
|
| 846 |
+
7020 train 3.918733 (lr=1.1650e-04) (hash(x)=141101046)
|
| 847 |
+
7030 train 4.062104 (lr=1.1587e-04) (hash(x)=152098342)
|
| 848 |
+
7040 train 3.916734 (lr=1.1525e-04) (hash(x)=169244968)
|
| 849 |
+
7050 train 4.041646 (lr=1.1463e-04) (hash(x)=149886680)
|
| 850 |
+
7060 train 4.208581 (lr=1.1401e-04) (hash(x)=144670487)
|
| 851 |
+
7070 train 4.230530 (lr=1.1339e-04) (hash(x)=148774474)
|
| 852 |
+
7080 train 4.061625 (lr=1.1278e-04) (hash(x)=155096286)
|
| 853 |
+
7090 train 4.037445 (lr=1.1217e-04) (hash(x)=145048246)
|
| 854 |
+
7100 val loss 4.0724
|
| 855 |
+
7100 val perplexity 58.7005
|
| 856 |
+
7100 train 4.182160 (lr=1.1157e-04) (hash(x)=162866028)
|
| 857 |
+
7110 train 4.045501 (lr=1.1096e-04) (hash(x)=162308558)
|
| 858 |
+
7120 train 4.019279 (lr=1.1036e-04) (hash(x)=153468309)
|
| 859 |
+
7130 train 4.038635 (lr=1.0977e-04) (hash(x)=144522880)
|
| 860 |
+
7140 train 4.189165 (lr=1.0917e-04) (hash(x)=153055749)
|
| 861 |
+
7150 train 4.003743 (lr=1.0858e-04) (hash(x)=171163513)
|
| 862 |
+
7160 train 4.081718 (lr=1.0800e-04) (hash(x)=138563864)
|
| 863 |
+
7170 train 4.215020 (lr=1.0741e-04) (hash(x)=166996657)
|
| 864 |
+
7180 train 3.910931 (lr=1.0683e-04) (hash(x)=154809376)
|
| 865 |
+
7190 train 3.989370 (lr=1.0626e-04) (hash(x)=147988525)
|
| 866 |
+
7200 val loss 4.0713
|
| 867 |
+
7200 val perplexity 58.6325
|
| 868 |
+
7200 train 4.039526 (lr=1.0568e-04) (hash(x)=142998115)
|
| 869 |
+
7210 train 4.027789 (lr=1.0511e-04) (hash(x)=145518575)
|
| 870 |
+
7220 train 4.053323 (lr=1.0454e-04) (hash(x)=145758781)
|
| 871 |
+
7230 train 4.155663 (lr=1.0398e-04) (hash(x)=148909637)
|
| 872 |
+
7240 train 4.015603 (lr=1.0342e-04) (hash(x)=144422691)
|
| 873 |
+
7250 train 3.865092 (lr=1.0286e-04) (hash(x)=140396153)
|
| 874 |
+
7260 train 4.101425 (lr=1.0231e-04) (hash(x)=153619124)
|
| 875 |
+
7270 train 4.005348 (lr=1.0176e-04) (hash(x)=161980521)
|
| 876 |
+
7280 train 3.834925 (lr=1.0121e-04) (hash(x)=135160527)
|
| 877 |
+
7290 train 3.962374 (lr=1.0066e-04) (hash(x)=167156181)
|
| 878 |
+
7300 val loss 4.0717
|
| 879 |
+
7300 val perplexity 58.6537
|
| 880 |
+
7300 train 3.901530 (lr=1.0012e-04) (hash(x)=145486999)
|
| 881 |
+
7310 train 3.977693 (lr=9.9586e-05) (hash(x)=155013351)
|
| 882 |
+
7320 train 4.023978 (lr=9.9052e-05) (hash(x)=169616991)
|
| 883 |
+
7330 train 3.901934 (lr=9.8521e-05) (hash(x)=146516856)
|
| 884 |
+
7340 train 4.229925 (lr=9.7993e-05) (hash(x)=171918417)
|
| 885 |
+
7350 train 4.244969 (lr=9.7469e-05) (hash(x)=160511891)
|
| 886 |
+
7360 train 4.087421 (lr=9.6948e-05) (hash(x)=150280167)
|
| 887 |
+
7370 train 4.164104 (lr=9.6431e-05) (hash(x)=151373787)
|
| 888 |
+
7380 train 4.267698 (lr=9.5917e-05) (hash(x)=163311616)
|
| 889 |
+
7390 train 4.250998 (lr=9.5406e-05) (hash(x)=164418521)
|
| 890 |
+
7400 val loss 4.0614
|
| 891 |
+
7400 val perplexity 58.0563
|
| 892 |
+
7400 train 4.013670 (lr=9.4899e-05) (hash(x)=155325873)
|
| 893 |
+
7410 train 4.068769 (lr=9.4395e-05) (hash(x)=135020608)
|
| 894 |
+
7420 train 3.958039 (lr=9.3894e-05) (hash(x)=142812793)
|
| 895 |
+
7430 train 4.070299 (lr=9.3397e-05) (hash(x)=149463283)
|
| 896 |
+
7440 train 4.032311 (lr=9.2904e-05) (hash(x)=170512352)
|
| 897 |
+
7450 train 4.000636 (lr=9.2413e-05) (hash(x)=146253806)
|
| 898 |
+
7460 train 4.657423 (lr=9.1927e-05) (hash(x)=133479853)
|
| 899 |
+
7470 train 4.010655 (lr=9.1443e-05) (hash(x)=138198949)
|
| 900 |
+
7480 train 3.998845 (lr=9.0964e-05) (hash(x)=143831431)
|
| 901 |
+
7490 train 4.159807 (lr=9.0487e-05) (hash(x)=148147144)
|
| 902 |
+
7500 val loss 4.0613
|
| 903 |
+
7500 val perplexity 58.0501
|
| 904 |
+
7500 train 4.057199 (lr=9.0014e-05) (hash(x)=145131256)
|
| 905 |
+
7510 train 4.096821 (lr=8.9545e-05) (hash(x)=150932291)
|
| 906 |
+
7520 train 4.058319 (lr=8.9079e-05) (hash(x)=147605934)
|
| 907 |
+
7530 train 4.011611 (lr=8.8617e-05) (hash(x)=154343507)
|
| 908 |
+
7540 train 4.150372 (lr=8.8158e-05) (hash(x)=148440064)
|
| 909 |
+
7550 train 3.961226 (lr=8.7702e-05) (hash(x)=140505990)
|
| 910 |
+
7560 train 3.936891 (lr=8.7251e-05) (hash(x)=177438878)
|
| 911 |
+
7570 train 4.017305 (lr=8.6802e-05) (hash(x)=154527960)
|
| 912 |
+
7580 train 4.104770 (lr=8.6357e-05) (hash(x)=153793091)
|
| 913 |
+
7590 train 3.928969 (lr=8.5916e-05) (hash(x)=163097232)
|
| 914 |
+
7600 val loss 4.0677
|
| 915 |
+
7600 val perplexity 58.4208
|
| 916 |
+
7600 train 4.060457 (lr=8.5478e-05) (hash(x)=144008365)
|
| 917 |
+
7610 train 4.048656 (lr=8.5044e-05) (hash(x)=139649886)
|
| 918 |
+
7620 train 3.993801 (lr=8.4613e-05) (hash(x)=145452123)
|
| 919 |
+
7630 train 4.018026 (lr=8.4186e-05) (hash(x)=132477285)
|
| 920 |
+
7640 train 3.926105 (lr=8.3763e-05) (hash(x)=139030720)
|
| 921 |
+
7650 train 4.058655 (lr=8.3343e-05) (hash(x)=157466085)
|
| 922 |
+
7660 train 4.067036 (lr=8.2926e-05) (hash(x)=153638385)
|
| 923 |
+
7670 train 4.124471 (lr=8.2514e-05) (hash(x)=154823094)
|
| 924 |
+
7680 train 4.028829 (lr=8.2104e-05) (hash(x)=153107930)
|
| 925 |
+
7690 train 4.012093 (lr=8.1699e-05) (hash(x)=153681916)
|
| 926 |
+
7700 val loss 4.0533
|
| 927 |
+
7700 val perplexity 57.5871
|
| 928 |
+
7700 train 4.325125 (lr=8.1297e-05) (hash(x)=148848532)
|
| 929 |
+
7710 train 4.049396 (lr=8.0898e-05) (hash(x)=156640791)
|
| 930 |
+
7720 train 4.291998 (lr=8.0503e-05) (hash(x)=152573035)
|
| 931 |
+
7730 train 4.129068 (lr=8.0112e-05) (hash(x)=147536091)
|
| 932 |
+
7740 train 4.289264 (lr=7.9725e-05) (hash(x)=175687483)
|
| 933 |
+
7750 train 4.039374 (lr=7.9341e-05) (hash(x)=143775898)
|
| 934 |
+
7760 train 4.064744 (lr=7.8960e-05) (hash(x)=146986193)
|
| 935 |
+
7770 train 3.973891 (lr=7.8584e-05) (hash(x)=160924293)
|
| 936 |
+
7780 train 4.018727 (lr=7.8211e-05) (hash(x)=132579169)
|
| 937 |
+
7790 train 4.244557 (lr=7.7841e-05) (hash(x)=147474225)
|
| 938 |
+
7800 val loss 4.0525
|
| 939 |
+
7800 val perplexity 57.5394
|
| 940 |
+
7800 train 4.086012 (lr=7.7476e-05) (hash(x)=150391642)
|
| 941 |
+
7810 train 4.254696 (lr=7.7114e-05) (hash(x)=156984481)
|
| 942 |
+
7820 train 3.973237 (lr=7.6755e-05) (hash(x)=139652488)
|
| 943 |
+
7830 train 3.994135 (lr=7.6400e-05) (hash(x)=169033643)
|
| 944 |
+
7840 train 3.966312 (lr=7.6049e-05) (hash(x)=152645857)
|
| 945 |
+
7850 train 4.117519 (lr=7.5702e-05) (hash(x)=150228075)
|
| 946 |
+
7860 train 4.032240 (lr=7.5358e-05) (hash(x)=178788133)
|
| 947 |
+
7870 train 4.018313 (lr=7.5018e-05) (hash(x)=149891068)
|
| 948 |
+
7880 train 3.961585 (lr=7.4682e-05) (hash(x)=157461488)
|
| 949 |
+
7890 train 3.927008 (lr=7.4350e-05) (hash(x)=142502446)
|
| 950 |
+
7900 val loss 4.0535
|
| 951 |
+
7900 val perplexity 57.5997
|
| 952 |
+
7900 train 3.859741 (lr=7.4021e-05) (hash(x)=152191414)
|
| 953 |
+
7910 train 4.037789 (lr=7.3696e-05) (hash(x)=151329539)
|
| 954 |
+
7920 train 3.916925 (lr=7.3374e-05) (hash(x)=134143115)
|
| 955 |
+
7930 train 3.921423 (lr=7.3056e-05) (hash(x)=155536160)
|
| 956 |
+
7940 train 3.993058 (lr=7.2742e-05) (hash(x)=146709034)
|
| 957 |
+
7950 train 4.072831 (lr=7.2432e-05) (hash(x)=141966330)
|
| 958 |
+
7960 train 4.045867 (lr=7.2126e-05) (hash(x)=152102296)
|
| 959 |
+
7970 train 4.215632 (lr=7.1823e-05) (hash(x)=150272684)
|
| 960 |
+
7980 train 4.195538 (lr=7.1524e-05) (hash(x)=150251784)
|
| 961 |
+
7990 train 4.177956 (lr=7.1228e-05) (hash(x)=151226159)
|
| 962 |
+
8000 val loss 4.0466
|
| 963 |
+
8000 val perplexity 57.2035
|
| 964 |
+
8000 train 4.068943 (lr=7.0937e-05) (hash(x)=159755587)
|
| 965 |
+
8010 train 4.072999 (lr=7.0649e-05) (hash(x)=137607202)
|
| 966 |
+
8020 train 4.072789 (lr=7.0365e-05) (hash(x)=149574886)
|
| 967 |
+
8030 train 3.981304 (lr=7.0085e-05) (hash(x)=151690927)
|
| 968 |
+
8040 train 4.008238 (lr=6.9808e-05) (hash(x)=151181288)
|
| 969 |
+
8050 train 4.125662 (lr=6.9536e-05) (hash(x)=150465993)
|
| 970 |
+
8060 train 4.140782 (lr=6.9267e-05) (hash(x)=134201020)
|
| 971 |
+
8070 train 4.037067 (lr=6.9002e-05) (hash(x)=157696045)
|
| 972 |
+
8080 train 4.035491 (lr=6.8740e-05) (hash(x)=135581816)
|
| 973 |
+
8090 train 3.966376 (lr=6.8483e-05) (hash(x)=153898278)
|
| 974 |
+
8100 val loss 4.0447
|
| 975 |
+
8100 val perplexity 57.0941
|
| 976 |
+
8100 train 4.095786 (lr=6.8229e-05) (hash(x)=156664468)
|
| 977 |
+
8110 train 4.196200 (lr=6.7979e-05) (hash(x)=161572593)
|
| 978 |
+
8120 train 3.939740 (lr=6.7733e-05) (hash(x)=154381651)
|
| 979 |
+
8130 train 4.201794 (lr=6.7490e-05) (hash(x)=149810514)
|
| 980 |
+
8140 train 4.063170 (lr=6.7252e-05) (hash(x)=145085369)
|
| 981 |
+
8150 train 3.909049 (lr=6.7017e-05) (hash(x)=149020616)
|
| 982 |
+
8160 train 3.791153 (lr=6.6786e-05) (hash(x)=146450644)
|
| 983 |
+
8170 train 3.993742 (lr=6.6559e-05) (hash(x)=148094499)
|
| 984 |
+
8180 train 3.864647 (lr=6.6335e-05) (hash(x)=140810897)
|
| 985 |
+
8190 train 3.968001 (lr=6.6116e-05) (hash(x)=142960309)
|
| 986 |
+
8200 val loss 4.0449
|
| 987 |
+
8200 val perplexity 57.1079
|
| 988 |
+
8200 train 4.004465 (lr=6.5900e-05) (hash(x)=139457379)
|
| 989 |
+
8210 train 3.985307 (lr=6.5688e-05) (hash(x)=145575125)
|
| 990 |
+
8220 train 3.781904 (lr=6.5480e-05) (hash(x)=152836922)
|
| 991 |
+
8230 train 3.868821 (lr=6.5276e-05) (hash(x)=135086706)
|
| 992 |
+
8240 train 3.840396 (lr=6.5076e-05) (hash(x)=153448613)
|
| 993 |
+
8250 train 3.801072 (lr=6.4879e-05) (hash(x)=138041222)
|
| 994 |
+
8260 train 3.861392 (lr=6.4687e-05) (hash(x)=158857968)
|
| 995 |
+
8270 train 3.871105 (lr=6.4498e-05) (hash(x)=157265315)
|
| 996 |
+
8280 train 3.717079 (lr=6.4313e-05) (hash(x)=136059191)
|
| 997 |
+
8290 train 3.791139 (lr=6.4132e-05) (hash(x)=162526712)
|
| 998 |
+
8300 val loss 4.0517
|
| 999 |
+
8300 val perplexity 57.4950
|
| 1000 |
+
8300 train 3.868982 (lr=6.3954e-05) (hash(x)=145478564)
|
| 1001 |
+
8310 train 3.521606 (lr=6.3781e-05) (hash(x)=157367549)
|
| 1002 |
+
8320 train 3.913667 (lr=6.3612e-05) (hash(x)=156085849)
|
| 1003 |
+
8330 train 4.191821 (lr=6.3446e-05) (hash(x)=154822719)
|
| 1004 |
+
8340 train 3.912183 (lr=6.3284e-05) (hash(x)=159688342)
|
| 1005 |
+
8350 train 4.039035 (lr=6.3126e-05) (hash(x)=144133822)
|
| 1006 |
+
8360 train 3.984049 (lr=6.2972e-05) (hash(x)=154155471)
|
| 1007 |
+
8370 train 3.992030 (lr=6.2822e-05) (hash(x)=157077804)
|
| 1008 |
+
8380 train 3.993416 (lr=6.2676e-05) (hash(x)=147796790)
|
| 1009 |
+
8390 train 4.002159 (lr=6.2533e-05) (hash(x)=145208254)
|
| 1010 |
+
8400 val loss 4.0401
|
| 1011 |
+
8400 val perplexity 56.8332
|
| 1012 |
+
8400 train 3.949569 (lr=6.2395e-05) (hash(x)=154982769)
|
| 1013 |
+
8410 train 4.153895 (lr=6.2260e-05) (hash(x)=138073867)
|
| 1014 |
+
8420 train 4.194023 (lr=6.2129e-05) (hash(x)=159521725)
|
| 1015 |
+
8430 train 4.111905 (lr=6.2002e-05) (hash(x)=149988578)
|
| 1016 |
+
8440 train 4.083979 (lr=6.1879e-05) (hash(x)=149457063)
|
| 1017 |
+
8450 train 3.947554 (lr=6.1760e-05) (hash(x)=134678896)
|
| 1018 |
+
8460 train 3.998334 (lr=6.1645e-05) (hash(x)=146122249)
|
| 1019 |
+
8470 train 4.021149 (lr=6.1533e-05) (hash(x)=141393319)
|
| 1020 |
+
8480 train 4.046381 (lr=6.1426e-05) (hash(x)=165396836)
|
| 1021 |
+
8490 train 4.119053 (lr=6.1322e-05) (hash(x)=140043806)
|
| 1022 |
+
8500 val loss 4.0371
|
| 1023 |
+
8500 val perplexity 56.6632
|
| 1024 |
+
8500 train 4.071108 (lr=6.1223e-05) (hash(x)=145798118)
|
| 1025 |
+
8510 train 4.018235 (lr=6.1127e-05) (hash(x)=150358299)
|
| 1026 |
+
8520 train 3.910102 (lr=6.1035e-05) (hash(x)=152572913)
|
| 1027 |
+
8530 train 3.981836 (lr=6.0947e-05) (hash(x)=148669229)
|
| 1028 |
+
8540 train 4.019794 (lr=6.0863e-05) (hash(x)=147078347)
|
| 1029 |
+
8550 train 3.996059 (lr=6.0783e-05) (hash(x)=146715166)
|
| 1030 |
+
8560 train 3.969763 (lr=6.0706e-05) (hash(x)=143176392)
|
| 1031 |
+
8570 train 4.090860 (lr=6.0634e-05) (hash(x)=153825484)
|
| 1032 |
+
8580 train 4.106710 (lr=6.0566e-05) (hash(x)=163379045)
|
| 1033 |
+
8590 train 3.902104 (lr=6.0501e-05) (hash(x)=140050873)
|
| 1034 |
+
8600 val loss 4.0393
|
| 1035 |
+
8600 val perplexity 56.7860
|
| 1036 |
+
8600 train 3.978757 (lr=6.0440e-05) (hash(x)=143231551)
|
| 1037 |
+
8610 train 3.838269 (lr=6.0384e-05) (hash(x)=164291360)
|
| 1038 |
+
8620 train 3.860530 (lr=6.0331e-05) (hash(x)=159122123)
|
| 1039 |
+
8630 train 3.836739 (lr=6.0282e-05) (hash(x)=140041695)
|
| 1040 |
+
8640 train 3.860580 (lr=6.0237e-05) (hash(x)=146010790)
|
| 1041 |
+
8650 train 3.774630 (lr=6.0196e-05) (hash(x)=146882123)
|
| 1042 |
+
8660 train 3.595366 (lr=6.0159e-05) (hash(x)=144933705)
|
| 1043 |
+
8670 train 3.822641 (lr=6.0125e-05) (hash(x)=145580075)
|
| 1044 |
+
8680 train 3.873542 (lr=6.0096e-05) (hash(x)=160982926)
|
| 1045 |
+
8690 train 3.746416 (lr=6.0070e-05) (hash(x)=140047108)
|
| 1046 |
+
8700 val loss 4.0492
|
| 1047 |
+
8700 val perplexity 57.3495
|
| 1048 |
+
8700 train 3.877881 (lr=6.0049e-05) (hash(x)=154780112)
|
| 1049 |
+
8710 train 3.961047 (lr=6.0031e-05) (hash(x)=157568560)
|
| 1050 |
+
8720 train 4.004353 (lr=6.0018e-05) (hash(x)=159587459)
|
| 1051 |
+
8730 train 4.001848 (lr=6.0008e-05) (hash(x)=152918103)
|
| 1052 |
+
8740 train 4.002712 (lr=6.0002e-05) (hash(x)=156928586)
|
| 1053 |
+
8749 val loss 4.0374
|
| 1054 |
+
8749 val perplexity 56.6812
|
lr6e-4_total_batch_size61440_baseline_seed1340/model_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e164078064d8faede2e3f19f43d8a8c8ef062dcaf29f983c8d0f7714b1e8c5b
|
| 3 |
+
size 92843394
|
lr6e-4_total_batch_size61440_baseline_seed1340/optimizer_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9286a5b0a808330739591424ae678a613538c8e3d7048be1a4112b396fbb76c
|
| 3 |
+
size 179406214
|