Upload folder using huggingface_hub
Browse files- lr2e-4_total_batch_size61440_baseline_seed1341/args.json +1 -0
- lr2e-4_total_batch_size61440_baseline_seed1341/dataloader_08749.pt +3 -0
- lr2e-4_total_batch_size61440_baseline_seed1341/log2.txt +1054 -0
- lr2e-4_total_batch_size61440_baseline_seed1341/model_08749.pt +3 -0
- lr2e-4_total_batch_size61440_baseline_seed1341/optimizer_08749.pt +3 -0
lr2e-4_total_batch_size61440_baseline_seed1341/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/lr2e-4_total_batch_size61440_baseline_seed1341", "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": 1341, "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": 0.0002, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "2e-4_61440", "n_embd": 256}
|
lr2e-4_total_batch_size61440_baseline_seed1341/dataloader_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:953385078aa3787b69fc6857dfd48b0a2cd2f4d27c6f8892e01211aca53d07f5
|
| 3 |
+
size 964
|
lr2e-4_total_batch_size61440_baseline_seed1341/log2.txt
ADDED
|
@@ -0,0 +1,1054 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
max_steps: 8750
|
| 2 |
+
0 val loss 11.3073
|
| 3 |
+
0 val perplexity 81417.6484
|
| 4 |
+
0 train 11.312944 (lr=4.0000e-07) (hash(x)=145079536)
|
| 5 |
+
10 train 10.787292 (lr=4.4000e-06) (hash(x)=142935521)
|
| 6 |
+
20 train 10.238597 (lr=8.4000e-06) (hash(x)=136726044)
|
| 7 |
+
30 train 9.943329 (lr=1.2400e-05) (hash(x)=149725277)
|
| 8 |
+
40 train 9.771080 (lr=1.6400e-05) (hash(x)=137527851)
|
| 9 |
+
50 train 9.589325 (lr=2.0400e-05) (hash(x)=154745873)
|
| 10 |
+
60 train 9.474270 (lr=2.4400e-05) (hash(x)=158449312)
|
| 11 |
+
70 train 9.238415 (lr=2.8400e-05) (hash(x)=158867878)
|
| 12 |
+
80 train 9.017046 (lr=3.2400e-05) (hash(x)=152693855)
|
| 13 |
+
90 train 8.756454 (lr=3.6400e-05) (hash(x)=152089620)
|
| 14 |
+
100 val loss 8.5253
|
| 15 |
+
100 val perplexity 5040.7900
|
| 16 |
+
100 train 8.576111 (lr=4.0400e-05) (hash(x)=155800595)
|
| 17 |
+
110 train 8.351785 (lr=4.4400e-05) (hash(x)=156324852)
|
| 18 |
+
120 train 8.029536 (lr=4.8400e-05) (hash(x)=148079157)
|
| 19 |
+
130 train 7.916952 (lr=5.2400e-05) (hash(x)=151260070)
|
| 20 |
+
140 train 7.779294 (lr=5.6400e-05) (hash(x)=161098412)
|
| 21 |
+
150 train 7.639030 (lr=6.0400e-05) (hash(x)=148595389)
|
| 22 |
+
160 train 7.552447 (lr=6.4400e-05) (hash(x)=162558230)
|
| 23 |
+
170 train 7.311265 (lr=6.8400e-05) (hash(x)=136967959)
|
| 24 |
+
180 train 7.534757 (lr=7.2400e-05) (hash(x)=160279390)
|
| 25 |
+
190 train 7.267334 (lr=7.6400e-05) (hash(x)=139072589)
|
| 26 |
+
200 val loss 7.3129
|
| 27 |
+
200 val perplexity 1499.5332
|
| 28 |
+
200 train 7.149296 (lr=8.0400e-05) (hash(x)=145606733)
|
| 29 |
+
210 train 7.242370 (lr=8.4400e-05) (hash(x)=151385952)
|
| 30 |
+
220 train 7.067221 (lr=8.8400e-05) (hash(x)=141556767)
|
| 31 |
+
230 train 7.266524 (lr=9.2400e-05) (hash(x)=159727511)
|
| 32 |
+
240 train 6.880598 (lr=9.6400e-05) (hash(x)=145935602)
|
| 33 |
+
250 train 6.913472 (lr=1.0040e-04) (hash(x)=138221231)
|
| 34 |
+
260 train 6.971863 (lr=1.0440e-04) (hash(x)=148897684)
|
| 35 |
+
270 train 7.453491 (lr=1.0840e-04) (hash(x)=155063217)
|
| 36 |
+
280 train 6.790126 (lr=1.1240e-04) (hash(x)=147263890)
|
| 37 |
+
290 train 6.814515 (lr=1.1640e-04) (hash(x)=151660516)
|
| 38 |
+
300 val loss 6.8247
|
| 39 |
+
300 val perplexity 920.2725
|
| 40 |
+
300 train 6.756166 (lr=1.2040e-04) (hash(x)=150367139)
|
| 41 |
+
310 train 6.500960 (lr=1.2440e-04) (hash(x)=151170832)
|
| 42 |
+
320 train 6.646540 (lr=1.2840e-04) (hash(x)=138591183)
|
| 43 |
+
330 train 6.431677 (lr=1.3240e-04) (hash(x)=143603420)
|
| 44 |
+
340 train 6.531790 (lr=1.3640e-04) (hash(x)=140068134)
|
| 45 |
+
350 train 6.572825 (lr=1.4040e-04) (hash(x)=155579314)
|
| 46 |
+
360 train 6.364952 (lr=1.4440e-04) (hash(x)=148496629)
|
| 47 |
+
370 train 6.486331 (lr=1.4840e-04) (hash(x)=156763563)
|
| 48 |
+
380 train 6.384606 (lr=1.5240e-04) (hash(x)=160905666)
|
| 49 |
+
390 train 6.676793 (lr=1.5640e-04) (hash(x)=157064388)
|
| 50 |
+
400 val loss 6.4644
|
| 51 |
+
400 val perplexity 641.8781
|
| 52 |
+
400 train 6.753990 (lr=1.6040e-04) (hash(x)=155747374)
|
| 53 |
+
410 train 6.403709 (lr=1.6440e-04) (hash(x)=147294192)
|
| 54 |
+
420 train 6.431028 (lr=1.6840e-04) (hash(x)=144952473)
|
| 55 |
+
430 train 6.377934 (lr=1.7240e-04) (hash(x)=160908725)
|
| 56 |
+
440 train 6.417203 (lr=1.7640e-04) (hash(x)=145105827)
|
| 57 |
+
450 train 6.498774 (lr=1.8040e-04) (hash(x)=159334575)
|
| 58 |
+
460 train 6.314979 (lr=1.8440e-04) (hash(x)=161983315)
|
| 59 |
+
470 train 6.211282 (lr=1.8840e-04) (hash(x)=138900475)
|
| 60 |
+
480 train 6.232064 (lr=1.9240e-04) (hash(x)=152981338)
|
| 61 |
+
490 train 6.144073 (lr=1.9640e-04) (hash(x)=143353862)
|
| 62 |
+
500 val loss 6.1796
|
| 63 |
+
500 val perplexity 482.7803
|
| 64 |
+
500 train 6.054503 (lr=2.0000e-04) (hash(x)=140604760)
|
| 65 |
+
510 train 6.010351 (lr=2.0000e-04) (hash(x)=150184398)
|
| 66 |
+
520 train 6.098560 (lr=2.0000e-04) (hash(x)=141863041)
|
| 67 |
+
530 train 6.067205 (lr=1.9999e-04) (hash(x)=153148776)
|
| 68 |
+
540 train 6.013084 (lr=1.9999e-04) (hash(x)=144196873)
|
| 69 |
+
550 train 6.007036 (lr=1.9998e-04) (hash(x)=146208052)
|
| 70 |
+
560 train 5.952679 (lr=1.9998e-04) (hash(x)=139601869)
|
| 71 |
+
570 train 6.092420 (lr=1.9997e-04) (hash(x)=150976466)
|
| 72 |
+
580 train 6.009419 (lr=1.9996e-04) (hash(x)=150801563)
|
| 73 |
+
590 train 6.157280 (lr=1.9995e-04) (hash(x)=167531956)
|
| 74 |
+
600 val loss 5.9907
|
| 75 |
+
600 val perplexity 399.7049
|
| 76 |
+
600 train 5.894267 (lr=1.9993e-04) (hash(x)=148404734)
|
| 77 |
+
610 train 5.861865 (lr=1.9992e-04) (hash(x)=146223362)
|
| 78 |
+
620 train 5.824109 (lr=1.9991e-04) (hash(x)=147269386)
|
| 79 |
+
630 train 5.767991 (lr=1.9989e-04) (hash(x)=149819251)
|
| 80 |
+
640 train 6.088976 (lr=1.9987e-04) (hash(x)=146771854)
|
| 81 |
+
650 train 6.001534 (lr=1.9985e-04) (hash(x)=155681970)
|
| 82 |
+
660 train 5.714972 (lr=1.9983e-04) (hash(x)=138901204)
|
| 83 |
+
670 train 5.537091 (lr=1.9981e-04) (hash(x)=143417964)
|
| 84 |
+
680 train 5.716635 (lr=1.9979e-04) (hash(x)=141188344)
|
| 85 |
+
690 train 5.824905 (lr=1.9976e-04) (hash(x)=148101044)
|
| 86 |
+
700 val loss 5.8238
|
| 87 |
+
700 val perplexity 338.2421
|
| 88 |
+
700 train 5.816873 (lr=1.9974e-04) (hash(x)=148115934)
|
| 89 |
+
710 train 5.838454 (lr=1.9971e-04) (hash(x)=142019907)
|
| 90 |
+
720 train 5.795227 (lr=1.9968e-04) (hash(x)=144138833)
|
| 91 |
+
730 train 5.837140 (lr=1.9966e-04) (hash(x)=150694253)
|
| 92 |
+
740 train 5.722432 (lr=1.9962e-04) (hash(x)=157694921)
|
| 93 |
+
750 train 5.816108 (lr=1.9959e-04) (hash(x)=157074034)
|
| 94 |
+
760 train 5.719412 (lr=1.9956e-04) (hash(x)=157110692)
|
| 95 |
+
770 train 5.648320 (lr=1.9952e-04) (hash(x)=145925624)
|
| 96 |
+
780 train 5.746453 (lr=1.9949e-04) (hash(x)=156687510)
|
| 97 |
+
790 train 5.708777 (lr=1.9945e-04) (hash(x)=150458321)
|
| 98 |
+
800 val loss 5.6768
|
| 99 |
+
800 val perplexity 292.0017
|
| 100 |
+
800 train 5.563079 (lr=1.9941e-04) (hash(x)=137464699)
|
| 101 |
+
810 train 5.607800 (lr=1.9937e-04) (hash(x)=158371555)
|
| 102 |
+
820 train 5.630704 (lr=1.9933e-04) (hash(x)=149201877)
|
| 103 |
+
830 train 5.553880 (lr=1.9929e-04) (hash(x)=143564160)
|
| 104 |
+
840 train 5.669060 (lr=1.9925e-04) (hash(x)=153937595)
|
| 105 |
+
850 train 5.607577 (lr=1.9920e-04) (hash(x)=166955614)
|
| 106 |
+
860 train 5.639781 (lr=1.9916e-04) (hash(x)=164264614)
|
| 107 |
+
870 train 5.421393 (lr=1.9911e-04) (hash(x)=139219447)
|
| 108 |
+
880 train 5.670821 (lr=1.9906e-04) (hash(x)=151139580)
|
| 109 |
+
890 train 5.514522 (lr=1.9901e-04) (hash(x)=147154297)
|
| 110 |
+
900 val loss 5.5801
|
| 111 |
+
900 val perplexity 265.1008
|
| 112 |
+
900 train 5.385757 (lr=1.9896e-04) (hash(x)=143886042)
|
| 113 |
+
910 train 5.330106 (lr=1.9891e-04) (hash(x)=144136925)
|
| 114 |
+
920 train 5.491297 (lr=1.9885e-04) (hash(x)=139764865)
|
| 115 |
+
930 train 5.465683 (lr=1.9880e-04) (hash(x)=137258208)
|
| 116 |
+
940 train 5.542581 (lr=1.9874e-04) (hash(x)=151881247)
|
| 117 |
+
950 train 5.595419 (lr=1.9868e-04) (hash(x)=193610391)
|
| 118 |
+
960 train 5.413902 (lr=1.9862e-04) (hash(x)=149364435)
|
| 119 |
+
970 train 5.437432 (lr=1.9856e-04) (hash(x)=148828909)
|
| 120 |
+
980 train 5.258856 (lr=1.9850e-04) (hash(x)=143605331)
|
| 121 |
+
990 train 5.289958 (lr=1.9844e-04) (hash(x)=153415979)
|
| 122 |
+
1000 val loss 5.4990
|
| 123 |
+
1000 val perplexity 244.4417
|
| 124 |
+
1000 train 5.553262 (lr=1.9837e-04) (hash(x)=163799796)
|
| 125 |
+
1010 train 5.524416 (lr=1.9831e-04) (hash(x)=145937873)
|
| 126 |
+
1020 train 5.443154 (lr=1.9824e-04) (hash(x)=165459184)
|
| 127 |
+
1030 train 5.535046 (lr=1.9817e-04) (hash(x)=150932001)
|
| 128 |
+
1040 train 5.490148 (lr=1.9810e-04) (hash(x)=162567106)
|
| 129 |
+
1050 train 5.482735 (lr=1.9803e-04) (hash(x)=154107339)
|
| 130 |
+
1060 train 5.477661 (lr=1.9796e-04) (hash(x)=149509318)
|
| 131 |
+
1070 train 5.326945 (lr=1.9789e-04) (hash(x)=152383589)
|
| 132 |
+
1080 train 5.383815 (lr=1.9781e-04) (hash(x)=155926898)
|
| 133 |
+
1090 train 5.367822 (lr=1.9774e-04) (hash(x)=150988851)
|
| 134 |
+
1100 val loss 5.4223
|
| 135 |
+
1100 val perplexity 226.3965
|
| 136 |
+
1100 train 5.481339 (lr=1.9766e-04) (hash(x)=144592844)
|
| 137 |
+
1110 train 5.390807 (lr=1.9758e-04) (hash(x)=146758273)
|
| 138 |
+
1120 train 5.290126 (lr=1.9750e-04) (hash(x)=146142884)
|
| 139 |
+
1130 train 5.410864 (lr=1.9742e-04) (hash(x)=152078362)
|
| 140 |
+
1140 train 5.468931 (lr=1.9734e-04) (hash(x)=160186914)
|
| 141 |
+
1150 train 5.438212 (lr=1.9726e-04) (hash(x)=154535861)
|
| 142 |
+
1160 train 5.285423 (lr=1.9717e-04) (hash(x)=151399108)
|
| 143 |
+
1170 train 5.348726 (lr=1.9709e-04) (hash(x)=146833517)
|
| 144 |
+
1180 train 5.238908 (lr=1.9700e-04) (hash(x)=140850009)
|
| 145 |
+
1190 train 5.288931 (lr=1.9691e-04) (hash(x)=160947364)
|
| 146 |
+
1200 val loss 5.3306
|
| 147 |
+
1200 val perplexity 206.5528
|
| 148 |
+
1200 train 5.341519 (lr=1.9682e-04) (hash(x)=204706354)
|
| 149 |
+
1210 train 5.128953 (lr=1.9673e-04) (hash(x)=145764356)
|
| 150 |
+
1220 train 5.237423 (lr=1.9664e-04) (hash(x)=153116691)
|
| 151 |
+
1230 train 5.245912 (lr=1.9654e-04) (hash(x)=160544763)
|
| 152 |
+
1240 train 5.125156 (lr=1.9645e-04) (hash(x)=137389978)
|
| 153 |
+
1250 train 5.202380 (lr=1.9635e-04) (hash(x)=143783202)
|
| 154 |
+
1260 train 5.102818 (lr=1.9626e-04) (hash(x)=153470243)
|
| 155 |
+
1270 train 5.123411 (lr=1.9616e-04) (hash(x)=149055354)
|
| 156 |
+
1280 train 5.128794 (lr=1.9606e-04) (hash(x)=146874658)
|
| 157 |
+
1290 train 5.309134 (lr=1.9596e-04) (hash(x)=152527777)
|
| 158 |
+
1300 val loss 5.2711
|
| 159 |
+
1300 val perplexity 194.6209
|
| 160 |
+
1300 train 5.230252 (lr=1.9586e-04) (hash(x)=150862210)
|
| 161 |
+
1310 train 5.241816 (lr=1.9575e-04) (hash(x)=157185054)
|
| 162 |
+
1320 train 5.302879 (lr=1.9565e-04) (hash(x)=153733727)
|
| 163 |
+
1330 train 5.309816 (lr=1.9554e-04) (hash(x)=163696656)
|
| 164 |
+
1340 train 5.393682 (lr=1.9543e-04) (hash(x)=156931875)
|
| 165 |
+
1350 train 5.198901 (lr=1.9533e-04) (hash(x)=150753426)
|
| 166 |
+
1360 train 5.176104 (lr=1.9522e-04) (hash(x)=145642367)
|
| 167 |
+
1370 train 5.244959 (lr=1.9511e-04) (hash(x)=159217491)
|
| 168 |
+
1380 train 5.211625 (lr=1.9499e-04) (hash(x)=151499952)
|
| 169 |
+
1390 train 5.313730 (lr=1.9488e-04) (hash(x)=147137197)
|
| 170 |
+
1400 val loss 5.2205
|
| 171 |
+
1400 val perplexity 185.0231
|
| 172 |
+
1400 train 5.246036 (lr=1.9477e-04) (hash(x)=147766811)
|
| 173 |
+
1410 train 5.204038 (lr=1.9465e-04) (hash(x)=155728979)
|
| 174 |
+
1420 train 5.048640 (lr=1.9453e-04) (hash(x)=151858238)
|
| 175 |
+
1430 train 5.132884 (lr=1.9441e-04) (hash(x)=147007501)
|
| 176 |
+
1440 train 5.008930 (lr=1.9430e-04) (hash(x)=123449142)
|
| 177 |
+
1450 train 5.060804 (lr=1.9417e-04) (hash(x)=151177814)
|
| 178 |
+
1460 train 5.237168 (lr=1.9405e-04) (hash(x)=151804649)
|
| 179 |
+
1470 train 5.145801 (lr=1.9393e-04) (hash(x)=140614506)
|
| 180 |
+
1480 train 5.362651 (lr=1.9381e-04) (hash(x)=154590935)
|
| 181 |
+
1490 train 5.133527 (lr=1.9368e-04) (hash(x)=155969672)
|
| 182 |
+
1500 val loss 5.1527
|
| 183 |
+
1500 val perplexity 172.8984
|
| 184 |
+
1500 train 5.042354 (lr=1.9355e-04) (hash(x)=135925327)
|
| 185 |
+
1510 train 5.036717 (lr=1.9343e-04) (hash(x)=140488340)
|
| 186 |
+
1520 train 5.081488 (lr=1.9330e-04) (hash(x)=150707923)
|
| 187 |
+
1530 train 4.962043 (lr=1.9317e-04) (hash(x)=141081542)
|
| 188 |
+
1540 train 5.024741 (lr=1.9303e-04) (hash(x)=150136050)
|
| 189 |
+
1550 train 4.871648 (lr=1.9290e-04) (hash(x)=124842353)
|
| 190 |
+
1560 train 5.011339 (lr=1.9277e-04) (hash(x)=142320279)
|
| 191 |
+
1570 train 4.930458 (lr=1.9263e-04) (hash(x)=137593322)
|
| 192 |
+
1580 train 5.010745 (lr=1.9250e-04) (hash(x)=154607301)
|
| 193 |
+
1590 train 4.932122 (lr=1.9236e-04) (hash(x)=143089856)
|
| 194 |
+
1600 val loss 5.1373
|
| 195 |
+
1600 val perplexity 170.2600
|
| 196 |
+
1600 train 5.107610 (lr=1.9222e-04) (hash(x)=160440642)
|
| 197 |
+
1610 train 5.316317 (lr=1.9208e-04) (hash(x)=150875971)
|
| 198 |
+
1620 train 5.246145 (lr=1.9194e-04) (hash(x)=165443918)
|
| 199 |
+
1630 train 5.062717 (lr=1.9180e-04) (hash(x)=145853246)
|
| 200 |
+
1640 train 5.034171 (lr=1.9165e-04) (hash(x)=120920586)
|
| 201 |
+
1650 train 5.020453 (lr=1.9151e-04) (hash(x)=146732412)
|
| 202 |
+
1660 train 5.081528 (lr=1.9136e-04) (hash(x)=148900413)
|
| 203 |
+
1670 train 5.342816 (lr=1.9121e-04) (hash(x)=147090451)
|
| 204 |
+
1680 train 5.102453 (lr=1.9107e-04) (hash(x)=153099341)
|
| 205 |
+
1690 train 5.083346 (lr=1.9092e-04) (hash(x)=148556821)
|
| 206 |
+
1700 val loss 5.0571
|
| 207 |
+
1700 val perplexity 157.1411
|
| 208 |
+
1700 train 5.208211 (lr=1.9077e-04) (hash(x)=151184106)
|
| 209 |
+
1710 train 5.173571 (lr=1.9061e-04) (hash(x)=151304861)
|
| 210 |
+
1720 train 4.853308 (lr=1.9046e-04) (hash(x)=147047589)
|
| 211 |
+
1730 train 4.985229 (lr=1.9031e-04) (hash(x)=157613602)
|
| 212 |
+
1740 train 4.890061 (lr=1.9015e-04) (hash(x)=145631923)
|
| 213 |
+
1750 train 4.888452 (lr=1.9000e-04) (hash(x)=153416694)
|
| 214 |
+
1760 train 5.077673 (lr=1.8984e-04) (hash(x)=143038500)
|
| 215 |
+
1770 train 4.996080 (lr=1.8968e-04) (hash(x)=163545780)
|
| 216 |
+
1780 train 4.966696 (lr=1.8952e-04) (hash(x)=159712197)
|
| 217 |
+
1790 train 4.866303 (lr=1.8936e-04) (hash(x)=146293430)
|
| 218 |
+
1800 val loss 5.0079
|
| 219 |
+
1800 val perplexity 149.5899
|
| 220 |
+
1800 train 4.951153 (lr=1.8920e-04) (hash(x)=148100580)
|
| 221 |
+
1810 train 4.902009 (lr=1.8903e-04) (hash(x)=137858614)
|
| 222 |
+
1820 train 4.951807 (lr=1.8887e-04) (hash(x)=155765218)
|
| 223 |
+
1830 train 4.869789 (lr=1.8870e-04) (hash(x)=147554787)
|
| 224 |
+
1840 train 4.928384 (lr=1.8854e-04) (hash(x)=160001341)
|
| 225 |
+
1850 train 4.841151 (lr=1.8837e-04) (hash(x)=153059873)
|
| 226 |
+
1860 train 4.802756 (lr=1.8820e-04) (hash(x)=155656532)
|
| 227 |
+
1870 train 4.832808 (lr=1.8803e-04) (hash(x)=146515528)
|
| 228 |
+
1880 train 4.698668 (lr=1.8786e-04) (hash(x)=144385809)
|
| 229 |
+
1890 train 4.754938 (lr=1.8768e-04) (hash(x)=150375274)
|
| 230 |
+
1900 val loss 5.0044
|
| 231 |
+
1900 val perplexity 149.0625
|
| 232 |
+
1900 train 4.790755 (lr=1.8751e-04) (hash(x)=149434659)
|
| 233 |
+
1910 train 4.856748 (lr=1.8734e-04) (hash(x)=140811482)
|
| 234 |
+
1920 train 4.918816 (lr=1.8716e-04) (hash(x)=149776450)
|
| 235 |
+
1930 train 4.903820 (lr=1.8698e-04) (hash(x)=153611546)
|
| 236 |
+
1940 train 4.976712 (lr=1.8680e-04) (hash(x)=152628274)
|
| 237 |
+
1950 train 4.995188 (lr=1.8663e-04) (hash(x)=151183924)
|
| 238 |
+
1960 train 4.960773 (lr=1.8645e-04) (hash(x)=143507386)
|
| 239 |
+
1970 train 5.040929 (lr=1.8626e-04) (hash(x)=180301574)
|
| 240 |
+
1980 train 4.932968 (lr=1.8608e-04) (hash(x)=156383157)
|
| 241 |
+
1990 train 4.933593 (lr=1.8590e-04) (hash(x)=145911523)
|
| 242 |
+
2000 val loss 4.9144
|
| 243 |
+
2000 val perplexity 136.2345
|
| 244 |
+
2000 train 4.925945 (lr=1.8571e-04) (hash(x)=152285486)
|
| 245 |
+
2010 train 4.907732 (lr=1.8553e-04) (hash(x)=154531062)
|
| 246 |
+
2020 train 4.849087 (lr=1.8534e-04) (hash(x)=154403800)
|
| 247 |
+
2030 train 4.973152 (lr=1.8515e-04) (hash(x)=153109813)
|
| 248 |
+
2040 train 4.978485 (lr=1.8496e-04) (hash(x)=149156251)
|
| 249 |
+
2050 train 4.774532 (lr=1.8477e-04) (hash(x)=147613800)
|
| 250 |
+
2060 train 4.898724 (lr=1.8458e-04) (hash(x)=160283013)
|
| 251 |
+
2070 train 4.853971 (lr=1.8439e-04) (hash(x)=141632717)
|
| 252 |
+
2080 train 4.813588 (lr=1.8420e-04) (hash(x)=147292755)
|
| 253 |
+
2090 train 4.985944 (lr=1.8400e-04) (hash(x)=146447537)
|
| 254 |
+
2100 val loss 4.8677
|
| 255 |
+
2100 val perplexity 130.0189
|
| 256 |
+
2100 train 4.607906 (lr=1.8381e-04) (hash(x)=144294295)
|
| 257 |
+
2110 train 5.173122 (lr=1.8361e-04) (hash(x)=142838110)
|
| 258 |
+
2120 train 5.134693 (lr=1.8341e-04) (hash(x)=151878701)
|
| 259 |
+
2130 train 4.725236 (lr=1.8321e-04) (hash(x)=143929160)
|
| 260 |
+
2140 train 4.701893 (lr=1.8301e-04) (hash(x)=149368632)
|
| 261 |
+
2150 train 4.746010 (lr=1.8281e-04) (hash(x)=141460003)
|
| 262 |
+
2160 train 4.719842 (lr=1.8261e-04) (hash(x)=140793655)
|
| 263 |
+
2170 train 4.875239 (lr=1.8241e-04) (hash(x)=146336171)
|
| 264 |
+
2180 train 4.703368 (lr=1.8220e-04) (hash(x)=150200011)
|
| 265 |
+
2190 train 4.752065 (lr=1.8200e-04) (hash(x)=155337561)
|
| 266 |
+
2200 val loss 4.8491
|
| 267 |
+
2200 val perplexity 127.6268
|
| 268 |
+
2200 train 4.763447 (lr=1.8179e-04) (hash(x)=175030215)
|
| 269 |
+
2210 train 4.883061 (lr=1.8158e-04) (hash(x)=139653118)
|
| 270 |
+
2220 train 4.780651 (lr=1.8138e-04) (hash(x)=151690464)
|
| 271 |
+
2230 train 4.530656 (lr=1.8117e-04) (hash(x)=164526288)
|
| 272 |
+
2240 train 4.609910 (lr=1.8096e-04) (hash(x)=141536420)
|
| 273 |
+
2250 train 4.849339 (lr=1.8074e-04) (hash(x)=167362519)
|
| 274 |
+
2260 train 4.566165 (lr=1.8053e-04) (hash(x)=142323410)
|
| 275 |
+
2270 train 4.968311 (lr=1.8032e-04) (hash(x)=176298114)
|
| 276 |
+
2280 train 4.645576 (lr=1.8010e-04) (hash(x)=149454362)
|
| 277 |
+
2290 train 4.792210 (lr=1.7989e-04) (hash(x)=172920991)
|
| 278 |
+
2300 val loss 4.8096
|
| 279 |
+
2300 val perplexity 122.6844
|
| 280 |
+
2300 train 4.726462 (lr=1.7967e-04) (hash(x)=150831428)
|
| 281 |
+
2310 train 4.582036 (lr=1.7946e-04) (hash(x)=141773102)
|
| 282 |
+
2320 train 4.551571 (lr=1.7924e-04) (hash(x)=146520072)
|
| 283 |
+
2330 train 4.749744 (lr=1.7902e-04) (hash(x)=174782103)
|
| 284 |
+
2340 train 4.866252 (lr=1.7880e-04) (hash(x)=171102155)
|
| 285 |
+
2350 train 4.701398 (lr=1.7858e-04) (hash(x)=152910404)
|
| 286 |
+
2360 train 4.680302 (lr=1.7835e-04) (hash(x)=147122637)
|
| 287 |
+
2370 train 4.636809 (lr=1.7813e-04) (hash(x)=152558061)
|
| 288 |
+
2380 train 4.682303 (lr=1.7791e-04) (hash(x)=146957950)
|
| 289 |
+
2390 train 4.730158 (lr=1.7768e-04) (hash(x)=156879460)
|
| 290 |
+
2400 val loss 4.7487
|
| 291 |
+
2400 val perplexity 115.4378
|
| 292 |
+
2400 train 4.897130 (lr=1.7745e-04) (hash(x)=140808297)
|
| 293 |
+
2410 train 4.625083 (lr=1.7723e-04) (hash(x)=164207329)
|
| 294 |
+
2420 train 4.822042 (lr=1.7700e-04) (hash(x)=150119566)
|
| 295 |
+
2430 train 4.662491 (lr=1.7677e-04) (hash(x)=148059202)
|
| 296 |
+
2440 train 4.738927 (lr=1.7654e-04) (hash(x)=144913823)
|
| 297 |
+
2450 train 4.873259 (lr=1.7631e-04) (hash(x)=155469321)
|
| 298 |
+
2460 train 4.945536 (lr=1.7607e-04) (hash(x)=147971519)
|
| 299 |
+
2470 train 4.641431 (lr=1.7584e-04) (hash(x)=151115240)
|
| 300 |
+
2480 train 4.643377 (lr=1.7561e-04) (hash(x)=158904683)
|
| 301 |
+
2490 train 4.647720 (lr=1.7537e-04) (hash(x)=153639698)
|
| 302 |
+
2500 val loss 4.7342
|
| 303 |
+
2500 val perplexity 113.7694
|
| 304 |
+
2500 train 4.689008 (lr=1.7514e-04) (hash(x)=153160275)
|
| 305 |
+
2510 train 4.529145 (lr=1.7490e-04) (hash(x)=166439539)
|
| 306 |
+
2520 train 4.763387 (lr=1.7466e-04) (hash(x)=141364112)
|
| 307 |
+
2530 train 4.733432 (lr=1.7442e-04) (hash(x)=147911679)
|
| 308 |
+
2540 train 4.666969 (lr=1.7418e-04) (hash(x)=152607847)
|
| 309 |
+
2550 train 4.596761 (lr=1.7394e-04) (hash(x)=149178874)
|
| 310 |
+
2560 train 4.780900 (lr=1.7370e-04) (hash(x)=158403024)
|
| 311 |
+
2570 train 4.611135 (lr=1.7346e-04) (hash(x)=148252901)
|
| 312 |
+
2580 train 4.651484 (lr=1.7321e-04) (hash(x)=156933994)
|
| 313 |
+
2590 train 4.583423 (lr=1.7297e-04) (hash(x)=147869995)
|
| 314 |
+
2600 val loss 4.7020
|
| 315 |
+
2600 val perplexity 110.1620
|
| 316 |
+
2600 train 4.583520 (lr=1.7272e-04) (hash(x)=133990623)
|
| 317 |
+
2610 train 4.668188 (lr=1.7248e-04) (hash(x)=155745526)
|
| 318 |
+
2620 train 4.577861 (lr=1.7223e-04) (hash(x)=145640933)
|
| 319 |
+
2630 train 4.654515 (lr=1.7198e-04) (hash(x)=144600130)
|
| 320 |
+
2640 train 4.496461 (lr=1.7173e-04) (hash(x)=143319960)
|
| 321 |
+
2650 train 4.413713 (lr=1.7148e-04) (hash(x)=144841781)
|
| 322 |
+
2660 train 4.657990 (lr=1.7123e-04) (hash(x)=142306571)
|
| 323 |
+
2670 train 4.604876 (lr=1.7098e-04) (hash(x)=150824939)
|
| 324 |
+
2680 train 4.651042 (lr=1.7073e-04) (hash(x)=144954685)
|
| 325 |
+
2690 train 4.434755 (lr=1.7048e-04) (hash(x)=144071317)
|
| 326 |
+
2700 val loss 4.7046
|
| 327 |
+
2700 val perplexity 110.4551
|
| 328 |
+
2700 train 4.606939 (lr=1.7022e-04) (hash(x)=142860944)
|
| 329 |
+
2710 train 4.568398 (lr=1.6997e-04) (hash(x)=150879676)
|
| 330 |
+
2720 train 4.455013 (lr=1.6971e-04) (hash(x)=158434126)
|
| 331 |
+
2730 train 4.721415 (lr=1.6945e-04) (hash(x)=149786230)
|
| 332 |
+
2740 train 4.739392 (lr=1.6920e-04) (hash(x)=148769334)
|
| 333 |
+
2750 train 4.670616 (lr=1.6894e-04) (hash(x)=151565702)
|
| 334 |
+
2760 train 4.806107 (lr=1.6868e-04) (hash(x)=162471158)
|
| 335 |
+
2770 train 4.716993 (lr=1.6842e-04) (hash(x)=168950803)
|
| 336 |
+
2780 train 4.726966 (lr=1.6816e-04) (hash(x)=149815020)
|
| 337 |
+
2790 train 4.725129 (lr=1.6789e-04) (hash(x)=161235555)
|
| 338 |
+
2800 val loss 4.6529
|
| 339 |
+
2800 val perplexity 104.8894
|
| 340 |
+
2800 train 4.580770 (lr=1.6763e-04) (hash(x)=137959511)
|
| 341 |
+
2810 train 4.739936 (lr=1.6737e-04) (hash(x)=148420397)
|
| 342 |
+
2820 train 4.619926 (lr=1.6710e-04) (hash(x)=151862744)
|
| 343 |
+
2830 train 4.721663 (lr=1.6684e-04) (hash(x)=153223303)
|
| 344 |
+
2840 train 4.660809 (lr=1.6657e-04) (hash(x)=149794760)
|
| 345 |
+
2850 train 4.721321 (lr=1.6631e-04) (hash(x)=153517085)
|
| 346 |
+
2860 train 4.695207 (lr=1.6604e-04) (hash(x)=152044927)
|
| 347 |
+
2870 train 4.841790 (lr=1.6577e-04) (hash(x)=139708749)
|
| 348 |
+
2880 train 4.621458 (lr=1.6550e-04) (hash(x)=152675893)
|
| 349 |
+
2890 train 4.657474 (lr=1.6523e-04) (hash(x)=153950772)
|
| 350 |
+
2900 val loss 4.6178
|
| 351 |
+
2900 val perplexity 101.2756
|
| 352 |
+
2900 train 4.583674 (lr=1.6496e-04) (hash(x)=147009873)
|
| 353 |
+
2910 train 4.659160 (lr=1.6469e-04) (hash(x)=147497701)
|
| 354 |
+
2920 train 4.807261 (lr=1.6441e-04) (hash(x)=149615491)
|
| 355 |
+
2930 train 4.790352 (lr=1.6414e-04) (hash(x)=152588554)
|
| 356 |
+
2940 train 4.613001 (lr=1.6387e-04) (hash(x)=147080266)
|
| 357 |
+
2950 train 4.845810 (lr=1.6359e-04) (hash(x)=149607682)
|
| 358 |
+
2960 train 4.553452 (lr=1.6332e-04) (hash(x)=156855690)
|
| 359 |
+
2970 train 4.539155 (lr=1.6304e-04) (hash(x)=147759452)
|
| 360 |
+
2980 train 4.510334 (lr=1.6276e-04) (hash(x)=168906251)
|
| 361 |
+
2990 train 4.608634 (lr=1.6248e-04) (hash(x)=159612982)
|
| 362 |
+
3000 val loss 4.6042
|
| 363 |
+
3000 val perplexity 99.9070
|
| 364 |
+
3000 train 4.449373 (lr=1.6221e-04) (hash(x)=158264841)
|
| 365 |
+
3010 train 4.475074 (lr=1.6193e-04) (hash(x)=151971710)
|
| 366 |
+
3020 train 4.521866 (lr=1.6165e-04) (hash(x)=148961929)
|
| 367 |
+
3030 train 4.346803 (lr=1.6136e-04) (hash(x)=144951447)
|
| 368 |
+
3040 train 4.361747 (lr=1.6108e-04) (hash(x)=137727737)
|
| 369 |
+
3050 train 4.405043 (lr=1.6080e-04) (hash(x)=145598216)
|
| 370 |
+
3060 train 4.419352 (lr=1.6052e-04) (hash(x)=139250122)
|
| 371 |
+
3070 train 4.470995 (lr=1.6023e-04) (hash(x)=158020889)
|
| 372 |
+
3080 train 4.292494 (lr=1.5995e-04) (hash(x)=147345272)
|
| 373 |
+
3090 train 4.321687 (lr=1.5966e-04) (hash(x)=148006111)
|
| 374 |
+
3100 val loss 4.5993
|
| 375 |
+
3100 val perplexity 99.4136
|
| 376 |
+
3100 train 4.451292 (lr=1.5938e-04) (hash(x)=139232251)
|
| 377 |
+
3110 train 4.287091 (lr=1.5909e-04) (hash(x)=164654684)
|
| 378 |
+
3120 train 4.513332 (lr=1.5880e-04) (hash(x)=153975753)
|
| 379 |
+
3130 train 4.641528 (lr=1.5851e-04) (hash(x)=154784772)
|
| 380 |
+
3140 train 4.516481 (lr=1.5822e-04) (hash(x)=156671619)
|
| 381 |
+
3150 train 4.501432 (lr=1.5793e-04) (hash(x)=158188012)
|
| 382 |
+
3160 train 4.589303 (lr=1.5764e-04) (hash(x)=152266024)
|
| 383 |
+
3170 train 4.554278 (lr=1.5735e-04) (hash(x)=145418617)
|
| 384 |
+
3180 train 4.609259 (lr=1.5706e-04) (hash(x)=149947903)
|
| 385 |
+
3190 train 4.502955 (lr=1.5677e-04) (hash(x)=156664225)
|
| 386 |
+
3200 val loss 4.5651
|
| 387 |
+
3200 val perplexity 96.0691
|
| 388 |
+
3200 train 4.598315 (lr=1.5648e-04) (hash(x)=153436104)
|
| 389 |
+
3210 train 4.694862 (lr=1.5618e-04) (hash(x)=163550923)
|
| 390 |
+
3220 train 4.435464 (lr=1.5589e-04) (hash(x)=144583183)
|
| 391 |
+
3230 train 4.511980 (lr=1.5559e-04) (hash(x)=165547680)
|
| 392 |
+
3240 train 4.799246 (lr=1.5530e-04) (hash(x)=152195267)
|
| 393 |
+
3250 train 4.570601 (lr=1.5500e-04) (hash(x)=150040234)
|
| 394 |
+
3260 train 4.541203 (lr=1.5470e-04) (hash(x)=143788229)
|
| 395 |
+
3270 train 4.429381 (lr=1.5441e-04) (hash(x)=154472495)
|
| 396 |
+
3280 train 4.449681 (lr=1.5411e-04) (hash(x)=162620947)
|
| 397 |
+
3290 train 4.551185 (lr=1.5381e-04) (hash(x)=157731379)
|
| 398 |
+
3300 val loss 4.5519
|
| 399 |
+
3300 val perplexity 94.8171
|
| 400 |
+
3300 train 4.407149 (lr=1.5351e-04) (hash(x)=149681831)
|
| 401 |
+
3310 train 4.482942 (lr=1.5321e-04) (hash(x)=147216223)
|
| 402 |
+
3320 train 4.418862 (lr=1.5291e-04) (hash(x)=146478959)
|
| 403 |
+
3330 train 4.525304 (lr=1.5261e-04) (hash(x)=161840239)
|
| 404 |
+
3340 train 4.477151 (lr=1.5230e-04) (hash(x)=141748686)
|
| 405 |
+
3350 train 4.328275 (lr=1.5200e-04) (hash(x)=134782096)
|
| 406 |
+
3360 train 4.535439 (lr=1.5170e-04) (hash(x)=164103503)
|
| 407 |
+
3370 train 4.365314 (lr=1.5139e-04) (hash(x)=144658716)
|
| 408 |
+
3380 train 4.291780 (lr=1.5109e-04) (hash(x)=144343513)
|
| 409 |
+
3390 train 4.380495 (lr=1.5078e-04) (hash(x)=145365816)
|
| 410 |
+
3400 val loss 4.5421
|
| 411 |
+
3400 val perplexity 93.8914
|
| 412 |
+
3400 train 4.378392 (lr=1.5048e-04) (hash(x)=168425516)
|
| 413 |
+
3410 train 4.355678 (lr=1.5017e-04) (hash(x)=139704073)
|
| 414 |
+
3420 train 4.469814 (lr=1.4986e-04) (hash(x)=156592513)
|
| 415 |
+
3430 train 4.438216 (lr=1.4956e-04) (hash(x)=136773063)
|
| 416 |
+
3440 train 4.467009 (lr=1.4925e-04) (hash(x)=162468494)
|
| 417 |
+
3450 train 4.507737 (lr=1.4894e-04) (hash(x)=143829320)
|
| 418 |
+
3460 train 4.585582 (lr=1.4863e-04) (hash(x)=158562331)
|
| 419 |
+
3470 train 4.573942 (lr=1.4832e-04) (hash(x)=151882487)
|
| 420 |
+
3480 train 4.566612 (lr=1.4801e-04) (hash(x)=150575558)
|
| 421 |
+
3490 train 4.474866 (lr=1.4770e-04) (hash(x)=147407741)
|
| 422 |
+
3500 val loss 4.5098
|
| 423 |
+
3500 val perplexity 90.9036
|
| 424 |
+
3500 train 4.561738 (lr=1.4739e-04) (hash(x)=163104338)
|
| 425 |
+
3510 train 4.487819 (lr=1.4708e-04) (hash(x)=157461421)
|
| 426 |
+
3520 train 4.458289 (lr=1.4676e-04) (hash(x)=151459070)
|
| 427 |
+
3530 train 4.382645 (lr=1.4645e-04) (hash(x)=150777169)
|
| 428 |
+
3540 train 4.522990 (lr=1.4614e-04) (hash(x)=159138904)
|
| 429 |
+
3550 train 4.422854 (lr=1.4582e-04) (hash(x)=157853732)
|
| 430 |
+
3560 train 4.408754 (lr=1.4551e-04) (hash(x)=152552809)
|
| 431 |
+
3570 train 4.455391 (lr=1.4519e-04) (hash(x)=145219770)
|
| 432 |
+
3580 train 4.404461 (lr=1.4488e-04) (hash(x)=147325185)
|
| 433 |
+
3590 train 4.494122 (lr=1.4456e-04) (hash(x)=146746257)
|
| 434 |
+
3600 val loss 4.4976
|
| 435 |
+
3600 val perplexity 89.8053
|
| 436 |
+
3600 train 4.355057 (lr=1.4424e-04) (hash(x)=165109772)
|
| 437 |
+
3610 train 4.306524 (lr=1.4393e-04) (hash(x)=142561694)
|
| 438 |
+
3620 train 4.450736 (lr=1.4361e-04) (hash(x)=153512606)
|
| 439 |
+
3630 train 4.285353 (lr=1.4329e-04) (hash(x)=140485474)
|
| 440 |
+
3640 train 4.412274 (lr=1.4297e-04) (hash(x)=142084056)
|
| 441 |
+
3650 train 4.463349 (lr=1.4265e-04) (hash(x)=143729044)
|
| 442 |
+
3660 train 4.371139 (lr=1.4233e-04) (hash(x)=129606055)
|
| 443 |
+
3670 train 4.415470 (lr=1.4201e-04) (hash(x)=149878209)
|
| 444 |
+
3680 train 4.218594 (lr=1.4169e-04) (hash(x)=147988422)
|
| 445 |
+
3690 train 4.317169 (lr=1.4137e-04) (hash(x)=134245780)
|
| 446 |
+
3700 val loss 4.5100
|
| 447 |
+
3700 val perplexity 90.9244
|
| 448 |
+
3700 train 4.239594 (lr=1.4105e-04) (hash(x)=153420306)
|
| 449 |
+
3710 train 4.412879 (lr=1.4073e-04) (hash(x)=151618359)
|
| 450 |
+
3720 train 4.306239 (lr=1.4041e-04) (hash(x)=146201479)
|
| 451 |
+
3730 train 4.396004 (lr=1.4008e-04) (hash(x)=173571327)
|
| 452 |
+
3740 train 4.501903 (lr=1.3976e-04) (hash(x)=141793500)
|
| 453 |
+
3750 train 4.614050 (lr=1.3944e-04) (hash(x)=184628568)
|
| 454 |
+
3760 train 4.554639 (lr=1.3911e-04) (hash(x)=143308741)
|
| 455 |
+
3770 train 4.664432 (lr=1.3879e-04) (hash(x)=151283512)
|
| 456 |
+
3780 train 4.468656 (lr=1.3846e-04) (hash(x)=143336516)
|
| 457 |
+
3790 train 4.479311 (lr=1.3814e-04) (hash(x)=148124486)
|
| 458 |
+
3800 val loss 4.4751
|
| 459 |
+
3800 val perplexity 87.8002
|
| 460 |
+
3800 train 4.395710 (lr=1.3781e-04) (hash(x)=160168863)
|
| 461 |
+
3810 train 4.421645 (lr=1.3749e-04) (hash(x)=155072251)
|
| 462 |
+
3820 train 4.589968 (lr=1.3716e-04) (hash(x)=153837417)
|
| 463 |
+
3830 train 4.632998 (lr=1.3683e-04) (hash(x)=130032011)
|
| 464 |
+
3840 train 4.363121 (lr=1.3650e-04) (hash(x)=146270962)
|
| 465 |
+
3850 train 4.374743 (lr=1.3618e-04) (hash(x)=146936934)
|
| 466 |
+
3860 train 4.314685 (lr=1.3585e-04) (hash(x)=165785187)
|
| 467 |
+
3870 train 4.498698 (lr=1.3552e-04) (hash(x)=145393482)
|
| 468 |
+
3880 train 4.431012 (lr=1.3519e-04) (hash(x)=148819533)
|
| 469 |
+
3890 train 4.317963 (lr=1.3486e-04) (hash(x)=146870411)
|
| 470 |
+
3900 val loss 4.4652
|
| 471 |
+
3900 val perplexity 86.9403
|
| 472 |
+
3900 train 4.331470 (lr=1.3453e-04) (hash(x)=153906073)
|
| 473 |
+
3910 train 4.421483 (lr=1.3420e-04) (hash(x)=135632500)
|
| 474 |
+
3920 train 4.430678 (lr=1.3387e-04) (hash(x)=147393037)
|
| 475 |
+
3930 train 4.598029 (lr=1.3354e-04) (hash(x)=150847811)
|
| 476 |
+
3940 train 4.379625 (lr=1.3321e-04) (hash(x)=150253123)
|
| 477 |
+
3950 train 4.234768 (lr=1.3288e-04) (hash(x)=162609448)
|
| 478 |
+
3960 train 4.257111 (lr=1.3255e-04) (hash(x)=150901940)
|
| 479 |
+
3970 train 4.287763 (lr=1.3222e-04) (hash(x)=149698197)
|
| 480 |
+
3980 train 4.397413 (lr=1.3188e-04) (hash(x)=156055141)
|
| 481 |
+
3990 train 4.349399 (lr=1.3155e-04) (hash(x)=163154074)
|
| 482 |
+
4000 val loss 4.4588
|
| 483 |
+
4000 val perplexity 86.3877
|
| 484 |
+
4000 train 4.404885 (lr=1.3122e-04) (hash(x)=151055067)
|
| 485 |
+
4010 train 4.362798 (lr=1.3089e-04) (hash(x)=149313232)
|
| 486 |
+
4020 train 4.389666 (lr=1.3055e-04) (hash(x)=144350585)
|
| 487 |
+
4030 train 4.399717 (lr=1.3022e-04) (hash(x)=143139724)
|
| 488 |
+
4040 train 4.454997 (lr=1.2988e-04) (hash(x)=137534879)
|
| 489 |
+
4050 train 4.345033 (lr=1.2955e-04) (hash(x)=147037092)
|
| 490 |
+
4060 train 4.379547 (lr=1.2921e-04) (hash(x)=149915654)
|
| 491 |
+
4070 train 4.331511 (lr=1.2888e-04) (hash(x)=148002990)
|
| 492 |
+
4080 train 4.320145 (lr=1.2854e-04) (hash(x)=141860091)
|
| 493 |
+
4090 train 4.467549 (lr=1.2821e-04) (hash(x)=152510717)
|
| 494 |
+
4100 val loss 4.4448
|
| 495 |
+
4100 val perplexity 85.1800
|
| 496 |
+
4100 train 4.350614 (lr=1.2787e-04) (hash(x)=149629830)
|
| 497 |
+
4110 train 4.328860 (lr=1.2754e-04) (hash(x)=143677674)
|
| 498 |
+
4120 train 4.449202 (lr=1.2720e-04) (hash(x)=143870235)
|
| 499 |
+
4130 train 4.448778 (lr=1.2686e-04) (hash(x)=156172240)
|
| 500 |
+
4140 train 4.270171 (lr=1.2653e-04) (hash(x)=142768662)
|
| 501 |
+
4150 train 4.329774 (lr=1.2619e-04) (hash(x)=151256841)
|
| 502 |
+
4160 train 4.395031 (lr=1.2585e-04) (hash(x)=144941747)
|
| 503 |
+
4170 train 4.300069 (lr=1.2552e-04) (hash(x)=143588738)
|
| 504 |
+
4180 train 4.270582 (lr=1.2518e-04) (hash(x)=145117633)
|
| 505 |
+
4190 train 4.438657 (lr=1.2484e-04) (hash(x)=158244222)
|
| 506 |
+
4200 val loss 4.4319
|
| 507 |
+
4200 val perplexity 84.0924
|
| 508 |
+
4200 train 4.334982 (lr=1.2450e-04) (hash(x)=143101381)
|
| 509 |
+
4210 train 4.353006 (lr=1.2416e-04) (hash(x)=148704069)
|
| 510 |
+
4220 train 4.267306 (lr=1.2383e-04) (hash(x)=165334845)
|
| 511 |
+
4230 train 4.326064 (lr=1.2349e-04) (hash(x)=147150527)
|
| 512 |
+
4240 train 4.302474 (lr=1.2315e-04) (hash(x)=150118606)
|
| 513 |
+
4250 train 4.259768 (lr=1.2281e-04) (hash(x)=142159597)
|
| 514 |
+
4260 train 4.305237 (lr=1.2247e-04) (hash(x)=152185092)
|
| 515 |
+
4270 train 4.301809 (lr=1.2213e-04) (hash(x)=148870173)
|
| 516 |
+
4280 train 4.332313 (lr=1.2179e-04) (hash(x)=150977664)
|
| 517 |
+
4290 train 4.405171 (lr=1.2145e-04) (hash(x)=165032394)
|
| 518 |
+
4300 val loss 4.4278
|
| 519 |
+
4300 val perplexity 83.7462
|
| 520 |
+
4300 train 4.408392 (lr=1.2111e-04) (hash(x)=149712044)
|
| 521 |
+
4310 train 4.330970 (lr=1.2077e-04) (hash(x)=151956500)
|
| 522 |
+
4320 train 4.490977 (lr=1.2043e-04) (hash(x)=136107484)
|
| 523 |
+
4330 train 4.580441 (lr=1.2009e-04) (hash(x)=160783980)
|
| 524 |
+
4340 train 4.475519 (lr=1.1975e-04) (hash(x)=151772228)
|
| 525 |
+
4350 train 4.319439 (lr=1.1941e-04) (hash(x)=176468048)
|
| 526 |
+
4360 train 4.357060 (lr=1.1907e-04) (hash(x)=144588632)
|
| 527 |
+
4370 train 4.396204 (lr=1.1873e-04) (hash(x)=156374655)
|
| 528 |
+
4380 train 4.490053 (lr=1.1838e-04) (hash(x)=148502191)
|
| 529 |
+
4390 train 4.569447 (lr=1.1804e-04) (hash(x)=152391280)
|
| 530 |
+
4400 val loss 4.4121
|
| 531 |
+
4400 val perplexity 82.4455
|
| 532 |
+
4400 train 4.318466 (lr=1.1770e-04) (hash(x)=153446449)
|
| 533 |
+
4410 train 4.421370 (lr=1.1736e-04) (hash(x)=167668129)
|
| 534 |
+
4420 train 4.328272 (lr=1.1702e-04) (hash(x)=148598803)
|
| 535 |
+
4430 train 4.412292 (lr=1.1668e-04) (hash(x)=161720292)
|
| 536 |
+
4440 train 4.229603 (lr=1.1634e-04) (hash(x)=144133776)
|
| 537 |
+
4450 train 4.337811 (lr=1.1599e-04) (hash(x)=151057987)
|
| 538 |
+
4460 train 4.375474 (lr=1.1565e-04) (hash(x)=151971972)
|
| 539 |
+
4470 train 4.205178 (lr=1.1531e-04) (hash(x)=140885903)
|
| 540 |
+
4480 train 4.239341 (lr=1.1497e-04) (hash(x)=138198700)
|
| 541 |
+
4490 train 4.344744 (lr=1.1462e-04) (hash(x)=150424198)
|
| 542 |
+
4500 val loss 4.4007
|
| 543 |
+
4500 val perplexity 81.5112
|
| 544 |
+
4500 train 4.339964 (lr=1.1428e-04) (hash(x)=146086947)
|
| 545 |
+
4510 train 4.321537 (lr=1.1394e-04) (hash(x)=145752336)
|
| 546 |
+
4520 train 4.274489 (lr=1.1360e-04) (hash(x)=145005726)
|
| 547 |
+
4530 train 4.285138 (lr=1.1326e-04) (hash(x)=144116626)
|
| 548 |
+
4540 train 4.347938 (lr=1.1291e-04) (hash(x)=164507978)
|
| 549 |
+
4550 train 4.218055 (lr=1.1257e-04) (hash(x)=154512699)
|
| 550 |
+
4560 train 4.311101 (lr=1.1223e-04) (hash(x)=149920787)
|
| 551 |
+
4570 train 4.240500 (lr=1.1188e-04) (hash(x)=155720526)
|
| 552 |
+
4580 train 4.296546 (lr=1.1154e-04) (hash(x)=149918053)
|
| 553 |
+
4590 train 4.437987 (lr=1.1120e-04) (hash(x)=170567219)
|
| 554 |
+
4600 val loss 4.4005
|
| 555 |
+
4600 val perplexity 81.4925
|
| 556 |
+
4600 train 4.271751 (lr=1.1086e-04) (hash(x)=153800173)
|
| 557 |
+
4610 train 4.264026 (lr=1.1051e-04) (hash(x)=151762309)
|
| 558 |
+
4620 train 4.317380 (lr=1.1017e-04) (hash(x)=138620238)
|
| 559 |
+
4630 train 4.274055 (lr=1.0983e-04) (hash(x)=155847676)
|
| 560 |
+
4640 train 4.402434 (lr=1.0949e-04) (hash(x)=159145173)
|
| 561 |
+
4650 train 4.450144 (lr=1.0914e-04) (hash(x)=163662103)
|
| 562 |
+
4660 train 4.332089 (lr=1.0880e-04) (hash(x)=168862572)
|
| 563 |
+
4670 train 5.453598 (lr=1.0846e-04) (hash(x)=128395338)
|
| 564 |
+
4680 train 4.256945 (lr=1.0812e-04) (hash(x)=149975049)
|
| 565 |
+
4690 train 4.284852 (lr=1.0777e-04) (hash(x)=156604294)
|
| 566 |
+
4700 val loss 4.3840
|
| 567 |
+
4700 val perplexity 80.1575
|
| 568 |
+
4700 train 4.545529 (lr=1.0743e-04) (hash(x)=155962726)
|
| 569 |
+
4710 train 4.285121 (lr=1.0709e-04) (hash(x)=160931268)
|
| 570 |
+
4720 train 4.350848 (lr=1.0674e-04) (hash(x)=173485074)
|
| 571 |
+
4730 train 4.467741 (lr=1.0640e-04) (hash(x)=146001945)
|
| 572 |
+
4740 train 4.403823 (lr=1.0606e-04) (hash(x)=148305446)
|
| 573 |
+
4750 train 4.310658 (lr=1.0572e-04) (hash(x)=131655842)
|
| 574 |
+
4760 train 4.281639 (lr=1.0538e-04) (hash(x)=145775165)
|
| 575 |
+
4770 train 4.207619 (lr=1.0503e-04) (hash(x)=146613282)
|
| 576 |
+
4780 train 4.221068 (lr=1.0469e-04) (hash(x)=137882160)
|
| 577 |
+
4790 train 4.386369 (lr=1.0435e-04) (hash(x)=157177531)
|
| 578 |
+
4800 val loss 4.3777
|
| 579 |
+
4800 val perplexity 79.6539
|
| 580 |
+
4800 train 4.684868 (lr=1.0401e-04) (hash(x)=142045616)
|
| 581 |
+
4810 train 4.534530 (lr=1.0366e-04) (hash(x)=153739117)
|
| 582 |
+
4820 train 4.184145 (lr=1.0332e-04) (hash(x)=159336444)
|
| 583 |
+
4830 train 4.364075 (lr=1.0298e-04) (hash(x)=162617526)
|
| 584 |
+
4840 train 4.389979 (lr=1.0264e-04) (hash(x)=141336820)
|
| 585 |
+
4850 train 4.327240 (lr=1.0230e-04) (hash(x)=146718291)
|
| 586 |
+
4860 train 4.286326 (lr=1.0196e-04) (hash(x)=159542679)
|
| 587 |
+
4870 train 4.283735 (lr=1.0162e-04) (hash(x)=149318029)
|
| 588 |
+
4880 train 4.285145 (lr=1.0127e-04) (hash(x)=150182605)
|
| 589 |
+
4890 train 4.264644 (lr=1.0093e-04) (hash(x)=139657075)
|
| 590 |
+
4900 val loss 4.3670
|
| 591 |
+
4900 val perplexity 78.8095
|
| 592 |
+
4900 train 4.389112 (lr=1.0059e-04) (hash(x)=143418248)
|
| 593 |
+
4910 train 4.450047 (lr=1.0025e-04) (hash(x)=153395021)
|
| 594 |
+
4920 train 4.359869 (lr=9.9911e-05) (hash(x)=144977141)
|
| 595 |
+
4930 train 4.373468 (lr=9.9571e-05) (hash(x)=143569016)
|
| 596 |
+
4940 train 4.347807 (lr=9.9230e-05) (hash(x)=145702219)
|
| 597 |
+
4950 train 4.327699 (lr=9.8890e-05) (hash(x)=150669399)
|
| 598 |
+
4960 train 4.319583 (lr=9.8550e-05) (hash(x)=147697301)
|
| 599 |
+
4970 train 4.236506 (lr=9.8210e-05) (hash(x)=158617139)
|
| 600 |
+
4980 train 4.361155 (lr=9.7870e-05) (hash(x)=158937457)
|
| 601 |
+
4990 train 4.248951 (lr=9.7531e-05) (hash(x)=155540364)
|
| 602 |
+
5000 val loss 4.3620
|
| 603 |
+
5000 val perplexity 78.4154
|
| 604 |
+
5000 train 4.290029 (lr=9.7192e-05) (hash(x)=145789790)
|
| 605 |
+
5010 train 4.310887 (lr=9.6853e-05) (hash(x)=142919148)
|
| 606 |
+
5020 train 4.207727 (lr=9.6514e-05) (hash(x)=149129606)
|
| 607 |
+
5030 train 4.447894 (lr=9.6175e-05) (hash(x)=143060789)
|
| 608 |
+
5040 train 4.271350 (lr=9.5836e-05) (hash(x)=145733956)
|
| 609 |
+
5050 train 4.081942 (lr=9.5498e-05) (hash(x)=139750793)
|
| 610 |
+
5060 train 4.366995 (lr=9.5160e-05) (hash(x)=148027808)
|
| 611 |
+
5070 train 4.199711 (lr=9.4822e-05) (hash(x)=149119770)
|
| 612 |
+
5080 train 4.235122 (lr=9.4484e-05) (hash(x)=145741944)
|
| 613 |
+
5090 train 4.305195 (lr=9.4147e-05) (hash(x)=144783744)
|
| 614 |
+
5100 val loss 4.3599
|
| 615 |
+
5100 val perplexity 78.2470
|
| 616 |
+
5100 train 4.330678 (lr=9.3809e-05) (hash(x)=137795633)
|
| 617 |
+
5110 train 4.231199 (lr=9.3472e-05) (hash(x)=145995538)
|
| 618 |
+
5120 train 4.238341 (lr=9.3136e-05) (hash(x)=143019933)
|
| 619 |
+
5130 train 4.188108 (lr=9.2799e-05) (hash(x)=143746069)
|
| 620 |
+
5140 train 4.007740 (lr=9.2463e-05) (hash(x)=152212469)
|
| 621 |
+
5150 train 4.306718 (lr=9.2127e-05) (hash(x)=152608669)
|
| 622 |
+
5160 train 4.335310 (lr=9.1791e-05) (hash(x)=150750381)
|
| 623 |
+
5170 train 4.284182 (lr=9.1456e-05) (hash(x)=133198485)
|
| 624 |
+
5180 train 4.347716 (lr=9.1120e-05) (hash(x)=142233420)
|
| 625 |
+
5190 train 4.220807 (lr=9.0785e-05) (hash(x)=147366941)
|
| 626 |
+
5200 val loss 4.3483
|
| 627 |
+
5200 val perplexity 77.3500
|
| 628 |
+
5200 train 4.438392 (lr=9.0451e-05) (hash(x)=148907132)
|
| 629 |
+
5210 train 4.474904 (lr=9.0116e-05) (hash(x)=154827138)
|
| 630 |
+
5220 train 4.469321 (lr=8.9782e-05) (hash(x)=148717408)
|
| 631 |
+
5230 train 4.635363 (lr=8.9448e-05) (hash(x)=147711017)
|
| 632 |
+
5240 train 4.292656 (lr=8.9115e-05) (hash(x)=141796353)
|
| 633 |
+
5250 train 4.212835 (lr=8.8782e-05) (hash(x)=143529037)
|
| 634 |
+
5260 train 4.232874 (lr=8.8449e-05) (hash(x)=144773706)
|
| 635 |
+
5270 train 4.271386 (lr=8.8116e-05) (hash(x)=148894329)
|
| 636 |
+
5280 train 4.239782 (lr=8.7784e-05) (hash(x)=153761544)
|
| 637 |
+
5290 train 4.276737 (lr=8.7452e-05) (hash(x)=150927827)
|
| 638 |
+
5300 val loss 4.3494
|
| 639 |
+
5300 val perplexity 77.4342
|
| 640 |
+
5300 train 4.179814 (lr=8.7120e-05) (hash(x)=152343580)
|
| 641 |
+
5310 train 4.104673 (lr=8.6789e-05) (hash(x)=146690249)
|
| 642 |
+
5320 train 4.268581 (lr=8.6458e-05) (hash(x)=155991065)
|
| 643 |
+
5330 train 4.260195 (lr=8.6127e-05) (hash(x)=151437785)
|
| 644 |
+
5340 train 4.208722 (lr=8.5797e-05) (hash(x)=158840015)
|
| 645 |
+
5350 train 4.181820 (lr=8.5467e-05) (hash(x)=155080378)
|
| 646 |
+
5360 train 4.222994 (lr=8.5138e-05) (hash(x)=150513270)
|
| 647 |
+
5370 train 4.136909 (lr=8.4809e-05) (hash(x)=147729861)
|
| 648 |
+
5380 train 4.156466 (lr=8.4480e-05) (hash(x)=149870080)
|
| 649 |
+
5390 train 3.989431 (lr=8.4151e-05) (hash(x)=137577816)
|
| 650 |
+
5400 val loss 4.3387
|
| 651 |
+
5400 val perplexity 76.6114
|
| 652 |
+
5400 train 4.194353 (lr=8.3823e-05) (hash(x)=148578264)
|
| 653 |
+
5410 train 4.293601 (lr=8.3495e-05) (hash(x)=149338196)
|
| 654 |
+
5420 train 4.192098 (lr=8.3168e-05) (hash(x)=149581125)
|
| 655 |
+
5430 train 4.240108 (lr=8.2841e-05) (hash(x)=146340931)
|
| 656 |
+
5440 train 4.126627 (lr=8.2515e-05) (hash(x)=127884657)
|
| 657 |
+
5450 train 4.290208 (lr=8.2188e-05) (hash(x)=142632516)
|
| 658 |
+
5460 train 4.279861 (lr=8.1863e-05) (hash(x)=146519487)
|
| 659 |
+
5470 train 4.376921 (lr=8.1537e-05) (hash(x)=136409211)
|
| 660 |
+
5480 train 4.303849 (lr=8.1212e-05) (hash(x)=158622762)
|
| 661 |
+
5490 train 4.242623 (lr=8.0888e-05) (hash(x)=154703333)
|
| 662 |
+
5500 val loss 4.3254
|
| 663 |
+
5500 val perplexity 75.5936
|
| 664 |
+
5500 train 4.389395 (lr=8.0564e-05) (hash(x)=145635833)
|
| 665 |
+
5510 train 4.356792 (lr=8.0240e-05) (hash(x)=158781769)
|
| 666 |
+
5520 train 4.333563 (lr=7.9917e-05) (hash(x)=142550540)
|
| 667 |
+
5530 train 4.335950 (lr=7.9594e-05) (hash(x)=152024805)
|
| 668 |
+
5540 train 4.157989 (lr=7.9272e-05) (hash(x)=145812646)
|
| 669 |
+
5550 train 4.245994 (lr=7.8950e-05) (hash(x)=144941659)
|
| 670 |
+
5560 train 4.196007 (lr=7.8629e-05) (hash(x)=145887398)
|
| 671 |
+
5570 train 4.203807 (lr=7.8308e-05) (hash(x)=144988642)
|
| 672 |
+
5580 train 4.221129 (lr=7.7987e-05) (hash(x)=146366431)
|
| 673 |
+
5590 train 4.146015 (lr=7.7667e-05) (hash(x)=143301433)
|
| 674 |
+
5600 val loss 4.3223
|
| 675 |
+
5600 val perplexity 75.3610
|
| 676 |
+
5600 train 4.337992 (lr=7.7347e-05) (hash(x)=156337844)
|
| 677 |
+
5610 train 4.295769 (lr=7.7028e-05) (hash(x)=146546252)
|
| 678 |
+
5620 train 4.232712 (lr=7.6710e-05) (hash(x)=137610278)
|
| 679 |
+
5630 train 4.253319 (lr=7.6391e-05) (hash(x)=148847402)
|
| 680 |
+
5640 train 4.232903 (lr=7.6074e-05) (hash(x)=148805052)
|
| 681 |
+
5650 train 4.130242 (lr=7.5756e-05) (hash(x)=144882220)
|
| 682 |
+
5660 train 4.211919 (lr=7.5440e-05) (hash(x)=147894128)
|
| 683 |
+
5670 train 4.132584 (lr=7.5124e-05) (hash(x)=144855036)
|
| 684 |
+
5680 train 4.114679 (lr=7.4808e-05) (hash(x)=148400728)
|
| 685 |
+
5690 train 4.098059 (lr=7.4493e-05) (hash(x)=138377186)
|
| 686 |
+
5700 val loss 4.3223
|
| 687 |
+
5700 val perplexity 75.3652
|
| 688 |
+
5700 train 4.123394 (lr=7.4178e-05) (hash(x)=147168506)
|
| 689 |
+
5710 train 4.253973 (lr=7.3864e-05) (hash(x)=150486217)
|
| 690 |
+
5720 train 4.223296 (lr=7.3550e-05) (hash(x)=148658778)
|
| 691 |
+
5730 train 4.029077 (lr=7.3237e-05) (hash(x)=148402206)
|
| 692 |
+
5740 train 4.077699 (lr=7.2925e-05) (hash(x)=144432268)
|
| 693 |
+
5750 train 4.445924 (lr=7.2613e-05) (hash(x)=153770308)
|
| 694 |
+
5760 train 4.198049 (lr=7.2301e-05) (hash(x)=150546820)
|
| 695 |
+
5770 train 4.375805 (lr=7.1990e-05) (hash(x)=140176711)
|
| 696 |
+
5780 train 4.276854 (lr=7.1680e-05) (hash(x)=164276928)
|
| 697 |
+
5790 train 4.181480 (lr=7.1370e-05) (hash(x)=153418756)
|
| 698 |
+
5800 val loss 4.3105
|
| 699 |
+
5800 val perplexity 74.4792
|
| 700 |
+
5800 train 4.278124 (lr=7.1061e-05) (hash(x)=159566920)
|
| 701 |
+
5810 train 4.214949 (lr=7.0752e-05) (hash(x)=161266875)
|
| 702 |
+
5820 train 4.239595 (lr=7.0444e-05) (hash(x)=156298754)
|
| 703 |
+
5830 train 4.367199 (lr=7.0136e-05) (hash(x)=139302200)
|
| 704 |
+
5840 train 4.244771 (lr=6.9829e-05) (hash(x)=126108545)
|
| 705 |
+
5850 train 4.161056 (lr=6.9523e-05) (hash(x)=147526120)
|
| 706 |
+
5860 train 4.259589 (lr=6.9217e-05) (hash(x)=150242466)
|
| 707 |
+
5870 train 4.215409 (lr=6.8912e-05) (hash(x)=139851741)
|
| 708 |
+
5880 train 4.144442 (lr=6.8607e-05) (hash(x)=159464779)
|
| 709 |
+
5890 train 4.295774 (lr=6.8303e-05) (hash(x)=148069134)
|
| 710 |
+
5900 val loss 4.3055
|
| 711 |
+
5900 val perplexity 74.1063
|
| 712 |
+
5900 train 4.296760 (lr=6.8000e-05) (hash(x)=158273929)
|
| 713 |
+
5910 train 3.985941 (lr=6.7697e-05) (hash(x)=150013727)
|
| 714 |
+
5920 train 4.260183 (lr=6.7395e-05) (hash(x)=148803117)
|
| 715 |
+
5930 train 4.281883 (lr=6.7093e-05) (hash(x)=156729045)
|
| 716 |
+
5940 train 4.204551 (lr=6.6792e-05) (hash(x)=150475415)
|
| 717 |
+
5950 train 4.083829 (lr=6.6492e-05) (hash(x)=153585864)
|
| 718 |
+
5960 train 4.085408 (lr=6.6192e-05) (hash(x)=145189335)
|
| 719 |
+
5970 train 4.098610 (lr=6.5893e-05) (hash(x)=145673396)
|
| 720 |
+
5980 train 4.238984 (lr=6.5595e-05) (hash(x)=138413986)
|
| 721 |
+
5990 train 4.042863 (lr=6.5297e-05) (hash(x)=140270894)
|
| 722 |
+
6000 val loss 4.3052
|
| 723 |
+
6000 val perplexity 74.0825
|
| 724 |
+
6000 train 4.232789 (lr=6.5000e-05) (hash(x)=156649749)
|
| 725 |
+
6010 train 4.340338 (lr=6.4704e-05) (hash(x)=154762134)
|
| 726 |
+
6020 train 4.333528 (lr=6.4408e-05) (hash(x)=156088385)
|
| 727 |
+
6030 train 4.200102 (lr=6.4113e-05) (hash(x)=152521323)
|
| 728 |
+
6040 train 4.108745 (lr=6.3818e-05) (hash(x)=148376918)
|
| 729 |
+
6050 train 4.282449 (lr=6.3524e-05) (hash(x)=155673243)
|
| 730 |
+
6060 train 4.253149 (lr=6.3231e-05) (hash(x)=142344218)
|
| 731 |
+
6070 train 4.402580 (lr=6.2939e-05) (hash(x)=151751923)
|
| 732 |
+
6080 train 4.202380 (lr=6.2647e-05) (hash(x)=156311576)
|
| 733 |
+
6090 train 4.160943 (lr=6.2356e-05) (hash(x)=145058290)
|
| 734 |
+
6100 val loss 4.2956
|
| 735 |
+
6100 val perplexity 73.3761
|
| 736 |
+
6100 train 4.254804 (lr=6.2065e-05) (hash(x)=146812388)
|
| 737 |
+
6110 train 4.292779 (lr=6.1776e-05) (hash(x)=138440385)
|
| 738 |
+
6120 train 4.299343 (lr=6.1487e-05) (hash(x)=162764713)
|
| 739 |
+
6130 train 4.212253 (lr=6.1198e-05) (hash(x)=156456134)
|
| 740 |
+
6140 train 4.121292 (lr=6.0911e-05) (hash(x)=136402679)
|
| 741 |
+
6150 train 4.134806 (lr=6.0624e-05) (hash(x)=153380357)
|
| 742 |
+
6160 train 3.835257 (lr=6.0338e-05) (hash(x)=170571967)
|
| 743 |
+
6170 train 4.178360 (lr=6.0052e-05) (hash(x)=150984242)
|
| 744 |
+
6180 train 4.276917 (lr=5.9767e-05) (hash(x)=143187168)
|
| 745 |
+
6190 train 4.268665 (lr=5.9483e-05) (hash(x)=142999298)
|
| 746 |
+
6200 val loss 4.2925
|
| 747 |
+
6200 val perplexity 73.1513
|
| 748 |
+
6200 train 4.197761 (lr=5.9200e-05) (hash(x)=143522146)
|
| 749 |
+
6210 train 4.195680 (lr=5.8918e-05) (hash(x)=141961270)
|
| 750 |
+
6220 train 4.231529 (lr=5.8636e-05) (hash(x)=158572673)
|
| 751 |
+
6230 train 4.186938 (lr=5.8355e-05) (hash(x)=142949296)
|
| 752 |
+
6240 train 4.038501 (lr=5.8074e-05) (hash(x)=128564671)
|
| 753 |
+
6250 train 4.132696 (lr=5.7795e-05) (hash(x)=147515208)
|
| 754 |
+
6260 train 4.095105 (lr=5.7516e-05) (hash(x)=151825788)
|
| 755 |
+
6270 train 4.059223 (lr=5.7238e-05) (hash(x)=136625971)
|
| 756 |
+
6280 train 4.124689 (lr=5.6961e-05) (hash(x)=146904276)
|
| 757 |
+
6290 train 4.022074 (lr=5.6684e-05) (hash(x)=151876945)
|
| 758 |
+
6300 val loss 4.2913
|
| 759 |
+
6300 val perplexity 73.0605
|
| 760 |
+
6300 train 4.076968 (lr=5.6409e-05) (hash(x)=150124474)
|
| 761 |
+
6310 train 4.032629 (lr=5.6134e-05) (hash(x)=139897420)
|
| 762 |
+
6320 train 4.172024 (lr=5.5859e-05) (hash(x)=136670912)
|
| 763 |
+
6330 train 4.057052 (lr=5.5586e-05) (hash(x)=140302490)
|
| 764 |
+
6340 train 4.001121 (lr=5.5313e-05) (hash(x)=145564419)
|
| 765 |
+
6350 train 4.359475 (lr=5.5042e-05) (hash(x)=164988308)
|
| 766 |
+
6360 train 4.302389 (lr=5.4771e-05) (hash(x)=139721549)
|
| 767 |
+
6370 train 4.256631 (lr=5.4500e-05) (hash(x)=154183530)
|
| 768 |
+
6380 train 4.250035 (lr=5.4231e-05) (hash(x)=143895743)
|
| 769 |
+
6390 train 4.287355 (lr=5.3963e-05) (hash(x)=153091571)
|
| 770 |
+
6400 val loss 4.2858
|
| 771 |
+
6400 val perplexity 72.6622
|
| 772 |
+
6400 train 4.259851 (lr=5.3695e-05) (hash(x)=141242117)
|
| 773 |
+
6410 train 4.330456 (lr=5.3428e-05) (hash(x)=146638943)
|
| 774 |
+
6420 train 4.229600 (lr=5.3162e-05) (hash(x)=147429468)
|
| 775 |
+
6430 train 4.323775 (lr=5.2896e-05) (hash(x)=147104391)
|
| 776 |
+
6440 train 4.183428 (lr=5.2632e-05) (hash(x)=147003825)
|
| 777 |
+
6450 train 4.139755 (lr=5.2368e-05) (hash(x)=150910430)
|
| 778 |
+
6460 train 4.083426 (lr=5.2105e-05) (hash(x)=165853376)
|
| 779 |
+
6470 train 4.248239 (lr=5.1843e-05) (hash(x)=145957630)
|
| 780 |
+
6480 train 4.150528 (lr=5.1582e-05) (hash(x)=155685880)
|
| 781 |
+
6490 train 4.393590 (lr=5.1322e-05) (hash(x)=150307407)
|
| 782 |
+
6500 val loss 4.2762
|
| 783 |
+
6500 val perplexity 71.9664
|
| 784 |
+
6500 train 4.068428 (lr=5.1063e-05) (hash(x)=143529762)
|
| 785 |
+
6510 train 4.306029 (lr=5.0804e-05) (hash(x)=145652949)
|
| 786 |
+
6520 train 4.180875 (lr=5.0546e-05) (hash(x)=142033417)
|
| 787 |
+
6530 train 4.187957 (lr=5.0289e-05) (hash(x)=131869635)
|
| 788 |
+
6540 train 4.245570 (lr=5.0033e-05) (hash(x)=151253771)
|
| 789 |
+
6550 train 4.302132 (lr=4.9778e-05) (hash(x)=164286071)
|
| 790 |
+
6560 train 4.174497 (lr=4.9524e-05) (hash(x)=148872674)
|
| 791 |
+
6570 train 4.174606 (lr=4.9271e-05) (hash(x)=147653143)
|
| 792 |
+
6580 train 4.181898 (lr=4.9018e-05) (hash(x)=151684539)
|
| 793 |
+
6590 train 4.184108 (lr=4.8767e-05) (hash(x)=143354218)
|
| 794 |
+
6600 val loss 4.2730
|
| 795 |
+
6600 val perplexity 71.7388
|
| 796 |
+
6600 train 4.164752 (lr=4.8516e-05) (hash(x)=136948374)
|
| 797 |
+
6610 train 4.156300 (lr=4.8266e-05) (hash(x)=149846260)
|
| 798 |
+
6620 train 4.062312 (lr=4.8017e-05) (hash(x)=151615259)
|
| 799 |
+
6630 train 4.015197 (lr=4.7769e-05) (hash(x)=153129395)
|
| 800 |
+
6640 train 4.191104 (lr=4.7522e-05) (hash(x)=133054060)
|
| 801 |
+
6650 train 4.234651 (lr=4.7276e-05) (hash(x)=149911476)
|
| 802 |
+
6660 train 4.186780 (lr=4.7030e-05) (hash(x)=150548384)
|
| 803 |
+
6670 train 4.059508 (lr=4.6786e-05) (hash(x)=154950888)
|
| 804 |
+
6680 train 3.922714 (lr=4.6542e-05) (hash(x)=145532409)
|
| 805 |
+
6690 train 4.125059 (lr=4.6300e-05) (hash(x)=151197362)
|
| 806 |
+
6700 val loss 4.2750
|
| 807 |
+
6700 val perplexity 71.8817
|
| 808 |
+
6700 train 4.037487 (lr=4.6058e-05) (hash(x)=146268592)
|
| 809 |
+
6710 train 4.020319 (lr=4.5817e-05) (hash(x)=154495757)
|
| 810 |
+
6720 train 4.106662 (lr=4.5578e-05) (hash(x)=168472165)
|
| 811 |
+
6730 train 4.089464 (lr=4.5339e-05) (hash(x)=150287847)
|
| 812 |
+
6740 train 4.052947 (lr=4.5101e-05) (hash(x)=151465011)
|
| 813 |
+
6750 train 4.008319 (lr=4.4864e-05) (hash(x)=151898013)
|
| 814 |
+
6760 train 4.138402 (lr=4.4628e-05) (hash(x)=145836413)
|
| 815 |
+
6770 train 4.385863 (lr=4.4393e-05) (hash(x)=146057728)
|
| 816 |
+
6780 train 4.278585 (lr=4.4159e-05) (hash(x)=160113674)
|
| 817 |
+
6790 train 4.285954 (lr=4.3926e-05) (hash(x)=149844509)
|
| 818 |
+
6800 val loss 4.2641
|
| 819 |
+
6800 val perplexity 71.1025
|
| 820 |
+
6800 train 4.237763 (lr=4.3693e-05) (hash(x)=152676836)
|
| 821 |
+
6810 train 4.311237 (lr=4.3462e-05) (hash(x)=147510869)
|
| 822 |
+
6820 train 4.391291 (lr=4.3232e-05) (hash(x)=149778657)
|
| 823 |
+
6830 train 4.328356 (lr=4.3002e-05) (hash(x)=143851885)
|
| 824 |
+
6840 train 4.334677 (lr=4.2774e-05) (hash(x)=158011137)
|
| 825 |
+
6850 train 4.396527 (lr=4.2547e-05) (hash(x)=171197810)
|
| 826 |
+
6860 train 4.314925 (lr=4.2320e-05) (hash(x)=151171060)
|
| 827 |
+
6870 train 4.233796 (lr=4.2095e-05) (hash(x)=135613468)
|
| 828 |
+
6880 train 4.274740 (lr=4.1870e-05) (hash(x)=153789364)
|
| 829 |
+
6890 train 4.211565 (lr=4.1647e-05) (hash(x)=128928397)
|
| 830 |
+
6900 val loss 4.2560
|
| 831 |
+
6900 val perplexity 70.5295
|
| 832 |
+
6900 train 4.260047 (lr=4.1425e-05) (hash(x)=134657776)
|
| 833 |
+
6910 train 4.239233 (lr=4.1203e-05) (hash(x)=170095944)
|
| 834 |
+
6920 train 4.466892 (lr=4.0983e-05) (hash(x)=147408535)
|
| 835 |
+
6930 train 4.247075 (lr=4.0763e-05) (hash(x)=141187674)
|
| 836 |
+
6940 train 4.140050 (lr=4.0545e-05) (hash(x)=157752429)
|
| 837 |
+
6950 train 4.323330 (lr=4.0327e-05) (hash(x)=149793159)
|
| 838 |
+
6960 train 4.202215 (lr=4.0111e-05) (hash(x)=135786360)
|
| 839 |
+
6970 train 4.102632 (lr=3.9895e-05) (hash(x)=144372214)
|
| 840 |
+
6980 train 4.222736 (lr=3.9681e-05) (hash(x)=147592321)
|
| 841 |
+
6990 train 4.150537 (lr=3.9468e-05) (hash(x)=135072395)
|
| 842 |
+
7000 val loss 4.2569
|
| 843 |
+
7000 val perplexity 70.5928
|
| 844 |
+
7000 train 4.153344 (lr=3.9255e-05) (hash(x)=166721861)
|
| 845 |
+
7010 train 4.232583 (lr=3.9044e-05) (hash(x)=157177130)
|
| 846 |
+
7020 train 4.124187 (lr=3.8834e-05) (hash(x)=154916555)
|
| 847 |
+
7030 train 4.229033 (lr=3.8624e-05) (hash(x)=139686863)
|
| 848 |
+
7040 train 4.204591 (lr=3.8416e-05) (hash(x)=139606744)
|
| 849 |
+
7050 train 4.262869 (lr=3.8209e-05) (hash(x)=149487898)
|
| 850 |
+
7060 train 4.156590 (lr=3.8003e-05) (hash(x)=162536366)
|
| 851 |
+
7070 train 4.202412 (lr=3.7798e-05) (hash(x)=144765016)
|
| 852 |
+
7080 train 4.181116 (lr=3.7593e-05) (hash(x)=146001679)
|
| 853 |
+
7090 train 3.980720 (lr=3.7390e-05) (hash(x)=153352255)
|
| 854 |
+
7100 val loss 4.2565
|
| 855 |
+
7100 val perplexity 70.5629
|
| 856 |
+
7100 train 4.012289 (lr=3.7188e-05) (hash(x)=135496702)
|
| 857 |
+
7110 train 4.198144 (lr=3.6988e-05) (hash(x)=155057574)
|
| 858 |
+
7120 train 4.158841 (lr=3.6788e-05) (hash(x)=145862613)
|
| 859 |
+
7130 train 3.870955 (lr=3.6589e-05) (hash(x)=160806933)
|
| 860 |
+
7140 train 3.940708 (lr=3.6391e-05) (hash(x)=154582601)
|
| 861 |
+
7150 train 4.024274 (lr=3.6195e-05) (hash(x)=151854329)
|
| 862 |
+
7160 train 4.187387 (lr=3.5999e-05) (hash(x)=153875998)
|
| 863 |
+
7170 train 3.981042 (lr=3.5804e-05) (hash(x)=143448354)
|
| 864 |
+
7180 train 4.014258 (lr=3.5611e-05) (hash(x)=141619703)
|
| 865 |
+
7190 train 4.273322 (lr=3.5419e-05) (hash(x)=154769375)
|
| 866 |
+
7200 val loss 4.2523
|
| 867 |
+
7200 val perplexity 70.2640
|
| 868 |
+
7200 train 4.386336 (lr=3.5227e-05) (hash(x)=155567461)
|
| 869 |
+
7210 train 4.323836 (lr=3.5037e-05) (hash(x)=160242796)
|
| 870 |
+
7220 train 4.349945 (lr=3.4848e-05) (hash(x)=146362204)
|
| 871 |
+
7230 train 4.097724 (lr=3.4660e-05) (hash(x)=155673641)
|
| 872 |
+
7240 train 4.309337 (lr=3.4473e-05) (hash(x)=163454733)
|
| 873 |
+
7250 train 4.316981 (lr=3.4287e-05) (hash(x)=166254410)
|
| 874 |
+
7260 train 4.163548 (lr=3.4102e-05) (hash(x)=149576049)
|
| 875 |
+
7270 train 4.218158 (lr=3.3919e-05) (hash(x)=159937867)
|
| 876 |
+
7280 train 4.347117 (lr=3.3736e-05) (hash(x)=151335403)
|
| 877 |
+
7290 train 4.194914 (lr=3.3555e-05) (hash(x)=148867536)
|
| 878 |
+
7300 val loss 4.2437
|
| 879 |
+
7300 val perplexity 69.6651
|
| 880 |
+
7300 train 4.131832 (lr=3.3375e-05) (hash(x)=142803829)
|
| 881 |
+
7310 train 4.289918 (lr=3.3195e-05) (hash(x)=149482208)
|
| 882 |
+
7320 train 4.059159 (lr=3.3017e-05) (hash(x)=144226264)
|
| 883 |
+
7330 train 4.176286 (lr=3.2840e-05) (hash(x)=189898865)
|
| 884 |
+
7340 train 4.286035 (lr=3.2664e-05) (hash(x)=153884999)
|
| 885 |
+
7350 train 4.199429 (lr=3.2490e-05) (hash(x)=149106483)
|
| 886 |
+
7360 train 4.230082 (lr=3.2316e-05) (hash(x)=146033939)
|
| 887 |
+
7370 train 4.208216 (lr=3.2144e-05) (hash(x)=133850095)
|
| 888 |
+
7380 train 4.225704 (lr=3.1972e-05) (hash(x)=155008783)
|
| 889 |
+
7390 train 4.232209 (lr=3.1802e-05) (hash(x)=158169816)
|
| 890 |
+
7400 val loss 4.2436
|
| 891 |
+
7400 val perplexity 69.6573
|
| 892 |
+
7400 train 4.147046 (lr=3.1633e-05) (hash(x)=145294178)
|
| 893 |
+
7410 train 4.102555 (lr=3.1465e-05) (hash(x)=156501995)
|
| 894 |
+
7420 train 4.185500 (lr=3.1298e-05) (hash(x)=154883953)
|
| 895 |
+
7430 train 4.216641 (lr=3.1132e-05) (hash(x)=153861637)
|
| 896 |
+
7440 train 4.138914 (lr=3.0968e-05) (hash(x)=143884732)
|
| 897 |
+
7450 train 4.271674 (lr=3.0804e-05) (hash(x)=147629077)
|
| 898 |
+
7460 train 4.162514 (lr=3.0642e-05) (hash(x)=147092200)
|
| 899 |
+
7470 train 4.034914 (lr=3.0481e-05) (hash(x)=153667163)
|
| 900 |
+
7480 train 4.097279 (lr=3.0321e-05) (hash(x)=155400781)
|
| 901 |
+
7490 train 4.343603 (lr=3.0162e-05) (hash(x)=158198995)
|
| 902 |
+
7500 val loss 4.2448
|
| 903 |
+
7500 val perplexity 69.7388
|
| 904 |
+
7500 train 4.042630 (lr=3.0005e-05) (hash(x)=150573713)
|
| 905 |
+
7510 train 4.132552 (lr=2.9848e-05) (hash(x)=159059956)
|
| 906 |
+
7520 train 3.950926 (lr=2.9693e-05) (hash(x)=148811846)
|
| 907 |
+
7530 train 4.093120 (lr=2.9539e-05) (hash(x)=143727205)
|
| 908 |
+
7540 train 3.990803 (lr=2.9386e-05) (hash(x)=154980561)
|
| 909 |
+
7550 train 4.123275 (lr=2.9234e-05) (hash(x)=151143108)
|
| 910 |
+
7560 train 4.049075 (lr=2.9084e-05) (hash(x)=133379076)
|
| 911 |
+
7570 train 4.007765 (lr=2.8934e-05) (hash(x)=148338995)
|
| 912 |
+
7580 train 4.083848 (lr=2.8786e-05) (hash(x)=156733806)
|
| 913 |
+
7590 train 4.200554 (lr=2.8639e-05) (hash(x)=159299389)
|
| 914 |
+
7600 val loss 4.2427
|
| 915 |
+
7600 val perplexity 69.5922
|
| 916 |
+
7600 train 4.231465 (lr=2.8493e-05) (hash(x)=142771511)
|
| 917 |
+
7610 train 4.327890 (lr=2.8348e-05) (hash(x)=157959626)
|
| 918 |
+
7620 train 4.359527 (lr=2.8204e-05) (hash(x)=156257811)
|
| 919 |
+
7630 train 4.384919 (lr=2.8062e-05) (hash(x)=142730196)
|
| 920 |
+
7640 train 4.219873 (lr=2.7921e-05) (hash(x)=147923184)
|
| 921 |
+
7650 train 4.130831 (lr=2.7781e-05) (hash(x)=154290536)
|
| 922 |
+
7660 train 4.143825 (lr=2.7642e-05) (hash(x)=156433550)
|
| 923 |
+
7670 train 4.282334 (lr=2.7505e-05) (hash(x)=146611945)
|
| 924 |
+
7680 train 4.108991 (lr=2.7368e-05) (hash(x)=153745377)
|
| 925 |
+
7690 train 4.214911 (lr=2.7233e-05) (hash(x)=150833491)
|
| 926 |
+
7700 val loss 4.2346
|
| 927 |
+
7700 val perplexity 69.0344
|
| 928 |
+
7700 train 4.209716 (lr=2.7099e-05) (hash(x)=143602175)
|
| 929 |
+
7710 train 4.109449 (lr=2.6966e-05) (hash(x)=137035204)
|
| 930 |
+
7720 train 4.537834 (lr=2.6834e-05) (hash(x)=157409397)
|
| 931 |
+
7730 train 4.168361 (lr=2.6704e-05) (hash(x)=180056426)
|
| 932 |
+
7740 train 4.288931 (lr=2.6575e-05) (hash(x)=167791182)
|
| 933 |
+
7750 train 4.099330 (lr=2.6447e-05) (hash(x)=144902753)
|
| 934 |
+
7760 train 4.197666 (lr=2.6320e-05) (hash(x)=142048628)
|
| 935 |
+
7770 train 4.413654 (lr=2.6195e-05) (hash(x)=150623561)
|
| 936 |
+
7780 train 4.005875 (lr=2.6070e-05) (hash(x)=133068514)
|
| 937 |
+
7790 train 4.230723 (lr=2.5947e-05) (hash(x)=144372942)
|
| 938 |
+
7800 val loss 4.2345
|
| 939 |
+
7800 val perplexity 69.0297
|
| 940 |
+
7800 train 4.394639 (lr=2.5825e-05) (hash(x)=152379862)
|
| 941 |
+
7810 train 4.085546 (lr=2.5705e-05) (hash(x)=149929258)
|
| 942 |
+
7820 train 4.263099 (lr=2.5585e-05) (hash(x)=147738863)
|
| 943 |
+
7830 train 4.271801 (lr=2.5467e-05) (hash(x)=150885420)
|
| 944 |
+
7840 train 4.290581 (lr=2.5350e-05) (hash(x)=151751961)
|
| 945 |
+
7850 train 4.178764 (lr=2.5234e-05) (hash(x)=141061924)
|
| 946 |
+
7860 train 4.262223 (lr=2.5119e-05) (hash(x)=165946945)
|
| 947 |
+
7870 train 4.133366 (lr=2.5006e-05) (hash(x)=162222193)
|
| 948 |
+
7880 train 4.253111 (lr=2.4894e-05) (hash(x)=156110032)
|
| 949 |
+
7890 train 4.021930 (lr=2.4783e-05) (hash(x)=155496634)
|
| 950 |
+
7900 val loss 4.2335
|
| 951 |
+
7900 val perplexity 68.9551
|
| 952 |
+
7900 train 3.966272 (lr=2.4674e-05) (hash(x)=146655921)
|
| 953 |
+
7910 train 3.956963 (lr=2.4565e-05) (hash(x)=154169360)
|
| 954 |
+
7920 train 4.130836 (lr=2.4458e-05) (hash(x)=164630987)
|
| 955 |
+
7930 train 4.043834 (lr=2.4352e-05) (hash(x)=157591350)
|
| 956 |
+
7940 train 4.042671 (lr=2.4247e-05) (hash(x)=147986483)
|
| 957 |
+
7950 train 3.881840 (lr=2.4144e-05) (hash(x)=153958875)
|
| 958 |
+
7960 train 4.157236 (lr=2.4042e-05) (hash(x)=149307886)
|
| 959 |
+
7970 train 3.987556 (lr=2.3941e-05) (hash(x)=143190550)
|
| 960 |
+
7980 train 4.005992 (lr=2.3841e-05) (hash(x)=157802449)
|
| 961 |
+
7990 train 4.029515 (lr=2.3743e-05) (hash(x)=167181278)
|
| 962 |
+
8000 val loss 4.2366
|
| 963 |
+
8000 val perplexity 69.1748
|
| 964 |
+
8000 train 4.387695 (lr=2.3646e-05) (hash(x)=148262482)
|
| 965 |
+
8010 train 4.205113 (lr=2.3550e-05) (hash(x)=151119149)
|
| 966 |
+
8020 train 4.181247 (lr=2.3455e-05) (hash(x)=149748120)
|
| 967 |
+
8030 train 4.150961 (lr=2.3362e-05) (hash(x)=152462863)
|
| 968 |
+
8040 train 4.286757 (lr=2.3269e-05) (hash(x)=148714170)
|
| 969 |
+
8050 train 4.327882 (lr=2.3179e-05) (hash(x)=150219940)
|
| 970 |
+
8060 train 4.226609 (lr=2.3089e-05) (hash(x)=154859690)
|
| 971 |
+
8070 train 4.206408 (lr=2.3001e-05) (hash(x)=156635675)
|
| 972 |
+
8080 train 4.349037 (lr=2.2913e-05) (hash(x)=164430040)
|
| 973 |
+
8090 train 4.153214 (lr=2.2828e-05) (hash(x)=149003088)
|
| 974 |
+
8100 val loss 4.2244
|
| 975 |
+
8100 val perplexity 68.3309
|
| 976 |
+
8100 train 4.256297 (lr=2.2743e-05) (hash(x)=147683655)
|
| 977 |
+
8110 train 4.137872 (lr=2.2660e-05) (hash(x)=142849380)
|
| 978 |
+
8120 train 4.169298 (lr=2.2578e-05) (hash(x)=147408722)
|
| 979 |
+
8130 train 4.205829 (lr=2.2497e-05) (hash(x)=154095861)
|
| 980 |
+
8140 train 4.246831 (lr=2.2417e-05) (hash(x)=154770296)
|
| 981 |
+
8150 train 4.251180 (lr=2.2339e-05) (hash(x)=150175912)
|
| 982 |
+
8160 train 4.249357 (lr=2.2262e-05) (hash(x)=146846694)
|
| 983 |
+
8170 train 3.634931 (lr=2.2186e-05) (hash(x)=161621402)
|
| 984 |
+
8180 train 4.422406 (lr=2.2112e-05) (hash(x)=151058128)
|
| 985 |
+
8190 train 4.356398 (lr=2.2039e-05) (hash(x)=157265995)
|
| 986 |
+
8200 val loss 4.2269
|
| 987 |
+
8200 val perplexity 68.5071
|
| 988 |
+
8200 train 4.408456 (lr=2.1967e-05) (hash(x)=157312987)
|
| 989 |
+
8210 train 4.234895 (lr=2.1896e-05) (hash(x)=158755166)
|
| 990 |
+
8220 train 4.069514 (lr=2.1827e-05) (hash(x)=137031704)
|
| 991 |
+
8230 train 4.043029 (lr=2.1759e-05) (hash(x)=133909075)
|
| 992 |
+
8240 train 4.235438 (lr=2.1692e-05) (hash(x)=152756441)
|
| 993 |
+
8250 train 4.187023 (lr=2.1626e-05) (hash(x)=161956442)
|
| 994 |
+
8260 train 4.237917 (lr=2.1562e-05) (hash(x)=146994534)
|
| 995 |
+
8270 train 4.284015 (lr=2.1499e-05) (hash(x)=154925653)
|
| 996 |
+
8280 train 4.262526 (lr=2.1438e-05) (hash(x)=141129942)
|
| 997 |
+
8290 train 4.168244 (lr=2.1377e-05) (hash(x)=146178879)
|
| 998 |
+
8300 val loss 4.2272
|
| 999 |
+
8300 val perplexity 68.5279
|
| 1000 |
+
8300 train 4.142079 (lr=2.1318e-05) (hash(x)=141107543)
|
| 1001 |
+
8310 train 4.163462 (lr=2.1260e-05) (hash(x)=153322855)
|
| 1002 |
+
8320 train 4.251170 (lr=2.1204e-05) (hash(x)=151197408)
|
| 1003 |
+
8330 train 4.416556 (lr=2.1149e-05) (hash(x)=155927851)
|
| 1004 |
+
8340 train 4.342586 (lr=2.1095e-05) (hash(x)=162484336)
|
| 1005 |
+
8350 train 4.218870 (lr=2.1042e-05) (hash(x)=144280686)
|
| 1006 |
+
8360 train 4.178376 (lr=2.0991e-05) (hash(x)=147652229)
|
| 1007 |
+
8370 train 4.310989 (lr=2.0941e-05) (hash(x)=150071753)
|
| 1008 |
+
8380 train 4.169306 (lr=2.0892e-05) (hash(x)=146798938)
|
| 1009 |
+
8390 train 4.197971 (lr=2.0844e-05) (hash(x)=149596226)
|
| 1010 |
+
8400 val loss 4.2197
|
| 1011 |
+
8400 val perplexity 68.0146
|
| 1012 |
+
8400 train 4.269833 (lr=2.0798e-05) (hash(x)=141323024)
|
| 1013 |
+
8410 train 4.564712 (lr=2.0753e-05) (hash(x)=154823144)
|
| 1014 |
+
8420 train 4.187773 (lr=2.0710e-05) (hash(x)=151554301)
|
| 1015 |
+
8430 train 4.210325 (lr=2.0667e-05) (hash(x)=153707717)
|
| 1016 |
+
8440 train 4.190251 (lr=2.0626e-05) (hash(x)=156091926)
|
| 1017 |
+
8450 train 4.224891 (lr=2.0587e-05) (hash(x)=147111043)
|
| 1018 |
+
8460 train 4.323840 (lr=2.0548e-05) (hash(x)=135068736)
|
| 1019 |
+
8470 train 4.249349 (lr=2.0511e-05) (hash(x)=142255415)
|
| 1020 |
+
8480 train 4.245239 (lr=2.0475e-05) (hash(x)=138060402)
|
| 1021 |
+
8490 train 4.164992 (lr=2.0441e-05) (hash(x)=154789940)
|
| 1022 |
+
8500 val loss 4.2183
|
| 1023 |
+
8500 val perplexity 67.9167
|
| 1024 |
+
8500 train 4.135272 (lr=2.0408e-05) (hash(x)=150696521)
|
| 1025 |
+
8510 train 4.053050 (lr=2.0376e-05) (hash(x)=139242896)
|
| 1026 |
+
8520 train 4.134508 (lr=2.0345e-05) (hash(x)=148066730)
|
| 1027 |
+
8530 train 4.246009 (lr=2.0316e-05) (hash(x)=148649882)
|
| 1028 |
+
8540 train 4.049674 (lr=2.0288e-05) (hash(x)=148449712)
|
| 1029 |
+
8550 train 4.229570 (lr=2.0261e-05) (hash(x)=157155657)
|
| 1030 |
+
8560 train 4.130078 (lr=2.0235e-05) (hash(x)=139492416)
|
| 1031 |
+
8570 train 4.043805 (lr=2.0211e-05) (hash(x)=157567533)
|
| 1032 |
+
8580 train 4.106028 (lr=2.0189e-05) (hash(x)=143032656)
|
| 1033 |
+
8590 train 4.231947 (lr=2.0167e-05) (hash(x)=150899581)
|
| 1034 |
+
8600 val loss 4.2207
|
| 1035 |
+
8600 val perplexity 68.0823
|
| 1036 |
+
8600 train 4.132443 (lr=2.0147e-05) (hash(x)=162288191)
|
| 1037 |
+
8610 train 4.284783 (lr=2.0128e-05) (hash(x)=148361485)
|
| 1038 |
+
8620 train 4.180494 (lr=2.0110e-05) (hash(x)=151740051)
|
| 1039 |
+
8630 train 4.133177 (lr=2.0094e-05) (hash(x)=154125251)
|
| 1040 |
+
8640 train 4.199639 (lr=2.0079e-05) (hash(x)=145857675)
|
| 1041 |
+
8650 train 4.079593 (lr=2.0065e-05) (hash(x)=148185775)
|
| 1042 |
+
8660 train 4.106319 (lr=2.0053e-05) (hash(x)=149778450)
|
| 1043 |
+
8670 train 4.368206 (lr=2.0042e-05) (hash(x)=148123274)
|
| 1044 |
+
8680 train 4.139586 (lr=2.0032e-05) (hash(x)=129372639)
|
| 1045 |
+
8690 train 4.232241 (lr=2.0023e-05) (hash(x)=145050683)
|
| 1046 |
+
8700 val loss 4.2154
|
| 1047 |
+
8700 val perplexity 67.7193
|
| 1048 |
+
8700 train 4.153593 (lr=2.0016e-05) (hash(x)=152860941)
|
| 1049 |
+
8710 train 4.255550 (lr=2.0010e-05) (hash(x)=156971498)
|
| 1050 |
+
8720 train 4.277818 (lr=2.0006e-05) (hash(x)=155513041)
|
| 1051 |
+
8730 train 4.194418 (lr=2.0003e-05) (hash(x)=167451760)
|
| 1052 |
+
8740 train 4.275624 (lr=2.0001e-05) (hash(x)=158768428)
|
| 1053 |
+
8749 val loss 4.2154
|
| 1054 |
+
8749 val perplexity 67.7199
|
lr2e-4_total_batch_size61440_baseline_seed1341/model_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9b5e43fb8ce8a33974cf0b306d216838015f81a9e857fc7b98129b801200ded
|
| 3 |
+
size 92843394
|
lr2e-4_total_batch_size61440_baseline_seed1341/optimizer_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e049eae718d7ac2c3a7c1b92a39ff27b461d9cab793c4cd64a0c78aeac0995a9
|
| 3 |
+
size 179406214
|