Upload folder using huggingface_hub
Browse files- n_heads12_lr14e-4_total_batch_size61440_seed1338/args.json +1 -0
- n_heads12_lr14e-4_total_batch_size61440_seed1338/dataloader_08749.pt +3 -0
- n_heads12_lr14e-4_total_batch_size61440_seed1338/log2.txt +1054 -0
- n_heads12_lr14e-4_total_batch_size61440_seed1338/model_08749.pt +3 -0
- n_heads12_lr14e-4_total_batch_size61440_seed1338/optimizer_08749.pt +3 -0
n_heads12_lr14e-4_total_batch_size61440_seed1338/args.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_11/n_heads12_lr14e-4_total_batch_size61440_seed1338", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 12, "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": 1338, "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": 60, "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": 0.0014, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "n_heads12_14e-4_61440", "n_embd": 768}
|
n_heads12_lr14e-4_total_batch_size61440_seed1338/dataloader_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9bbf5b7eef643b850a13a0d3881f9b995af25e67f722e39e1853dc09a06ad1dc
|
| 3 |
+
size 964
|
n_heads12_lr14e-4_total_batch_size61440_seed1338/log2.txt
ADDED
|
@@ -0,0 +1,1054 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
max_steps: 8750
|
| 2 |
+
0 val loss 10.9899
|
| 3 |
+
0 val perplexity 59271.4727
|
| 4 |
+
0 train 10.996866 (lr=2.8000e-06) (hash(x)=73026074)
|
| 5 |
+
10 train 9.771564 (lr=3.0800e-05) (hash(x)=81408174)
|
| 6 |
+
20 train 9.314571 (lr=5.8800e-05) (hash(x)=73752420)
|
| 7 |
+
30 train 8.757971 (lr=8.6800e-05) (hash(x)=77647299)
|
| 8 |
+
40 train 8.255341 (lr=1.1480e-04) (hash(x)=76104329)
|
| 9 |
+
50 train 7.819843 (lr=1.4280e-04) (hash(x)=69335494)
|
| 10 |
+
60 train 7.643772 (lr=1.7080e-04) (hash(x)=72716449)
|
| 11 |
+
70 train 7.554301 (lr=1.9880e-04) (hash(x)=71125105)
|
| 12 |
+
80 train 7.389592 (lr=2.2680e-04) (hash(x)=74407559)
|
| 13 |
+
90 train 7.210872 (lr=2.5480e-04) (hash(x)=78245164)
|
| 14 |
+
100 val loss 7.1842
|
| 15 |
+
100 val perplexity 1318.3804
|
| 16 |
+
100 train 7.161286 (lr=2.8280e-04) (hash(x)=66975638)
|
| 17 |
+
110 train 6.923670 (lr=3.1080e-04) (hash(x)=72573036)
|
| 18 |
+
120 train 6.852287 (lr=3.3880e-04) (hash(x)=72876251)
|
| 19 |
+
130 train 6.815176 (lr=3.6680e-04) (hash(x)=79969711)
|
| 20 |
+
140 train 6.711454 (lr=3.9480e-04) (hash(x)=76473027)
|
| 21 |
+
150 train 6.588482 (lr=4.2280e-04) (hash(x)=72199769)
|
| 22 |
+
160 train 6.424819 (lr=4.5080e-04) (hash(x)=79575844)
|
| 23 |
+
170 train 6.323165 (lr=4.7880e-04) (hash(x)=65860740)
|
| 24 |
+
180 train 6.304462 (lr=5.0680e-04) (hash(x)=79880170)
|
| 25 |
+
190 train 6.206789 (lr=5.3480e-04) (hash(x)=78465601)
|
| 26 |
+
200 val loss 6.3661
|
| 27 |
+
200 val perplexity 581.7808
|
| 28 |
+
200 train 6.099457 (lr=5.6280e-04) (hash(x)=72711480)
|
| 29 |
+
210 train 6.060147 (lr=5.9080e-04) (hash(x)=71194058)
|
| 30 |
+
220 train 6.237395 (lr=6.1880e-04) (hash(x)=92783037)
|
| 31 |
+
230 train 5.992590 (lr=6.4680e-04) (hash(x)=73611161)
|
| 32 |
+
240 train 5.792938 (lr=6.7480e-04) (hash(x)=74427917)
|
| 33 |
+
250 train 5.812118 (lr=7.0280e-04) (hash(x)=65305370)
|
| 34 |
+
260 train 5.907798 (lr=7.3080e-04) (hash(x)=81378465)
|
| 35 |
+
270 train 5.668375 (lr=7.5880e-04) (hash(x)=72738754)
|
| 36 |
+
280 train 6.009033 (lr=7.8680e-04) (hash(x)=82172761)
|
| 37 |
+
290 train 6.174212 (lr=8.1480e-04) (hash(x)=70024961)
|
| 38 |
+
300 val loss 5.9553
|
| 39 |
+
300 val perplexity 385.8067
|
| 40 |
+
300 train 5.900202 (lr=8.4280e-04) (hash(x)=74275986)
|
| 41 |
+
310 train 5.991466 (lr=8.7080e-04) (hash(x)=76295038)
|
| 42 |
+
320 train 5.974309 (lr=8.9880e-04) (hash(x)=74741156)
|
| 43 |
+
330 train 5.748874 (lr=9.2680e-04) (hash(x)=71941043)
|
| 44 |
+
340 train 5.839636 (lr=9.5480e-04) (hash(x)=75020464)
|
| 45 |
+
350 train 5.862627 (lr=9.8280e-04) (hash(x)=64712930)
|
| 46 |
+
360 train 6.138801 (lr=1.0108e-03) (hash(x)=68519941)
|
| 47 |
+
370 train 5.727160 (lr=1.0388e-03) (hash(x)=75355289)
|
| 48 |
+
380 train 5.719408 (lr=1.0668e-03) (hash(x)=66796276)
|
| 49 |
+
390 train 5.781187 (lr=1.0948e-03) (hash(x)=81002698)
|
| 50 |
+
400 val loss 5.6218
|
| 51 |
+
400 val perplexity 276.3919
|
| 52 |
+
400 train 5.460107 (lr=1.1228e-03) (hash(x)=76354037)
|
| 53 |
+
410 train 5.746010 (lr=1.1508e-03) (hash(x)=82242082)
|
| 54 |
+
420 train 5.462119 (lr=1.1788e-03) (hash(x)=77642063)
|
| 55 |
+
430 train 5.387616 (lr=1.2068e-03) (hash(x)=72281244)
|
| 56 |
+
440 train 5.570218 (lr=1.2348e-03) (hash(x)=79666830)
|
| 57 |
+
450 train 5.450129 (lr=1.2628e-03) (hash(x)=76014248)
|
| 58 |
+
460 train 5.630340 (lr=1.2908e-03) (hash(x)=81129062)
|
| 59 |
+
470 train 5.428161 (lr=1.3188e-03) (hash(x)=69374949)
|
| 60 |
+
480 train 5.418986 (lr=1.3468e-03) (hash(x)=72137706)
|
| 61 |
+
490 train 5.440450 (lr=1.3748e-03) (hash(x)=74403781)
|
| 62 |
+
500 val loss 5.4235
|
| 63 |
+
500 val perplexity 226.6691
|
| 64 |
+
500 train 5.327272 (lr=1.4000e-03) (hash(x)=78816019)
|
| 65 |
+
510 train 5.359278 (lr=1.4000e-03) (hash(x)=72193331)
|
| 66 |
+
520 train 5.365542 (lr=1.4000e-03) (hash(x)=79344438)
|
| 67 |
+
530 train 5.344696 (lr=1.4000e-03) (hash(x)=78092289)
|
| 68 |
+
540 train 5.338777 (lr=1.3999e-03) (hash(x)=74733731)
|
| 69 |
+
550 train 5.292069 (lr=1.3999e-03) (hash(x)=69920645)
|
| 70 |
+
560 train 5.120903 (lr=1.3998e-03) (hash(x)=76131735)
|
| 71 |
+
570 train 5.201076 (lr=1.3998e-03) (hash(x)=78418293)
|
| 72 |
+
580 train 5.024049 (lr=1.3997e-03) (hash(x)=73699256)
|
| 73 |
+
590 train 5.027842 (lr=1.3996e-03) (hash(x)=77505664)
|
| 74 |
+
600 val loss 5.2316
|
| 75 |
+
600 val perplexity 187.0873
|
| 76 |
+
600 train 4.951847 (lr=1.3995e-03) (hash(x)=71823058)
|
| 77 |
+
610 train 4.995150 (lr=1.3994e-03) (hash(x)=78450672)
|
| 78 |
+
620 train 4.902871 (lr=1.3993e-03) (hash(x)=75456256)
|
| 79 |
+
630 train 4.862373 (lr=1.3992e-03) (hash(x)=68515074)
|
| 80 |
+
640 train 4.927540 (lr=1.3991e-03) (hash(x)=71968654)
|
| 81 |
+
650 train 4.841162 (lr=1.3990e-03) (hash(x)=72586963)
|
| 82 |
+
660 train 4.793467 (lr=1.3988e-03) (hash(x)=67398625)
|
| 83 |
+
670 train 4.809901 (lr=1.3987e-03) (hash(x)=82737129)
|
| 84 |
+
680 train 5.069919 (lr=1.3985e-03) (hash(x)=78920914)
|
| 85 |
+
690 train 5.081184 (lr=1.3984e-03) (hash(x)=59859106)
|
| 86 |
+
700 val loss 5.0590
|
| 87 |
+
700 val perplexity 157.4337
|
| 88 |
+
700 train 4.985450 (lr=1.3982e-03) (hash(x)=76856514)
|
| 89 |
+
710 train 5.012370 (lr=1.3980e-03) (hash(x)=73530680)
|
| 90 |
+
720 train 5.079854 (lr=1.3978e-03) (hash(x)=74382454)
|
| 91 |
+
730 train 4.932901 (lr=1.3976e-03) (hash(x)=80641678)
|
| 92 |
+
740 train 5.062945 (lr=1.3974e-03) (hash(x)=82419404)
|
| 93 |
+
750 train 5.005933 (lr=1.3971e-03) (hash(x)=70898356)
|
| 94 |
+
760 train 4.992117 (lr=1.3969e-03) (hash(x)=76422181)
|
| 95 |
+
770 train 4.943111 (lr=1.3967e-03) (hash(x)=83711024)
|
| 96 |
+
780 train 4.870038 (lr=1.3964e-03) (hash(x)=74365746)
|
| 97 |
+
790 train 4.816755 (lr=1.3962e-03) (hash(x)=73104424)
|
| 98 |
+
800 val loss 4.8498
|
| 99 |
+
800 val perplexity 127.7133
|
| 100 |
+
800 train 4.985498 (lr=1.3959e-03) (hash(x)=73589083)
|
| 101 |
+
810 train 4.783382 (lr=1.3956e-03) (hash(x)=69132306)
|
| 102 |
+
820 train 4.828509 (lr=1.3953e-03) (hash(x)=75056169)
|
| 103 |
+
830 train 4.731311 (lr=1.3950e-03) (hash(x)=70032437)
|
| 104 |
+
840 train 4.740376 (lr=1.3947e-03) (hash(x)=73178845)
|
| 105 |
+
850 train 4.745549 (lr=1.3944e-03) (hash(x)=69463059)
|
| 106 |
+
860 train 4.830619 (lr=1.3941e-03) (hash(x)=74635339)
|
| 107 |
+
870 train 4.990119 (lr=1.3938e-03) (hash(x)=73964810)
|
| 108 |
+
880 train 4.729581 (lr=1.3934e-03) (hash(x)=79912096)
|
| 109 |
+
890 train 4.692575 (lr=1.3931e-03) (hash(x)=84832110)
|
| 110 |
+
900 val loss 4.7287
|
| 111 |
+
900 val perplexity 113.1432
|
| 112 |
+
900 train 4.702148 (lr=1.3927e-03) (hash(x)=71823388)
|
| 113 |
+
910 train 4.660867 (lr=1.3923e-03) (hash(x)=69541745)
|
| 114 |
+
920 train 4.603755 (lr=1.3920e-03) (hash(x)=67018969)
|
| 115 |
+
930 train 4.554440 (lr=1.3916e-03) (hash(x)=73346110)
|
| 116 |
+
940 train 4.624906 (lr=1.3912e-03) (hash(x)=72105603)
|
| 117 |
+
950 train 4.528996 (lr=1.3908e-03) (hash(x)=71803352)
|
| 118 |
+
960 train 4.682458 (lr=1.3904e-03) (hash(x)=75135621)
|
| 119 |
+
970 train 4.471962 (lr=1.3899e-03) (hash(x)=73334774)
|
| 120 |
+
980 train 4.437955 (lr=1.3895e-03) (hash(x)=80861275)
|
| 121 |
+
990 train 4.508183 (lr=1.3891e-03) (hash(x)=77124068)
|
| 122 |
+
1000 val loss 4.6650
|
| 123 |
+
1000 val perplexity 106.1609
|
| 124 |
+
1000 train 4.391986 (lr=1.3886e-03) (hash(x)=78127630)
|
| 125 |
+
1010 train 4.343163 (lr=1.3882e-03) (hash(x)=73670162)
|
| 126 |
+
1020 train 4.453861 (lr=1.3877e-03) (hash(x)=77732174)
|
| 127 |
+
1030 train 4.304074 (lr=1.3872e-03) (hash(x)=79418857)
|
| 128 |
+
1040 train 4.341743 (lr=1.3867e-03) (hash(x)=77732710)
|
| 129 |
+
1050 train 4.308478 (lr=1.3862e-03) (hash(x)=69751515)
|
| 130 |
+
1060 train 4.304161 (lr=1.3857e-03) (hash(x)=71788756)
|
| 131 |
+
1070 train 4.506732 (lr=1.3852e-03) (hash(x)=72944541)
|
| 132 |
+
1080 train 4.586261 (lr=1.3847e-03) (hash(x)=75792144)
|
| 133 |
+
1090 train 4.520107 (lr=1.3842e-03) (hash(x)=78746868)
|
| 134 |
+
1100 val loss 4.5809
|
| 135 |
+
1100 val perplexity 97.6018
|
| 136 |
+
1100 train 4.597838 (lr=1.3836e-03) (hash(x)=70879945)
|
| 137 |
+
1110 train 4.598518 (lr=1.3831e-03) (hash(x)=77572329)
|
| 138 |
+
1120 train 5.063550 (lr=1.3825e-03) (hash(x)=68749754)
|
| 139 |
+
1130 train 4.579143 (lr=1.3820e-03) (hash(x)=78632440)
|
| 140 |
+
1140 train 4.515252 (lr=1.3814e-03) (hash(x)=73565080)
|
| 141 |
+
1150 train 4.580207 (lr=1.3808e-03) (hash(x)=81594247)
|
| 142 |
+
1160 train 4.569056 (lr=1.3802e-03) (hash(x)=63200457)
|
| 143 |
+
1170 train 4.624634 (lr=1.3796e-03) (hash(x)=75895085)
|
| 144 |
+
1180 train 4.637632 (lr=1.3790e-03) (hash(x)=71402056)
|
| 145 |
+
1190 train 4.582512 (lr=1.3784e-03) (hash(x)=75263955)
|
| 146 |
+
1200 val loss 4.5174
|
| 147 |
+
1200 val perplexity 91.5952
|
| 148 |
+
1200 train 4.575253 (lr=1.3778e-03) (hash(x)=73579627)
|
| 149 |
+
1210 train 4.573855 (lr=1.3771e-03) (hash(x)=79106153)
|
| 150 |
+
1220 train 4.628505 (lr=1.3765e-03) (hash(x)=78475838)
|
| 151 |
+
1230 train 4.696779 (lr=1.3758e-03) (hash(x)=72129619)
|
| 152 |
+
1240 train 4.611730 (lr=1.3752e-03) (hash(x)=58067507)
|
| 153 |
+
1250 train 4.417848 (lr=1.3745e-03) (hash(x)=78666722)
|
| 154 |
+
1260 train 4.498178 (lr=1.3738e-03) (hash(x)=74964097)
|
| 155 |
+
1270 train 4.474730 (lr=1.3731e-03) (hash(x)=72774823)
|
| 156 |
+
1280 train 4.514427 (lr=1.3724e-03) (hash(x)=69959055)
|
| 157 |
+
1290 train 4.351562 (lr=1.3717e-03) (hash(x)=76629730)
|
| 158 |
+
1300 val loss 4.4706
|
| 159 |
+
1300 val perplexity 87.4106
|
| 160 |
+
1300 train 4.390519 (lr=1.3710e-03) (hash(x)=72183284)
|
| 161 |
+
1310 train 4.587470 (lr=1.3703e-03) (hash(x)=83147722)
|
| 162 |
+
1320 train 4.557054 (lr=1.3695e-03) (hash(x)=69788032)
|
| 163 |
+
1330 train 4.271809 (lr=1.3688e-03) (hash(x)=66629732)
|
| 164 |
+
1340 train 4.336130 (lr=1.3680e-03) (hash(x)=72542387)
|
| 165 |
+
1350 train 4.388614 (lr=1.3673e-03) (hash(x)=76517375)
|
| 166 |
+
1360 train 4.377736 (lr=1.3665e-03) (hash(x)=74401519)
|
| 167 |
+
1370 train 4.200589 (lr=1.3657e-03) (hash(x)=78409356)
|
| 168 |
+
1380 train 4.165195 (lr=1.3650e-03) (hash(x)=77626183)
|
| 169 |
+
1390 train 4.116465 (lr=1.3642e-03) (hash(x)=72496033)
|
| 170 |
+
1400 val loss 4.4347
|
| 171 |
+
1400 val perplexity 84.3302
|
| 172 |
+
1400 train 4.193803 (lr=1.3634e-03) (hash(x)=73708759)
|
| 173 |
+
1410 train 4.194789 (lr=1.3626e-03) (hash(x)=73706234)
|
| 174 |
+
1420 train 4.065205 (lr=1.3617e-03) (hash(x)=75918530)
|
| 175 |
+
1430 train 3.956392 (lr=1.3609e-03) (hash(x)=76806498)
|
| 176 |
+
1440 train 4.178704 (lr=1.3601e-03) (hash(x)=77310523)
|
| 177 |
+
1450 train 4.155601 (lr=1.3592e-03) (hash(x)=75964248)
|
| 178 |
+
1460 train 3.999964 (lr=1.3584e-03) (hash(x)=88549155)
|
| 179 |
+
1470 train 4.481586 (lr=1.3575e-03) (hash(x)=78654420)
|
| 180 |
+
1480 train 4.350353 (lr=1.3566e-03) (hash(x)=53433500)
|
| 181 |
+
1490 train 4.363974 (lr=1.3558e-03) (hash(x)=75862069)
|
| 182 |
+
1500 val loss 4.4237
|
| 183 |
+
1500 val perplexity 83.4054
|
| 184 |
+
1500 train 4.477839 (lr=1.3549e-03) (hash(x)=76037312)
|
| 185 |
+
1510 train 4.218045 (lr=1.3540e-03) (hash(x)=78654499)
|
| 186 |
+
1520 train 4.314601 (lr=1.3531e-03) (hash(x)=72319550)
|
| 187 |
+
1530 train 4.286132 (lr=1.3522e-03) (hash(x)=78901150)
|
| 188 |
+
1540 train 4.335060 (lr=1.3512e-03) (hash(x)=74848448)
|
| 189 |
+
1550 train 4.380551 (lr=1.3503e-03) (hash(x)=77019229)
|
| 190 |
+
1560 train 4.459638 (lr=1.3494e-03) (hash(x)=74200996)
|
| 191 |
+
1570 train 4.349664 (lr=1.3484e-03) (hash(x)=76717682)
|
| 192 |
+
1580 train 4.240232 (lr=1.3475e-03) (hash(x)=74662317)
|
| 193 |
+
1590 train 4.423012 (lr=1.3465e-03) (hash(x)=76066438)
|
| 194 |
+
1600 val loss 4.3573
|
| 195 |
+
1600 val perplexity 78.0477
|
| 196 |
+
1600 train 4.552492 (lr=1.3455e-03) (hash(x)=76123488)
|
| 197 |
+
1610 train 4.438817 (lr=1.3446e-03) (hash(x)=73211013)
|
| 198 |
+
1620 train 4.320731 (lr=1.3436e-03) (hash(x)=80389252)
|
| 199 |
+
1630 train 4.317528 (lr=1.3426e-03) (hash(x)=73449158)
|
| 200 |
+
1640 train 4.335651 (lr=1.3416e-03) (hash(x)=75060644)
|
| 201 |
+
1650 train 4.273685 (lr=1.3406e-03) (hash(x)=68966493)
|
| 202 |
+
1660 train 4.349126 (lr=1.3395e-03) (hash(x)=73496223)
|
| 203 |
+
1670 train 4.204063 (lr=1.3385e-03) (hash(x)=86210083)
|
| 204 |
+
1680 train 4.298112 (lr=1.3375e-03) (hash(x)=74053619)
|
| 205 |
+
1690 train 4.372838 (lr=1.3364e-03) (hash(x)=65910719)
|
| 206 |
+
1700 val loss 4.3430
|
| 207 |
+
1700 val perplexity 76.9411
|
| 208 |
+
1700 train 4.222454 (lr=1.3354e-03) (hash(x)=66496027)
|
| 209 |
+
1710 train 4.394483 (lr=1.3343e-03) (hash(x)=75485845)
|
| 210 |
+
1720 train 4.216524 (lr=1.3332e-03) (hash(x)=72417899)
|
| 211 |
+
1730 train 4.252777 (lr=1.3321e-03) (hash(x)=73593240)
|
| 212 |
+
1740 train 4.155072 (lr=1.3311e-03) (hash(x)=74832651)
|
| 213 |
+
1750 train 4.132105 (lr=1.3300e-03) (hash(x)=73364724)
|
| 214 |
+
1760 train 4.298528 (lr=1.3289e-03) (hash(x)=75903856)
|
| 215 |
+
1770 train 4.357667 (lr=1.3278e-03) (hash(x)=78183744)
|
| 216 |
+
1780 train 4.268876 (lr=1.3266e-03) (hash(x)=69993946)
|
| 217 |
+
1790 train 4.342669 (lr=1.3255e-03) (hash(x)=101032039)
|
| 218 |
+
1800 val loss 4.3020
|
| 219 |
+
1800 val perplexity 73.8508
|
| 220 |
+
1800 train 4.360539 (lr=1.3244e-03) (hash(x)=86018370)
|
| 221 |
+
1810 train 3.919175 (lr=1.3232e-03) (hash(x)=84205841)
|
| 222 |
+
1820 train 4.250078 (lr=1.3221e-03) (hash(x)=81735398)
|
| 223 |
+
1830 train 4.148702 (lr=1.3209e-03) (hash(x)=77297230)
|
| 224 |
+
1840 train 4.103259 (lr=1.3197e-03) (hash(x)=75535995)
|
| 225 |
+
1850 train 4.122056 (lr=1.3186e-03) (hash(x)=73930506)
|
| 226 |
+
1860 train 4.088635 (lr=1.3174e-03) (hash(x)=79524567)
|
| 227 |
+
1870 train 4.020846 (lr=1.3162e-03) (hash(x)=80242414)
|
| 228 |
+
1880 train 3.944268 (lr=1.3150e-03) (hash(x)=71395200)
|
| 229 |
+
1890 train 4.062816 (lr=1.3138e-03) (hash(x)=78269507)
|
| 230 |
+
1900 val loss 4.2839
|
| 231 |
+
1900 val perplexity 72.5212
|
| 232 |
+
1900 train 4.297132 (lr=1.3126e-03) (hash(x)=78327979)
|
| 233 |
+
1910 train 4.227680 (lr=1.3113e-03) (hash(x)=75760901)
|
| 234 |
+
1920 train 4.284264 (lr=1.3101e-03) (hash(x)=81699948)
|
| 235 |
+
1930 train 4.331404 (lr=1.3089e-03) (hash(x)=75326527)
|
| 236 |
+
1940 train 4.149306 (lr=1.3076e-03) (hash(x)=73660211)
|
| 237 |
+
1950 train 4.287952 (lr=1.3064e-03) (hash(x)=66946100)
|
| 238 |
+
1960 train 4.112702 (lr=1.3051e-03) (hash(x)=45679133)
|
| 239 |
+
1970 train 4.380965 (lr=1.3038e-03) (hash(x)=73647298)
|
| 240 |
+
1980 train 4.284623 (lr=1.3026e-03) (hash(x)=72037215)
|
| 241 |
+
1990 train 4.286258 (lr=1.3013e-03) (hash(x)=77640675)
|
| 242 |
+
2000 val loss 4.2286
|
| 243 |
+
2000 val perplexity 68.6217
|
| 244 |
+
2000 train 4.265011 (lr=1.3000e-03) (hash(x)=78165118)
|
| 245 |
+
2010 train 4.327103 (lr=1.2987e-03) (hash(x)=80013879)
|
| 246 |
+
2020 train 4.584308 (lr=1.2974e-03) (hash(x)=73778143)
|
| 247 |
+
2030 train 4.099406 (lr=1.2961e-03) (hash(x)=71396805)
|
| 248 |
+
2040 train 4.307127 (lr=1.2947e-03) (hash(x)=83350333)
|
| 249 |
+
2050 train 4.805965 (lr=1.2934e-03) (hash(x)=67399515)
|
| 250 |
+
2060 train 4.291596 (lr=1.2921e-03) (hash(x)=66586949)
|
| 251 |
+
2070 train 4.155853 (lr=1.2907e-03) (hash(x)=77648913)
|
| 252 |
+
2080 train 4.298593 (lr=1.2894e-03) (hash(x)=80507387)
|
| 253 |
+
2090 train 4.327189 (lr=1.2880e-03) (hash(x)=67897081)
|
| 254 |
+
2100 val loss 4.2058
|
| 255 |
+
2100 val perplexity 67.0716
|
| 256 |
+
2100 train 4.087350 (lr=1.2866e-03) (hash(x)=78780658)
|
| 257 |
+
2110 train 4.007301 (lr=1.2853e-03) (hash(x)=71002379)
|
| 258 |
+
2120 train 4.259398 (lr=1.2839e-03) (hash(x)=78941785)
|
| 259 |
+
2130 train 4.206317 (lr=1.2825e-03) (hash(x)=77840641)
|
| 260 |
+
2140 train 4.059928 (lr=1.2811e-03) (hash(x)=79973491)
|
| 261 |
+
2150 train 4.144462 (lr=1.2797e-03) (hash(x)=70608621)
|
| 262 |
+
2160 train 4.166276 (lr=1.2783e-03) (hash(x)=77576393)
|
| 263 |
+
2170 train 4.132263 (lr=1.2768e-03) (hash(x)=79697380)
|
| 264 |
+
2180 train 4.094278 (lr=1.2754e-03) (hash(x)=68595163)
|
| 265 |
+
2190 train 4.151357 (lr=1.2740e-03) (hash(x)=78393072)
|
| 266 |
+
2200 val loss 4.1944
|
| 267 |
+
2200 val perplexity 66.3133
|
| 268 |
+
2200 train 3.914644 (lr=1.2725e-03) (hash(x)=68832590)
|
| 269 |
+
2210 train 3.935787 (lr=1.2711e-03) (hash(x)=73069430)
|
| 270 |
+
2220 train 3.868876 (lr=1.2696e-03) (hash(x)=78403531)
|
| 271 |
+
2230 train 3.952213 (lr=1.2682e-03) (hash(x)=84282001)
|
| 272 |
+
2240 train 3.986274 (lr=1.2667e-03) (hash(x)=68791263)
|
| 273 |
+
2250 train 3.910575 (lr=1.2652e-03) (hash(x)=78324357)
|
| 274 |
+
2260 train 3.754681 (lr=1.2637e-03) (hash(x)=70769509)
|
| 275 |
+
2270 train 3.969132 (lr=1.2622e-03) (hash(x)=76496955)
|
| 276 |
+
2280 train 3.999491 (lr=1.2607e-03) (hash(x)=76607207)
|
| 277 |
+
2290 train 4.005269 (lr=1.2592e-03) (hash(x)=76204997)
|
| 278 |
+
2300 val loss 4.1839
|
| 279 |
+
2300 val perplexity 65.6181
|
| 280 |
+
2300 train 4.188251 (lr=1.2577e-03) (hash(x)=78644214)
|
| 281 |
+
2310 train 4.174999 (lr=1.2562e-03) (hash(x)=69825560)
|
| 282 |
+
2320 train 4.221283 (lr=1.2547e-03) (hash(x)=67690675)
|
| 283 |
+
2330 train 4.238270 (lr=1.2531e-03) (hash(x)=72400089)
|
| 284 |
+
2340 train 4.121150 (lr=1.2516e-03) (hash(x)=87283063)
|
| 285 |
+
2350 train 4.112997 (lr=1.2500e-03) (hash(x)=74257442)
|
| 286 |
+
2360 train 4.206243 (lr=1.2485e-03) (hash(x)=76596934)
|
| 287 |
+
2370 train 4.188194 (lr=1.2469e-03) (hash(x)=76345291)
|
| 288 |
+
2380 train 4.103049 (lr=1.2453e-03) (hash(x)=65495351)
|
| 289 |
+
2390 train 4.417353 (lr=1.2438e-03) (hash(x)=75795347)
|
| 290 |
+
2400 val loss 4.1619
|
| 291 |
+
2400 val perplexity 64.1919
|
| 292 |
+
2400 train 4.143678 (lr=1.2422e-03) (hash(x)=85857583)
|
| 293 |
+
2410 train 4.131220 (lr=1.2406e-03) (hash(x)=72305336)
|
| 294 |
+
2420 train 4.084464 (lr=1.2390e-03) (hash(x)=76780579)
|
| 295 |
+
2430 train 4.068442 (lr=1.2374e-03) (hash(x)=85532225)
|
| 296 |
+
2440 train 4.189742 (lr=1.2358e-03) (hash(x)=74308641)
|
| 297 |
+
2450 train 4.170431 (lr=1.2341e-03) (hash(x)=75205457)
|
| 298 |
+
2460 train 4.195348 (lr=1.2325e-03) (hash(x)=65232578)
|
| 299 |
+
2470 train 4.216726 (lr=1.2309e-03) (hash(x)=77893584)
|
| 300 |
+
2480 train 4.061187 (lr=1.2293e-03) (hash(x)=63720791)
|
| 301 |
+
2490 train 4.115158 (lr=1.2276e-03) (hash(x)=74714745)
|
| 302 |
+
2500 val loss 4.1203
|
| 303 |
+
2500 val perplexity 61.5767
|
| 304 |
+
2500 train 4.231857 (lr=1.2260e-03) (hash(x)=72628906)
|
| 305 |
+
2510 train 3.989606 (lr=1.2243e-03) (hash(x)=69138578)
|
| 306 |
+
2520 train 4.164144 (lr=1.2226e-03) (hash(x)=65122039)
|
| 307 |
+
2530 train 4.150131 (lr=1.2210e-03) (hash(x)=74953183)
|
| 308 |
+
2540 train 4.063065 (lr=1.2193e-03) (hash(x)=75410150)
|
| 309 |
+
2550 train 3.967093 (lr=1.2176e-03) (hash(x)=80818592)
|
| 310 |
+
2560 train 3.989614 (lr=1.2159e-03) (hash(x)=73726201)
|
| 311 |
+
2570 train 4.159233 (lr=1.2142e-03) (hash(x)=78518314)
|
| 312 |
+
2580 train 4.132399 (lr=1.2125e-03) (hash(x)=70758402)
|
| 313 |
+
2590 train 4.100067 (lr=1.2108e-03) (hash(x)=70495389)
|
| 314 |
+
2600 val loss 4.1223
|
| 315 |
+
2600 val perplexity 61.6998
|
| 316 |
+
2600 train 4.007177 (lr=1.2091e-03) (hash(x)=82589673)
|
| 317 |
+
2610 train 3.883005 (lr=1.2073e-03) (hash(x)=70936243)
|
| 318 |
+
2620 train 3.815397 (lr=1.2056e-03) (hash(x)=83907871)
|
| 319 |
+
2630 train 3.908987 (lr=1.2039e-03) (hash(x)=70432517)
|
| 320 |
+
2640 train 3.898691 (lr=1.2021e-03) (hash(x)=73683605)
|
| 321 |
+
2650 train 3.762873 (lr=1.2004e-03) (hash(x)=81772854)
|
| 322 |
+
2660 train 3.897710 (lr=1.1986e-03) (hash(x)=68745438)
|
| 323 |
+
2670 train 3.950229 (lr=1.1969e-03) (hash(x)=77762278)
|
| 324 |
+
2680 train 3.823393 (lr=1.1951e-03) (hash(x)=70738783)
|
| 325 |
+
2690 train 3.831790 (lr=1.1933e-03) (hash(x)=74534099)
|
| 326 |
+
2700 val loss 4.1103
|
| 327 |
+
2700 val perplexity 60.9634
|
| 328 |
+
2700 train 4.027716 (lr=1.1916e-03) (hash(x)=81441457)
|
| 329 |
+
2710 train 4.073734 (lr=1.1898e-03) (hash(x)=76167134)
|
| 330 |
+
2720 train 4.129731 (lr=1.1880e-03) (hash(x)=79036746)
|
| 331 |
+
2730 train 4.112789 (lr=1.1862e-03) (hash(x)=76789597)
|
| 332 |
+
2740 train 4.059471 (lr=1.1844e-03) (hash(x)=67515905)
|
| 333 |
+
2750 train 4.178035 (lr=1.1826e-03) (hash(x)=81464179)
|
| 334 |
+
2760 train 4.068268 (lr=1.1807e-03) (hash(x)=83004450)
|
| 335 |
+
2770 train 4.059431 (lr=1.1789e-03) (hash(x)=74647329)
|
| 336 |
+
2780 train 4.229542 (lr=1.1771e-03) (hash(x)=65160033)
|
| 337 |
+
2790 train 4.275399 (lr=1.1753e-03) (hash(x)=77815496)
|
| 338 |
+
2800 val loss 4.0807
|
| 339 |
+
2800 val perplexity 59.1864
|
| 340 |
+
2800 train 4.085231 (lr=1.1734e-03) (hash(x)=68408565)
|
| 341 |
+
2810 train 4.116845 (lr=1.1716e-03) (hash(x)=71127979)
|
| 342 |
+
2820 train 4.133625 (lr=1.1697e-03) (hash(x)=74423275)
|
| 343 |
+
2830 train 3.979952 (lr=1.1679e-03) (hash(x)=80921960)
|
| 344 |
+
2840 train 4.276316 (lr=1.1660e-03) (hash(x)=79254948)
|
| 345 |
+
2850 train 3.997131 (lr=1.1641e-03) (hash(x)=73700080)
|
| 346 |
+
2860 train 4.176439 (lr=1.1623e-03) (hash(x)=74328412)
|
| 347 |
+
2870 train 4.024445 (lr=1.1604e-03) (hash(x)=74359259)
|
| 348 |
+
2880 train 4.226033 (lr=1.1585e-03) (hash(x)=74479648)
|
| 349 |
+
2890 train 3.877771 (lr=1.1566e-03) (hash(x)=80993243)
|
| 350 |
+
2900 val loss 4.0561
|
| 351 |
+
2900 val perplexity 57.7482
|
| 352 |
+
2900 train 4.102278 (lr=1.1547e-03) (hash(x)=72286433)
|
| 353 |
+
2910 train 4.233449 (lr=1.1528e-03) (hash(x)=81081634)
|
| 354 |
+
2920 train 3.995192 (lr=1.1509e-03) (hash(x)=71809089)
|
| 355 |
+
2930 train 4.031490 (lr=1.1490e-03) (hash(x)=72668797)
|
| 356 |
+
2940 train 4.205581 (lr=1.1471e-03) (hash(x)=80472855)
|
| 357 |
+
2950 train 4.073675 (lr=1.1451e-03) (hash(x)=67764485)
|
| 358 |
+
2960 train 3.906231 (lr=1.1432e-03) (hash(x)=78590385)
|
| 359 |
+
2970 train 4.168770 (lr=1.1413e-03) (hash(x)=86434483)
|
| 360 |
+
2980 train 4.081698 (lr=1.1393e-03) (hash(x)=83872337)
|
| 361 |
+
2990 train 4.006442 (lr=1.1374e-03) (hash(x)=76866271)
|
| 362 |
+
3000 val loss 4.0585
|
| 363 |
+
3000 val perplexity 57.8901
|
| 364 |
+
3000 train 3.940145 (lr=1.1354e-03) (hash(x)=98889128)
|
| 365 |
+
3010 train 3.829953 (lr=1.1335e-03) (hash(x)=77128749)
|
| 366 |
+
3020 train 3.933706 (lr=1.1315e-03) (hash(x)=74427692)
|
| 367 |
+
3030 train 4.166184 (lr=1.1295e-03) (hash(x)=75155791)
|
| 368 |
+
3040 train 3.773731 (lr=1.1276e-03) (hash(x)=80384590)
|
| 369 |
+
3050 train 3.839023 (lr=1.1256e-03) (hash(x)=69078039)
|
| 370 |
+
3060 train 3.856933 (lr=1.1236e-03) (hash(x)=75289844)
|
| 371 |
+
3070 train 3.660677 (lr=1.1216e-03) (hash(x)=73776546)
|
| 372 |
+
3080 train 3.965837 (lr=1.1196e-03) (hash(x)=83632057)
|
| 373 |
+
3090 train 4.114454 (lr=1.1176e-03) (hash(x)=83219352)
|
| 374 |
+
3100 val loss 4.0804
|
| 375 |
+
3100 val perplexity 59.1700
|
| 376 |
+
3100 train 3.720832 (lr=1.1156e-03) (hash(x)=72986705)
|
| 377 |
+
3110 train 4.154251 (lr=1.1136e-03) (hash(x)=79090150)
|
| 378 |
+
3120 train 3.910917 (lr=1.1116e-03) (hash(x)=82919368)
|
| 379 |
+
3130 train 4.173755 (lr=1.1096e-03) (hash(x)=82752003)
|
| 380 |
+
3140 train 4.196417 (lr=1.1076e-03) (hash(x)=83049547)
|
| 381 |
+
3150 train 4.126358 (lr=1.1055e-03) (hash(x)=76384003)
|
| 382 |
+
3160 train 4.099752 (lr=1.1035e-03) (hash(x)=83069546)
|
| 383 |
+
3170 train 4.130581 (lr=1.1015e-03) (hash(x)=77245329)
|
| 384 |
+
3180 train 3.919234 (lr=1.0994e-03) (hash(x)=80555988)
|
| 385 |
+
3190 train 4.023094 (lr=1.0974e-03) (hash(x)=79506138)
|
| 386 |
+
3200 val loss 4.0164
|
| 387 |
+
3200 val perplexity 55.5022
|
| 388 |
+
3200 train 3.969558 (lr=1.0953e-03) (hash(x)=74238910)
|
| 389 |
+
3210 train 4.106257 (lr=1.0933e-03) (hash(x)=87246704)
|
| 390 |
+
3220 train 3.939982 (lr=1.0912e-03) (hash(x)=76286650)
|
| 391 |
+
3230 train 3.918532 (lr=1.0891e-03) (hash(x)=74128366)
|
| 392 |
+
3240 train 3.936282 (lr=1.0871e-03) (hash(x)=83686518)
|
| 393 |
+
3250 train 4.062903 (lr=1.0850e-03) (hash(x)=79334239)
|
| 394 |
+
3260 train 3.848360 (lr=1.0829e-03) (hash(x)=76991073)
|
| 395 |
+
3270 train 3.867776 (lr=1.0808e-03) (hash(x)=79645991)
|
| 396 |
+
3280 train 3.901214 (lr=1.0787e-03) (hash(x)=79642315)
|
| 397 |
+
3290 train 4.046340 (lr=1.0767e-03) (hash(x)=71394376)
|
| 398 |
+
3300 val loss 4.0155
|
| 399 |
+
3300 val perplexity 55.4514
|
| 400 |
+
3300 train 3.771777 (lr=1.0746e-03) (hash(x)=74335345)
|
| 401 |
+
3310 train 3.851122 (lr=1.0725e-03) (hash(x)=77872427)
|
| 402 |
+
3320 train 3.766908 (lr=1.0703e-03) (hash(x)=65707820)
|
| 403 |
+
3330 train 4.076684 (lr=1.0682e-03) (hash(x)=69949153)
|
| 404 |
+
3340 train 3.950765 (lr=1.0661e-03) (hash(x)=74387552)
|
| 405 |
+
3350 train 4.084062 (lr=1.0640e-03) (hash(x)=80320961)
|
| 406 |
+
3360 train 4.121322 (lr=1.0619e-03) (hash(x)=75749684)
|
| 407 |
+
3370 train 3.964305 (lr=1.0597e-03) (hash(x)=78610377)
|
| 408 |
+
3380 train 4.016845 (lr=1.0576e-03) (hash(x)=73378836)
|
| 409 |
+
3390 train 4.063757 (lr=1.0555e-03) (hash(x)=73242237)
|
| 410 |
+
3400 val loss 3.9942
|
| 411 |
+
3400 val perplexity 54.2809
|
| 412 |
+
3400 train 4.040943 (lr=1.0533e-03) (hash(x)=68240434)
|
| 413 |
+
3410 train 3.853514 (lr=1.0512e-03) (hash(x)=80087540)
|
| 414 |
+
3420 train 4.308168 (lr=1.0490e-03) (hash(x)=73529815)
|
| 415 |
+
3430 train 4.195695 (lr=1.0469e-03) (hash(x)=73306132)
|
| 416 |
+
3440 train 4.331204 (lr=1.0447e-03) (hash(x)=70500493)
|
| 417 |
+
3450 train 4.084498 (lr=1.0426e-03) (hash(x)=70007926)
|
| 418 |
+
3460 train 4.034789 (lr=1.0404e-03) (hash(x)=70024043)
|
| 419 |
+
3470 train 4.115597 (lr=1.0382e-03) (hash(x)=81605862)
|
| 420 |
+
3480 train 3.976336 (lr=1.0361e-03) (hash(x)=67926367)
|
| 421 |
+
3490 train 4.029185 (lr=1.0339e-03) (hash(x)=75476865)
|
| 422 |
+
3500 val loss 3.9781
|
| 423 |
+
3500 val perplexity 53.4161
|
| 424 |
+
3500 train 3.878834 (lr=1.0317e-03) (hash(x)=76813079)
|
| 425 |
+
3510 train 4.049231 (lr=1.0295e-03) (hash(x)=83284900)
|
| 426 |
+
3520 train 4.035156 (lr=1.0273e-03) (hash(x)=72533794)
|
| 427 |
+
3530 train 3.946522 (lr=1.0251e-03) (hash(x)=71496579)
|
| 428 |
+
3540 train 3.918399 (lr=1.0230e-03) (hash(x)=70592072)
|
| 429 |
+
3550 train 4.054464 (lr=1.0208e-03) (hash(x)=76928693)
|
| 430 |
+
3560 train 3.902505 (lr=1.0186e-03) (hash(x)=74503456)
|
| 431 |
+
3570 train 3.997057 (lr=1.0163e-03) (hash(x)=70401709)
|
| 432 |
+
3580 train 4.067445 (lr=1.0141e-03) (hash(x)=71225931)
|
| 433 |
+
3590 train 4.064699 (lr=1.0119e-03) (hash(x)=79059680)
|
| 434 |
+
3600 val loss 3.9713
|
| 435 |
+
3600 val perplexity 53.0533
|
| 436 |
+
3600 train 3.790024 (lr=1.0097e-03) (hash(x)=71650260)
|
| 437 |
+
3610 train 3.758954 (lr=1.0075e-03) (hash(x)=72844252)
|
| 438 |
+
3620 train 3.794262 (lr=1.0053e-03) (hash(x)=69989480)
|
| 439 |
+
3630 train 3.758096 (lr=1.0030e-03) (hash(x)=74336566)
|
| 440 |
+
3640 train 3.898620 (lr=1.0008e-03) (hash(x)=60353606)
|
| 441 |
+
3650 train 3.840124 (lr=9.9857e-04) (hash(x)=79424308)
|
| 442 |
+
3660 train 3.975306 (lr=9.9633e-04) (hash(x)=79976421)
|
| 443 |
+
3670 train 3.737185 (lr=9.9409e-04) (hash(x)=76742350)
|
| 444 |
+
3680 train 3.869838 (lr=9.9185e-04) (hash(x)=73225113)
|
| 445 |
+
3690 train 3.852203 (lr=9.8960e-04) (hash(x)=68913999)
|
| 446 |
+
3700 val loss 3.9822
|
| 447 |
+
3700 val perplexity 53.6373
|
| 448 |
+
3700 train 3.805826 (lr=9.8735e-04) (hash(x)=73295022)
|
| 449 |
+
3710 train 3.689860 (lr=9.8510e-04) (hash(x)=71341621)
|
| 450 |
+
3720 train 3.860883 (lr=9.8284e-04) (hash(x)=76711631)
|
| 451 |
+
3730 train 3.891455 (lr=9.8058e-04) (hash(x)=76294927)
|
| 452 |
+
3740 train 4.061416 (lr=9.7832e-04) (hash(x)=81950742)
|
| 453 |
+
3750 train 3.864950 (lr=9.7605e-04) (hash(x)=74343197)
|
| 454 |
+
3760 train 4.054268 (lr=9.7378e-04) (hash(x)=71093207)
|
| 455 |
+
3770 train 4.010053 (lr=9.7151e-04) (hash(x)=80706801)
|
| 456 |
+
3780 train 4.011995 (lr=9.6924e-04) (hash(x)=75011477)
|
| 457 |
+
3790 train 4.025649 (lr=9.6696e-04) (hash(x)=72248633)
|
| 458 |
+
3800 val loss 3.9514
|
| 459 |
+
3800 val perplexity 52.0095
|
| 460 |
+
3800 train 3.948764 (lr=9.6468e-04) (hash(x)=69136971)
|
| 461 |
+
3810 train 4.004524 (lr=9.6240e-04) (hash(x)=77522202)
|
| 462 |
+
3820 train 3.787655 (lr=9.6011e-04) (hash(x)=67695101)
|
| 463 |
+
3830 train 3.975669 (lr=9.5782e-04) (hash(x)=80051551)
|
| 464 |
+
3840 train 3.894165 (lr=9.5553e-04) (hash(x)=74191647)
|
| 465 |
+
3850 train 3.844064 (lr=9.5324e-04) (hash(x)=68047171)
|
| 466 |
+
3860 train 3.943107 (lr=9.5094e-04) (hash(x)=70045527)
|
| 467 |
+
3870 train 3.968391 (lr=9.4864e-04) (hash(x)=73697380)
|
| 468 |
+
3880 train 3.932387 (lr=9.4634e-04) (hash(x)=72925733)
|
| 469 |
+
3890 train 3.989490 (lr=9.4404e-04) (hash(x)=67921912)
|
| 470 |
+
3900 val loss 3.9451
|
| 471 |
+
3900 val perplexity 51.6798
|
| 472 |
+
3900 train 4.048510 (lr=9.4173e-04) (hash(x)=78738821)
|
| 473 |
+
3910 train 3.956899 (lr=9.3942e-04) (hash(x)=74879268)
|
| 474 |
+
3920 train 4.037656 (lr=9.3711e-04) (hash(x)=67259305)
|
| 475 |
+
3930 train 3.935457 (lr=9.3479e-04) (hash(x)=75660066)
|
| 476 |
+
3940 train 3.999289 (lr=9.3248e-04) (hash(x)=74572945)
|
| 477 |
+
3950 train 3.892618 (lr=9.3016e-04) (hash(x)=82744417)
|
| 478 |
+
3960 train 3.776059 (lr=9.2784e-04) (hash(x)=76023482)
|
| 479 |
+
3970 train 3.828683 (lr=9.2551e-04) (hash(x)=74825305)
|
| 480 |
+
3980 train 4.284902 (lr=9.2319e-04) (hash(x)=75541926)
|
| 481 |
+
3990 train 3.954080 (lr=9.2086e-04) (hash(x)=85705002)
|
| 482 |
+
4000 val loss 3.9352
|
| 483 |
+
4000 val perplexity 51.1704
|
| 484 |
+
4000 train 3.925224 (lr=9.1853e-04) (hash(x)=75062236)
|
| 485 |
+
4010 train 3.752635 (lr=9.1620e-04) (hash(x)=70532791)
|
| 486 |
+
4020 train 4.004159 (lr=9.1386e-04) (hash(x)=70014939)
|
| 487 |
+
4030 train 3.794122 (lr=9.1152e-04) (hash(x)=75315953)
|
| 488 |
+
4040 train 3.886059 (lr=9.0919e-04) (hash(x)=74206277)
|
| 489 |
+
4050 train 4.026473 (lr=9.0684e-04) (hash(x)=71678934)
|
| 490 |
+
4060 train 3.988764 (lr=9.0450e-04) (hash(x)=68169236)
|
| 491 |
+
4070 train 3.876254 (lr=9.0216e-04) (hash(x)=73440206)
|
| 492 |
+
4080 train 3.824935 (lr=8.9981e-04) (hash(x)=77333014)
|
| 493 |
+
4090 train 3.770763 (lr=8.9746e-04) (hash(x)=74699627)
|
| 494 |
+
4100 val loss 3.9517
|
| 495 |
+
4100 val perplexity 52.0257
|
| 496 |
+
4100 train 3.912891 (lr=8.9511e-04) (hash(x)=72537935)
|
| 497 |
+
4110 train 3.759724 (lr=8.9276e-04) (hash(x)=59960704)
|
| 498 |
+
4120 train 3.808694 (lr=8.9041e-04) (hash(x)=79504237)
|
| 499 |
+
4130 train 3.938347 (lr=8.8805e-04) (hash(x)=76513677)
|
| 500 |
+
4140 train 4.053992 (lr=8.8569e-04) (hash(x)=81601150)
|
| 501 |
+
4150 train 3.889416 (lr=8.8333e-04) (hash(x)=75447369)
|
| 502 |
+
4160 train 3.920038 (lr=8.8097e-04) (hash(x)=84280875)
|
| 503 |
+
4170 train 4.079497 (lr=8.7861e-04) (hash(x)=88239587)
|
| 504 |
+
4180 train 4.120664 (lr=8.7625e-04) (hash(x)=69175879)
|
| 505 |
+
4190 train 4.012738 (lr=8.7388e-04) (hash(x)=71751070)
|
| 506 |
+
4200 val loss 3.9053
|
| 507 |
+
4200 val perplexity 49.6642
|
| 508 |
+
4200 train 3.893376 (lr=8.7151e-04) (hash(x)=63098921)
|
| 509 |
+
4210 train 4.593186 (lr=8.6915e-04) (hash(x)=99229132)
|
| 510 |
+
4220 train 3.889224 (lr=8.6678e-04) (hash(x)=74149842)
|
| 511 |
+
4230 train 3.861020 (lr=8.6440e-04) (hash(x)=76000344)
|
| 512 |
+
4240 train 3.824990 (lr=8.6203e-04) (hash(x)=73033995)
|
| 513 |
+
4250 train 3.839565 (lr=8.5966e-04) (hash(x)=79351091)
|
| 514 |
+
4260 train 3.855897 (lr=8.5728e-04) (hash(x)=67587193)
|
| 515 |
+
4270 train 3.896488 (lr=8.5491e-04) (hash(x)=80457848)
|
| 516 |
+
4280 train 3.760039 (lr=8.5253e-04) (hash(x)=79619630)
|
| 517 |
+
4290 train 3.943452 (lr=8.5015e-04) (hash(x)=76884730)
|
| 518 |
+
4300 val loss 3.8894
|
| 519 |
+
4300 val perplexity 48.8794
|
| 520 |
+
4300 train 3.799943 (lr=8.4777e-04) (hash(x)=79561461)
|
| 521 |
+
4310 train 3.871903 (lr=8.4539e-04) (hash(x)=80388562)
|
| 522 |
+
4320 train 3.911822 (lr=8.4301e-04) (hash(x)=78267207)
|
| 523 |
+
4330 train 3.811517 (lr=8.4062e-04) (hash(x)=71301301)
|
| 524 |
+
4340 train 3.987394 (lr=8.3824e-04) (hash(x)=79138432)
|
| 525 |
+
4350 train 3.900640 (lr=8.3585e-04) (hash(x)=70573215)
|
| 526 |
+
4360 train 3.864222 (lr=8.3347e-04) (hash(x)=71492280)
|
| 527 |
+
4370 train 3.850598 (lr=8.3108e-04) (hash(x)=74751091)
|
| 528 |
+
4380 train 3.872483 (lr=8.2869e-04) (hash(x)=72310274)
|
| 529 |
+
4390 train 3.828910 (lr=8.2630e-04) (hash(x)=100848104)
|
| 530 |
+
4400 val loss 3.8936
|
| 531 |
+
4400 val perplexity 49.0892
|
| 532 |
+
4400 train 4.096554 (lr=8.2391e-04) (hash(x)=71196998)
|
| 533 |
+
4410 train 3.839976 (lr=8.2152e-04) (hash(x)=74205624)
|
| 534 |
+
4420 train 3.796909 (lr=8.1913e-04) (hash(x)=68781238)
|
| 535 |
+
4430 train 3.831192 (lr=8.1674e-04) (hash(x)=76961213)
|
| 536 |
+
4440 train 3.804280 (lr=8.1435e-04) (hash(x)=80006728)
|
| 537 |
+
4450 train 3.760864 (lr=8.1195e-04) (hash(x)=71337984)
|
| 538 |
+
4460 train 3.906299 (lr=8.0956e-04) (hash(x)=70514009)
|
| 539 |
+
4470 train 3.748165 (lr=8.0716e-04) (hash(x)=76238847)
|
| 540 |
+
4480 train 3.793830 (lr=8.0477e-04) (hash(x)=67646620)
|
| 541 |
+
4490 train 3.832017 (lr=8.0237e-04) (hash(x)=76200722)
|
| 542 |
+
4500 val loss 3.9079
|
| 543 |
+
4500 val perplexity 49.7960
|
| 544 |
+
4500 train 4.007967 (lr=7.9998e-04) (hash(x)=80808376)
|
| 545 |
+
4510 train 3.855742 (lr=7.9758e-04) (hash(x)=79670252)
|
| 546 |
+
4520 train 3.675745 (lr=7.9518e-04) (hash(x)=71430874)
|
| 547 |
+
4530 train 3.852278 (lr=7.9279e-04) (hash(x)=74443861)
|
| 548 |
+
4540 train 3.951531 (lr=7.9039e-04) (hash(x)=84712247)
|
| 549 |
+
4550 train 3.927889 (lr=7.8799e-04) (hash(x)=76296651)
|
| 550 |
+
4560 train 3.975340 (lr=7.8559e-04) (hash(x)=76991150)
|
| 551 |
+
4570 train 3.908820 (lr=7.8319e-04) (hash(x)=68369483)
|
| 552 |
+
4580 train 3.868467 (lr=7.8080e-04) (hash(x)=68777608)
|
| 553 |
+
4590 train 4.026533 (lr=7.7840e-04) (hash(x)=73172132)
|
| 554 |
+
4600 val loss 3.8651
|
| 555 |
+
4600 val perplexity 47.7100
|
| 556 |
+
4600 train 3.963861 (lr=7.7600e-04) (hash(x)=78012961)
|
| 557 |
+
4610 train 3.979441 (lr=7.7360e-04) (hash(x)=76270600)
|
| 558 |
+
4620 train 3.786058 (lr=7.7120e-04) (hash(x)=82249567)
|
| 559 |
+
4630 train 3.852795 (lr=7.6880e-04) (hash(x)=67713698)
|
| 560 |
+
4640 train 4.020086 (lr=7.6640e-04) (hash(x)=54462324)
|
| 561 |
+
4650 train 3.953427 (lr=7.6400e-04) (hash(x)=70825784)
|
| 562 |
+
4660 train 4.048795 (lr=7.6160e-04) (hash(x)=69303774)
|
| 563 |
+
4670 train 3.950115 (lr=7.5920e-04) (hash(x)=84155716)
|
| 564 |
+
4680 train 4.063286 (lr=7.5681e-04) (hash(x)=76409129)
|
| 565 |
+
4690 train 4.041676 (lr=7.5441e-04) (hash(x)=73455805)
|
| 566 |
+
4700 val loss 3.8597
|
| 567 |
+
4700 val perplexity 47.4512
|
| 568 |
+
4700 train 3.874127 (lr=7.5201e-04) (hash(x)=82597167)
|
| 569 |
+
4710 train 3.820425 (lr=7.4961e-04) (hash(x)=72293751)
|
| 570 |
+
4720 train 3.831949 (lr=7.4721e-04) (hash(x)=77164608)
|
| 571 |
+
4730 train 3.853704 (lr=7.4482e-04) (hash(x)=73781035)
|
| 572 |
+
4740 train 3.815444 (lr=7.4242e-04) (hash(x)=69426231)
|
| 573 |
+
4750 train 3.834098 (lr=7.4002e-04) (hash(x)=77279505)
|
| 574 |
+
4760 train 3.883283 (lr=7.3763e-04) (hash(x)=83295292)
|
| 575 |
+
4770 train 3.759319 (lr=7.3523e-04) (hash(x)=76817389)
|
| 576 |
+
4780 train 3.810769 (lr=7.3284e-04) (hash(x)=76512693)
|
| 577 |
+
4790 train 3.730307 (lr=7.3044e-04) (hash(x)=76469593)
|
| 578 |
+
4800 val loss 3.8548
|
| 579 |
+
4800 val perplexity 47.2170
|
| 580 |
+
4800 train 3.883495 (lr=7.2805e-04) (hash(x)=75145475)
|
| 581 |
+
4810 train 3.802714 (lr=7.2565e-04) (hash(x)=72173130)
|
| 582 |
+
4820 train 3.991738 (lr=7.2326e-04) (hash(x)=74239752)
|
| 583 |
+
4830 train 3.852606 (lr=7.2087e-04) (hash(x)=71592976)
|
| 584 |
+
4840 train 3.800125 (lr=7.1848e-04) (hash(x)=70809896)
|
| 585 |
+
4850 train 3.771695 (lr=7.1609e-04) (hash(x)=72044692)
|
| 586 |
+
4860 train 3.666957 (lr=7.1370e-04) (hash(x)=71791939)
|
| 587 |
+
4870 train 3.680958 (lr=7.1131e-04) (hash(x)=72591349)
|
| 588 |
+
4880 train 3.837759 (lr=7.0892e-04) (hash(x)=83767466)
|
| 589 |
+
4890 train 3.838070 (lr=7.0653e-04) (hash(x)=76798861)
|
| 590 |
+
4900 val loss 3.8535
|
| 591 |
+
4900 val perplexity 47.1601
|
| 592 |
+
4900 train 3.872662 (lr=7.0415e-04) (hash(x)=74555348)
|
| 593 |
+
4910 train 3.839625 (lr=7.0176e-04) (hash(x)=75894534)
|
| 594 |
+
4920 train 3.911762 (lr=6.9938e-04) (hash(x)=84093086)
|
| 595 |
+
4930 train 3.885873 (lr=6.9699e-04) (hash(x)=77209960)
|
| 596 |
+
4940 train 3.875981 (lr=6.9461e-04) (hash(x)=78448905)
|
| 597 |
+
4950 train 3.750815 (lr=6.9223e-04) (hash(x)=74336001)
|
| 598 |
+
4960 train 3.898740 (lr=6.8985e-04) (hash(x)=80895303)
|
| 599 |
+
4970 train 3.860817 (lr=6.8747e-04) (hash(x)=71883820)
|
| 600 |
+
4980 train 3.776807 (lr=6.8509e-04) (hash(x)=69893767)
|
| 601 |
+
4990 train 3.729636 (lr=6.8272e-04) (hash(x)=74660816)
|
| 602 |
+
5000 val loss 3.8450
|
| 603 |
+
5000 val perplexity 46.7588
|
| 604 |
+
5000 train 4.052056 (lr=6.8034e-04) (hash(x)=78057851)
|
| 605 |
+
5010 train 3.844744 (lr=6.7797e-04) (hash(x)=64235135)
|
| 606 |
+
5020 train 3.813846 (lr=6.7560e-04) (hash(x)=70397272)
|
| 607 |
+
5030 train 3.833499 (lr=6.7322e-04) (hash(x)=67171768)
|
| 608 |
+
5040 train 3.753875 (lr=6.7085e-04) (hash(x)=70332490)
|
| 609 |
+
5050 train 3.903934 (lr=6.6849e-04) (hash(x)=83006433)
|
| 610 |
+
5060 train 3.868190 (lr=6.6612e-04) (hash(x)=76575571)
|
| 611 |
+
5070 train 3.865733 (lr=6.6375e-04) (hash(x)=77691770)
|
| 612 |
+
5080 train 4.241705 (lr=6.6139e-04) (hash(x)=57255988)
|
| 613 |
+
5090 train 3.876283 (lr=6.5903e-04) (hash(x)=80108944)
|
| 614 |
+
5100 val loss 3.8362
|
| 615 |
+
5100 val perplexity 46.3503
|
| 616 |
+
5100 train 3.797772 (lr=6.5667e-04) (hash(x)=74319538)
|
| 617 |
+
5110 train 3.780989 (lr=6.5431e-04) (hash(x)=72033986)
|
| 618 |
+
5120 train 3.890330 (lr=6.5195e-04) (hash(x)=75108610)
|
| 619 |
+
5130 train 3.938090 (lr=6.4959e-04) (hash(x)=77275710)
|
| 620 |
+
5140 train 3.818253 (lr=6.4724e-04) (hash(x)=69731327)
|
| 621 |
+
5150 train 3.815135 (lr=6.4489e-04) (hash(x)=74809572)
|
| 622 |
+
5160 train 3.905946 (lr=6.4254e-04) (hash(x)=80883994)
|
| 623 |
+
5170 train 3.820192 (lr=6.4019e-04) (hash(x)=70390166)
|
| 624 |
+
5180 train 3.986687 (lr=6.3784e-04) (hash(x)=75614276)
|
| 625 |
+
5190 train 3.940828 (lr=6.3550e-04) (hash(x)=75431389)
|
| 626 |
+
5200 val loss 3.8231
|
| 627 |
+
5200 val perplexity 45.7467
|
| 628 |
+
5200 train 3.829622 (lr=6.3316e-04) (hash(x)=76239548)
|
| 629 |
+
5210 train 3.932194 (lr=6.3081e-04) (hash(x)=71988616)
|
| 630 |
+
5220 train 3.871243 (lr=6.2848e-04) (hash(x)=70379284)
|
| 631 |
+
5230 train 3.785189 (lr=6.2614e-04) (hash(x)=65937825)
|
| 632 |
+
5240 train 3.741514 (lr=6.2380e-04) (hash(x)=75529148)
|
| 633 |
+
5250 train 3.644694 (lr=6.2147e-04) (hash(x)=73186404)
|
| 634 |
+
5260 train 3.814425 (lr=6.1914e-04) (hash(x)=64888735)
|
| 635 |
+
5270 train 3.735037 (lr=6.1681e-04) (hash(x)=61047756)
|
| 636 |
+
5280 train 3.743160 (lr=6.1449e-04) (hash(x)=72880277)
|
| 637 |
+
5290 train 3.867976 (lr=6.1216e-04) (hash(x)=76312981)
|
| 638 |
+
5300 val loss 3.8137
|
| 639 |
+
5300 val perplexity 45.3158
|
| 640 |
+
5300 train 3.781587 (lr=6.0984e-04) (hash(x)=79007351)
|
| 641 |
+
5310 train 3.645232 (lr=6.0752e-04) (hash(x)=69120142)
|
| 642 |
+
5320 train 3.776814 (lr=6.0521e-04) (hash(x)=69703140)
|
| 643 |
+
5330 train 3.896553 (lr=6.0289e-04) (hash(x)=67515165)
|
| 644 |
+
5340 train 3.808111 (lr=6.0058e-04) (hash(x)=57789017)
|
| 645 |
+
5350 train 3.811449 (lr=5.9827e-04) (hash(x)=79434224)
|
| 646 |
+
5360 train 3.930370 (lr=5.9596e-04) (hash(x)=81242201)
|
| 647 |
+
5370 train 3.805331 (lr=5.9366e-04) (hash(x)=69646244)
|
| 648 |
+
5380 train 3.768810 (lr=5.9136e-04) (hash(x)=81911267)
|
| 649 |
+
5390 train 3.948683 (lr=5.8906e-04) (hash(x)=79086963)
|
| 650 |
+
5400 val loss 3.7922
|
| 651 |
+
5400 val perplexity 44.3526
|
| 652 |
+
5400 train 3.932899 (lr=5.8676e-04) (hash(x)=73958857)
|
| 653 |
+
5410 train 3.815255 (lr=5.8447e-04) (hash(x)=76487961)
|
| 654 |
+
5420 train 3.779708 (lr=5.8218e-04) (hash(x)=75600730)
|
| 655 |
+
5430 train 3.868872 (lr=5.7989e-04) (hash(x)=74683606)
|
| 656 |
+
5440 train 3.807911 (lr=5.7760e-04) (hash(x)=70879811)
|
| 657 |
+
5450 train 3.908965 (lr=5.7532e-04) (hash(x)=71654210)
|
| 658 |
+
5460 train 3.917725 (lr=5.7304e-04) (hash(x)=75386299)
|
| 659 |
+
5470 train 3.863113 (lr=5.7076e-04) (hash(x)=74366370)
|
| 660 |
+
5480 train 3.769624 (lr=5.6849e-04) (hash(x)=74267515)
|
| 661 |
+
5490 train 3.740507 (lr=5.6622e-04) (hash(x)=74435634)
|
| 662 |
+
5500 val loss 3.7844
|
| 663 |
+
5500 val perplexity 44.0101
|
| 664 |
+
5500 train 3.895658 (lr=5.6395e-04) (hash(x)=72000576)
|
| 665 |
+
5510 train 3.837529 (lr=5.6168e-04) (hash(x)=63862668)
|
| 666 |
+
5520 train 3.809539 (lr=5.5942e-04) (hash(x)=70635849)
|
| 667 |
+
5530 train 3.840061 (lr=5.5716e-04) (hash(x)=71094650)
|
| 668 |
+
5540 train 3.696723 (lr=5.5490e-04) (hash(x)=75063498)
|
| 669 |
+
5550 train 4.216496 (lr=5.5265e-04) (hash(x)=71807986)
|
| 670 |
+
5560 train 3.664000 (lr=5.5040e-04) (hash(x)=66657503)
|
| 671 |
+
5570 train 3.806708 (lr=5.4815e-04) (hash(x)=66345202)
|
| 672 |
+
5580 train 3.857528 (lr=5.4591e-04) (hash(x)=71849709)
|
| 673 |
+
5590 train 3.688624 (lr=5.4367e-04) (hash(x)=84702367)
|
| 674 |
+
5600 val loss 3.7861
|
| 675 |
+
5600 val perplexity 44.0825
|
| 676 |
+
5600 train 3.826987 (lr=5.4143e-04) (hash(x)=79109269)
|
| 677 |
+
5610 train 3.617103 (lr=5.3920e-04) (hash(x)=79051062)
|
| 678 |
+
5620 train 3.775809 (lr=5.3697e-04) (hash(x)=69861636)
|
| 679 |
+
5630 train 3.824331 (lr=5.3474e-04) (hash(x)=71549879)
|
| 680 |
+
5640 train 3.789013 (lr=5.3252e-04) (hash(x)=71055279)
|
| 681 |
+
5650 train 3.839933 (lr=5.3030e-04) (hash(x)=76078002)
|
| 682 |
+
5660 train 3.752698 (lr=5.2808e-04) (hash(x)=74335344)
|
| 683 |
+
5670 train 3.927066 (lr=5.2587e-04) (hash(x)=71454672)
|
| 684 |
+
5680 train 3.703490 (lr=5.2366e-04) (hash(x)=68592905)
|
| 685 |
+
5690 train 4.015969 (lr=5.2145e-04) (hash(x)=72942879)
|
| 686 |
+
5700 val loss 3.7741
|
| 687 |
+
5700 val perplexity 43.5587
|
| 688 |
+
5700 train 3.723315 (lr=5.1925e-04) (hash(x)=65107511)
|
| 689 |
+
5710 train 3.783133 (lr=5.1705e-04) (hash(x)=82256907)
|
| 690 |
+
5720 train 3.859226 (lr=5.1485e-04) (hash(x)=83454690)
|
| 691 |
+
5730 train 3.935768 (lr=5.1266e-04) (hash(x)=74898617)
|
| 692 |
+
5740 train 3.744740 (lr=5.1047e-04) (hash(x)=74025050)
|
| 693 |
+
5750 train 3.868131 (lr=5.0829e-04) (hash(x)=75125832)
|
| 694 |
+
5760 train 3.889325 (lr=5.0611e-04) (hash(x)=82640580)
|
| 695 |
+
5770 train 3.692922 (lr=5.0393e-04) (hash(x)=72169296)
|
| 696 |
+
5780 train 3.711759 (lr=5.0176e-04) (hash(x)=76860672)
|
| 697 |
+
5790 train 3.798202 (lr=4.9959e-04) (hash(x)=75858612)
|
| 698 |
+
5800 val loss 3.7778
|
| 699 |
+
5800 val perplexity 43.7213
|
| 700 |
+
5800 train 3.876491 (lr=4.9743e-04) (hash(x)=68326653)
|
| 701 |
+
5810 train 3.839730 (lr=4.9526e-04) (hash(x)=72416288)
|
| 702 |
+
5820 train 3.728068 (lr=4.9311e-04) (hash(x)=75390575)
|
| 703 |
+
5830 train 3.780803 (lr=4.9095e-04) (hash(x)=74274453)
|
| 704 |
+
5840 train 3.734004 (lr=4.8881e-04) (hash(x)=80802684)
|
| 705 |
+
5850 train 3.759165 (lr=4.8666e-04) (hash(x)=71662543)
|
| 706 |
+
5860 train 3.712453 (lr=4.8452e-04) (hash(x)=78625653)
|
| 707 |
+
5870 train 3.630938 (lr=4.8238e-04) (hash(x)=68749216)
|
| 708 |
+
5880 train 3.732988 (lr=4.8025e-04) (hash(x)=71914758)
|
| 709 |
+
5890 train 3.550396 (lr=4.7812e-04) (hash(x)=73589958)
|
| 710 |
+
5900 val loss 3.7618
|
| 711 |
+
5900 val perplexity 43.0266
|
| 712 |
+
5900 train 3.705497 (lr=4.7600e-04) (hash(x)=72500673)
|
| 713 |
+
5910 train 3.756702 (lr=4.7388e-04) (hash(x)=80963753)
|
| 714 |
+
5920 train 3.895484 (lr=4.7176e-04) (hash(x)=89659262)
|
| 715 |
+
5930 train 3.644107 (lr=4.6965e-04) (hash(x)=81546346)
|
| 716 |
+
5940 train 3.797551 (lr=4.6755e-04) (hash(x)=80298540)
|
| 717 |
+
5950 train 3.796628 (lr=4.6544e-04) (hash(x)=89054333)
|
| 718 |
+
5960 train 3.984135 (lr=4.6335e-04) (hash(x)=76292207)
|
| 719 |
+
5970 train 3.725864 (lr=4.6125e-04) (hash(x)=68170337)
|
| 720 |
+
5980 train 3.906424 (lr=4.5916e-04) (hash(x)=71621953)
|
| 721 |
+
5990 train 3.843237 (lr=4.5708e-04) (hash(x)=76140989)
|
| 722 |
+
6000 val loss 3.7482
|
| 723 |
+
6000 val perplexity 42.4453
|
| 724 |
+
6000 train 3.815634 (lr=4.5500e-04) (hash(x)=68289397)
|
| 725 |
+
6010 train 3.910280 (lr=4.5292e-04) (hash(x)=84960372)
|
| 726 |
+
6020 train 3.752624 (lr=4.5085e-04) (hash(x)=66386985)
|
| 727 |
+
6030 train 3.833742 (lr=4.4879e-04) (hash(x)=76775258)
|
| 728 |
+
6040 train 3.643125 (lr=4.4673e-04) (hash(x)=75546290)
|
| 729 |
+
6050 train 3.620050 (lr=4.4467e-04) (hash(x)=78124215)
|
| 730 |
+
6060 train 3.750315 (lr=4.4262e-04) (hash(x)=70915016)
|
| 731 |
+
6070 train 3.561512 (lr=4.4057e-04) (hash(x)=72509951)
|
| 732 |
+
6080 train 3.734336 (lr=4.3853e-04) (hash(x)=74035398)
|
| 733 |
+
6090 train 3.706557 (lr=4.3649e-04) (hash(x)=69655876)
|
| 734 |
+
6100 val loss 3.7407
|
| 735 |
+
6100 val perplexity 42.1280
|
| 736 |
+
6100 train 3.733545 (lr=4.3446e-04) (hash(x)=77457733)
|
| 737 |
+
6110 train 3.721175 (lr=4.3243e-04) (hash(x)=77781793)
|
| 738 |
+
6120 train 3.788910 (lr=4.3041e-04) (hash(x)=72732355)
|
| 739 |
+
6130 train 3.631293 (lr=4.2839e-04) (hash(x)=74032186)
|
| 740 |
+
6140 train 3.823047 (lr=4.2637e-04) (hash(x)=72856829)
|
| 741 |
+
6150 train 3.904289 (lr=4.2437e-04) (hash(x)=78098180)
|
| 742 |
+
6160 train 3.844294 (lr=4.2236e-04) (hash(x)=73990956)
|
| 743 |
+
6170 train 3.625069 (lr=4.2036e-04) (hash(x)=72632038)
|
| 744 |
+
6180 train 3.670491 (lr=4.1837e-04) (hash(x)=74962782)
|
| 745 |
+
6190 train 3.498389 (lr=4.1638e-04) (hash(x)=81020086)
|
| 746 |
+
6200 val loss 3.7384
|
| 747 |
+
6200 val perplexity 42.0299
|
| 748 |
+
6200 train 3.607941 (lr=4.1440e-04) (hash(x)=87386066)
|
| 749 |
+
6210 train 3.772514 (lr=4.1242e-04) (hash(x)=72382254)
|
| 750 |
+
6220 train 3.925740 (lr=4.1045e-04) (hash(x)=72565426)
|
| 751 |
+
6230 train 3.621914 (lr=4.0848e-04) (hash(x)=71306401)
|
| 752 |
+
6240 train 3.702580 (lr=4.0652e-04) (hash(x)=78821984)
|
| 753 |
+
6250 train 3.673308 (lr=4.0456e-04) (hash(x)=72065375)
|
| 754 |
+
6260 train 3.717981 (lr=4.0261e-04) (hash(x)=76100774)
|
| 755 |
+
6270 train 3.643425 (lr=4.0067e-04) (hash(x)=81896940)
|
| 756 |
+
6280 train 3.547706 (lr=3.9873e-04) (hash(x)=84686730)
|
| 757 |
+
6290 train 3.926195 (lr=3.9679e-04) (hash(x)=78981822)
|
| 758 |
+
6300 val loss 3.7209
|
| 759 |
+
6300 val perplexity 41.3013
|
| 760 |
+
6300 train 3.739345 (lr=3.9486e-04) (hash(x)=72121153)
|
| 761 |
+
6310 train 3.724002 (lr=3.9294e-04) (hash(x)=77322622)
|
| 762 |
+
6320 train 3.752290 (lr=3.9102e-04) (hash(x)=70252312)
|
| 763 |
+
6330 train 3.665483 (lr=3.8910e-04) (hash(x)=71167900)
|
| 764 |
+
6340 train 3.668996 (lr=3.8719e-04) (hash(x)=67968536)
|
| 765 |
+
6350 train 3.732727 (lr=3.8529e-04) (hash(x)=61548959)
|
| 766 |
+
6360 train 3.751049 (lr=3.8339e-04) (hash(x)=76835899)
|
| 767 |
+
6370 train 3.687469 (lr=3.8150e-04) (hash(x)=64612286)
|
| 768 |
+
6380 train 3.751592 (lr=3.7962e-04) (hash(x)=68631278)
|
| 769 |
+
6390 train 3.785738 (lr=3.7774e-04) (hash(x)=76777733)
|
| 770 |
+
6400 val loss 3.7209
|
| 771 |
+
6400 val perplexity 41.3034
|
| 772 |
+
6400 train 3.742000 (lr=3.7586e-04) (hash(x)=74010176)
|
| 773 |
+
6410 train 3.720256 (lr=3.7399e-04) (hash(x)=69368960)
|
| 774 |
+
6420 train 3.741033 (lr=3.7213e-04) (hash(x)=79065455)
|
| 775 |
+
6430 train 3.653737 (lr=3.7027e-04) (hash(x)=77858574)
|
| 776 |
+
6440 train 3.726258 (lr=3.6842e-04) (hash(x)=80136197)
|
| 777 |
+
6450 train 3.695586 (lr=3.6658e-04) (hash(x)=69612887)
|
| 778 |
+
6460 train 3.710082 (lr=3.6474e-04) (hash(x)=75004840)
|
| 779 |
+
6470 train 3.709419 (lr=3.6290e-04) (hash(x)=76506546)
|
| 780 |
+
6480 train 3.687003 (lr=3.6108e-04) (hash(x)=65864207)
|
| 781 |
+
6490 train 3.663729 (lr=3.5925e-04) (hash(x)=89740434)
|
| 782 |
+
6500 val loss 3.7158
|
| 783 |
+
6500 val perplexity 41.0908
|
| 784 |
+
6500 train 3.767954 (lr=3.5744e-04) (hash(x)=74487846)
|
| 785 |
+
6510 train 3.855637 (lr=3.5563e-04) (hash(x)=77407358)
|
| 786 |
+
6520 train 3.510503 (lr=3.5382e-04) (hash(x)=71194824)
|
| 787 |
+
6530 train 3.624622 (lr=3.5203e-04) (hash(x)=67104645)
|
| 788 |
+
6540 train 3.703190 (lr=3.5023e-04) (hash(x)=76148759)
|
| 789 |
+
6550 train 3.676487 (lr=3.4845e-04) (hash(x)=75122573)
|
| 790 |
+
6560 train 3.605250 (lr=3.4667e-04) (hash(x)=78906454)
|
| 791 |
+
6570 train 3.606991 (lr=3.4489e-04) (hash(x)=74690995)
|
| 792 |
+
6580 train 3.637324 (lr=3.4313e-04) (hash(x)=76927297)
|
| 793 |
+
6590 train 3.529497 (lr=3.4137e-04) (hash(x)=68404949)
|
| 794 |
+
6600 val loss 3.7160
|
| 795 |
+
6600 val perplexity 41.0993
|
| 796 |
+
6600 train 3.559453 (lr=3.3961e-04) (hash(x)=70670064)
|
| 797 |
+
6610 train 3.676680 (lr=3.3786e-04) (hash(x)=78186701)
|
| 798 |
+
6620 train 3.623041 (lr=3.3612e-04) (hash(x)=83774586)
|
| 799 |
+
6630 train 3.688436 (lr=3.3438e-04) (hash(x)=75862939)
|
| 800 |
+
6640 train 3.653182 (lr=3.3265e-04) (hash(x)=69660197)
|
| 801 |
+
6650 train 3.659436 (lr=3.3093e-04) (hash(x)=75014695)
|
| 802 |
+
6660 train 3.610671 (lr=3.2921e-04) (hash(x)=72303759)
|
| 803 |
+
6670 train 3.719573 (lr=3.2750e-04) (hash(x)=70865288)
|
| 804 |
+
6680 train 3.748780 (lr=3.2580e-04) (hash(x)=83055332)
|
| 805 |
+
6690 train 3.985394 (lr=3.2410e-04) (hash(x)=74803557)
|
| 806 |
+
6700 val loss 3.7035
|
| 807 |
+
6700 val perplexity 40.5896
|
| 808 |
+
6700 train 3.779985 (lr=3.2241e-04) (hash(x)=72939636)
|
| 809 |
+
6710 train 3.678798 (lr=3.2072e-04) (hash(x)=71614446)
|
| 810 |
+
6720 train 3.801291 (lr=3.1904e-04) (hash(x)=67555962)
|
| 811 |
+
6730 train 3.683165 (lr=3.1737e-04) (hash(x)=65536201)
|
| 812 |
+
6740 train 3.764827 (lr=3.1571e-04) (hash(x)=72108773)
|
| 813 |
+
6750 train 3.549830 (lr=3.1405e-04) (hash(x)=76780817)
|
| 814 |
+
6760 train 3.861091 (lr=3.1240e-04) (hash(x)=78273645)
|
| 815 |
+
6770 train 3.658759 (lr=3.1075e-04) (hash(x)=79605660)
|
| 816 |
+
6780 train 3.716903 (lr=3.0911e-04) (hash(x)=80009237)
|
| 817 |
+
6790 train 3.692259 (lr=3.0748e-04) (hash(x)=74533044)
|
| 818 |
+
6800 val loss 3.6985
|
| 819 |
+
6800 val perplexity 40.3851
|
| 820 |
+
6800 train 3.906264 (lr=3.0585e-04) (hash(x)=80026711)
|
| 821 |
+
6810 train 3.652435 (lr=3.0423e-04) (hash(x)=70338829)
|
| 822 |
+
6820 train 3.618076 (lr=3.0262e-04) (hash(x)=76285045)
|
| 823 |
+
6830 train 3.583163 (lr=3.0102e-04) (hash(x)=74154856)
|
| 824 |
+
6840 train 3.436554 (lr=2.9942e-04) (hash(x)=77355534)
|
| 825 |
+
6850 train 3.622731 (lr=2.9783e-04) (hash(x)=70624171)
|
| 826 |
+
6860 train 3.589119 (lr=2.9624e-04) (hash(x)=67874910)
|
| 827 |
+
6870 train 3.562646 (lr=2.9466e-04) (hash(x)=67935593)
|
| 828 |
+
6880 train 3.669332 (lr=2.9309e-04) (hash(x)=73054733)
|
| 829 |
+
6890 train 3.568755 (lr=2.9153e-04) (hash(x)=71490380)
|
| 830 |
+
6900 val loss 3.6981
|
| 831 |
+
6900 val perplexity 40.3698
|
| 832 |
+
6900 train 3.732036 (lr=2.8997e-04) (hash(x)=77350548)
|
| 833 |
+
6910 train 3.498166 (lr=2.8842e-04) (hash(x)=69026579)
|
| 834 |
+
6920 train 3.802921 (lr=2.8688e-04) (hash(x)=73897252)
|
| 835 |
+
6930 train 3.858767 (lr=2.8534e-04) (hash(x)=79854836)
|
| 836 |
+
6940 train 3.709643 (lr=2.8381e-04) (hash(x)=72056411)
|
| 837 |
+
6950 train 3.664470 (lr=2.8229e-04) (hash(x)=80112011)
|
| 838 |
+
6960 train 3.722703 (lr=2.8078e-04) (hash(x)=57470392)
|
| 839 |
+
6970 train 3.640669 (lr=2.7927e-04) (hash(x)=70347716)
|
| 840 |
+
6980 train 3.819735 (lr=2.7777e-04) (hash(x)=82281687)
|
| 841 |
+
6990 train 3.723538 (lr=2.7627e-04) (hash(x)=74694435)
|
| 842 |
+
7000 val loss 3.6846
|
| 843 |
+
7000 val perplexity 39.8295
|
| 844 |
+
7000 train 3.697739 (lr=2.7479e-04) (hash(x)=81879854)
|
| 845 |
+
7010 train 3.729430 (lr=2.7331e-04) (hash(x)=69165491)
|
| 846 |
+
7020 train 3.632441 (lr=2.7183e-04) (hash(x)=75320851)
|
| 847 |
+
7030 train 3.691919 (lr=2.7037e-04) (hash(x)=77375045)
|
| 848 |
+
7040 train 3.692253 (lr=2.6891e-04) (hash(x)=80127540)
|
| 849 |
+
7050 train 3.521776 (lr=2.6746e-04) (hash(x)=74056421)
|
| 850 |
+
7060 train 3.641689 (lr=2.6602e-04) (hash(x)=76784854)
|
| 851 |
+
7070 train 3.574489 (lr=2.6458e-04) (hash(x)=78122882)
|
| 852 |
+
7080 train 3.551386 (lr=2.6315e-04) (hash(x)=75152878)
|
| 853 |
+
7090 train 3.604688 (lr=2.6173e-04) (hash(x)=79990627)
|
| 854 |
+
7100 val loss 3.6797
|
| 855 |
+
7100 val perplexity 39.6331
|
| 856 |
+
7100 train 3.675170 (lr=2.6032e-04) (hash(x)=70855798)
|
| 857 |
+
7110 train 3.632467 (lr=2.5891e-04) (hash(x)=73080655)
|
| 858 |
+
7120 train 3.652696 (lr=2.5751e-04) (hash(x)=69035245)
|
| 859 |
+
7130 train 3.412613 (lr=2.5612e-04) (hash(x)=64555239)
|
| 860 |
+
7140 train 3.560743 (lr=2.5474e-04) (hash(x)=71734297)
|
| 861 |
+
7150 train 3.595245 (lr=2.5336e-04) (hash(x)=82270339)
|
| 862 |
+
7160 train 3.472945 (lr=2.5199e-04) (hash(x)=76649629)
|
| 863 |
+
7170 train 3.561399 (lr=2.5063e-04) (hash(x)=73236815)
|
| 864 |
+
7180 train 3.463922 (lr=2.4928e-04) (hash(x)=73739555)
|
| 865 |
+
7190 train 3.564978 (lr=2.4793e-04) (hash(x)=65956849)
|
| 866 |
+
7200 val loss 3.6806
|
| 867 |
+
7200 val perplexity 39.6710
|
| 868 |
+
7200 train 3.477367 (lr=2.4659e-04) (hash(x)=76352714)
|
| 869 |
+
7210 train 3.606181 (lr=2.4526e-04) (hash(x)=77523342)
|
| 870 |
+
7220 train 3.666519 (lr=2.4394e-04) (hash(x)=75193820)
|
| 871 |
+
7230 train 3.598501 (lr=2.4262e-04) (hash(x)=72975023)
|
| 872 |
+
7240 train 3.592944 (lr=2.4131e-04) (hash(x)=70152397)
|
| 873 |
+
7250 train 3.591011 (lr=2.4001e-04) (hash(x)=78430692)
|
| 874 |
+
7260 train 3.709949 (lr=2.3872e-04) (hash(x)=73137006)
|
| 875 |
+
7270 train 3.631072 (lr=2.3743e-04) (hash(x)=76187623)
|
| 876 |
+
7280 train 3.658927 (lr=2.3615e-04) (hash(x)=76534981)
|
| 877 |
+
7290 train 3.839647 (lr=2.3488e-04) (hash(x)=77886369)
|
| 878 |
+
7300 val loss 3.6717
|
| 879 |
+
7300 val perplexity 39.3191
|
| 880 |
+
7300 train 3.689902 (lr=2.3362e-04) (hash(x)=73689615)
|
| 881 |
+
7310 train 3.699527 (lr=2.3237e-04) (hash(x)=73658711)
|
| 882 |
+
7320 train 3.559494 (lr=2.3112e-04) (hash(x)=81887275)
|
| 883 |
+
7330 train 3.572438 (lr=2.2988e-04) (hash(x)=69952710)
|
| 884 |
+
7340 train 3.717915 (lr=2.2865e-04) (hash(x)=74902753)
|
| 885 |
+
7350 train 3.614307 (lr=2.2743e-04) (hash(x)=76782773)
|
| 886 |
+
7360 train 3.534500 (lr=2.2621e-04) (hash(x)=73139063)
|
| 887 |
+
7370 train 3.705810 (lr=2.2500e-04) (hash(x)=83222375)
|
| 888 |
+
7380 train 3.634892 (lr=2.2381e-04) (hash(x)=74319008)
|
| 889 |
+
7390 train 3.487841 (lr=2.2261e-04) (hash(x)=79871842)
|
| 890 |
+
7400 val loss 3.6668
|
| 891 |
+
7400 val perplexity 39.1250
|
| 892 |
+
7400 train 3.661918 (lr=2.2143e-04) (hash(x)=74853788)
|
| 893 |
+
7410 train 3.619586 (lr=2.2025e-04) (hash(x)=79119178)
|
| 894 |
+
7420 train 3.566223 (lr=2.1909e-04) (hash(x)=71667016)
|
| 895 |
+
7430 train 3.547193 (lr=2.1793e-04) (hash(x)=70294981)
|
| 896 |
+
7440 train 3.610118 (lr=2.1678e-04) (hash(x)=76749861)
|
| 897 |
+
7450 train 3.431506 (lr=2.1563e-04) (hash(x)=67932362)
|
| 898 |
+
7460 train 3.577607 (lr=2.1450e-04) (hash(x)=71646803)
|
| 899 |
+
7470 train 3.505394 (lr=2.1337e-04) (hash(x)=76832078)
|
| 900 |
+
7480 train 3.541753 (lr=2.1225e-04) (hash(x)=78030848)
|
| 901 |
+
7490 train 3.527037 (lr=2.1114e-04) (hash(x)=79789250)
|
| 902 |
+
7500 val loss 3.6653
|
| 903 |
+
7500 val perplexity 39.0680
|
| 904 |
+
7500 train 3.576111 (lr=2.1003e-04) (hash(x)=74510406)
|
| 905 |
+
7510 train 3.604684 (lr=2.0894e-04) (hash(x)=75354875)
|
| 906 |
+
7520 train 3.634070 (lr=2.0785e-04) (hash(x)=75617004)
|
| 907 |
+
7530 train 3.646052 (lr=2.0677e-04) (hash(x)=70778712)
|
| 908 |
+
7540 train 3.847727 (lr=2.0570e-04) (hash(x)=65376419)
|
| 909 |
+
7550 train 3.655178 (lr=2.0464e-04) (hash(x)=72248752)
|
| 910 |
+
7560 train 3.668872 (lr=2.0358e-04) (hash(x)=73692695)
|
| 911 |
+
7570 train 3.660278 (lr=2.0254e-04) (hash(x)=74787412)
|
| 912 |
+
7580 train 3.739914 (lr=2.0150e-04) (hash(x)=78483480)
|
| 913 |
+
7590 train 3.621348 (lr=2.0047e-04) (hash(x)=73847861)
|
| 914 |
+
7600 val loss 3.6529
|
| 915 |
+
7600 val perplexity 38.5879
|
| 916 |
+
7600 train 3.608267 (lr=1.9945e-04) (hash(x)=71708692)
|
| 917 |
+
7610 train 3.676996 (lr=1.9844e-04) (hash(x)=79089102)
|
| 918 |
+
7620 train 3.556099 (lr=1.9743e-04) (hash(x)=71032201)
|
| 919 |
+
7630 train 3.717488 (lr=1.9643e-04) (hash(x)=71581412)
|
| 920 |
+
7640 train 3.647498 (lr=1.9545e-04) (hash(x)=79778462)
|
| 921 |
+
7650 train 3.839658 (lr=1.9447e-04) (hash(x)=71585919)
|
| 922 |
+
7660 train 3.731389 (lr=1.9349e-04) (hash(x)=72139974)
|
| 923 |
+
7670 train 3.502422 (lr=1.9253e-04) (hash(x)=75620437)
|
| 924 |
+
7680 train 3.744277 (lr=1.9158e-04) (hash(x)=71555417)
|
| 925 |
+
7690 train 3.512033 (lr=1.9063e-04) (hash(x)=63719984)
|
| 926 |
+
7700 val loss 3.6533
|
| 927 |
+
7700 val perplexity 38.6029
|
| 928 |
+
7700 train 3.567901 (lr=1.8969e-04) (hash(x)=74494467)
|
| 929 |
+
7710 train 3.498682 (lr=1.8876e-04) (hash(x)=76821872)
|
| 930 |
+
7720 train 3.515217 (lr=1.8784e-04) (hash(x)=76993734)
|
| 931 |
+
7730 train 3.406214 (lr=1.8693e-04) (hash(x)=74942500)
|
| 932 |
+
7740 train 3.640529 (lr=1.8602e-04) (hash(x)=72583299)
|
| 933 |
+
7750 train 3.400465 (lr=1.8513e-04) (hash(x)=68209490)
|
| 934 |
+
7760 train 3.463583 (lr=1.8424e-04) (hash(x)=69093853)
|
| 935 |
+
7770 train 3.650626 (lr=1.8336e-04) (hash(x)=74460215)
|
| 936 |
+
7780 train 3.579513 (lr=1.8249e-04) (hash(x)=68407147)
|
| 937 |
+
7790 train 3.396707 (lr=1.8163e-04) (hash(x)=76153902)
|
| 938 |
+
7800 val loss 3.6539
|
| 939 |
+
7800 val perplexity 38.6253
|
| 940 |
+
7800 train 3.443745 (lr=1.8078e-04) (hash(x)=84446164)
|
| 941 |
+
7810 train 3.654004 (lr=1.7993e-04) (hash(x)=73681777)
|
| 942 |
+
7820 train 3.673856 (lr=1.7910e-04) (hash(x)=79064713)
|
| 943 |
+
7830 train 3.636933 (lr=1.7827e-04) (hash(x)=72713629)
|
| 944 |
+
7840 train 3.849604 (lr=1.7745e-04) (hash(x)=83622263)
|
| 945 |
+
7850 train 3.613157 (lr=1.7664e-04) (hash(x)=73574852)
|
| 946 |
+
7860 train 3.730911 (lr=1.7584e-04) (hash(x)=67285714)
|
| 947 |
+
7870 train 3.718651 (lr=1.7504e-04) (hash(x)=77222135)
|
| 948 |
+
7880 train 3.746972 (lr=1.7426e-04) (hash(x)=73302118)
|
| 949 |
+
7890 train 3.607560 (lr=1.7348e-04) (hash(x)=61831553)
|
| 950 |
+
7900 val loss 3.6449
|
| 951 |
+
7900 val perplexity 38.2785
|
| 952 |
+
7900 train 3.566517 (lr=1.7271e-04) (hash(x)=70412356)
|
| 953 |
+
7910 train 3.737230 (lr=1.7196e-04) (hash(x)=69987458)
|
| 954 |
+
7920 train 3.709064 (lr=1.7121e-04) (hash(x)=72573609)
|
| 955 |
+
7930 train 3.454720 (lr=1.7046e-04) (hash(x)=67089230)
|
| 956 |
+
7940 train 3.548358 (lr=1.6973e-04) (hash(x)=66572639)
|
| 957 |
+
7950 train 3.661102 (lr=1.6901e-04) (hash(x)=66156765)
|
| 958 |
+
7960 train 3.621139 (lr=1.6829e-04) (hash(x)=75055477)
|
| 959 |
+
7970 train 3.733837 (lr=1.6759e-04) (hash(x)=65210584)
|
| 960 |
+
7980 train 3.539434 (lr=1.6689e-04) (hash(x)=63802308)
|
| 961 |
+
7990 train 3.625554 (lr=1.6620e-04) (hash(x)=76379611)
|
| 962 |
+
8000 val loss 3.6423
|
| 963 |
+
8000 val perplexity 38.1783
|
| 964 |
+
8000 train 3.609458 (lr=1.6552e-04) (hash(x)=68482769)
|
| 965 |
+
8010 train 3.448157 (lr=1.6485e-04) (hash(x)=77934567)
|
| 966 |
+
8020 train 3.536312 (lr=1.6419e-04) (hash(x)=78687646)
|
| 967 |
+
8030 train 3.440270 (lr=1.6353e-04) (hash(x)=73044643)
|
| 968 |
+
8040 train 3.495602 (lr=1.6289e-04) (hash(x)=68884035)
|
| 969 |
+
8050 train 3.321629 (lr=1.6225e-04) (hash(x)=74889605)
|
| 970 |
+
8060 train 3.515282 (lr=1.6162e-04) (hash(x)=76659012)
|
| 971 |
+
8070 train 3.611347 (lr=1.6100e-04) (hash(x)=70403652)
|
| 972 |
+
8080 train 3.466627 (lr=1.6039e-04) (hash(x)=78290747)
|
| 973 |
+
8090 train 3.526579 (lr=1.5979e-04) (hash(x)=81394906)
|
| 974 |
+
8100 val loss 3.6447
|
| 975 |
+
8100 val perplexity 38.2721
|
| 976 |
+
8100 train 3.511029 (lr=1.5920e-04) (hash(x)=67764612)
|
| 977 |
+
8110 train 3.520666 (lr=1.5862e-04) (hash(x)=76334301)
|
| 978 |
+
8120 train 3.581636 (lr=1.5804e-04) (hash(x)=75412335)
|
| 979 |
+
8130 train 3.516148 (lr=1.5748e-04) (hash(x)=74846582)
|
| 980 |
+
8140 train 3.563849 (lr=1.5692e-04) (hash(x)=80776867)
|
| 981 |
+
8150 train 3.596975 (lr=1.5637e-04) (hash(x)=74759651)
|
| 982 |
+
8160 train 3.600293 (lr=1.5583e-04) (hash(x)=75135943)
|
| 983 |
+
8170 train 3.602058 (lr=1.5530e-04) (hash(x)=88547984)
|
| 984 |
+
8180 train 3.677267 (lr=1.5478e-04) (hash(x)=67588135)
|
| 985 |
+
8190 train 3.526083 (lr=1.5427e-04) (hash(x)=75190422)
|
| 986 |
+
8200 val loss 3.6376
|
| 987 |
+
8200 val perplexity 37.9992
|
| 988 |
+
8200 train 3.519204 (lr=1.5377e-04) (hash(x)=79609415)
|
| 989 |
+
8210 train 3.641994 (lr=1.5327e-04) (hash(x)=73291424)
|
| 990 |
+
8220 train 3.640909 (lr=1.5279e-04) (hash(x)=76172342)
|
| 991 |
+
8230 train 3.566577 (lr=1.5231e-04) (hash(x)=72229274)
|
| 992 |
+
8240 train 3.520736 (lr=1.5184e-04) (hash(x)=70393283)
|
| 993 |
+
8250 train 3.593691 (lr=1.5138e-04) (hash(x)=76659165)
|
| 994 |
+
8260 train 3.578496 (lr=1.5094e-04) (hash(x)=75284539)
|
| 995 |
+
8270 train 3.355545 (lr=1.5049e-04) (hash(x)=71495901)
|
| 996 |
+
8280 train 3.584828 (lr=1.5006e-04) (hash(x)=74742339)
|
| 997 |
+
8290 train 3.686184 (lr=1.4964e-04) (hash(x)=73748628)
|
| 998 |
+
8300 val loss 3.6373
|
| 999 |
+
8300 val perplexity 37.9896
|
| 1000 |
+
8300 train 3.665058 (lr=1.4923e-04) (hash(x)=75127072)
|
| 1001 |
+
8310 train 3.507266 (lr=1.4882e-04) (hash(x)=69260375)
|
| 1002 |
+
8320 train 3.530530 (lr=1.4843e-04) (hash(x)=77687746)
|
| 1003 |
+
8330 train 3.460413 (lr=1.4804e-04) (hash(x)=69399950)
|
| 1004 |
+
8340 train 3.678315 (lr=1.4766e-04) (hash(x)=79102450)
|
| 1005 |
+
8350 train 3.615826 (lr=1.4729e-04) (hash(x)=74555497)
|
| 1006 |
+
8360 train 3.882697 (lr=1.4693e-04) (hash(x)=65367249)
|
| 1007 |
+
8370 train 3.789770 (lr=1.4658e-04) (hash(x)=78862619)
|
| 1008 |
+
8380 train 3.586155 (lr=1.4624e-04) (hash(x)=70989373)
|
| 1009 |
+
8390 train 3.698113 (lr=1.4591e-04) (hash(x)=80650634)
|
| 1010 |
+
8400 val loss 3.6287
|
| 1011 |
+
8400 val perplexity 37.6656
|
| 1012 |
+
8400 train 3.570008 (lr=1.4559e-04) (hash(x)=74168359)
|
| 1013 |
+
8410 train 3.666315 (lr=1.4527e-04) (hash(x)=80595745)
|
| 1014 |
+
8420 train 3.621702 (lr=1.4497e-04) (hash(x)=74239969)
|
| 1015 |
+
8430 train 3.633392 (lr=1.4467e-04) (hash(x)=72089061)
|
| 1016 |
+
8440 train 3.645253 (lr=1.4438e-04) (hash(x)=79471873)
|
| 1017 |
+
8450 train 3.639676 (lr=1.4411e-04) (hash(x)=70814855)
|
| 1018 |
+
8460 train 3.593217 (lr=1.4384e-04) (hash(x)=72900750)
|
| 1019 |
+
8470 train 3.736093 (lr=1.4358e-04) (hash(x)=73723555)
|
| 1020 |
+
8480 train 3.568275 (lr=1.4333e-04) (hash(x)=77750599)
|
| 1021 |
+
8490 train 3.571191 (lr=1.4309e-04) (hash(x)=82913770)
|
| 1022 |
+
8500 val loss 3.6268
|
| 1023 |
+
8500 val perplexity 37.5908
|
| 1024 |
+
8500 train 3.746575 (lr=1.4285e-04) (hash(x)=58014966)
|
| 1025 |
+
8510 train 3.598013 (lr=1.4263e-04) (hash(x)=78964850)
|
| 1026 |
+
8520 train 3.823884 (lr=1.4241e-04) (hash(x)=79106421)
|
| 1027 |
+
8530 train 3.610446 (lr=1.4221e-04) (hash(x)=75662227)
|
| 1028 |
+
8540 train 3.489582 (lr=1.4201e-04) (hash(x)=73928041)
|
| 1029 |
+
8550 train 3.640761 (lr=1.4183e-04) (hash(x)=73949623)
|
| 1030 |
+
8560 train 3.487322 (lr=1.4165e-04) (hash(x)=66671864)
|
| 1031 |
+
8570 train 3.550586 (lr=1.4148e-04) (hash(x)=73819413)
|
| 1032 |
+
8580 train 3.498424 (lr=1.4132e-04) (hash(x)=71116170)
|
| 1033 |
+
8590 train 3.505502 (lr=1.4117e-04) (hash(x)=75792152)
|
| 1034 |
+
8600 val loss 3.6306
|
| 1035 |
+
8600 val perplexity 37.7336
|
| 1036 |
+
8600 train 3.467249 (lr=1.4103e-04) (hash(x)=76199926)
|
| 1037 |
+
8610 train 3.446555 (lr=1.4090e-04) (hash(x)=72341134)
|
| 1038 |
+
8620 train 3.492881 (lr=1.4077e-04) (hash(x)=73669000)
|
| 1039 |
+
8630 train 3.578854 (lr=1.4066e-04) (hash(x)=76197518)
|
| 1040 |
+
8640 train 3.575135 (lr=1.4055e-04) (hash(x)=77186492)
|
| 1041 |
+
8650 train 3.572467 (lr=1.4046e-04) (hash(x)=72965152)
|
| 1042 |
+
8660 train 3.598943 (lr=1.4037e-04) (hash(x)=66084492)
|
| 1043 |
+
8670 train 3.688285 (lr=1.4029e-04) (hash(x)=77644722)
|
| 1044 |
+
8680 train 3.515165 (lr=1.4022e-04) (hash(x)=70362453)
|
| 1045 |
+
8690 train 3.628487 (lr=1.4016e-04) (hash(x)=77832585)
|
| 1046 |
+
8700 val loss 3.6215
|
| 1047 |
+
8700 val perplexity 37.3922
|
| 1048 |
+
8700 train 3.715358 (lr=1.4011e-04) (hash(x)=93249763)
|
| 1049 |
+
8710 train 3.667107 (lr=1.4007e-04) (hash(x)=88835109)
|
| 1050 |
+
8720 train 3.780373 (lr=1.4004e-04) (hash(x)=76773769)
|
| 1051 |
+
8730 train 3.585804 (lr=1.4002e-04) (hash(x)=78751013)
|
| 1052 |
+
8740 train 3.518092 (lr=1.4000e-04) (hash(x)=72850278)
|
| 1053 |
+
8749 val loss 3.6205
|
| 1054 |
+
8749 val perplexity 37.3548
|
n_heads12_lr14e-4_total_batch_size61440_seed1338/model_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a11690aa83ea0d0bc7cb2c7a48913df4968b297f6536821c245ead2308a6c22a
|
| 3 |
+
size 498607490
|
n_heads12_lr14e-4_total_batch_size61440_seed1338/optimizer_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab05f27039e020c4bdd7fe3f26813af076aa152b9a072800c7c3e55a6126bb58
|
| 3 |
+
size 990934406
|