Upload folder using huggingface_hub
Browse files- n_heads12_lr10e-4_total_batch_size61440_seed1338/args.json +1 -0
- n_heads12_lr10e-4_total_batch_size61440_seed1338/dataloader_08749.pt +3 -0
- n_heads12_lr10e-4_total_batch_size61440_seed1338/log2.txt +1054 -0
- n_heads12_lr10e-4_total_batch_size61440_seed1338/model_08749.pt +3 -0
- n_heads12_lr10e-4_total_batch_size61440_seed1338/optimizer_08749.pt +3 -0
n_heads12_lr10e-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_lr10e-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": 0.001, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "n_heads12_10e-4_61440", "n_embd": 768}
|
n_heads12_lr10e-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_lr10e-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 59271.3594
|
| 4 |
+
0 train 10.996866 (lr=2.0000e-06) (hash(x)=73026074)
|
| 5 |
+
10 train 9.878416 (lr=2.2000e-05) (hash(x)=81408174)
|
| 6 |
+
20 train 9.437363 (lr=4.2000e-05) (hash(x)=73752420)
|
| 7 |
+
30 train 9.008246 (lr=6.2000e-05) (hash(x)=77647299)
|
| 8 |
+
40 train 8.589837 (lr=8.2000e-05) (hash(x)=76104329)
|
| 9 |
+
50 train 8.106159 (lr=1.0200e-04) (hash(x)=69335494)
|
| 10 |
+
60 train 7.770695 (lr=1.2200e-04) (hash(x)=72716449)
|
| 11 |
+
70 train 7.610338 (lr=1.4200e-04) (hash(x)=71125105)
|
| 12 |
+
80 train 7.492192 (lr=1.6200e-04) (hash(x)=74407559)
|
| 13 |
+
90 train 7.331594 (lr=1.8200e-04) (hash(x)=78245164)
|
| 14 |
+
100 val loss 7.3030
|
| 15 |
+
100 val perplexity 1484.6930
|
| 16 |
+
100 train 7.273099 (lr=2.0200e-04) (hash(x)=66975638)
|
| 17 |
+
110 train 7.049629 (lr=2.2200e-04) (hash(x)=72573036)
|
| 18 |
+
120 train 6.967019 (lr=2.4200e-04) (hash(x)=72876251)
|
| 19 |
+
130 train 6.937787 (lr=2.6200e-04) (hash(x)=79969711)
|
| 20 |
+
140 train 6.822509 (lr=2.8200e-04) (hash(x)=76473027)
|
| 21 |
+
150 train 6.709403 (lr=3.0200e-04) (hash(x)=72199769)
|
| 22 |
+
160 train 6.526720 (lr=3.2200e-04) (hash(x)=79575844)
|
| 23 |
+
170 train 6.426419 (lr=3.4200e-04) (hash(x)=65860740)
|
| 24 |
+
180 train 6.399967 (lr=3.6200e-04) (hash(x)=79880170)
|
| 25 |
+
190 train 6.293023 (lr=3.8200e-04) (hash(x)=78465601)
|
| 26 |
+
200 val loss 6.4675
|
| 27 |
+
200 val perplexity 643.8866
|
| 28 |
+
200 train 6.200614 (lr=4.0200e-04) (hash(x)=72711480)
|
| 29 |
+
210 train 6.149294 (lr=4.2200e-04) (hash(x)=71194058)
|
| 30 |
+
220 train 6.341296 (lr=4.4200e-04) (hash(x)=92783037)
|
| 31 |
+
230 train 6.089361 (lr=4.6200e-04) (hash(x)=73611161)
|
| 32 |
+
240 train 5.879336 (lr=4.8200e-04) (hash(x)=74427917)
|
| 33 |
+
250 train 5.897179 (lr=5.0200e-04) (hash(x)=65305370)
|
| 34 |
+
260 train 5.989441 (lr=5.2200e-04) (hash(x)=81378465)
|
| 35 |
+
270 train 5.762376 (lr=5.4200e-04) (hash(x)=72738754)
|
| 36 |
+
280 train 6.082486 (lr=5.6200e-04) (hash(x)=82172761)
|
| 37 |
+
290 train 6.245429 (lr=5.8200e-04) (hash(x)=70024961)
|
| 38 |
+
300 val loss 6.0048
|
| 39 |
+
300 val perplexity 405.3865
|
| 40 |
+
300 train 5.950438 (lr=6.0200e-04) (hash(x)=74275986)
|
| 41 |
+
310 train 6.054238 (lr=6.2200e-04) (hash(x)=76295038)
|
| 42 |
+
320 train 6.028188 (lr=6.4200e-04) (hash(x)=74741156)
|
| 43 |
+
330 train 5.800296 (lr=6.6200e-04) (hash(x)=71941043)
|
| 44 |
+
340 train 5.910799 (lr=6.8200e-04) (hash(x)=75020464)
|
| 45 |
+
350 train 5.912827 (lr=7.0200e-04) (hash(x)=64712930)
|
| 46 |
+
360 train 6.203627 (lr=7.2200e-04) (hash(x)=68519941)
|
| 47 |
+
370 train 5.780438 (lr=7.4200e-04) (hash(x)=75355289)
|
| 48 |
+
380 train 5.762534 (lr=7.6200e-04) (hash(x)=66796276)
|
| 49 |
+
390 train 5.833106 (lr=7.8200e-04) (hash(x)=81002698)
|
| 50 |
+
400 val loss 5.6717
|
| 51 |
+
400 val perplexity 290.5200
|
| 52 |
+
400 train 5.530255 (lr=8.0200e-04) (hash(x)=76354037)
|
| 53 |
+
410 train 5.791349 (lr=8.2200e-04) (hash(x)=82242082)
|
| 54 |
+
420 train 5.521913 (lr=8.4200e-04) (hash(x)=77642063)
|
| 55 |
+
430 train 5.445039 (lr=8.6200e-04) (hash(x)=72281244)
|
| 56 |
+
440 train 5.613255 (lr=8.8200e-04) (hash(x)=79666830)
|
| 57 |
+
450 train 5.516862 (lr=9.0200e-04) (hash(x)=76014248)
|
| 58 |
+
460 train 5.642200 (lr=9.2200e-04) (hash(x)=81129062)
|
| 59 |
+
470 train 5.453766 (lr=9.4200e-04) (hash(x)=69374949)
|
| 60 |
+
480 train 5.465752 (lr=9.6200e-04) (hash(x)=72137706)
|
| 61 |
+
490 train 5.478282 (lr=9.8200e-04) (hash(x)=74403781)
|
| 62 |
+
500 val loss 5.4677
|
| 63 |
+
500 val perplexity 236.9264
|
| 64 |
+
500 train 5.378243 (lr=1.0000e-03) (hash(x)=78816019)
|
| 65 |
+
510 train 5.413568 (lr=1.0000e-03) (hash(x)=72193331)
|
| 66 |
+
520 train 5.407718 (lr=9.9999e-04) (hash(x)=79344438)
|
| 67 |
+
530 train 5.388953 (lr=9.9997e-04) (hash(x)=78092289)
|
| 68 |
+
540 train 5.375805 (lr=9.9995e-04) (hash(x)=74733731)
|
| 69 |
+
550 train 5.339091 (lr=9.9992e-04) (hash(x)=69920645)
|
| 70 |
+
560 train 5.170500 (lr=9.9988e-04) (hash(x)=76131735)
|
| 71 |
+
570 train 5.257198 (lr=9.9984e-04) (hash(x)=78418293)
|
| 72 |
+
580 train 5.068708 (lr=9.9979e-04) (hash(x)=73699256)
|
| 73 |
+
590 train 5.071388 (lr=9.9974e-04) (hash(x)=77505664)
|
| 74 |
+
600 val loss 5.2762
|
| 75 |
+
600 val perplexity 195.6227
|
| 76 |
+
600 train 5.010283 (lr=9.9967e-04) (hash(x)=71823058)
|
| 77 |
+
610 train 5.046767 (lr=9.9961e-04) (hash(x)=78450672)
|
| 78 |
+
620 train 4.940421 (lr=9.9953e-04) (hash(x)=75456256)
|
| 79 |
+
630 train 4.914159 (lr=9.9945e-04) (hash(x)=68515074)
|
| 80 |
+
640 train 4.981021 (lr=9.9936e-04) (hash(x)=71968654)
|
| 81 |
+
650 train 4.904106 (lr=9.9927e-04) (hash(x)=72586963)
|
| 82 |
+
660 train 4.861030 (lr=9.9917e-04) (hash(x)=67398625)
|
| 83 |
+
670 train 4.874962 (lr=9.9906e-04) (hash(x)=82737129)
|
| 84 |
+
680 train 5.116166 (lr=9.9894e-04) (hash(x)=78920914)
|
| 85 |
+
690 train 5.134350 (lr=9.9882e-04) (hash(x)=59859106)
|
| 86 |
+
700 val loss 5.1209
|
| 87 |
+
700 val perplexity 167.4840
|
| 88 |
+
700 train 5.015658 (lr=9.9870e-04) (hash(x)=76856514)
|
| 89 |
+
710 train 5.064469 (lr=9.9856e-04) (hash(x)=73530680)
|
| 90 |
+
720 train 5.152414 (lr=9.9842e-04) (hash(x)=74382454)
|
| 91 |
+
730 train 5.000039 (lr=9.9828e-04) (hash(x)=80641678)
|
| 92 |
+
740 train 5.147391 (lr=9.9812e-04) (hash(x)=82419404)
|
| 93 |
+
750 train 5.105894 (lr=9.9796e-04) (hash(x)=70898356)
|
| 94 |
+
760 train 5.059180 (lr=9.9780e-04) (hash(x)=76422181)
|
| 95 |
+
770 train 5.017817 (lr=9.9762e-04) (hash(x)=83711024)
|
| 96 |
+
780 train 4.930438 (lr=9.9744e-04) (hash(x)=74365746)
|
| 97 |
+
790 train 4.879364 (lr=9.9726e-04) (hash(x)=73104424)
|
| 98 |
+
800 val loss 4.9352
|
| 99 |
+
800 val perplexity 139.0995
|
| 100 |
+
800 train 5.063818 (lr=9.9707e-04) (hash(x)=73589083)
|
| 101 |
+
810 train 4.866418 (lr=9.9687e-04) (hash(x)=69132306)
|
| 102 |
+
820 train 4.920248 (lr=9.9666e-04) (hash(x)=75056169)
|
| 103 |
+
830 train 4.800425 (lr=9.9645e-04) (hash(x)=70032437)
|
| 104 |
+
840 train 4.815232 (lr=9.9623e-04) (hash(x)=73178845)
|
| 105 |
+
850 train 4.810349 (lr=9.9601e-04) (hash(x)=69463059)
|
| 106 |
+
860 train 4.890258 (lr=9.9578e-04) (hash(x)=74635339)
|
| 107 |
+
870 train 5.054516 (lr=9.9554e-04) (hash(x)=73964810)
|
| 108 |
+
880 train 4.784234 (lr=9.9530e-04) (hash(x)=79912096)
|
| 109 |
+
890 train 4.762746 (lr=9.9505e-04) (hash(x)=84832110)
|
| 110 |
+
900 val loss 4.7948
|
| 111 |
+
900 val perplexity 120.8858
|
| 112 |
+
900 train 4.760536 (lr=9.9479e-04) (hash(x)=71823388)
|
| 113 |
+
910 train 4.714203 (lr=9.9453e-04) (hash(x)=69541745)
|
| 114 |
+
920 train 4.652880 (lr=9.9426e-04) (hash(x)=67018969)
|
| 115 |
+
930 train 4.604096 (lr=9.9398e-04) (hash(x)=73346110)
|
| 116 |
+
940 train 4.674997 (lr=9.9370e-04) (hash(x)=72105603)
|
| 117 |
+
950 train 4.572038 (lr=9.9341e-04) (hash(x)=71803352)
|
| 118 |
+
960 train 4.733698 (lr=9.9311e-04) (hash(x)=75135621)
|
| 119 |
+
970 train 4.526983 (lr=9.9281e-04) (hash(x)=73334774)
|
| 120 |
+
980 train 4.494944 (lr=9.9250e-04) (hash(x)=80861275)
|
| 121 |
+
990 train 4.545967 (lr=9.9219e-04) (hash(x)=77124068)
|
| 122 |
+
1000 val loss 4.7104
|
| 123 |
+
1000 val perplexity 111.0961
|
| 124 |
+
1000 train 4.433167 (lr=9.9187e-04) (hash(x)=78127630)
|
| 125 |
+
1010 train 4.377372 (lr=9.9154e-04) (hash(x)=73670162)
|
| 126 |
+
1020 train 4.492303 (lr=9.9121e-04) (hash(x)=77732174)
|
| 127 |
+
1030 train 4.339534 (lr=9.9087e-04) (hash(x)=79418857)
|
| 128 |
+
1040 train 4.380932 (lr=9.9052e-04) (hash(x)=77732710)
|
| 129 |
+
1050 train 4.353242 (lr=9.9017e-04) (hash(x)=69751515)
|
| 130 |
+
1060 train 4.329982 (lr=9.8981e-04) (hash(x)=71788756)
|
| 131 |
+
1070 train 4.538909 (lr=9.8944e-04) (hash(x)=72944541)
|
| 132 |
+
1080 train 4.613462 (lr=9.8907e-04) (hash(x)=75792144)
|
| 133 |
+
1090 train 4.559521 (lr=9.8869e-04) (hash(x)=78746868)
|
| 134 |
+
1100 val loss 4.6126
|
| 135 |
+
1100 val perplexity 100.7506
|
| 136 |
+
1100 train 4.641263 (lr=9.8831e-04) (hash(x)=70879945)
|
| 137 |
+
1110 train 4.627989 (lr=9.8791e-04) (hash(x)=77572329)
|
| 138 |
+
1120 train 5.074735 (lr=9.8752e-04) (hash(x)=68749754)
|
| 139 |
+
1130 train 4.607842 (lr=9.8711e-04) (hash(x)=78632440)
|
| 140 |
+
1140 train 4.538727 (lr=9.8670e-04) (hash(x)=73565080)
|
| 141 |
+
1150 train 4.599656 (lr=9.8629e-04) (hash(x)=81594247)
|
| 142 |
+
1160 train 4.592741 (lr=9.8586e-04) (hash(x)=63200457)
|
| 143 |
+
1170 train 4.641797 (lr=9.8543e-04) (hash(x)=75895085)
|
| 144 |
+
1180 train 4.660604 (lr=9.8500e-04) (hash(x)=71402056)
|
| 145 |
+
1190 train 4.604215 (lr=9.8456e-04) (hash(x)=75263955)
|
| 146 |
+
1200 val loss 4.5415
|
| 147 |
+
1200 val perplexity 93.8294
|
| 148 |
+
1200 train 4.598189 (lr=9.8411e-04) (hash(x)=73579627)
|
| 149 |
+
1210 train 4.595469 (lr=9.8365e-04) (hash(x)=79106153)
|
| 150 |
+
1220 train 4.654102 (lr=9.8319e-04) (hash(x)=78475838)
|
| 151 |
+
1230 train 4.711361 (lr=9.8272e-04) (hash(x)=72129619)
|
| 152 |
+
1240 train 4.623965 (lr=9.8225e-04) (hash(x)=58067507)
|
| 153 |
+
1250 train 4.432872 (lr=9.8177e-04) (hash(x)=78666722)
|
| 154 |
+
1260 train 4.517206 (lr=9.8129e-04) (hash(x)=74964097)
|
| 155 |
+
1270 train 4.492600 (lr=9.8079e-04) (hash(x)=72774823)
|
| 156 |
+
1280 train 4.535753 (lr=9.8030e-04) (hash(x)=69959055)
|
| 157 |
+
1290 train 4.377015 (lr=9.7979e-04) (hash(x)=76629730)
|
| 158 |
+
1300 val loss 4.4865
|
| 159 |
+
1300 val perplexity 88.8096
|
| 160 |
+
1300 train 4.413606 (lr=9.7928e-04) (hash(x)=72183284)
|
| 161 |
+
1310 train 4.612698 (lr=9.7876e-04) (hash(x)=83147722)
|
| 162 |
+
1320 train 4.556895 (lr=9.7824e-04) (hash(x)=69788032)
|
| 163 |
+
1330 train 4.280804 (lr=9.7771e-04) (hash(x)=66629732)
|
| 164 |
+
1340 train 4.352966 (lr=9.7717e-04) (hash(x)=72542387)
|
| 165 |
+
1350 train 4.410437 (lr=9.7663e-04) (hash(x)=76517375)
|
| 166 |
+
1360 train 4.400441 (lr=9.7608e-04) (hash(x)=74401519)
|
| 167 |
+
1370 train 4.214207 (lr=9.7553e-04) (hash(x)=78409356)
|
| 168 |
+
1380 train 4.176289 (lr=9.7497e-04) (hash(x)=77626183)
|
| 169 |
+
1390 train 4.131511 (lr=9.7440e-04) (hash(x)=72496033)
|
| 170 |
+
1400 val loss 4.4495
|
| 171 |
+
1400 val perplexity 85.5817
|
| 172 |
+
1400 train 4.209481 (lr=9.7383e-04) (hash(x)=73708759)
|
| 173 |
+
1410 train 4.204567 (lr=9.7325e-04) (hash(x)=73706234)
|
| 174 |
+
1420 train 4.080943 (lr=9.7267e-04) (hash(x)=75918530)
|
| 175 |
+
1430 train 3.968040 (lr=9.7207e-04) (hash(x)=76806498)
|
| 176 |
+
1440 train 4.197422 (lr=9.7148e-04) (hash(x)=77310523)
|
| 177 |
+
1450 train 4.163127 (lr=9.7087e-04) (hash(x)=75964248)
|
| 178 |
+
1460 train 4.009656 (lr=9.7026e-04) (hash(x)=88549155)
|
| 179 |
+
1470 train 4.501064 (lr=9.6965e-04) (hash(x)=78654420)
|
| 180 |
+
1480 train 4.328508 (lr=9.6903e-04) (hash(x)=53433500)
|
| 181 |
+
1490 train 4.356875 (lr=9.6840e-04) (hash(x)=75862069)
|
| 182 |
+
1500 val loss 4.4325
|
| 183 |
+
1500 val perplexity 84.1424
|
| 184 |
+
1500 train 4.486503 (lr=9.6777e-04) (hash(x)=76037312)
|
| 185 |
+
1510 train 4.237566 (lr=9.6713e-04) (hash(x)=78654499)
|
| 186 |
+
1520 train 4.323501 (lr=9.6648e-04) (hash(x)=72319550)
|
| 187 |
+
1530 train 4.301304 (lr=9.6583e-04) (hash(x)=78901150)
|
| 188 |
+
1540 train 4.351883 (lr=9.6517e-04) (hash(x)=74848448)
|
| 189 |
+
1550 train 4.381869 (lr=9.6451e-04) (hash(x)=77019229)
|
| 190 |
+
1560 train 4.462658 (lr=9.6384e-04) (hash(x)=74200996)
|
| 191 |
+
1570 train 4.365378 (lr=9.6316e-04) (hash(x)=76717682)
|
| 192 |
+
1580 train 4.258556 (lr=9.6248e-04) (hash(x)=74662317)
|
| 193 |
+
1590 train 4.428346 (lr=9.6179e-04) (hash(x)=76066438)
|
| 194 |
+
1600 val loss 4.3716
|
| 195 |
+
1600 val perplexity 79.1720
|
| 196 |
+
1600 train 4.570393 (lr=9.6110e-04) (hash(x)=76123488)
|
| 197 |
+
1610 train 4.458269 (lr=9.6040e-04) (hash(x)=73211013)
|
| 198 |
+
1620 train 4.327531 (lr=9.5969e-04) (hash(x)=80389252)
|
| 199 |
+
1630 train 4.320492 (lr=9.5898e-04) (hash(x)=73449158)
|
| 200 |
+
1640 train 4.350865 (lr=9.5826e-04) (hash(x)=75060644)
|
| 201 |
+
1650 train 4.280501 (lr=9.5754e-04) (hash(x)=68966493)
|
| 202 |
+
1660 train 4.356060 (lr=9.5681e-04) (hash(x)=73496223)
|
| 203 |
+
1670 train 4.223134 (lr=9.5607e-04) (hash(x)=86210083)
|
| 204 |
+
1680 train 4.313195 (lr=9.5533e-04) (hash(x)=74053619)
|
| 205 |
+
1690 train 4.375888 (lr=9.5458e-04) (hash(x)=65910719)
|
| 206 |
+
1700 val loss 4.3444
|
| 207 |
+
1700 val perplexity 77.0446
|
| 208 |
+
1700 train 4.226717 (lr=9.5383e-04) (hash(x)=66496027)
|
| 209 |
+
1710 train 4.402362 (lr=9.5307e-04) (hash(x)=75485845)
|
| 210 |
+
1720 train 4.217659 (lr=9.5231e-04) (hash(x)=72417899)
|
| 211 |
+
1730 train 4.262581 (lr=9.5153e-04) (hash(x)=73593240)
|
| 212 |
+
1740 train 4.163245 (lr=9.5076e-04) (hash(x)=74832651)
|
| 213 |
+
1750 train 4.145145 (lr=9.4998e-04) (hash(x)=73364724)
|
| 214 |
+
1760 train 4.311493 (lr=9.4919e-04) (hash(x)=75903856)
|
| 215 |
+
1770 train 4.368478 (lr=9.4839e-04) (hash(x)=78183744)
|
| 216 |
+
1780 train 4.272652 (lr=9.4759e-04) (hash(x)=69993946)
|
| 217 |
+
1790 train 4.349708 (lr=9.4679e-04) (hash(x)=101032039)
|
| 218 |
+
1800 val loss 4.3083
|
| 219 |
+
1800 val perplexity 74.3140
|
| 220 |
+
1800 train 4.363121 (lr=9.4598e-04) (hash(x)=86018370)
|
| 221 |
+
1810 train 3.936415 (lr=9.4516e-04) (hash(x)=84205841)
|
| 222 |
+
1820 train 4.260911 (lr=9.4434e-04) (hash(x)=81735398)
|
| 223 |
+
1830 train 4.155681 (lr=9.4351e-04) (hash(x)=77297230)
|
| 224 |
+
1840 train 4.104589 (lr=9.4268e-04) (hash(x)=75535995)
|
| 225 |
+
1850 train 4.126131 (lr=9.4184e-04) (hash(x)=73930506)
|
| 226 |
+
1860 train 4.089764 (lr=9.4099e-04) (hash(x)=79524567)
|
| 227 |
+
1870 train 4.028361 (lr=9.4014e-04) (hash(x)=80242414)
|
| 228 |
+
1880 train 3.939608 (lr=9.3928e-04) (hash(x)=71395200)
|
| 229 |
+
1890 train 4.069571 (lr=9.3842e-04) (hash(x)=78269507)
|
| 230 |
+
1900 val loss 4.2854
|
| 231 |
+
1900 val perplexity 72.6328
|
| 232 |
+
1900 train 4.301173 (lr=9.3755e-04) (hash(x)=78327979)
|
| 233 |
+
1910 train 4.234810 (lr=9.3668e-04) (hash(x)=75760901)
|
| 234 |
+
1920 train 4.283826 (lr=9.3580e-04) (hash(x)=81699948)
|
| 235 |
+
1930 train 4.332939 (lr=9.3491e-04) (hash(x)=75326527)
|
| 236 |
+
1940 train 4.159538 (lr=9.3402e-04) (hash(x)=73660211)
|
| 237 |
+
1950 train 4.297121 (lr=9.3313e-04) (hash(x)=66946100)
|
| 238 |
+
1960 train 4.112506 (lr=9.3223e-04) (hash(x)=45679133)
|
| 239 |
+
1970 train 4.387769 (lr=9.3132e-04) (hash(x)=73647298)
|
| 240 |
+
1980 train 4.293734 (lr=9.3041e-04) (hash(x)=72037215)
|
| 241 |
+
1990 train 4.289642 (lr=9.2949e-04) (hash(x)=77640675)
|
| 242 |
+
2000 val loss 4.2333
|
| 243 |
+
2000 val perplexity 68.9460
|
| 244 |
+
2000 train 4.270515 (lr=9.2856e-04) (hash(x)=78165118)
|
| 245 |
+
2010 train 4.329265 (lr=9.2763e-04) (hash(x)=80013879)
|
| 246 |
+
2020 train 4.595002 (lr=9.2670e-04) (hash(x)=73778143)
|
| 247 |
+
2030 train 4.104586 (lr=9.2576e-04) (hash(x)=71396805)
|
| 248 |
+
2040 train 4.312492 (lr=9.2481e-04) (hash(x)=83350333)
|
| 249 |
+
2050 train 4.805598 (lr=9.2386e-04) (hash(x)=67399515)
|
| 250 |
+
2060 train 4.292835 (lr=9.2291e-04) (hash(x)=66586949)
|
| 251 |
+
2070 train 4.159421 (lr=9.2195e-04) (hash(x)=77648913)
|
| 252 |
+
2080 train 4.301746 (lr=9.2098e-04) (hash(x)=80507387)
|
| 253 |
+
2090 train 4.327753 (lr=9.2001e-04) (hash(x)=67897081)
|
| 254 |
+
2100 val loss 4.2106
|
| 255 |
+
2100 val perplexity 67.3954
|
| 256 |
+
2100 train 4.091825 (lr=9.1903e-04) (hash(x)=78780658)
|
| 257 |
+
2110 train 4.011407 (lr=9.1804e-04) (hash(x)=71002379)
|
| 258 |
+
2120 train 4.254676 (lr=9.1706e-04) (hash(x)=78941785)
|
| 259 |
+
2130 train 4.209649 (lr=9.1606e-04) (hash(x)=77840641)
|
| 260 |
+
2140 train 4.070915 (lr=9.1506e-04) (hash(x)=79973491)
|
| 261 |
+
2150 train 4.150324 (lr=9.1406e-04) (hash(x)=70608621)
|
| 262 |
+
2160 train 4.166295 (lr=9.1305e-04) (hash(x)=77576393)
|
| 263 |
+
2170 train 4.131457 (lr=9.1203e-04) (hash(x)=79697380)
|
| 264 |
+
2180 train 4.089016 (lr=9.1101e-04) (hash(x)=68595163)
|
| 265 |
+
2190 train 4.154143 (lr=9.0999e-04) (hash(x)=78393072)
|
| 266 |
+
2200 val loss 4.1962
|
| 267 |
+
2200 val perplexity 66.4345
|
| 268 |
+
2200 train 3.913084 (lr=9.0896e-04) (hash(x)=68832590)
|
| 269 |
+
2210 train 3.941712 (lr=9.0792e-04) (hash(x)=73069430)
|
| 270 |
+
2220 train 3.871197 (lr=9.0688e-04) (hash(x)=78403531)
|
| 271 |
+
2230 train 3.948771 (lr=9.0583e-04) (hash(x)=84282001)
|
| 272 |
+
2240 train 3.987436 (lr=9.0478e-04) (hash(x)=68791263)
|
| 273 |
+
2250 train 3.916656 (lr=9.0372e-04) (hash(x)=78324357)
|
| 274 |
+
2260 train 3.761337 (lr=9.0266e-04) (hash(x)=70769509)
|
| 275 |
+
2270 train 3.969893 (lr=9.0160e-04) (hash(x)=76496955)
|
| 276 |
+
2280 train 3.995754 (lr=9.0052e-04) (hash(x)=76607207)
|
| 277 |
+
2290 train 4.006201 (lr=8.9945e-04) (hash(x)=76204997)
|
| 278 |
+
2300 val loss 4.1834
|
| 279 |
+
2300 val perplexity 65.5914
|
| 280 |
+
2300 train 4.198055 (lr=8.9836e-04) (hash(x)=78644214)
|
| 281 |
+
2310 train 4.174824 (lr=8.9728e-04) (hash(x)=69825560)
|
| 282 |
+
2320 train 4.221256 (lr=8.9618e-04) (hash(x)=67690675)
|
| 283 |
+
2330 train 4.235637 (lr=8.9509e-04) (hash(x)=72400089)
|
| 284 |
+
2340 train 4.130614 (lr=8.9398e-04) (hash(x)=87283063)
|
| 285 |
+
2350 train 4.110744 (lr=8.9288e-04) (hash(x)=74257442)
|
| 286 |
+
2360 train 4.209318 (lr=8.9177e-04) (hash(x)=76596934)
|
| 287 |
+
2370 train 4.187047 (lr=8.9065e-04) (hash(x)=76345291)
|
| 288 |
+
2380 train 4.102968 (lr=8.8953e-04) (hash(x)=65495351)
|
| 289 |
+
2390 train 4.403996 (lr=8.8840e-04) (hash(x)=75795347)
|
| 290 |
+
2400 val loss 4.1627
|
| 291 |
+
2400 val perplexity 64.2450
|
| 292 |
+
2400 train 4.142782 (lr=8.8727e-04) (hash(x)=85857583)
|
| 293 |
+
2410 train 4.133544 (lr=8.8613e-04) (hash(x)=72305336)
|
| 294 |
+
2420 train 4.091039 (lr=8.8499e-04) (hash(x)=76780579)
|
| 295 |
+
2430 train 4.070125 (lr=8.8384e-04) (hash(x)=85532225)
|
| 296 |
+
2440 train 4.194731 (lr=8.8269e-04) (hash(x)=74308641)
|
| 297 |
+
2450 train 4.177492 (lr=8.8153e-04) (hash(x)=75205457)
|
| 298 |
+
2460 train 4.192517 (lr=8.8037e-04) (hash(x)=65232578)
|
| 299 |
+
2470 train 4.219167 (lr=8.7921e-04) (hash(x)=77893584)
|
| 300 |
+
2480 train 4.063409 (lr=8.7804e-04) (hash(x)=63720791)
|
| 301 |
+
2490 train 4.115788 (lr=8.7686e-04) (hash(x)=74714745)
|
| 302 |
+
2500 val loss 4.1203
|
| 303 |
+
2500 val perplexity 61.5781
|
| 304 |
+
2500 train 4.230078 (lr=8.7568e-04) (hash(x)=72628906)
|
| 305 |
+
2510 train 3.996775 (lr=8.7450e-04) (hash(x)=69138578)
|
| 306 |
+
2520 train 4.160388 (lr=8.7331e-04) (hash(x)=65122039)
|
| 307 |
+
2530 train 4.152202 (lr=8.7211e-04) (hash(x)=74953183)
|
| 308 |
+
2540 train 4.064047 (lr=8.7091e-04) (hash(x)=75410150)
|
| 309 |
+
2550 train 3.964011 (lr=8.6971e-04) (hash(x)=80818592)
|
| 310 |
+
2560 train 3.985015 (lr=8.6850e-04) (hash(x)=73726201)
|
| 311 |
+
2570 train 4.164223 (lr=8.6729e-04) (hash(x)=78518314)
|
| 312 |
+
2580 train 4.135106 (lr=8.6607e-04) (hash(x)=70758402)
|
| 313 |
+
2590 train 4.096724 (lr=8.6485e-04) (hash(x)=70495389)
|
| 314 |
+
2600 val loss 4.1246
|
| 315 |
+
2600 val perplexity 61.8457
|
| 316 |
+
2600 train 4.010422 (lr=8.6362e-04) (hash(x)=82589673)
|
| 317 |
+
2610 train 3.881681 (lr=8.6239e-04) (hash(x)=70936243)
|
| 318 |
+
2620 train 3.823193 (lr=8.6116e-04) (hash(x)=83907871)
|
| 319 |
+
2630 train 3.905704 (lr=8.5991e-04) (hash(x)=70432517)
|
| 320 |
+
2640 train 3.895645 (lr=8.5867e-04) (hash(x)=73683605)
|
| 321 |
+
2650 train 3.757514 (lr=8.5742e-04) (hash(x)=81772854)
|
| 322 |
+
2660 train 3.899540 (lr=8.5617e-04) (hash(x)=68745438)
|
| 323 |
+
2670 train 3.953605 (lr=8.5491e-04) (hash(x)=77762278)
|
| 324 |
+
2680 train 3.827003 (lr=8.5365e-04) (hash(x)=70738783)
|
| 325 |
+
2690 train 3.829181 (lr=8.5238e-04) (hash(x)=74534099)
|
| 326 |
+
2700 val loss 4.1121
|
| 327 |
+
2700 val perplexity 61.0766
|
| 328 |
+
2700 train 4.027041 (lr=8.5111e-04) (hash(x)=81441457)
|
| 329 |
+
2710 train 4.074991 (lr=8.4983e-04) (hash(x)=76167134)
|
| 330 |
+
2720 train 4.137204 (lr=8.4855e-04) (hash(x)=79036746)
|
| 331 |
+
2730 train 4.112680 (lr=8.4727e-04) (hash(x)=76789597)
|
| 332 |
+
2740 train 4.071312 (lr=8.4598e-04) (hash(x)=67515905)
|
| 333 |
+
2750 train 4.182815 (lr=8.4469e-04) (hash(x)=81464179)
|
| 334 |
+
2760 train 4.066255 (lr=8.4339e-04) (hash(x)=83004450)
|
| 335 |
+
2770 train 4.057415 (lr=8.4209e-04) (hash(x)=74647329)
|
| 336 |
+
2780 train 4.233309 (lr=8.4078e-04) (hash(x)=65160033)
|
| 337 |
+
2790 train 4.274494 (lr=8.3947e-04) (hash(x)=77815496)
|
| 338 |
+
2800 val loss 4.0825
|
| 339 |
+
2800 val perplexity 59.2950
|
| 340 |
+
2800 train 4.089107 (lr=8.3816e-04) (hash(x)=68408565)
|
| 341 |
+
2810 train 4.116800 (lr=8.3684e-04) (hash(x)=71127979)
|
| 342 |
+
2820 train 4.139161 (lr=8.3552e-04) (hash(x)=74423275)
|
| 343 |
+
2830 train 3.975283 (lr=8.3419e-04) (hash(x)=80921960)
|
| 344 |
+
2840 train 4.271122 (lr=8.3286e-04) (hash(x)=79254948)
|
| 345 |
+
2850 train 3.994164 (lr=8.3153e-04) (hash(x)=73700080)
|
| 346 |
+
2860 train 4.181685 (lr=8.3019e-04) (hash(x)=74328412)
|
| 347 |
+
2870 train 4.021886 (lr=8.2884e-04) (hash(x)=74359259)
|
| 348 |
+
2880 train 4.224534 (lr=8.2750e-04) (hash(x)=74479648)
|
| 349 |
+
2890 train 3.880072 (lr=8.2615e-04) (hash(x)=80993243)
|
| 350 |
+
2900 val loss 4.0558
|
| 351 |
+
2900 val perplexity 57.7324
|
| 352 |
+
2900 train 4.103284 (lr=8.2479e-04) (hash(x)=72286433)
|
| 353 |
+
2910 train 4.238840 (lr=8.2343e-04) (hash(x)=81081634)
|
| 354 |
+
2920 train 4.000973 (lr=8.2207e-04) (hash(x)=71809089)
|
| 355 |
+
2930 train 4.024914 (lr=8.2070e-04) (hash(x)=72668797)
|
| 356 |
+
2940 train 4.210224 (lr=8.1933e-04) (hash(x)=80472855)
|
| 357 |
+
2950 train 4.069575 (lr=8.1796e-04) (hash(x)=67764485)
|
| 358 |
+
2960 train 3.910945 (lr=8.1658e-04) (hash(x)=78590385)
|
| 359 |
+
2970 train 4.170641 (lr=8.1520e-04) (hash(x)=86434483)
|
| 360 |
+
2980 train 4.078607 (lr=8.1381e-04) (hash(x)=83872337)
|
| 361 |
+
2990 train 4.006131 (lr=8.1242e-04) (hash(x)=76866271)
|
| 362 |
+
3000 val loss 4.0597
|
| 363 |
+
3000 val perplexity 57.9593
|
| 364 |
+
3000 train 3.944339 (lr=8.1103e-04) (hash(x)=98889128)
|
| 365 |
+
3010 train 3.832625 (lr=8.0963e-04) (hash(x)=77128749)
|
| 366 |
+
3020 train 3.935330 (lr=8.0823e-04) (hash(x)=74427692)
|
| 367 |
+
3030 train 4.167792 (lr=8.0682e-04) (hash(x)=75155791)
|
| 368 |
+
3040 train 3.775682 (lr=8.0541e-04) (hash(x)=80384590)
|
| 369 |
+
3050 train 3.838757 (lr=8.0400e-04) (hash(x)=69078039)
|
| 370 |
+
3060 train 3.852766 (lr=8.0258e-04) (hash(x)=75289844)
|
| 371 |
+
3070 train 3.658214 (lr=8.0116e-04) (hash(x)=73776546)
|
| 372 |
+
3080 train 3.964492 (lr=7.9974e-04) (hash(x)=83632057)
|
| 373 |
+
3090 train 4.115197 (lr=7.9831e-04) (hash(x)=83219352)
|
| 374 |
+
3100 val loss 4.0759
|
| 375 |
+
3100 val perplexity 58.9055
|
| 376 |
+
3100 train 3.719601 (lr=7.9688e-04) (hash(x)=72986705)
|
| 377 |
+
3110 train 4.151382 (lr=7.9545e-04) (hash(x)=79090150)
|
| 378 |
+
3120 train 3.909985 (lr=7.9401e-04) (hash(x)=82919368)
|
| 379 |
+
3130 train 4.170765 (lr=7.9257e-04) (hash(x)=82752003)
|
| 380 |
+
3140 train 4.190866 (lr=7.9112e-04) (hash(x)=83049547)
|
| 381 |
+
3150 train 4.122423 (lr=7.8967e-04) (hash(x)=76384003)
|
| 382 |
+
3160 train 4.094978 (lr=7.8822e-04) (hash(x)=83069546)
|
| 383 |
+
3170 train 4.135162 (lr=7.8677e-04) (hash(x)=77245329)
|
| 384 |
+
3180 train 3.917817 (lr=7.8531e-04) (hash(x)=80555988)
|
| 385 |
+
3190 train 4.017499 (lr=7.8384e-04) (hash(x)=79506138)
|
| 386 |
+
3200 val loss 4.0148
|
| 387 |
+
3200 val perplexity 55.4118
|
| 388 |
+
3200 train 3.972687 (lr=7.8238e-04) (hash(x)=74238910)
|
| 389 |
+
3210 train 4.109450 (lr=7.8091e-04) (hash(x)=87246704)
|
| 390 |
+
3220 train 3.943842 (lr=7.7944e-04) (hash(x)=76286650)
|
| 391 |
+
3230 train 3.915924 (lr=7.7796e-04) (hash(x)=74128366)
|
| 392 |
+
3240 train 3.933496 (lr=7.7648e-04) (hash(x)=83686518)
|
| 393 |
+
3250 train 4.060529 (lr=7.7500e-04) (hash(x)=79334239)
|
| 394 |
+
3260 train 3.840187 (lr=7.7351e-04) (hash(x)=76991073)
|
| 395 |
+
3270 train 3.861104 (lr=7.7203e-04) (hash(x)=79645991)
|
| 396 |
+
3280 train 3.894144 (lr=7.7053e-04) (hash(x)=79642315)
|
| 397 |
+
3290 train 4.040541 (lr=7.6904e-04) (hash(x)=71394376)
|
| 398 |
+
3300 val loss 4.0149
|
| 399 |
+
3300 val perplexity 55.4187
|
| 400 |
+
3300 train 3.768021 (lr=7.6754e-04) (hash(x)=74335345)
|
| 401 |
+
3310 train 3.848773 (lr=7.6604e-04) (hash(x)=77872427)
|
| 402 |
+
3320 train 3.770685 (lr=7.6453e-04) (hash(x)=65707820)
|
| 403 |
+
3330 train 4.078043 (lr=7.6303e-04) (hash(x)=69949153)
|
| 404 |
+
3340 train 3.953735 (lr=7.6151e-04) (hash(x)=74387552)
|
| 405 |
+
3350 train 4.086580 (lr=7.6000e-04) (hash(x)=80320961)
|
| 406 |
+
3360 train 4.120649 (lr=7.5848e-04) (hash(x)=75749684)
|
| 407 |
+
3370 train 3.964887 (lr=7.5696e-04) (hash(x)=78610377)
|
| 408 |
+
3380 train 4.018353 (lr=7.5544e-04) (hash(x)=73378836)
|
| 409 |
+
3390 train 4.063714 (lr=7.5391e-04) (hash(x)=73242237)
|
| 410 |
+
3400 val loss 3.9934
|
| 411 |
+
3400 val perplexity 54.2374
|
| 412 |
+
3400 train 4.040217 (lr=7.5238e-04) (hash(x)=68240434)
|
| 413 |
+
3410 train 3.855806 (lr=7.5085e-04) (hash(x)=80087540)
|
| 414 |
+
3420 train 4.308402 (lr=7.4932e-04) (hash(x)=73529815)
|
| 415 |
+
3430 train 4.190962 (lr=7.4778e-04) (hash(x)=73306132)
|
| 416 |
+
3440 train 4.333542 (lr=7.4624e-04) (hash(x)=70500493)
|
| 417 |
+
3450 train 4.080011 (lr=7.4470e-04) (hash(x)=70007926)
|
| 418 |
+
3460 train 4.030647 (lr=7.4315e-04) (hash(x)=70024043)
|
| 419 |
+
3470 train 4.116632 (lr=7.4160e-04) (hash(x)=81605862)
|
| 420 |
+
3480 train 3.982551 (lr=7.4005e-04) (hash(x)=67926367)
|
| 421 |
+
3490 train 4.030613 (lr=7.3849e-04) (hash(x)=75476865)
|
| 422 |
+
3500 val loss 3.9769
|
| 423 |
+
3500 val perplexity 53.3505
|
| 424 |
+
3500 train 3.875070 (lr=7.3694e-04) (hash(x)=76813079)
|
| 425 |
+
3510 train 4.042651 (lr=7.3538e-04) (hash(x)=83284900)
|
| 426 |
+
3520 train 4.031151 (lr=7.3381e-04) (hash(x)=72533794)
|
| 427 |
+
3530 train 3.948003 (lr=7.3225e-04) (hash(x)=71496579)
|
| 428 |
+
3540 train 3.915365 (lr=7.3068e-04) (hash(x)=70592072)
|
| 429 |
+
3550 train 4.046032 (lr=7.2911e-04) (hash(x)=76928693)
|
| 430 |
+
3560 train 3.906385 (lr=7.2754e-04) (hash(x)=74503456)
|
| 431 |
+
3570 train 3.997969 (lr=7.2596e-04) (hash(x)=70401709)
|
| 432 |
+
3580 train 4.073640 (lr=7.2438e-04) (hash(x)=71225931)
|
| 433 |
+
3590 train 4.064763 (lr=7.2280e-04) (hash(x)=79059680)
|
| 434 |
+
3600 val loss 3.9706
|
| 435 |
+
3600 val perplexity 53.0154
|
| 436 |
+
3600 train 3.792188 (lr=7.2122e-04) (hash(x)=71650260)
|
| 437 |
+
3610 train 3.757215 (lr=7.1963e-04) (hash(x)=72844252)
|
| 438 |
+
3620 train 3.790796 (lr=7.1804e-04) (hash(x)=69989480)
|
| 439 |
+
3630 train 3.760044 (lr=7.1645e-04) (hash(x)=74336566)
|
| 440 |
+
3640 train 3.895407 (lr=7.1486e-04) (hash(x)=60353606)
|
| 441 |
+
3650 train 3.834394 (lr=7.1326e-04) (hash(x)=79424308)
|
| 442 |
+
3660 train 3.973822 (lr=7.1167e-04) (hash(x)=79976421)
|
| 443 |
+
3670 train 3.734791 (lr=7.1007e-04) (hash(x)=76742350)
|
| 444 |
+
3680 train 3.870183 (lr=7.0846e-04) (hash(x)=73225113)
|
| 445 |
+
3690 train 3.849728 (lr=7.0686e-04) (hash(x)=68913999)
|
| 446 |
+
3700 val loss 3.9799
|
| 447 |
+
3700 val perplexity 53.5128
|
| 448 |
+
3700 train 3.800503 (lr=7.0525e-04) (hash(x)=73295022)
|
| 449 |
+
3710 train 3.684872 (lr=7.0364e-04) (hash(x)=71341621)
|
| 450 |
+
3720 train 3.859540 (lr=7.0203e-04) (hash(x)=76711631)
|
| 451 |
+
3730 train 3.879754 (lr=7.0041e-04) (hash(x)=76294927)
|
| 452 |
+
3740 train 4.059006 (lr=6.9880e-04) (hash(x)=81950742)
|
| 453 |
+
3750 train 3.859149 (lr=6.9718e-04) (hash(x)=74343197)
|
| 454 |
+
3760 train 4.047125 (lr=6.9556e-04) (hash(x)=71093207)
|
| 455 |
+
3770 train 4.007198 (lr=6.9394e-04) (hash(x)=80706801)
|
| 456 |
+
3780 train 4.013706 (lr=6.9231e-04) (hash(x)=75011477)
|
| 457 |
+
3790 train 4.024292 (lr=6.9069e-04) (hash(x)=72248633)
|
| 458 |
+
3800 val loss 3.9522
|
| 459 |
+
3800 val perplexity 52.0493
|
| 460 |
+
3800 train 3.947133 (lr=6.8906e-04) (hash(x)=69136971)
|
| 461 |
+
3810 train 4.006185 (lr=6.8743e-04) (hash(x)=77522202)
|
| 462 |
+
3820 train 3.796079 (lr=6.8579e-04) (hash(x)=67695101)
|
| 463 |
+
3830 train 3.979883 (lr=6.8416e-04) (hash(x)=80051551)
|
| 464 |
+
3840 train 3.894925 (lr=6.8252e-04) (hash(x)=74191647)
|
| 465 |
+
3850 train 3.842561 (lr=6.8088e-04) (hash(x)=68047171)
|
| 466 |
+
3860 train 3.939575 (lr=6.7924e-04) (hash(x)=70045527)
|
| 467 |
+
3870 train 3.970980 (lr=6.7760e-04) (hash(x)=73697380)
|
| 468 |
+
3880 train 3.936097 (lr=6.7596e-04) (hash(x)=72925733)
|
| 469 |
+
3890 train 3.989959 (lr=6.7431e-04) (hash(x)=67921912)
|
| 470 |
+
3900 val loss 3.9443
|
| 471 |
+
3900 val perplexity 51.6376
|
| 472 |
+
3900 train 4.042893 (lr=6.7266e-04) (hash(x)=78738821)
|
| 473 |
+
3910 train 3.953716 (lr=6.7101e-04) (hash(x)=74879268)
|
| 474 |
+
3920 train 4.032205 (lr=6.6936e-04) (hash(x)=67259305)
|
| 475 |
+
3930 train 3.929624 (lr=6.6771e-04) (hash(x)=75660066)
|
| 476 |
+
3940 train 4.000964 (lr=6.6605e-04) (hash(x)=74572945)
|
| 477 |
+
3950 train 3.896636 (lr=6.6440e-04) (hash(x)=82744417)
|
| 478 |
+
3960 train 3.777549 (lr=6.6274e-04) (hash(x)=76023482)
|
| 479 |
+
3970 train 3.821615 (lr=6.6108e-04) (hash(x)=74825305)
|
| 480 |
+
3980 train 4.284239 (lr=6.5942e-04) (hash(x)=75541926)
|
| 481 |
+
3990 train 3.950650 (lr=6.5776e-04) (hash(x)=85705002)
|
| 482 |
+
4000 val loss 3.9330
|
| 483 |
+
4000 val perplexity 51.0578
|
| 484 |
+
4000 train 3.919246 (lr=6.5609e-04) (hash(x)=75062236)
|
| 485 |
+
4010 train 3.747275 (lr=6.5443e-04) (hash(x)=70532791)
|
| 486 |
+
4020 train 3.999386 (lr=6.5276e-04) (hash(x)=70014939)
|
| 487 |
+
4030 train 3.789179 (lr=6.5109e-04) (hash(x)=75315953)
|
| 488 |
+
4040 train 3.883654 (lr=6.4942e-04) (hash(x)=74206277)
|
| 489 |
+
4050 train 4.032455 (lr=6.4775e-04) (hash(x)=71678934)
|
| 490 |
+
4060 train 3.981370 (lr=6.4607e-04) (hash(x)=68169236)
|
| 491 |
+
4070 train 3.881472 (lr=6.4440e-04) (hash(x)=73440206)
|
| 492 |
+
4080 train 3.821061 (lr=6.4272e-04) (hash(x)=77333014)
|
| 493 |
+
4090 train 3.768232 (lr=6.4104e-04) (hash(x)=74699627)
|
| 494 |
+
4100 val loss 3.9492
|
| 495 |
+
4100 val perplexity 51.8932
|
| 496 |
+
4100 train 3.907684 (lr=6.3937e-04) (hash(x)=72537935)
|
| 497 |
+
4110 train 3.764351 (lr=6.3769e-04) (hash(x)=59960704)
|
| 498 |
+
4120 train 3.805621 (lr=6.3600e-04) (hash(x)=79504237)
|
| 499 |
+
4130 train 3.934200 (lr=6.3432e-04) (hash(x)=76513677)
|
| 500 |
+
4140 train 4.047119 (lr=6.3264e-04) (hash(x)=81601150)
|
| 501 |
+
4150 train 3.885881 (lr=6.3095e-04) (hash(x)=75447369)
|
| 502 |
+
4160 train 3.919132 (lr=6.2927e-04) (hash(x)=84280875)
|
| 503 |
+
4170 train 4.074868 (lr=6.2758e-04) (hash(x)=88239587)
|
| 504 |
+
4180 train 4.120585 (lr=6.2589e-04) (hash(x)=69175879)
|
| 505 |
+
4190 train 4.011086 (lr=6.2420e-04) (hash(x)=71751070)
|
| 506 |
+
4200 val loss 3.9037
|
| 507 |
+
4200 val perplexity 49.5859
|
| 508 |
+
4200 train 3.894259 (lr=6.2251e-04) (hash(x)=63098921)
|
| 509 |
+
4210 train 4.599662 (lr=6.2082e-04) (hash(x)=99229132)
|
| 510 |
+
4220 train 3.885340 (lr=6.1913e-04) (hash(x)=74149842)
|
| 511 |
+
4230 train 3.855309 (lr=6.1743e-04) (hash(x)=76000344)
|
| 512 |
+
4240 train 3.828799 (lr=6.1574e-04) (hash(x)=73033995)
|
| 513 |
+
4250 train 3.835330 (lr=6.1404e-04) (hash(x)=79351091)
|
| 514 |
+
4260 train 3.858554 (lr=6.1235e-04) (hash(x)=67587193)
|
| 515 |
+
4270 train 3.892792 (lr=6.1065e-04) (hash(x)=80457848)
|
| 516 |
+
4280 train 3.753623 (lr=6.0895e-04) (hash(x)=79619630)
|
| 517 |
+
4290 train 3.937970 (lr=6.0725e-04) (hash(x)=76884730)
|
| 518 |
+
4300 val loss 3.8907
|
| 519 |
+
4300 val perplexity 48.9431
|
| 520 |
+
4300 train 3.796824 (lr=6.0555e-04) (hash(x)=79561461)
|
| 521 |
+
4310 train 3.871387 (lr=6.0385e-04) (hash(x)=80388562)
|
| 522 |
+
4320 train 3.909767 (lr=6.0215e-04) (hash(x)=78267207)
|
| 523 |
+
4330 train 3.806445 (lr=6.0044e-04) (hash(x)=71301301)
|
| 524 |
+
4340 train 3.984378 (lr=5.9874e-04) (hash(x)=79138432)
|
| 525 |
+
4350 train 3.899526 (lr=5.9704e-04) (hash(x)=70573215)
|
| 526 |
+
4360 train 3.862393 (lr=5.9533e-04) (hash(x)=71492280)
|
| 527 |
+
4370 train 3.847002 (lr=5.9363e-04) (hash(x)=74751091)
|
| 528 |
+
4380 train 3.865252 (lr=5.9192e-04) (hash(x)=72310274)
|
| 529 |
+
4390 train 3.826214 (lr=5.9022e-04) (hash(x)=100848104)
|
| 530 |
+
4400 val loss 3.8941
|
| 531 |
+
4400 val perplexity 49.1098
|
| 532 |
+
4400 train 4.095712 (lr=5.8851e-04) (hash(x)=71196998)
|
| 533 |
+
4410 train 3.843240 (lr=5.8680e-04) (hash(x)=74205624)
|
| 534 |
+
4420 train 3.801262 (lr=5.8509e-04) (hash(x)=68781238)
|
| 535 |
+
4430 train 3.821807 (lr=5.8338e-04) (hash(x)=76961213)
|
| 536 |
+
4440 train 3.804907 (lr=5.8168e-04) (hash(x)=80006728)
|
| 537 |
+
4450 train 3.761891 (lr=5.7997e-04) (hash(x)=71337984)
|
| 538 |
+
4460 train 3.900461 (lr=5.7826e-04) (hash(x)=70514009)
|
| 539 |
+
4470 train 3.746757 (lr=5.7655e-04) (hash(x)=76238847)
|
| 540 |
+
4480 train 3.792816 (lr=5.7483e-04) (hash(x)=67646620)
|
| 541 |
+
4490 train 3.825971 (lr=5.7312e-04) (hash(x)=76200722)
|
| 542 |
+
4500 val loss 3.9071
|
| 543 |
+
4500 val perplexity 49.7544
|
| 544 |
+
4500 train 4.010912 (lr=5.7141e-04) (hash(x)=80808376)
|
| 545 |
+
4510 train 3.851734 (lr=5.6970e-04) (hash(x)=79670252)
|
| 546 |
+
4520 train 3.671171 (lr=5.6799e-04) (hash(x)=71430874)
|
| 547 |
+
4530 train 3.850690 (lr=5.6628e-04) (hash(x)=74443861)
|
| 548 |
+
4540 train 3.942988 (lr=5.6456e-04) (hash(x)=84712247)
|
| 549 |
+
4550 train 3.925770 (lr=5.6285e-04) (hash(x)=76296651)
|
| 550 |
+
4560 train 3.972735 (lr=5.6114e-04) (hash(x)=76991150)
|
| 551 |
+
4570 train 3.908177 (lr=5.5942e-04) (hash(x)=68369483)
|
| 552 |
+
4580 train 3.874026 (lr=5.5771e-04) (hash(x)=68777608)
|
| 553 |
+
4590 train 4.022985 (lr=5.5600e-04) (hash(x)=73172132)
|
| 554 |
+
4600 val loss 3.8655
|
| 555 |
+
4600 val perplexity 47.7254
|
| 556 |
+
4600 train 3.964214 (lr=5.5428e-04) (hash(x)=78012961)
|
| 557 |
+
4610 train 3.984230 (lr=5.5257e-04) (hash(x)=76270600)
|
| 558 |
+
4620 train 3.786776 (lr=5.5086e-04) (hash(x)=82249567)
|
| 559 |
+
4630 train 3.853583 (lr=5.4914e-04) (hash(x)=67713698)
|
| 560 |
+
4640 train 4.017416 (lr=5.4743e-04) (hash(x)=54462324)
|
| 561 |
+
4650 train 3.951839 (lr=5.4572e-04) (hash(x)=70825784)
|
| 562 |
+
4660 train 4.053914 (lr=5.4400e-04) (hash(x)=69303774)
|
| 563 |
+
4670 train 3.946279 (lr=5.4229e-04) (hash(x)=84155716)
|
| 564 |
+
4680 train 4.065107 (lr=5.4058e-04) (hash(x)=76409129)
|
| 565 |
+
4690 train 4.036716 (lr=5.3886e-04) (hash(x)=73455805)
|
| 566 |
+
4700 val loss 3.8598
|
| 567 |
+
4700 val perplexity 47.4577
|
| 568 |
+
4700 train 3.867461 (lr=5.3715e-04) (hash(x)=82597167)
|
| 569 |
+
4710 train 3.819057 (lr=5.3544e-04) (hash(x)=72293751)
|
| 570 |
+
4720 train 3.828080 (lr=5.3372e-04) (hash(x)=77164608)
|
| 571 |
+
4730 train 3.846265 (lr=5.3201e-04) (hash(x)=73781035)
|
| 572 |
+
4740 train 3.820917 (lr=5.3030e-04) (hash(x)=69426231)
|
| 573 |
+
4750 train 3.831368 (lr=5.2859e-04) (hash(x)=77279505)
|
| 574 |
+
4760 train 3.882812 (lr=5.2688e-04) (hash(x)=83295292)
|
| 575 |
+
4770 train 3.757774 (lr=5.2517e-04) (hash(x)=76817389)
|
| 576 |
+
4780 train 3.807137 (lr=5.2345e-04) (hash(x)=76512693)
|
| 577 |
+
4790 train 3.734309 (lr=5.2174e-04) (hash(x)=76469593)
|
| 578 |
+
4800 val loss 3.8544
|
| 579 |
+
4800 val perplexity 47.2018
|
| 580 |
+
4800 train 3.876808 (lr=5.2003e-04) (hash(x)=75145475)
|
| 581 |
+
4810 train 3.806314 (lr=5.1832e-04) (hash(x)=72173130)
|
| 582 |
+
4820 train 3.986234 (lr=5.1662e-04) (hash(x)=74239752)
|
| 583 |
+
4830 train 3.865416 (lr=5.1491e-04) (hash(x)=71592976)
|
| 584 |
+
4840 train 3.802169 (lr=5.1320e-04) (hash(x)=70809896)
|
| 585 |
+
4850 train 3.773521 (lr=5.1149e-04) (hash(x)=72044692)
|
| 586 |
+
4860 train 3.668056 (lr=5.0978e-04) (hash(x)=71791939)
|
| 587 |
+
4870 train 3.682731 (lr=5.0808e-04) (hash(x)=72591349)
|
| 588 |
+
4880 train 3.838553 (lr=5.0637e-04) (hash(x)=83767466)
|
| 589 |
+
4890 train 3.836951 (lr=5.0467e-04) (hash(x)=76798861)
|
| 590 |
+
4900 val loss 3.8541
|
| 591 |
+
4900 val perplexity 47.1845
|
| 592 |
+
4900 train 3.870109 (lr=5.0296e-04) (hash(x)=74555348)
|
| 593 |
+
4910 train 3.843024 (lr=5.0126e-04) (hash(x)=75894534)
|
| 594 |
+
4920 train 3.912025 (lr=4.9956e-04) (hash(x)=84093086)
|
| 595 |
+
4930 train 3.880683 (lr=4.9785e-04) (hash(x)=77209960)
|
| 596 |
+
4940 train 3.876299 (lr=4.9615e-04) (hash(x)=78448905)
|
| 597 |
+
4950 train 3.744340 (lr=4.9445e-04) (hash(x)=74336001)
|
| 598 |
+
4960 train 3.902538 (lr=4.9275e-04) (hash(x)=80895303)
|
| 599 |
+
4970 train 3.851454 (lr=4.9105e-04) (hash(x)=71883820)
|
| 600 |
+
4980 train 3.774611 (lr=4.8935e-04) (hash(x)=69893767)
|
| 601 |
+
4990 train 3.733173 (lr=4.8765e-04) (hash(x)=74660816)
|
| 602 |
+
5000 val loss 3.8430
|
| 603 |
+
5000 val perplexity 46.6632
|
| 604 |
+
5000 train 4.046066 (lr=4.8596e-04) (hash(x)=78057851)
|
| 605 |
+
5010 train 3.846924 (lr=4.8426e-04) (hash(x)=64235135)
|
| 606 |
+
5020 train 3.811690 (lr=4.8257e-04) (hash(x)=70397272)
|
| 607 |
+
5030 train 3.832674 (lr=4.8087e-04) (hash(x)=67171768)
|
| 608 |
+
5040 train 3.750823 (lr=4.7918e-04) (hash(x)=70332490)
|
| 609 |
+
5050 train 3.911602 (lr=4.7749e-04) (hash(x)=83006433)
|
| 610 |
+
5060 train 3.858884 (lr=4.7580e-04) (hash(x)=76575571)
|
| 611 |
+
5070 train 3.861979 (lr=4.7411e-04) (hash(x)=77691770)
|
| 612 |
+
5080 train 4.246527 (lr=4.7242e-04) (hash(x)=57255988)
|
| 613 |
+
5090 train 3.869546 (lr=4.7073e-04) (hash(x)=80108944)
|
| 614 |
+
5100 val loss 3.8350
|
| 615 |
+
5100 val perplexity 46.2952
|
| 616 |
+
5100 train 3.795482 (lr=4.6905e-04) (hash(x)=74319538)
|
| 617 |
+
5110 train 3.778503 (lr=4.6736e-04) (hash(x)=72033986)
|
| 618 |
+
5120 train 3.887837 (lr=4.6568e-04) (hash(x)=75108610)
|
| 619 |
+
5130 train 3.939145 (lr=4.6400e-04) (hash(x)=77275710)
|
| 620 |
+
5140 train 3.820401 (lr=4.6231e-04) (hash(x)=69731327)
|
| 621 |
+
5150 train 3.809475 (lr=4.6063e-04) (hash(x)=74809572)
|
| 622 |
+
5160 train 3.903027 (lr=4.5896e-04) (hash(x)=80883994)
|
| 623 |
+
5170 train 3.824394 (lr=4.5728e-04) (hash(x)=70390166)
|
| 624 |
+
5180 train 3.989122 (lr=4.5560e-04) (hash(x)=75614276)
|
| 625 |
+
5190 train 3.937702 (lr=4.5393e-04) (hash(x)=75431389)
|
| 626 |
+
5200 val loss 3.8235
|
| 627 |
+
5200 val perplexity 45.7637
|
| 628 |
+
5200 train 3.832769 (lr=4.5225e-04) (hash(x)=76239548)
|
| 629 |
+
5210 train 3.934418 (lr=4.5058e-04) (hash(x)=71988616)
|
| 630 |
+
5220 train 3.874312 (lr=4.4891e-04) (hash(x)=70379284)
|
| 631 |
+
5230 train 3.785838 (lr=4.4724e-04) (hash(x)=65937825)
|
| 632 |
+
5240 train 3.739779 (lr=4.4557e-04) (hash(x)=75529148)
|
| 633 |
+
5250 train 3.630558 (lr=4.4391e-04) (hash(x)=73186404)
|
| 634 |
+
5260 train 3.811908 (lr=4.4224e-04) (hash(x)=64888735)
|
| 635 |
+
5270 train 3.739376 (lr=4.4058e-04) (hash(x)=61047756)
|
| 636 |
+
5280 train 3.747163 (lr=4.3892e-04) (hash(x)=72880277)
|
| 637 |
+
5290 train 3.876018 (lr=4.3726e-04) (hash(x)=76312981)
|
| 638 |
+
5300 val loss 3.8136
|
| 639 |
+
5300 val perplexity 45.3148
|
| 640 |
+
5300 train 3.784813 (lr=4.3560e-04) (hash(x)=79007351)
|
| 641 |
+
5310 train 3.652471 (lr=4.3395e-04) (hash(x)=69120142)
|
| 642 |
+
5320 train 3.773806 (lr=4.3229e-04) (hash(x)=69703140)
|
| 643 |
+
5330 train 3.898447 (lr=4.3064e-04) (hash(x)=67515165)
|
| 644 |
+
5340 train 3.808012 (lr=4.2899e-04) (hash(x)=57789017)
|
| 645 |
+
5350 train 3.818954 (lr=4.2734e-04) (hash(x)=79434224)
|
| 646 |
+
5360 train 3.929090 (lr=4.2569e-04) (hash(x)=81242201)
|
| 647 |
+
5370 train 3.804117 (lr=4.2404e-04) (hash(x)=69646244)
|
| 648 |
+
5380 train 3.773784 (lr=4.2240e-04) (hash(x)=81911267)
|
| 649 |
+
5390 train 3.944068 (lr=4.2076e-04) (hash(x)=79086963)
|
| 650 |
+
5400 val loss 3.7944
|
| 651 |
+
5400 val perplexity 44.4533
|
| 652 |
+
5400 train 3.937070 (lr=4.1912e-04) (hash(x)=73958857)
|
| 653 |
+
5410 train 3.823615 (lr=4.1748e-04) (hash(x)=76487961)
|
| 654 |
+
5420 train 3.781365 (lr=4.1584e-04) (hash(x)=75600730)
|
| 655 |
+
5430 train 3.867919 (lr=4.1421e-04) (hash(x)=74683606)
|
| 656 |
+
5440 train 3.801675 (lr=4.1257e-04) (hash(x)=70879811)
|
| 657 |
+
5450 train 3.903506 (lr=4.1094e-04) (hash(x)=71654210)
|
| 658 |
+
5460 train 3.920702 (lr=4.0931e-04) (hash(x)=75386299)
|
| 659 |
+
5470 train 3.860429 (lr=4.0769e-04) (hash(x)=74366370)
|
| 660 |
+
5480 train 3.767556 (lr=4.0606e-04) (hash(x)=74267515)
|
| 661 |
+
5490 train 3.740665 (lr=4.0444e-04) (hash(x)=74435634)
|
| 662 |
+
5500 val loss 3.7846
|
| 663 |
+
5500 val perplexity 44.0182
|
| 664 |
+
5500 train 3.896198 (lr=4.0282e-04) (hash(x)=72000576)
|
| 665 |
+
5510 train 3.840467 (lr=4.0120e-04) (hash(x)=63862668)
|
| 666 |
+
5520 train 3.809402 (lr=3.9959e-04) (hash(x)=70635849)
|
| 667 |
+
5530 train 3.842788 (lr=3.9797e-04) (hash(x)=71094650)
|
| 668 |
+
5540 train 3.694288 (lr=3.9636e-04) (hash(x)=75063498)
|
| 669 |
+
5550 train 4.222892 (lr=3.9475e-04) (hash(x)=71807986)
|
| 670 |
+
5560 train 3.668620 (lr=3.9314e-04) (hash(x)=66657503)
|
| 671 |
+
5570 train 3.800491 (lr=3.9154e-04) (hash(x)=66345202)
|
| 672 |
+
5580 train 3.862552 (lr=3.8993e-04) (hash(x)=71849709)
|
| 673 |
+
5590 train 3.692991 (lr=3.8833e-04) (hash(x)=84702367)
|
| 674 |
+
5600 val loss 3.7873
|
| 675 |
+
5600 val perplexity 44.1373
|
| 676 |
+
5600 train 3.827471 (lr=3.8674e-04) (hash(x)=79109269)
|
| 677 |
+
5610 train 3.621153 (lr=3.8514e-04) (hash(x)=79051062)
|
| 678 |
+
5620 train 3.778828 (lr=3.8355e-04) (hash(x)=69861636)
|
| 679 |
+
5630 train 3.818734 (lr=3.8196e-04) (hash(x)=71549879)
|
| 680 |
+
5640 train 3.782319 (lr=3.8037e-04) (hash(x)=71055279)
|
| 681 |
+
5650 train 3.833887 (lr=3.7878e-04) (hash(x)=76078002)
|
| 682 |
+
5660 train 3.749403 (lr=3.7720e-04) (hash(x)=74335344)
|
| 683 |
+
5670 train 3.928575 (lr=3.7562e-04) (hash(x)=71454672)
|
| 684 |
+
5680 train 3.703824 (lr=3.7404e-04) (hash(x)=68592905)
|
| 685 |
+
5690 train 4.021572 (lr=3.7246e-04) (hash(x)=72942879)
|
| 686 |
+
5700 val loss 3.7753
|
| 687 |
+
5700 val perplexity 43.6123
|
| 688 |
+
5700 train 3.722890 (lr=3.7089e-04) (hash(x)=65107511)
|
| 689 |
+
5710 train 3.779541 (lr=3.6932e-04) (hash(x)=82256907)
|
| 690 |
+
5720 train 3.858287 (lr=3.6775e-04) (hash(x)=83454690)
|
| 691 |
+
5730 train 3.933522 (lr=3.6619e-04) (hash(x)=74898617)
|
| 692 |
+
5740 train 3.745434 (lr=3.6462e-04) (hash(x)=74025050)
|
| 693 |
+
5750 train 3.871578 (lr=3.6306e-04) (hash(x)=75125832)
|
| 694 |
+
5760 train 3.891793 (lr=3.6151e-04) (hash(x)=82640580)
|
| 695 |
+
5770 train 3.703700 (lr=3.5995e-04) (hash(x)=72169296)
|
| 696 |
+
5780 train 3.707704 (lr=3.5840e-04) (hash(x)=76860672)
|
| 697 |
+
5790 train 3.796941 (lr=3.5685e-04) (hash(x)=75858612)
|
| 698 |
+
5800 val loss 3.7796
|
| 699 |
+
5800 val perplexity 43.8007
|
| 700 |
+
5800 train 3.877649 (lr=3.5530e-04) (hash(x)=68326653)
|
| 701 |
+
5810 train 3.839266 (lr=3.5376e-04) (hash(x)=72416288)
|
| 702 |
+
5820 train 3.726267 (lr=3.5222e-04) (hash(x)=75390575)
|
| 703 |
+
5830 train 3.779076 (lr=3.5068e-04) (hash(x)=74274453)
|
| 704 |
+
5840 train 3.732033 (lr=3.4915e-04) (hash(x)=80802684)
|
| 705 |
+
5850 train 3.768624 (lr=3.4762e-04) (hash(x)=71662543)
|
| 706 |
+
5860 train 3.717766 (lr=3.4609e-04) (hash(x)=78625653)
|
| 707 |
+
5870 train 3.630612 (lr=3.4456e-04) (hash(x)=68749216)
|
| 708 |
+
5880 train 3.733176 (lr=3.4304e-04) (hash(x)=71914758)
|
| 709 |
+
5890 train 3.552140 (lr=3.4152e-04) (hash(x)=73589958)
|
| 710 |
+
5900 val loss 3.7644
|
| 711 |
+
5900 val perplexity 43.1383
|
| 712 |
+
5900 train 3.702776 (lr=3.4000e-04) (hash(x)=72500673)
|
| 713 |
+
5910 train 3.755464 (lr=3.3849e-04) (hash(x)=80963753)
|
| 714 |
+
5920 train 3.892773 (lr=3.3697e-04) (hash(x)=89659262)
|
| 715 |
+
5930 train 3.648090 (lr=3.3547e-04) (hash(x)=81546346)
|
| 716 |
+
5940 train 3.794944 (lr=3.3396e-04) (hash(x)=80298540)
|
| 717 |
+
5950 train 3.798755 (lr=3.3246e-04) (hash(x)=89054333)
|
| 718 |
+
5960 train 3.986171 (lr=3.3096e-04) (hash(x)=76292207)
|
| 719 |
+
5970 train 3.731558 (lr=3.2947e-04) (hash(x)=68170337)
|
| 720 |
+
5980 train 3.912131 (lr=3.2797e-04) (hash(x)=71621953)
|
| 721 |
+
5990 train 3.845360 (lr=3.2649e-04) (hash(x)=76140989)
|
| 722 |
+
6000 val loss 3.7512
|
| 723 |
+
6000 val perplexity 42.5728
|
| 724 |
+
6000 train 3.814646 (lr=3.2500e-04) (hash(x)=68289397)
|
| 725 |
+
6010 train 3.913048 (lr=3.2352e-04) (hash(x)=84960372)
|
| 726 |
+
6020 train 3.757274 (lr=3.2204e-04) (hash(x)=66386985)
|
| 727 |
+
6030 train 3.833538 (lr=3.2056e-04) (hash(x)=76775258)
|
| 728 |
+
6040 train 3.647609 (lr=3.1909e-04) (hash(x)=75546290)
|
| 729 |
+
6050 train 3.622678 (lr=3.1762e-04) (hash(x)=78124215)
|
| 730 |
+
6060 train 3.751968 (lr=3.1616e-04) (hash(x)=70915016)
|
| 731 |
+
6070 train 3.562368 (lr=3.1469e-04) (hash(x)=72509951)
|
| 732 |
+
6080 train 3.737669 (lr=3.1323e-04) (hash(x)=74035398)
|
| 733 |
+
6090 train 3.708034 (lr=3.1178e-04) (hash(x)=69655876)
|
| 734 |
+
6100 val loss 3.7436
|
| 735 |
+
6100 val perplexity 42.2505
|
| 736 |
+
6100 train 3.739582 (lr=3.1033e-04) (hash(x)=77457733)
|
| 737 |
+
6110 train 3.725136 (lr=3.0888e-04) (hash(x)=77781793)
|
| 738 |
+
6120 train 3.792169 (lr=3.0743e-04) (hash(x)=72732355)
|
| 739 |
+
6130 train 3.633420 (lr=3.0599e-04) (hash(x)=74032186)
|
| 740 |
+
6140 train 3.822440 (lr=3.0455e-04) (hash(x)=72856829)
|
| 741 |
+
6150 train 3.903080 (lr=3.0312e-04) (hash(x)=78098180)
|
| 742 |
+
6160 train 3.839728 (lr=3.0169e-04) (hash(x)=73990956)
|
| 743 |
+
6170 train 3.628723 (lr=3.0026e-04) (hash(x)=72632038)
|
| 744 |
+
6180 train 3.670879 (lr=2.9884e-04) (hash(x)=74962782)
|
| 745 |
+
6190 train 3.498938 (lr=2.9742e-04) (hash(x)=81020086)
|
| 746 |
+
6200 val loss 3.7424
|
| 747 |
+
6200 val perplexity 42.1974
|
| 748 |
+
6200 train 3.609439 (lr=2.9600e-04) (hash(x)=87386066)
|
| 749 |
+
6210 train 3.774763 (lr=2.9459e-04) (hash(x)=72382254)
|
| 750 |
+
6220 train 3.938158 (lr=2.9318e-04) (hash(x)=72565426)
|
| 751 |
+
6230 train 3.632704 (lr=2.9177e-04) (hash(x)=71306401)
|
| 752 |
+
6240 train 3.708160 (lr=2.9037e-04) (hash(x)=78821984)
|
| 753 |
+
6250 train 3.678956 (lr=2.8897e-04) (hash(x)=72065375)
|
| 754 |
+
6260 train 3.722539 (lr=2.8758e-04) (hash(x)=76100774)
|
| 755 |
+
6270 train 3.652312 (lr=2.8619e-04) (hash(x)=81896940)
|
| 756 |
+
6280 train 3.548109 (lr=2.8480e-04) (hash(x)=84686730)
|
| 757 |
+
6290 train 3.929032 (lr=2.8342e-04) (hash(x)=78981822)
|
| 758 |
+
6300 val loss 3.7263
|
| 759 |
+
6300 val perplexity 41.5268
|
| 760 |
+
6300 train 3.746358 (lr=2.8204e-04) (hash(x)=72121153)
|
| 761 |
+
6310 train 3.725568 (lr=2.8067e-04) (hash(x)=77322622)
|
| 762 |
+
6320 train 3.758033 (lr=2.7930e-04) (hash(x)=70252312)
|
| 763 |
+
6330 train 3.666236 (lr=2.7793e-04) (hash(x)=71167900)
|
| 764 |
+
6340 train 3.668889 (lr=2.7657e-04) (hash(x)=67968536)
|
| 765 |
+
6350 train 3.731276 (lr=2.7521e-04) (hash(x)=61548959)
|
| 766 |
+
6360 train 3.750206 (lr=2.7385e-04) (hash(x)=76835899)
|
| 767 |
+
6370 train 3.697115 (lr=2.7250e-04) (hash(x)=64612286)
|
| 768 |
+
6380 train 3.755551 (lr=2.7116e-04) (hash(x)=68631278)
|
| 769 |
+
6390 train 3.791091 (lr=2.6981e-04) (hash(x)=76777733)
|
| 770 |
+
6400 val loss 3.7256
|
| 771 |
+
6400 val perplexity 41.4942
|
| 772 |
+
6400 train 3.748106 (lr=2.6847e-04) (hash(x)=74010176)
|
| 773 |
+
6410 train 3.723989 (lr=2.6714e-04) (hash(x)=69368960)
|
| 774 |
+
6420 train 3.751087 (lr=2.6581e-04) (hash(x)=79065455)
|
| 775 |
+
6430 train 3.657869 (lr=2.6448e-04) (hash(x)=77858574)
|
| 776 |
+
6440 train 3.733454 (lr=2.6316e-04) (hash(x)=80136197)
|
| 777 |
+
6450 train 3.699249 (lr=2.6184e-04) (hash(x)=69612887)
|
| 778 |
+
6460 train 3.714925 (lr=2.6053e-04) (hash(x)=75004840)
|
| 779 |
+
6470 train 3.710804 (lr=2.5922e-04) (hash(x)=76506546)
|
| 780 |
+
6480 train 3.692865 (lr=2.5791e-04) (hash(x)=65864207)
|
| 781 |
+
6490 train 3.669059 (lr=2.5661e-04) (hash(x)=89740434)
|
| 782 |
+
6500 val loss 3.7220
|
| 783 |
+
6500 val perplexity 41.3463
|
| 784 |
+
6500 train 3.771426 (lr=2.5531e-04) (hash(x)=74487846)
|
| 785 |
+
6510 train 3.866957 (lr=2.5402e-04) (hash(x)=77407358)
|
| 786 |
+
6520 train 3.513989 (lr=2.5273e-04) (hash(x)=71194824)
|
| 787 |
+
6530 train 3.627395 (lr=2.5145e-04) (hash(x)=67104645)
|
| 788 |
+
6540 train 3.705828 (lr=2.5017e-04) (hash(x)=76148759)
|
| 789 |
+
6550 train 3.681012 (lr=2.4889e-04) (hash(x)=75122573)
|
| 790 |
+
6560 train 3.607277 (lr=2.4762e-04) (hash(x)=78906454)
|
| 791 |
+
6570 train 3.617641 (lr=2.4635e-04) (hash(x)=74690995)
|
| 792 |
+
6580 train 3.640659 (lr=2.4509e-04) (hash(x)=76927297)
|
| 793 |
+
6590 train 3.529907 (lr=2.4383e-04) (hash(x)=68404949)
|
| 794 |
+
6600 val loss 3.7210
|
| 795 |
+
6600 val perplexity 41.3074
|
| 796 |
+
6600 train 3.563925 (lr=2.4258e-04) (hash(x)=70670064)
|
| 797 |
+
6610 train 3.680010 (lr=2.4133e-04) (hash(x)=78186701)
|
| 798 |
+
6620 train 3.629197 (lr=2.4009e-04) (hash(x)=83774586)
|
| 799 |
+
6630 train 3.690136 (lr=2.3884e-04) (hash(x)=75862939)
|
| 800 |
+
6640 train 3.654970 (lr=2.3761e-04) (hash(x)=69660197)
|
| 801 |
+
6650 train 3.670448 (lr=2.3638e-04) (hash(x)=75014695)
|
| 802 |
+
6660 train 3.615069 (lr=2.3515e-04) (hash(x)=72303759)
|
| 803 |
+
6670 train 3.718569 (lr=2.3393e-04) (hash(x)=70865288)
|
| 804 |
+
6680 train 3.758483 (lr=2.3271e-04) (hash(x)=83055332)
|
| 805 |
+
6690 train 3.993328 (lr=2.3150e-04) (hash(x)=74803557)
|
| 806 |
+
6700 val loss 3.7102
|
| 807 |
+
6700 val perplexity 40.8617
|
| 808 |
+
6700 train 3.784678 (lr=2.3029e-04) (hash(x)=72939636)
|
| 809 |
+
6710 train 3.677957 (lr=2.2909e-04) (hash(x)=71614446)
|
| 810 |
+
6720 train 3.808560 (lr=2.2789e-04) (hash(x)=67555962)
|
| 811 |
+
6730 train 3.689415 (lr=2.2669e-04) (hash(x)=65536201)
|
| 812 |
+
6740 train 3.774216 (lr=2.2550e-04) (hash(x)=72108773)
|
| 813 |
+
6750 train 3.552342 (lr=2.2432e-04) (hash(x)=76780817)
|
| 814 |
+
6760 train 3.864172 (lr=2.2314e-04) (hash(x)=78273645)
|
| 815 |
+
6770 train 3.662185 (lr=2.2196e-04) (hash(x)=79605660)
|
| 816 |
+
6780 train 3.716326 (lr=2.2079e-04) (hash(x)=80009237)
|
| 817 |
+
6790 train 3.696447 (lr=2.1963e-04) (hash(x)=74533044)
|
| 818 |
+
6800 val loss 3.7053
|
| 819 |
+
6800 val perplexity 40.6634
|
| 820 |
+
6800 train 3.909237 (lr=2.1847e-04) (hash(x)=80026711)
|
| 821 |
+
6810 train 3.658340 (lr=2.1731e-04) (hash(x)=70338829)
|
| 822 |
+
6820 train 3.619749 (lr=2.1616e-04) (hash(x)=76285045)
|
| 823 |
+
6830 train 3.589653 (lr=2.1501e-04) (hash(x)=74154856)
|
| 824 |
+
6840 train 3.439939 (lr=2.1387e-04) (hash(x)=77355534)
|
| 825 |
+
6850 train 3.623714 (lr=2.1273e-04) (hash(x)=70624171)
|
| 826 |
+
6860 train 3.600848 (lr=2.1160e-04) (hash(x)=67874910)
|
| 827 |
+
6870 train 3.563037 (lr=2.1047e-04) (hash(x)=67935593)
|
| 828 |
+
6880 train 3.671965 (lr=2.0935e-04) (hash(x)=73054733)
|
| 829 |
+
6890 train 3.575140 (lr=2.0823e-04) (hash(x)=71490380)
|
| 830 |
+
6900 val loss 3.7051
|
| 831 |
+
6900 val perplexity 40.6540
|
| 832 |
+
6900 train 3.737555 (lr=2.0712e-04) (hash(x)=77350548)
|
| 833 |
+
6910 train 3.498832 (lr=2.0602e-04) (hash(x)=69026579)
|
| 834 |
+
6920 train 3.804343 (lr=2.0491e-04) (hash(x)=73897252)
|
| 835 |
+
6930 train 3.872568 (lr=2.0382e-04) (hash(x)=79854836)
|
| 836 |
+
6940 train 3.714164 (lr=2.0272e-04) (hash(x)=72056411)
|
| 837 |
+
6950 train 3.666084 (lr=2.0164e-04) (hash(x)=80112011)
|
| 838 |
+
6960 train 3.727348 (lr=2.0055e-04) (hash(x)=57470392)
|
| 839 |
+
6970 train 3.647768 (lr=1.9948e-04) (hash(x)=70347716)
|
| 840 |
+
6980 train 3.832709 (lr=1.9840e-04) (hash(x)=82281687)
|
| 841 |
+
6990 train 3.730881 (lr=1.9734e-04) (hash(x)=74694435)
|
| 842 |
+
7000 val loss 3.6920
|
| 843 |
+
7000 val perplexity 40.1236
|
| 844 |
+
7000 train 3.707842 (lr=1.9628e-04) (hash(x)=81879854)
|
| 845 |
+
7010 train 3.738984 (lr=1.9522e-04) (hash(x)=69165491)
|
| 846 |
+
7020 train 3.639477 (lr=1.9417e-04) (hash(x)=75320851)
|
| 847 |
+
7030 train 3.695074 (lr=1.9312e-04) (hash(x)=77375045)
|
| 848 |
+
7040 train 3.701880 (lr=1.9208e-04) (hash(x)=80127540)
|
| 849 |
+
7050 train 3.523616 (lr=1.9104e-04) (hash(x)=74056421)
|
| 850 |
+
7060 train 3.650572 (lr=1.9001e-04) (hash(x)=76784854)
|
| 851 |
+
7070 train 3.583572 (lr=1.8899e-04) (hash(x)=78122882)
|
| 852 |
+
7080 train 3.558365 (lr=1.8797e-04) (hash(x)=75152878)
|
| 853 |
+
7090 train 3.613444 (lr=1.8695e-04) (hash(x)=79990627)
|
| 854 |
+
7100 val loss 3.6875
|
| 855 |
+
7100 val perplexity 39.9429
|
| 856 |
+
7100 train 3.680961 (lr=1.8594e-04) (hash(x)=70855798)
|
| 857 |
+
7110 train 3.641630 (lr=1.8494e-04) (hash(x)=73080655)
|
| 858 |
+
7120 train 3.653192 (lr=1.8394e-04) (hash(x)=69035245)
|
| 859 |
+
7130 train 3.419105 (lr=1.8294e-04) (hash(x)=64555239)
|
| 860 |
+
7140 train 3.565031 (lr=1.8196e-04) (hash(x)=71734297)
|
| 861 |
+
7150 train 3.605397 (lr=1.8097e-04) (hash(x)=82270339)
|
| 862 |
+
7160 train 3.483007 (lr=1.7999e-04) (hash(x)=76649629)
|
| 863 |
+
7170 train 3.568456 (lr=1.7902e-04) (hash(x)=73236815)
|
| 864 |
+
7180 train 3.475019 (lr=1.7805e-04) (hash(x)=73739555)
|
| 865 |
+
7190 train 3.571881 (lr=1.7709e-04) (hash(x)=65956849)
|
| 866 |
+
7200 val loss 3.6881
|
| 867 |
+
7200 val perplexity 39.9672
|
| 868 |
+
7200 train 3.490003 (lr=1.7614e-04) (hash(x)=76352714)
|
| 869 |
+
7210 train 3.612017 (lr=1.7519e-04) (hash(x)=77523342)
|
| 870 |
+
7220 train 3.680798 (lr=1.7424e-04) (hash(x)=75193820)
|
| 871 |
+
7230 train 3.608378 (lr=1.7330e-04) (hash(x)=72975023)
|
| 872 |
+
7240 train 3.599787 (lr=1.7237e-04) (hash(x)=70152397)
|
| 873 |
+
7250 train 3.595937 (lr=1.7144e-04) (hash(x)=78430692)
|
| 874 |
+
7260 train 3.722397 (lr=1.7051e-04) (hash(x)=73137006)
|
| 875 |
+
7270 train 3.642856 (lr=1.6959e-04) (hash(x)=76187623)
|
| 876 |
+
7280 train 3.667818 (lr=1.6868e-04) (hash(x)=76534981)
|
| 877 |
+
7290 train 3.850186 (lr=1.6777e-04) (hash(x)=77886369)
|
| 878 |
+
7300 val loss 3.6800
|
| 879 |
+
7300 val perplexity 39.6466
|
| 880 |
+
7300 train 3.701992 (lr=1.6687e-04) (hash(x)=73689615)
|
| 881 |
+
7310 train 3.707147 (lr=1.6598e-04) (hash(x)=73658711)
|
| 882 |
+
7320 train 3.566888 (lr=1.6509e-04) (hash(x)=81887275)
|
| 883 |
+
7330 train 3.583124 (lr=1.6420e-04) (hash(x)=69952710)
|
| 884 |
+
7340 train 3.728839 (lr=1.6332e-04) (hash(x)=74902753)
|
| 885 |
+
7350 train 3.621027 (lr=1.6245e-04) (hash(x)=76782773)
|
| 886 |
+
7360 train 3.545683 (lr=1.6158e-04) (hash(x)=73139063)
|
| 887 |
+
7370 train 3.713619 (lr=1.6072e-04) (hash(x)=83222375)
|
| 888 |
+
7380 train 3.646162 (lr=1.5986e-04) (hash(x)=74319008)
|
| 889 |
+
7390 train 3.495682 (lr=1.5901e-04) (hash(x)=79871842)
|
| 890 |
+
7400 val loss 3.6763
|
| 891 |
+
7400 val perplexity 39.4997
|
| 892 |
+
7400 train 3.673163 (lr=1.5816e-04) (hash(x)=74853788)
|
| 893 |
+
7410 train 3.626156 (lr=1.5732e-04) (hash(x)=79119178)
|
| 894 |
+
7420 train 3.576902 (lr=1.5649e-04) (hash(x)=71667016)
|
| 895 |
+
7430 train 3.557258 (lr=1.5566e-04) (hash(x)=70294981)
|
| 896 |
+
7440 train 3.618197 (lr=1.5484e-04) (hash(x)=76749861)
|
| 897 |
+
7450 train 3.441044 (lr=1.5402e-04) (hash(x)=67932362)
|
| 898 |
+
7460 train 3.588001 (lr=1.5321e-04) (hash(x)=71646803)
|
| 899 |
+
7470 train 3.512496 (lr=1.5241e-04) (hash(x)=76832078)
|
| 900 |
+
7480 train 3.549227 (lr=1.5161e-04) (hash(x)=78030848)
|
| 901 |
+
7490 train 3.532587 (lr=1.5081e-04) (hash(x)=79789250)
|
| 902 |
+
7500 val loss 3.6748
|
| 903 |
+
7500 val perplexity 39.4388
|
| 904 |
+
7500 train 3.584122 (lr=1.5002e-04) (hash(x)=74510406)
|
| 905 |
+
7510 train 3.614876 (lr=1.4924e-04) (hash(x)=75354875)
|
| 906 |
+
7520 train 3.647244 (lr=1.4847e-04) (hash(x)=75617004)
|
| 907 |
+
7530 train 3.654216 (lr=1.4769e-04) (hash(x)=70778712)
|
| 908 |
+
7540 train 3.853573 (lr=1.4693e-04) (hash(x)=65376419)
|
| 909 |
+
7550 train 3.667049 (lr=1.4617e-04) (hash(x)=72248752)
|
| 910 |
+
7560 train 3.673964 (lr=1.4542e-04) (hash(x)=73692695)
|
| 911 |
+
7570 train 3.667290 (lr=1.4467e-04) (hash(x)=74787412)
|
| 912 |
+
7580 train 3.748160 (lr=1.4393e-04) (hash(x)=78483480)
|
| 913 |
+
7590 train 3.630544 (lr=1.4319e-04) (hash(x)=73847861)
|
| 914 |
+
7600 val loss 3.6631
|
| 915 |
+
7600 val perplexity 38.9807
|
| 916 |
+
7600 train 3.615480 (lr=1.4246e-04) (hash(x)=71708692)
|
| 917 |
+
7610 train 3.681148 (lr=1.4174e-04) (hash(x)=79089102)
|
| 918 |
+
7620 train 3.566707 (lr=1.4102e-04) (hash(x)=71032201)
|
| 919 |
+
7630 train 3.727380 (lr=1.4031e-04) (hash(x)=71581412)
|
| 920 |
+
7640 train 3.652773 (lr=1.3960e-04) (hash(x)=79778462)
|
| 921 |
+
7650 train 3.850894 (lr=1.3890e-04) (hash(x)=71585919)
|
| 922 |
+
7660 train 3.748426 (lr=1.3821e-04) (hash(x)=72139974)
|
| 923 |
+
7670 train 3.512209 (lr=1.3752e-04) (hash(x)=75620437)
|
| 924 |
+
7680 train 3.754200 (lr=1.3684e-04) (hash(x)=71555417)
|
| 925 |
+
7690 train 3.520679 (lr=1.3616e-04) (hash(x)=63719984)
|
| 926 |
+
7700 val loss 3.6634
|
| 927 |
+
7700 val perplexity 38.9919
|
| 928 |
+
7700 train 3.576077 (lr=1.3549e-04) (hash(x)=74494467)
|
| 929 |
+
7710 train 3.500696 (lr=1.3483e-04) (hash(x)=76821872)
|
| 930 |
+
7720 train 3.523712 (lr=1.3417e-04) (hash(x)=76993734)
|
| 931 |
+
7730 train 3.411629 (lr=1.3352e-04) (hash(x)=74942500)
|
| 932 |
+
7740 train 3.652175 (lr=1.3287e-04) (hash(x)=72583299)
|
| 933 |
+
7750 train 3.415241 (lr=1.3223e-04) (hash(x)=68209490)
|
| 934 |
+
7760 train 3.475169 (lr=1.3160e-04) (hash(x)=69093853)
|
| 935 |
+
7770 train 3.661530 (lr=1.3097e-04) (hash(x)=74460215)
|
| 936 |
+
7780 train 3.589716 (lr=1.3035e-04) (hash(x)=68407147)
|
| 937 |
+
7790 train 3.401358 (lr=1.2974e-04) (hash(x)=76153902)
|
| 938 |
+
7800 val loss 3.6641
|
| 939 |
+
7800 val perplexity 39.0217
|
| 940 |
+
7800 train 3.446107 (lr=1.2913e-04) (hash(x)=84446164)
|
| 941 |
+
7810 train 3.659085 (lr=1.2852e-04) (hash(x)=73681777)
|
| 942 |
+
7820 train 3.678928 (lr=1.2793e-04) (hash(x)=79064713)
|
| 943 |
+
7830 train 3.647282 (lr=1.2733e-04) (hash(x)=72713629)
|
| 944 |
+
7840 train 3.862705 (lr=1.2675e-04) (hash(x)=83622263)
|
| 945 |
+
7850 train 3.620575 (lr=1.2617e-04) (hash(x)=73574852)
|
| 946 |
+
7860 train 3.744040 (lr=1.2560e-04) (hash(x)=67285714)
|
| 947 |
+
7870 train 3.727985 (lr=1.2503e-04) (hash(x)=77222135)
|
| 948 |
+
7880 train 3.757639 (lr=1.2447e-04) (hash(x)=73302118)
|
| 949 |
+
7890 train 3.617595 (lr=1.2392e-04) (hash(x)=61831553)
|
| 950 |
+
7900 val loss 3.6559
|
| 951 |
+
7900 val perplexity 38.7007
|
| 952 |
+
7900 train 3.576494 (lr=1.2337e-04) (hash(x)=70412356)
|
| 953 |
+
7910 train 3.749923 (lr=1.2283e-04) (hash(x)=69987458)
|
| 954 |
+
7920 train 3.720824 (lr=1.2229e-04) (hash(x)=72573609)
|
| 955 |
+
7930 train 3.469617 (lr=1.2176e-04) (hash(x)=67089230)
|
| 956 |
+
7940 train 3.561009 (lr=1.2124e-04) (hash(x)=66572639)
|
| 957 |
+
7950 train 3.671310 (lr=1.2072e-04) (hash(x)=66156765)
|
| 958 |
+
7960 train 3.632620 (lr=1.2021e-04) (hash(x)=75055477)
|
| 959 |
+
7970 train 3.742238 (lr=1.1970e-04) (hash(x)=65210584)
|
| 960 |
+
7980 train 3.549185 (lr=1.1921e-04) (hash(x)=63802308)
|
| 961 |
+
7990 train 3.641349 (lr=1.1871e-04) (hash(x)=76379611)
|
| 962 |
+
8000 val loss 3.6532
|
| 963 |
+
8000 val perplexity 38.5991
|
| 964 |
+
8000 train 3.620697 (lr=1.1823e-04) (hash(x)=68482769)
|
| 965 |
+
8010 train 3.457824 (lr=1.1775e-04) (hash(x)=77934567)
|
| 966 |
+
8020 train 3.546319 (lr=1.1728e-04) (hash(x)=78687646)
|
| 967 |
+
8030 train 3.450068 (lr=1.1681e-04) (hash(x)=73044643)
|
| 968 |
+
8040 train 3.506208 (lr=1.1635e-04) (hash(x)=68884035)
|
| 969 |
+
8050 train 3.331522 (lr=1.1589e-04) (hash(x)=74889605)
|
| 970 |
+
8060 train 3.522590 (lr=1.1544e-04) (hash(x)=76659012)
|
| 971 |
+
8070 train 3.617751 (lr=1.1500e-04) (hash(x)=70403652)
|
| 972 |
+
8080 train 3.471875 (lr=1.1457e-04) (hash(x)=78290747)
|
| 973 |
+
8090 train 3.539882 (lr=1.1414e-04) (hash(x)=81394906)
|
| 974 |
+
8100 val loss 3.6557
|
| 975 |
+
8100 val perplexity 38.6956
|
| 976 |
+
8100 train 3.519256 (lr=1.1371e-04) (hash(x)=67764612)
|
| 977 |
+
8110 train 3.531004 (lr=1.1330e-04) (hash(x)=76334301)
|
| 978 |
+
8120 train 3.595799 (lr=1.1289e-04) (hash(x)=75412335)
|
| 979 |
+
8130 train 3.525723 (lr=1.1248e-04) (hash(x)=74846582)
|
| 980 |
+
8140 train 3.580329 (lr=1.1209e-04) (hash(x)=80776867)
|
| 981 |
+
8150 train 3.605508 (lr=1.1169e-04) (hash(x)=74759651)
|
| 982 |
+
8160 train 3.614704 (lr=1.1131e-04) (hash(x)=75135943)
|
| 983 |
+
8170 train 3.608184 (lr=1.1093e-04) (hash(x)=88547984)
|
| 984 |
+
8180 train 3.686463 (lr=1.1056e-04) (hash(x)=67588135)
|
| 985 |
+
8190 train 3.530904 (lr=1.1019e-04) (hash(x)=75190422)
|
| 986 |
+
8200 val loss 3.6496
|
| 987 |
+
8200 val perplexity 38.4602
|
| 988 |
+
8200 train 3.528319 (lr=1.0983e-04) (hash(x)=79609415)
|
| 989 |
+
8210 train 3.654772 (lr=1.0948e-04) (hash(x)=73291424)
|
| 990 |
+
8220 train 3.646541 (lr=1.0913e-04) (hash(x)=76172342)
|
| 991 |
+
8230 train 3.579254 (lr=1.0879e-04) (hash(x)=72229274)
|
| 992 |
+
8240 train 3.529650 (lr=1.0846e-04) (hash(x)=70393283)
|
| 993 |
+
8250 train 3.603933 (lr=1.0813e-04) (hash(x)=76659165)
|
| 994 |
+
8260 train 3.586252 (lr=1.0781e-04) (hash(x)=75284539)
|
| 995 |
+
8270 train 3.363368 (lr=1.0750e-04) (hash(x)=71495901)
|
| 996 |
+
8280 train 3.597528 (lr=1.0719e-04) (hash(x)=74742339)
|
| 997 |
+
8290 train 3.697159 (lr=1.0689e-04) (hash(x)=73748628)
|
| 998 |
+
8300 val loss 3.6489
|
| 999 |
+
8300 val perplexity 38.4340
|
| 1000 |
+
8300 train 3.675335 (lr=1.0659e-04) (hash(x)=75127072)
|
| 1001 |
+
8310 train 3.515121 (lr=1.0630e-04) (hash(x)=69260375)
|
| 1002 |
+
8320 train 3.535777 (lr=1.0602e-04) (hash(x)=77687746)
|
| 1003 |
+
8330 train 3.466916 (lr=1.0574e-04) (hash(x)=69399950)
|
| 1004 |
+
8340 train 3.686811 (lr=1.0547e-04) (hash(x)=79102450)
|
| 1005 |
+
8350 train 3.624428 (lr=1.0521e-04) (hash(x)=74555497)
|
| 1006 |
+
8360 train 3.891762 (lr=1.0495e-04) (hash(x)=65367249)
|
| 1007 |
+
8370 train 3.801868 (lr=1.0470e-04) (hash(x)=78862619)
|
| 1008 |
+
8380 train 3.592611 (lr=1.0446e-04) (hash(x)=70989373)
|
| 1009 |
+
8390 train 3.709392 (lr=1.0422e-04) (hash(x)=80650634)
|
| 1010 |
+
8400 val loss 3.6415
|
| 1011 |
+
8400 val perplexity 38.1496
|
| 1012 |
+
8400 train 3.582992 (lr=1.0399e-04) (hash(x)=74168359)
|
| 1013 |
+
8410 train 3.681455 (lr=1.0377e-04) (hash(x)=80595745)
|
| 1014 |
+
8420 train 3.633307 (lr=1.0355e-04) (hash(x)=74239969)
|
| 1015 |
+
8430 train 3.648883 (lr=1.0334e-04) (hash(x)=72089061)
|
| 1016 |
+
8440 train 3.654515 (lr=1.0313e-04) (hash(x)=79471873)
|
| 1017 |
+
8450 train 3.650064 (lr=1.0293e-04) (hash(x)=70814855)
|
| 1018 |
+
8460 train 3.604288 (lr=1.0274e-04) (hash(x)=72900750)
|
| 1019 |
+
8470 train 3.747749 (lr=1.0256e-04) (hash(x)=73723555)
|
| 1020 |
+
8480 train 3.577946 (lr=1.0238e-04) (hash(x)=77750599)
|
| 1021 |
+
8490 train 3.581637 (lr=1.0220e-04) (hash(x)=82913770)
|
| 1022 |
+
8500 val loss 3.6393
|
| 1023 |
+
8500 val perplexity 38.0661
|
| 1024 |
+
8500 train 3.756681 (lr=1.0204e-04) (hash(x)=58014966)
|
| 1025 |
+
8510 train 3.610657 (lr=1.0188e-04) (hash(x)=78964850)
|
| 1026 |
+
8520 train 3.834585 (lr=1.0172e-04) (hash(x)=79106421)
|
| 1027 |
+
8530 train 3.616487 (lr=1.0158e-04) (hash(x)=75662227)
|
| 1028 |
+
8540 train 3.494726 (lr=1.0144e-04) (hash(x)=73928041)
|
| 1029 |
+
8550 train 3.649508 (lr=1.0130e-04) (hash(x)=73949623)
|
| 1030 |
+
8560 train 3.492687 (lr=1.0118e-04) (hash(x)=66671864)
|
| 1031 |
+
8570 train 3.556316 (lr=1.0106e-04) (hash(x)=73819413)
|
| 1032 |
+
8580 train 3.510235 (lr=1.0094e-04) (hash(x)=71116170)
|
| 1033 |
+
8590 train 3.515911 (lr=1.0083e-04) (hash(x)=75792152)
|
| 1034 |
+
8600 val loss 3.6428
|
| 1035 |
+
8600 val perplexity 38.1969
|
| 1036 |
+
8600 train 3.480192 (lr=1.0073e-04) (hash(x)=76199926)
|
| 1037 |
+
8610 train 3.456395 (lr=1.0064e-04) (hash(x)=72341134)
|
| 1038 |
+
8620 train 3.504135 (lr=1.0055e-04) (hash(x)=73669000)
|
| 1039 |
+
8630 train 3.587533 (lr=1.0047e-04) (hash(x)=76197518)
|
| 1040 |
+
8640 train 3.586119 (lr=1.0039e-04) (hash(x)=77186492)
|
| 1041 |
+
8650 train 3.582174 (lr=1.0033e-04) (hash(x)=72965152)
|
| 1042 |
+
8660 train 3.615223 (lr=1.0026e-04) (hash(x)=66084492)
|
| 1043 |
+
8670 train 3.700512 (lr=1.0021e-04) (hash(x)=77644722)
|
| 1044 |
+
8680 train 3.528085 (lr=1.0016e-04) (hash(x)=70362453)
|
| 1045 |
+
8690 train 3.636358 (lr=1.0012e-04) (hash(x)=77832585)
|
| 1046 |
+
8700 val loss 3.6341
|
| 1047 |
+
8700 val perplexity 37.8663
|
| 1048 |
+
8700 train 3.727388 (lr=1.0008e-04) (hash(x)=93249763)
|
| 1049 |
+
8710 train 3.673150 (lr=1.0005e-04) (hash(x)=88835109)
|
| 1050 |
+
8720 train 3.790535 (lr=1.0003e-04) (hash(x)=76773769)
|
| 1051 |
+
8730 train 3.594644 (lr=1.0001e-04) (hash(x)=78751013)
|
| 1052 |
+
8740 train 3.531550 (lr=1.0000e-04) (hash(x)=72850278)
|
| 1053 |
+
8749 val loss 3.6338
|
| 1054 |
+
8749 val perplexity 37.8553
|
n_heads12_lr10e-4_total_batch_size61440_seed1338/model_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:183aa52ffebb0e9ab4fb70fc95d04d38858a385e34e7028d3e6c59d89765c190
|
| 3 |
+
size 498607490
|
n_heads12_lr10e-4_total_batch_size61440_seed1338/optimizer_08749.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9715ac7ee691eea63cfda50c5bfa9079aa910e7ebec5c1b69acb5b7bb68e42cf
|
| 3 |
+
size 990934406
|