lewtun HF staff commited on
Commit
e849365
1 Parent(s): 744636b

Model save

Browse files
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ datasets:
9
+ - generator
10
+ base_model: mistralai/Mistral-7B-v0.1
11
+ model-index:
12
+ - name: zephyr-7b-sft-qlora
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # zephyr-7b-sft-qlora
20
+
21
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.9502
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 4
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - gradient_accumulation_steps: 2
49
+ - total_train_batch_size: 64
50
+ - total_eval_batch_size: 64
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:-----:|:----:|:---------------:|
60
+ | 0.9428 | 1.0 | 2179 | 0.9502 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.7.1
66
+ - Transformers 4.36.2
67
+ - Pytorch 2.1.2+cu121
68
+ - Datasets 2.14.6
69
+ - Tokenizers 0.15.0
adapter_config.json CHANGED
@@ -19,13 +19,13 @@
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
22
- "k_proj",
23
- "q_proj",
24
- "o_proj",
25
- "gate_proj",
26
  "down_proj",
 
 
27
  "v_proj",
28
- "up_proj"
 
 
29
  ],
30
  "task_type": "CAUSAL_LM"
31
  }
 
19
  "rank_pattern": {},
20
  "revision": null,
21
  "target_modules": [
 
 
 
 
22
  "down_proj",
23
+ "up_proj",
24
+ "q_proj",
25
  "v_proj",
26
+ "k_proj",
27
+ "gate_proj",
28
+ "o_proj"
29
  ],
30
  "task_type": "CAUSAL_LM"
31
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:32ec71851adec8e267d3f0c19a58d1cbf8a543868ed469f6e9a3993b61ef1b7d
3
  size 83946192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd1a9b20633bf82bd4d577e00e9dca91e0aa5ca500846bc0806c8c201e25c0b7
3
  size 83946192
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9502018094062805,
4
+ "eval_runtime": 365.0951,
5
+ "eval_samples": 23110,
6
+ "eval_samples_per_second": 42.266,
7
+ "eval_steps_per_second": 0.663,
8
+ "train_loss": 0.03432130868306889,
9
+ "train_runtime": 782.9437,
10
+ "train_samples": 207865,
11
+ "train_samples_per_second": 178.079,
12
+ "train_steps_per_second": 2.783
13
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 0.9502018094062805,
4
+ "eval_runtime": 365.0951,
5
+ "eval_samples": 23110,
6
+ "eval_samples_per_second": 42.266,
7
+ "eval_steps_per_second": 0.663
8
+ }
runs/Jan04_06-33-14_ip-26-0-171-168/events.out.tfevents.1704350705.ip-26-0-171-168.2251825.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7e9cc9682d4c16e148cbb2f22938e33846433bd42f668c37854d67dc0f958bdb
3
- size 70841
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75b09bee8cb1327361ccda08e54bb180418a466a3e21440ecfacd68ba2c715f7
3
+ size 73821
runs/Jan04_06-33-14_ip-26-0-171-168/events.out.tfevents.1704362888.ip-26-0-171-168.2251825.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84f28eed4146bfaf4a04c1ac01e9a580b9b59ade4fc093191c5481edd68ce0c9
3
+ size 359
runs/Jan04_10-08-35_ip-26-0-171-168/events.out.tfevents.1704362977.ip-26-0-171-168.2277076.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18746eb9cfeb4581ff9a4457cd0dcc96f5f10c7547844f64ef609bf33741cf40
3
+ size 7802
runs/Jan04_10-08-35_ip-26-0-171-168/events.out.tfevents.1704364125.ip-26-0-171-168.2277076.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cabaa14dc4ea96ce0a7197934bd39bf1f8c1b7325cb5b70032f764f792b5691d
3
+ size 359
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "train_loss": 0.03432130868306889,
4
+ "train_runtime": 782.9437,
5
+ "train_samples": 207865,
6
+ "train_samples_per_second": 178.079,
7
+ "train_steps_per_second": 2.783
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,2654 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 2179,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 9.174311926605506e-07,
14
+ "loss": 1.136,
15
+ "step": 1
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 4.587155963302753e-06,
20
+ "loss": 1.1485,
21
+ "step": 5
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 9.174311926605506e-06,
26
+ "loss": 1.1204,
27
+ "step": 10
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 1.3761467889908258e-05,
32
+ "loss": 1.1304,
33
+ "step": 15
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 1.834862385321101e-05,
38
+ "loss": 1.0939,
39
+ "step": 20
40
+ },
41
+ {
42
+ "epoch": 0.01,
43
+ "learning_rate": 2.2935779816513765e-05,
44
+ "loss": 1.0726,
45
+ "step": 25
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "learning_rate": 2.7522935779816515e-05,
50
+ "loss": 1.0481,
51
+ "step": 30
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 3.211009174311927e-05,
56
+ "loss": 1.0549,
57
+ "step": 35
58
+ },
59
+ {
60
+ "epoch": 0.02,
61
+ "learning_rate": 3.669724770642202e-05,
62
+ "loss": 1.0569,
63
+ "step": 40
64
+ },
65
+ {
66
+ "epoch": 0.02,
67
+ "learning_rate": 4.1284403669724776e-05,
68
+ "loss": 1.0303,
69
+ "step": 45
70
+ },
71
+ {
72
+ "epoch": 0.02,
73
+ "learning_rate": 4.587155963302753e-05,
74
+ "loss": 1.0319,
75
+ "step": 50
76
+ },
77
+ {
78
+ "epoch": 0.03,
79
+ "learning_rate": 5.0458715596330276e-05,
80
+ "loss": 1.0037,
81
+ "step": 55
82
+ },
83
+ {
84
+ "epoch": 0.03,
85
+ "learning_rate": 5.504587155963303e-05,
86
+ "loss": 1.007,
87
+ "step": 60
88
+ },
89
+ {
90
+ "epoch": 0.03,
91
+ "learning_rate": 5.9633027522935784e-05,
92
+ "loss": 1.0205,
93
+ "step": 65
94
+ },
95
+ {
96
+ "epoch": 0.03,
97
+ "learning_rate": 6.422018348623854e-05,
98
+ "loss": 0.9893,
99
+ "step": 70
100
+ },
101
+ {
102
+ "epoch": 0.03,
103
+ "learning_rate": 6.880733944954129e-05,
104
+ "loss": 1.0094,
105
+ "step": 75
106
+ },
107
+ {
108
+ "epoch": 0.04,
109
+ "learning_rate": 7.339449541284404e-05,
110
+ "loss": 1.0048,
111
+ "step": 80
112
+ },
113
+ {
114
+ "epoch": 0.04,
115
+ "learning_rate": 7.79816513761468e-05,
116
+ "loss": 1.0111,
117
+ "step": 85
118
+ },
119
+ {
120
+ "epoch": 0.04,
121
+ "learning_rate": 8.256880733944955e-05,
122
+ "loss": 1.0033,
123
+ "step": 90
124
+ },
125
+ {
126
+ "epoch": 0.04,
127
+ "learning_rate": 8.715596330275229e-05,
128
+ "loss": 1.0049,
129
+ "step": 95
130
+ },
131
+ {
132
+ "epoch": 0.05,
133
+ "learning_rate": 9.174311926605506e-05,
134
+ "loss": 0.9886,
135
+ "step": 100
136
+ },
137
+ {
138
+ "epoch": 0.05,
139
+ "learning_rate": 9.63302752293578e-05,
140
+ "loss": 0.9863,
141
+ "step": 105
142
+ },
143
+ {
144
+ "epoch": 0.05,
145
+ "learning_rate": 0.00010091743119266055,
146
+ "loss": 0.9712,
147
+ "step": 110
148
+ },
149
+ {
150
+ "epoch": 0.05,
151
+ "learning_rate": 0.00010550458715596329,
152
+ "loss": 1.0071,
153
+ "step": 115
154
+ },
155
+ {
156
+ "epoch": 0.06,
157
+ "learning_rate": 0.00011009174311926606,
158
+ "loss": 1.0044,
159
+ "step": 120
160
+ },
161
+ {
162
+ "epoch": 0.06,
163
+ "learning_rate": 0.00011467889908256881,
164
+ "loss": 0.9665,
165
+ "step": 125
166
+ },
167
+ {
168
+ "epoch": 0.06,
169
+ "learning_rate": 0.00011926605504587157,
170
+ "loss": 0.9981,
171
+ "step": 130
172
+ },
173
+ {
174
+ "epoch": 0.06,
175
+ "learning_rate": 0.00012385321100917432,
176
+ "loss": 0.9578,
177
+ "step": 135
178
+ },
179
+ {
180
+ "epoch": 0.06,
181
+ "learning_rate": 0.00012844036697247707,
182
+ "loss": 0.986,
183
+ "step": 140
184
+ },
185
+ {
186
+ "epoch": 0.07,
187
+ "learning_rate": 0.00013302752293577983,
188
+ "loss": 0.9935,
189
+ "step": 145
190
+ },
191
+ {
192
+ "epoch": 0.07,
193
+ "learning_rate": 0.00013761467889908258,
194
+ "loss": 0.953,
195
+ "step": 150
196
+ },
197
+ {
198
+ "epoch": 0.07,
199
+ "learning_rate": 0.0001422018348623853,
200
+ "loss": 0.9565,
201
+ "step": 155
202
+ },
203
+ {
204
+ "epoch": 0.07,
205
+ "learning_rate": 0.0001467889908256881,
206
+ "loss": 0.9737,
207
+ "step": 160
208
+ },
209
+ {
210
+ "epoch": 0.08,
211
+ "learning_rate": 0.00015137614678899084,
212
+ "loss": 0.9819,
213
+ "step": 165
214
+ },
215
+ {
216
+ "epoch": 0.08,
217
+ "learning_rate": 0.0001559633027522936,
218
+ "loss": 0.9927,
219
+ "step": 170
220
+ },
221
+ {
222
+ "epoch": 0.08,
223
+ "learning_rate": 0.00016055045871559632,
224
+ "loss": 0.9803,
225
+ "step": 175
226
+ },
227
+ {
228
+ "epoch": 0.08,
229
+ "learning_rate": 0.0001651376146788991,
230
+ "loss": 0.9545,
231
+ "step": 180
232
+ },
233
+ {
234
+ "epoch": 0.08,
235
+ "learning_rate": 0.00016972477064220186,
236
+ "loss": 0.9927,
237
+ "step": 185
238
+ },
239
+ {
240
+ "epoch": 0.09,
241
+ "learning_rate": 0.00017431192660550458,
242
+ "loss": 0.9737,
243
+ "step": 190
244
+ },
245
+ {
246
+ "epoch": 0.09,
247
+ "learning_rate": 0.00017889908256880734,
248
+ "loss": 0.9959,
249
+ "step": 195
250
+ },
251
+ {
252
+ "epoch": 0.09,
253
+ "learning_rate": 0.00018348623853211012,
254
+ "loss": 0.9655,
255
+ "step": 200
256
+ },
257
+ {
258
+ "epoch": 0.09,
259
+ "learning_rate": 0.00018807339449541284,
260
+ "loss": 0.951,
261
+ "step": 205
262
+ },
263
+ {
264
+ "epoch": 0.1,
265
+ "learning_rate": 0.0001926605504587156,
266
+ "loss": 0.9531,
267
+ "step": 210
268
+ },
269
+ {
270
+ "epoch": 0.1,
271
+ "learning_rate": 0.00019724770642201835,
272
+ "loss": 0.9682,
273
+ "step": 215
274
+ },
275
+ {
276
+ "epoch": 0.1,
277
+ "learning_rate": 0.00019999948669655127,
278
+ "loss": 0.9663,
279
+ "step": 220
280
+ },
281
+ {
282
+ "epoch": 0.1,
283
+ "learning_rate": 0.00019999371209327089,
284
+ "loss": 0.9875,
285
+ "step": 225
286
+ },
287
+ {
288
+ "epoch": 0.11,
289
+ "learning_rate": 0.00019998152162914806,
290
+ "loss": 0.9624,
291
+ "step": 230
292
+ },
293
+ {
294
+ "epoch": 0.11,
295
+ "learning_rate": 0.00019996291608635525,
296
+ "loss": 0.9615,
297
+ "step": 235
298
+ },
299
+ {
300
+ "epoch": 0.11,
301
+ "learning_rate": 0.00019993789665867314,
302
+ "loss": 0.9677,
303
+ "step": 240
304
+ },
305
+ {
306
+ "epoch": 0.11,
307
+ "learning_rate": 0.00019990646495141444,
308
+ "loss": 0.961,
309
+ "step": 245
310
+ },
311
+ {
312
+ "epoch": 0.11,
313
+ "learning_rate": 0.0001998686229813205,
314
+ "loss": 0.9984,
315
+ "step": 250
316
+ },
317
+ {
318
+ "epoch": 0.12,
319
+ "learning_rate": 0.00019982437317643217,
320
+ "loss": 1.0021,
321
+ "step": 255
322
+ },
323
+ {
324
+ "epoch": 0.12,
325
+ "learning_rate": 0.0001997737183759338,
326
+ "loss": 0.9611,
327
+ "step": 260
328
+ },
329
+ {
330
+ "epoch": 0.12,
331
+ "learning_rate": 0.00019971666182997137,
332
+ "loss": 0.9813,
333
+ "step": 265
334
+ },
335
+ {
336
+ "epoch": 0.12,
337
+ "learning_rate": 0.00019965320719944366,
338
+ "loss": 0.9875,
339
+ "step": 270
340
+ },
341
+ {
342
+ "epoch": 0.13,
343
+ "learning_rate": 0.00019958335855576738,
344
+ "loss": 0.992,
345
+ "step": 275
346
+ },
347
+ {
348
+ "epoch": 0.13,
349
+ "learning_rate": 0.00019950712038061617,
350
+ "loss": 0.9735,
351
+ "step": 280
352
+ },
353
+ {
354
+ "epoch": 0.13,
355
+ "learning_rate": 0.00019942449756563279,
356
+ "loss": 0.9527,
357
+ "step": 285
358
+ },
359
+ {
360
+ "epoch": 0.13,
361
+ "learning_rate": 0.00019933549541211552,
362
+ "loss": 0.9663,
363
+ "step": 290
364
+ },
365
+ {
366
+ "epoch": 0.14,
367
+ "learning_rate": 0.00019924011963067765,
368
+ "loss": 0.9567,
369
+ "step": 295
370
+ },
371
+ {
372
+ "epoch": 0.14,
373
+ "learning_rate": 0.00019913837634088144,
374
+ "loss": 0.9592,
375
+ "step": 300
376
+ },
377
+ {
378
+ "epoch": 0.14,
379
+ "learning_rate": 0.00019903027207084523,
380
+ "loss": 0.9665,
381
+ "step": 305
382
+ },
383
+ {
384
+ "epoch": 0.14,
385
+ "learning_rate": 0.00019891581375682473,
386
+ "loss": 0.9899,
387
+ "step": 310
388
+ },
389
+ {
390
+ "epoch": 0.14,
391
+ "learning_rate": 0.00019879500874276786,
392
+ "loss": 0.9579,
393
+ "step": 315
394
+ },
395
+ {
396
+ "epoch": 0.15,
397
+ "learning_rate": 0.00019866786477984357,
398
+ "loss": 0.9722,
399
+ "step": 320
400
+ },
401
+ {
402
+ "epoch": 0.15,
403
+ "learning_rate": 0.0001985343900259446,
404
+ "loss": 0.9859,
405
+ "step": 325
406
+ },
407
+ {
408
+ "epoch": 0.15,
409
+ "learning_rate": 0.0001983945930451639,
410
+ "loss": 0.9507,
411
+ "step": 330
412
+ },
413
+ {
414
+ "epoch": 0.15,
415
+ "learning_rate": 0.0001982484828072452,
416
+ "loss": 0.954,
417
+ "step": 335
418
+ },
419
+ {
420
+ "epoch": 0.16,
421
+ "learning_rate": 0.00019809606868700756,
422
+ "loss": 0.9514,
423
+ "step": 340
424
+ },
425
+ {
426
+ "epoch": 0.16,
427
+ "learning_rate": 0.00019793736046374373,
428
+ "loss": 0.9837,
429
+ "step": 345
430
+ },
431
+ {
432
+ "epoch": 0.16,
433
+ "learning_rate": 0.0001977723683205928,
434
+ "loss": 0.9614,
435
+ "step": 350
436
+ },
437
+ {
438
+ "epoch": 0.16,
439
+ "learning_rate": 0.00019760110284388667,
440
+ "loss": 0.9599,
441
+ "step": 355
442
+ },
443
+ {
444
+ "epoch": 0.17,
445
+ "learning_rate": 0.00019742357502247103,
446
+ "loss": 0.9687,
447
+ "step": 360
448
+ },
449
+ {
450
+ "epoch": 0.17,
451
+ "learning_rate": 0.00019723979624700004,
452
+ "loss": 0.9341,
453
+ "step": 365
454
+ },
455
+ {
456
+ "epoch": 0.17,
457
+ "learning_rate": 0.0001970497783092057,
458
+ "loss": 0.9717,
459
+ "step": 370
460
+ },
461
+ {
462
+ "epoch": 0.17,
463
+ "learning_rate": 0.00019685353340114103,
464
+ "loss": 0.9618,
465
+ "step": 375
466
+ },
467
+ {
468
+ "epoch": 0.17,
469
+ "learning_rate": 0.00019665107411439805,
470
+ "loss": 0.9694,
471
+ "step": 380
472
+ },
473
+ {
474
+ "epoch": 0.18,
475
+ "learning_rate": 0.00019644241343929966,
476
+ "loss": 0.9594,
477
+ "step": 385
478
+ },
479
+ {
480
+ "epoch": 0.18,
481
+ "learning_rate": 0.00019622756476406626,
482
+ "loss": 0.9628,
483
+ "step": 390
484
+ },
485
+ {
486
+ "epoch": 0.18,
487
+ "learning_rate": 0.00019600654187395663,
488
+ "loss": 0.9898,
489
+ "step": 395
490
+ },
491
+ {
492
+ "epoch": 0.18,
493
+ "learning_rate": 0.00019577935895038361,
494
+ "loss": 0.9763,
495
+ "step": 400
496
+ },
497
+ {
498
+ "epoch": 0.19,
499
+ "learning_rate": 0.00019554603057000397,
500
+ "loss": 0.9506,
501
+ "step": 405
502
+ },
503
+ {
504
+ "epoch": 0.19,
505
+ "learning_rate": 0.0001953065717037832,
506
+ "loss": 0.9376,
507
+ "step": 410
508
+ },
509
+ {
510
+ "epoch": 0.19,
511
+ "learning_rate": 0.00019506099771603513,
512
+ "loss": 0.9545,
513
+ "step": 415
514
+ },
515
+ {
516
+ "epoch": 0.19,
517
+ "learning_rate": 0.00019480932436343582,
518
+ "loss": 0.9778,
519
+ "step": 420
520
+ },
521
+ {
522
+ "epoch": 0.2,
523
+ "learning_rate": 0.00019455156779401265,
524
+ "loss": 0.9516,
525
+ "step": 425
526
+ },
527
+ {
528
+ "epoch": 0.2,
529
+ "learning_rate": 0.00019428774454610843,
530
+ "loss": 0.9429,
531
+ "step": 430
532
+ },
533
+ {
534
+ "epoch": 0.2,
535
+ "learning_rate": 0.00019401787154731993,
536
+ "loss": 0.9667,
537
+ "step": 435
538
+ },
539
+ {
540
+ "epoch": 0.2,
541
+ "learning_rate": 0.0001937419661134121,
542
+ "loss": 0.9919,
543
+ "step": 440
544
+ },
545
+ {
546
+ "epoch": 0.2,
547
+ "learning_rate": 0.00019346004594720669,
548
+ "loss": 0.9647,
549
+ "step": 445
550
+ },
551
+ {
552
+ "epoch": 0.21,
553
+ "learning_rate": 0.0001931721291374467,
554
+ "loss": 0.9802,
555
+ "step": 450
556
+ },
557
+ {
558
+ "epoch": 0.21,
559
+ "learning_rate": 0.00019287823415763553,
560
+ "loss": 0.9597,
561
+ "step": 455
562
+ },
563
+ {
564
+ "epoch": 0.21,
565
+ "learning_rate": 0.00019257837986485187,
566
+ "loss": 0.9517,
567
+ "step": 460
568
+ },
569
+ {
570
+ "epoch": 0.21,
571
+ "learning_rate": 0.0001922725854985396,
572
+ "loss": 0.9479,
573
+ "step": 465
574
+ },
575
+ {
576
+ "epoch": 0.22,
577
+ "learning_rate": 0.0001919608706792735,
578
+ "loss": 0.9598,
579
+ "step": 470
580
+ },
581
+ {
582
+ "epoch": 0.22,
583
+ "learning_rate": 0.00019164325540750016,
584
+ "loss": 0.9674,
585
+ "step": 475
586
+ },
587
+ {
588
+ "epoch": 0.22,
589
+ "learning_rate": 0.00019131976006225488,
590
+ "loss": 0.9816,
591
+ "step": 480
592
+ },
593
+ {
594
+ "epoch": 0.22,
595
+ "learning_rate": 0.00019099040539985394,
596
+ "loss": 0.9858,
597
+ "step": 485
598
+ },
599
+ {
600
+ "epoch": 0.22,
601
+ "learning_rate": 0.00019065521255256298,
602
+ "loss": 0.9755,
603
+ "step": 490
604
+ },
605
+ {
606
+ "epoch": 0.23,
607
+ "learning_rate": 0.00019031420302724096,
608
+ "loss": 0.9582,
609
+ "step": 495
610
+ },
611
+ {
612
+ "epoch": 0.23,
613
+ "learning_rate": 0.00018996739870396026,
614
+ "loss": 0.9595,
615
+ "step": 500
616
+ },
617
+ {
618
+ "epoch": 0.23,
619
+ "learning_rate": 0.00018961482183460282,
620
+ "loss": 0.9523,
621
+ "step": 505
622
+ },
623
+ {
624
+ "epoch": 0.23,
625
+ "learning_rate": 0.00018925649504143244,
626
+ "loss": 0.9531,
627
+ "step": 510
628
+ },
629
+ {
630
+ "epoch": 0.24,
631
+ "learning_rate": 0.0001888924413156432,
632
+ "loss": 0.9659,
633
+ "step": 515
634
+ },
635
+ {
636
+ "epoch": 0.24,
637
+ "learning_rate": 0.0001885226840158843,
638
+ "loss": 0.9642,
639
+ "step": 520
640
+ },
641
+ {
642
+ "epoch": 0.24,
643
+ "learning_rate": 0.00018814724686676133,
644
+ "loss": 0.9894,
645
+ "step": 525
646
+ },
647
+ {
648
+ "epoch": 0.24,
649
+ "learning_rate": 0.00018776615395731398,
650
+ "loss": 0.9611,
651
+ "step": 530
652
+ },
653
+ {
654
+ "epoch": 0.25,
655
+ "learning_rate": 0.0001873794297394706,
656
+ "loss": 0.934,
657
+ "step": 535
658
+ },
659
+ {
660
+ "epoch": 0.25,
661
+ "learning_rate": 0.00018698709902647902,
662
+ "loss": 0.9633,
663
+ "step": 540
664
+ },
665
+ {
666
+ "epoch": 0.25,
667
+ "learning_rate": 0.00018658918699131468,
668
+ "loss": 0.9509,
669
+ "step": 545
670
+ },
671
+ {
672
+ "epoch": 0.25,
673
+ "learning_rate": 0.00018618571916506545,
674
+ "loss": 0.9642,
675
+ "step": 550
676
+ },
677
+ {
678
+ "epoch": 0.25,
679
+ "learning_rate": 0.00018577672143529336,
680
+ "loss": 0.9582,
681
+ "step": 555
682
+ },
683
+ {
684
+ "epoch": 0.26,
685
+ "learning_rate": 0.00018536222004437368,
686
+ "loss": 0.9632,
687
+ "step": 560
688
+ },
689
+ {
690
+ "epoch": 0.26,
691
+ "learning_rate": 0.0001849422415878112,
692
+ "loss": 0.9604,
693
+ "step": 565
694
+ },
695
+ {
696
+ "epoch": 0.26,
697
+ "learning_rate": 0.00018451681301253362,
698
+ "loss": 0.9754,
699
+ "step": 570
700
+ },
701
+ {
702
+ "epoch": 0.26,
703
+ "learning_rate": 0.00018408596161516267,
704
+ "loss": 0.9724,
705
+ "step": 575
706
+ },
707
+ {
708
+ "epoch": 0.27,
709
+ "learning_rate": 0.00018364971504026273,
710
+ "loss": 0.9722,
711
+ "step": 580
712
+ },
713
+ {
714
+ "epoch": 0.27,
715
+ "learning_rate": 0.00018320810127856705,
716
+ "loss": 0.9498,
717
+ "step": 585
718
+ },
719
+ {
720
+ "epoch": 0.27,
721
+ "learning_rate": 0.00018276114866518168,
722
+ "loss": 0.9548,
723
+ "step": 590
724
+ },
725
+ {
726
+ "epoch": 0.27,
727
+ "learning_rate": 0.00018230888587776755,
728
+ "loss": 0.9414,
729
+ "step": 595
730
+ },
731
+ {
732
+ "epoch": 0.28,
733
+ "learning_rate": 0.00018185134193470043,
734
+ "loss": 0.9492,
735
+ "step": 600
736
+ },
737
+ {
738
+ "epoch": 0.28,
739
+ "learning_rate": 0.00018138854619320893,
740
+ "loss": 0.9346,
741
+ "step": 605
742
+ },
743
+ {
744
+ "epoch": 0.28,
745
+ "learning_rate": 0.00018092052834749094,
746
+ "loss": 0.9561,
747
+ "step": 610
748
+ },
749
+ {
750
+ "epoch": 0.28,
751
+ "learning_rate": 0.0001804473184268084,
752
+ "loss": 0.9477,
753
+ "step": 615
754
+ },
755
+ {
756
+ "epoch": 0.28,
757
+ "learning_rate": 0.0001799689467935604,
758
+ "loss": 0.9432,
759
+ "step": 620
760
+ },
761
+ {
762
+ "epoch": 0.29,
763
+ "learning_rate": 0.00017948544414133534,
764
+ "loss": 0.93,
765
+ "step": 625
766
+ },
767
+ {
768
+ "epoch": 0.29,
769
+ "learning_rate": 0.00017899684149294117,
770
+ "loss": 0.9643,
771
+ "step": 630
772
+ },
773
+ {
774
+ "epoch": 0.29,
775
+ "learning_rate": 0.00017850317019841514,
776
+ "loss": 0.9441,
777
+ "step": 635
778
+ },
779
+ {
780
+ "epoch": 0.29,
781
+ "learning_rate": 0.00017800446193301227,
782
+ "loss": 0.9684,
783
+ "step": 640
784
+ },
785
+ {
786
+ "epoch": 0.3,
787
+ "learning_rate": 0.00017750074869517284,
788
+ "loss": 0.9372,
789
+ "step": 645
790
+ },
791
+ {
792
+ "epoch": 0.3,
793
+ "learning_rate": 0.00017699206280446953,
794
+ "loss": 0.9582,
795
+ "step": 650
796
+ },
797
+ {
798
+ "epoch": 0.3,
799
+ "learning_rate": 0.00017647843689953352,
800
+ "loss": 0.9602,
801
+ "step": 655
802
+ },
803
+ {
804
+ "epoch": 0.3,
805
+ "learning_rate": 0.00017595990393596026,
806
+ "loss": 0.9721,
807
+ "step": 660
808
+ },
809
+ {
810
+ "epoch": 0.31,
811
+ "learning_rate": 0.0001754364971841952,
812
+ "loss": 0.9657,
813
+ "step": 665
814
+ },
815
+ {
816
+ "epoch": 0.31,
817
+ "learning_rate": 0.0001749082502273988,
818
+ "loss": 0.9369,
819
+ "step": 670
820
+ },
821
+ {
822
+ "epoch": 0.31,
823
+ "learning_rate": 0.00017437519695929194,
824
+ "loss": 0.9591,
825
+ "step": 675
826
+ },
827
+ {
828
+ "epoch": 0.31,
829
+ "learning_rate": 0.0001738373715819811,
830
+ "loss": 0.9628,
831
+ "step": 680
832
+ },
833
+ {
834
+ "epoch": 0.31,
835
+ "learning_rate": 0.00017329480860376391,
836
+ "loss": 0.9611,
837
+ "step": 685
838
+ },
839
+ {
840
+ "epoch": 0.32,
841
+ "learning_rate": 0.00017274754283691504,
842
+ "loss": 0.9724,
843
+ "step": 690
844
+ },
845
+ {
846
+ "epoch": 0.32,
847
+ "learning_rate": 0.00017219560939545246,
848
+ "loss": 0.9514,
849
+ "step": 695
850
+ },
851
+ {
852
+ "epoch": 0.32,
853
+ "learning_rate": 0.00017163904369288445,
854
+ "loss": 0.9536,
855
+ "step": 700
856
+ },
857
+ {
858
+ "epoch": 0.32,
859
+ "learning_rate": 0.00017107788143993743,
860
+ "loss": 0.942,
861
+ "step": 705
862
+ },
863
+ {
864
+ "epoch": 0.33,
865
+ "learning_rate": 0.00017051215864226469,
866
+ "loss": 0.9705,
867
+ "step": 710
868
+ },
869
+ {
870
+ "epoch": 0.33,
871
+ "learning_rate": 0.0001699419115981361,
872
+ "loss": 0.9683,
873
+ "step": 715
874
+ },
875
+ {
876
+ "epoch": 0.33,
877
+ "learning_rate": 0.00016936717689610903,
878
+ "loss": 0.9672,
879
+ "step": 720
880
+ },
881
+ {
882
+ "epoch": 0.33,
883
+ "learning_rate": 0.00016878799141268106,
884
+ "loss": 0.9224,
885
+ "step": 725
886
+ },
887
+ {
888
+ "epoch": 0.34,
889
+ "learning_rate": 0.00016820439230992343,
890
+ "loss": 0.9482,
891
+ "step": 730
892
+ },
893
+ {
894
+ "epoch": 0.34,
895
+ "learning_rate": 0.00016761641703309704,
896
+ "loss": 0.9733,
897
+ "step": 735
898
+ },
899
+ {
900
+ "epoch": 0.34,
901
+ "learning_rate": 0.00016702410330824962,
902
+ "loss": 0.9445,
903
+ "step": 740
904
+ },
905
+ {
906
+ "epoch": 0.34,
907
+ "learning_rate": 0.00016642748913979513,
908
+ "loss": 0.9564,
909
+ "step": 745
910
+ },
911
+ {
912
+ "epoch": 0.34,
913
+ "learning_rate": 0.00016582661280807552,
914
+ "loss": 0.9623,
915
+ "step": 750
916
+ },
917
+ {
918
+ "epoch": 0.35,
919
+ "learning_rate": 0.0001652215128669042,
920
+ "loss": 0.9501,
921
+ "step": 755
922
+ },
923
+ {
924
+ "epoch": 0.35,
925
+ "learning_rate": 0.00016461222814109268,
926
+ "loss": 0.9746,
927
+ "step": 760
928
+ },
929
+ {
930
+ "epoch": 0.35,
931
+ "learning_rate": 0.00016399879772395915,
932
+ "loss": 0.9298,
933
+ "step": 765
934
+ },
935
+ {
936
+ "epoch": 0.35,
937
+ "learning_rate": 0.00016338126097482057,
938
+ "loss": 0.9564,
939
+ "step": 770
940
+ },
941
+ {
942
+ "epoch": 0.36,
943
+ "learning_rate": 0.0001627596575164668,
944
+ "loss": 0.9743,
945
+ "step": 775
946
+ },
947
+ {
948
+ "epoch": 0.36,
949
+ "learning_rate": 0.0001621340272326185,
950
+ "loss": 0.9626,
951
+ "step": 780
952
+ },
953
+ {
954
+ "epoch": 0.36,
955
+ "learning_rate": 0.00016150441026536823,
956
+ "loss": 0.9584,
957
+ "step": 785
958
+ },
959
+ {
960
+ "epoch": 0.36,
961
+ "learning_rate": 0.00016087084701260466,
962
+ "loss": 0.9679,
963
+ "step": 790
964
+ },
965
+ {
966
+ "epoch": 0.36,
967
+ "learning_rate": 0.00016023337812542048,
968
+ "loss": 0.9611,
969
+ "step": 795
970
+ },
971
+ {
972
+ "epoch": 0.37,
973
+ "learning_rate": 0.00015959204450550428,
974
+ "loss": 0.9511,
975
+ "step": 800
976
+ },
977
+ {
978
+ "epoch": 0.37,
979
+ "learning_rate": 0.00015894688730251614,
980
+ "loss": 0.9421,
981
+ "step": 805
982
+ },
983
+ {
984
+ "epoch": 0.37,
985
+ "learning_rate": 0.0001582979479114472,
986
+ "loss": 0.9358,
987
+ "step": 810
988
+ },
989
+ {
990
+ "epoch": 0.37,
991
+ "learning_rate": 0.000157645267969964,
992
+ "loss": 0.9657,
993
+ "step": 815
994
+ },
995
+ {
996
+ "epoch": 0.38,
997
+ "learning_rate": 0.0001569888893557365,
998
+ "loss": 0.9834,
999
+ "step": 820
1000
+ },
1001
+ {
1002
+ "epoch": 0.38,
1003
+ "learning_rate": 0.00015632885418375136,
1004
+ "loss": 0.9345,
1005
+ "step": 825
1006
+ },
1007
+ {
1008
+ "epoch": 0.38,
1009
+ "learning_rate": 0.00015566520480360957,
1010
+ "loss": 0.938,
1011
+ "step": 830
1012
+ },
1013
+ {
1014
+ "epoch": 0.38,
1015
+ "learning_rate": 0.0001549979837968094,
1016
+ "loss": 0.9689,
1017
+ "step": 835
1018
+ },
1019
+ {
1020
+ "epoch": 0.39,
1021
+ "learning_rate": 0.00015432723397401405,
1022
+ "loss": 0.949,
1023
+ "step": 840
1024
+ },
1025
+ {
1026
+ "epoch": 0.39,
1027
+ "learning_rate": 0.00015365299837230483,
1028
+ "loss": 0.9793,
1029
+ "step": 845
1030
+ },
1031
+ {
1032
+ "epoch": 0.39,
1033
+ "learning_rate": 0.00015297532025241992,
1034
+ "loss": 0.9817,
1035
+ "step": 850
1036
+ },
1037
+ {
1038
+ "epoch": 0.39,
1039
+ "learning_rate": 0.00015229424309597852,
1040
+ "loss": 0.9594,
1041
+ "step": 855
1042
+ },
1043
+ {
1044
+ "epoch": 0.39,
1045
+ "learning_rate": 0.00015160981060269107,
1046
+ "loss": 0.9492,
1047
+ "step": 860
1048
+ },
1049
+ {
1050
+ "epoch": 0.4,
1051
+ "learning_rate": 0.00015092206668755517,
1052
+ "loss": 0.9577,
1053
+ "step": 865
1054
+ },
1055
+ {
1056
+ "epoch": 0.4,
1057
+ "learning_rate": 0.00015023105547803806,
1058
+ "loss": 0.9571,
1059
+ "step": 870
1060
+ },
1061
+ {
1062
+ "epoch": 0.4,
1063
+ "learning_rate": 0.00014953682131124528,
1064
+ "loss": 0.9305,
1065
+ "step": 875
1066
+ },
1067
+ {
1068
+ "epoch": 0.4,
1069
+ "learning_rate": 0.0001488394087310757,
1070
+ "loss": 0.9605,
1071
+ "step": 880
1072
+ },
1073
+ {
1074
+ "epoch": 0.41,
1075
+ "learning_rate": 0.00014813886248536375,
1076
+ "loss": 0.9463,
1077
+ "step": 885
1078
+ },
1079
+ {
1080
+ "epoch": 0.41,
1081
+ "learning_rate": 0.00014743522752300793,
1082
+ "loss": 0.9414,
1083
+ "step": 890
1084
+ },
1085
+ {
1086
+ "epoch": 0.41,
1087
+ "learning_rate": 0.00014672854899108718,
1088
+ "loss": 0.9486,
1089
+ "step": 895
1090
+ },
1091
+ {
1092
+ "epoch": 0.41,
1093
+ "learning_rate": 0.00014601887223196372,
1094
+ "loss": 0.9595,
1095
+ "step": 900
1096
+ },
1097
+ {
1098
+ "epoch": 0.42,
1099
+ "learning_rate": 0.00014530624278037406,
1100
+ "loss": 0.9755,
1101
+ "step": 905
1102
+ },
1103
+ {
1104
+ "epoch": 0.42,
1105
+ "learning_rate": 0.0001445907063605072,
1106
+ "loss": 0.9635,
1107
+ "step": 910
1108
+ },
1109
+ {
1110
+ "epoch": 0.42,
1111
+ "learning_rate": 0.00014387230888307097,
1112
+ "loss": 0.9518,
1113
+ "step": 915
1114
+ },
1115
+ {
1116
+ "epoch": 0.42,
1117
+ "learning_rate": 0.0001431510964423462,
1118
+ "loss": 0.9446,
1119
+ "step": 920
1120
+ },
1121
+ {
1122
+ "epoch": 0.42,
1123
+ "learning_rate": 0.0001424271153132291,
1124
+ "loss": 0.9636,
1125
+ "step": 925
1126
+ },
1127
+ {
1128
+ "epoch": 0.43,
1129
+ "learning_rate": 0.00014170041194826248,
1130
+ "loss": 0.941,
1131
+ "step": 930
1132
+ },
1133
+ {
1134
+ "epoch": 0.43,
1135
+ "learning_rate": 0.0001409710329746547,
1136
+ "loss": 0.9571,
1137
+ "step": 935
1138
+ },
1139
+ {
1140
+ "epoch": 0.43,
1141
+ "learning_rate": 0.0001402390251912885,
1142
+ "loss": 0.9402,
1143
+ "step": 940
1144
+ },
1145
+ {
1146
+ "epoch": 0.43,
1147
+ "learning_rate": 0.00013950443556571778,
1148
+ "loss": 0.9489,
1149
+ "step": 945
1150
+ },
1151
+ {
1152
+ "epoch": 0.44,
1153
+ "learning_rate": 0.0001387673112311545,
1154
+ "loss": 0.9711,
1155
+ "step": 950
1156
+ },
1157
+ {
1158
+ "epoch": 0.44,
1159
+ "learning_rate": 0.00013802769948344405,
1160
+ "loss": 0.9624,
1161
+ "step": 955
1162
+ },
1163
+ {
1164
+ "epoch": 0.44,
1165
+ "learning_rate": 0.00013728564777803088,
1166
+ "loss": 0.9483,
1167
+ "step": 960
1168
+ },
1169
+ {
1170
+ "epoch": 0.44,
1171
+ "learning_rate": 0.0001365412037269136,
1172
+ "loss": 0.966,
1173
+ "step": 965
1174
+ },
1175
+ {
1176
+ "epoch": 0.45,
1177
+ "learning_rate": 0.00013579441509559004,
1178
+ "loss": 0.9386,
1179
+ "step": 970
1180
+ },
1181
+ {
1182
+ "epoch": 0.45,
1183
+ "learning_rate": 0.0001350453297999925,
1184
+ "loss": 0.9489,
1185
+ "step": 975
1186
+ },
1187
+ {
1188
+ "epoch": 0.45,
1189
+ "learning_rate": 0.00013429399590341324,
1190
+ "loss": 0.9417,
1191
+ "step": 980
1192
+ },
1193
+ {
1194
+ "epoch": 0.45,
1195
+ "learning_rate": 0.00013354046161342085,
1196
+ "loss": 0.9699,
1197
+ "step": 985
1198
+ },
1199
+ {
1200
+ "epoch": 0.45,
1201
+ "learning_rate": 0.0001327847752787669,
1202
+ "loss": 0.9395,
1203
+ "step": 990
1204
+ },
1205
+ {
1206
+ "epoch": 0.46,
1207
+ "learning_rate": 0.00013202698538628376,
1208
+ "loss": 0.9344,
1209
+ "step": 995
1210
+ },
1211
+ {
1212
+ "epoch": 0.46,
1213
+ "learning_rate": 0.00013126714055777377,
1214
+ "loss": 0.9291,
1215
+ "step": 1000
1216
+ },
1217
+ {
1218
+ "epoch": 0.46,
1219
+ "learning_rate": 0.0001305052895468893,
1220
+ "loss": 0.9535,
1221
+ "step": 1005
1222
+ },
1223
+ {
1224
+ "epoch": 0.46,
1225
+ "learning_rate": 0.00012974148123600477,
1226
+ "loss": 0.9784,
1227
+ "step": 1010
1228
+ },
1229
+ {
1230
+ "epoch": 0.47,
1231
+ "learning_rate": 0.00012897576463307997,
1232
+ "loss": 0.9547,
1233
+ "step": 1015
1234
+ },
1235
+ {
1236
+ "epoch": 0.47,
1237
+ "learning_rate": 0.00012820818886851598,
1238
+ "loss": 0.9349,
1239
+ "step": 1020
1240
+ },
1241
+ {
1242
+ "epoch": 0.47,
1243
+ "learning_rate": 0.00012743880319200242,
1244
+ "loss": 0.9619,
1245
+ "step": 1025
1246
+ },
1247
+ {
1248
+ "epoch": 0.47,
1249
+ "learning_rate": 0.00012666765696935773,
1250
+ "loss": 0.9487,
1251
+ "step": 1030
1252
+ },
1253
+ {
1254
+ "epoch": 0.47,
1255
+ "learning_rate": 0.0001258947996793616,
1256
+ "loss": 0.9415,
1257
+ "step": 1035
1258
+ },
1259
+ {
1260
+ "epoch": 0.48,
1261
+ "learning_rate": 0.00012512028091058044,
1262
+ "loss": 0.9346,
1263
+ "step": 1040
1264
+ },
1265
+ {
1266
+ "epoch": 0.48,
1267
+ "learning_rate": 0.00012434415035818535,
1268
+ "loss": 0.9423,
1269
+ "step": 1045
1270
+ },
1271
+ {
1272
+ "epoch": 0.48,
1273
+ "learning_rate": 0.00012356645782076383,
1274
+ "loss": 0.9546,
1275
+ "step": 1050
1276
+ },
1277
+ {
1278
+ "epoch": 0.48,
1279
+ "learning_rate": 0.00012278725319712447,
1280
+ "loss": 0.9534,
1281
+ "step": 1055
1282
+ },
1283
+ {
1284
+ "epoch": 0.49,
1285
+ "learning_rate": 0.00012200658648309531,
1286
+ "loss": 0.9592,
1287
+ "step": 1060
1288
+ },
1289
+ {
1290
+ "epoch": 0.49,
1291
+ "learning_rate": 0.00012122450776831594,
1292
+ "loss": 0.9446,
1293
+ "step": 1065
1294
+ },
1295
+ {
1296
+ "epoch": 0.49,
1297
+ "learning_rate": 0.00012044106723302364,
1298
+ "loss": 0.9418,
1299
+ "step": 1070
1300
+ },
1301
+ {
1302
+ "epoch": 0.49,
1303
+ "learning_rate": 0.00011965631514483375,
1304
+ "loss": 0.9591,
1305
+ "step": 1075
1306
+ },
1307
+ {
1308
+ "epoch": 0.5,
1309
+ "learning_rate": 0.00011887030185551426,
1310
+ "loss": 0.9674,
1311
+ "step": 1080
1312
+ },
1313
+ {
1314
+ "epoch": 0.5,
1315
+ "learning_rate": 0.00011808307779775518,
1316
+ "loss": 0.9651,
1317
+ "step": 1085
1318
+ },
1319
+ {
1320
+ "epoch": 0.5,
1321
+ "learning_rate": 0.00011729469348193262,
1322
+ "loss": 0.9482,
1323
+ "step": 1090
1324
+ },
1325
+ {
1326
+ "epoch": 0.5,
1327
+ "learning_rate": 0.00011650519949286795,
1328
+ "loss": 0.9484,
1329
+ "step": 1095
1330
+ },
1331
+ {
1332
+ "epoch": 0.5,
1333
+ "learning_rate": 0.00011571464648658201,
1334
+ "loss": 0.9306,
1335
+ "step": 1100
1336
+ },
1337
+ {
1338
+ "epoch": 0.51,
1339
+ "learning_rate": 0.00011492308518704506,
1340
+ "loss": 0.9342,
1341
+ "step": 1105
1342
+ },
1343
+ {
1344
+ "epoch": 0.51,
1345
+ "learning_rate": 0.00011413056638292215,
1346
+ "loss": 0.9506,
1347
+ "step": 1110
1348
+ },
1349
+ {
1350
+ "epoch": 0.51,
1351
+ "learning_rate": 0.00011333714092431423,
1352
+ "loss": 0.9625,
1353
+ "step": 1115
1354
+ },
1355
+ {
1356
+ "epoch": 0.51,
1357
+ "learning_rate": 0.00011254285971949574,
1358
+ "loss": 0.9709,
1359
+ "step": 1120
1360
+ },
1361
+ {
1362
+ "epoch": 0.52,
1363
+ "learning_rate": 0.00011174777373164798,
1364
+ "loss": 0.9229,
1365
+ "step": 1125
1366
+ },
1367
+ {
1368
+ "epoch": 0.52,
1369
+ "learning_rate": 0.0001109519339755893,
1370
+ "loss": 0.9306,
1371
+ "step": 1130
1372
+ },
1373
+ {
1374
+ "epoch": 0.52,
1375
+ "learning_rate": 0.00011015539151450172,
1376
+ "loss": 0.9394,
1377
+ "step": 1135
1378
+ },
1379
+ {
1380
+ "epoch": 0.52,
1381
+ "learning_rate": 0.00010935819745665477,
1382
+ "loss": 0.9445,
1383
+ "step": 1140
1384
+ },
1385
+ {
1386
+ "epoch": 0.53,
1387
+ "learning_rate": 0.00010856040295212612,
1388
+ "loss": 0.9618,
1389
+ "step": 1145
1390
+ },
1391
+ {
1392
+ "epoch": 0.53,
1393
+ "learning_rate": 0.00010776205918951969,
1394
+ "loss": 0.9556,
1395
+ "step": 1150
1396
+ },
1397
+ {
1398
+ "epoch": 0.53,
1399
+ "learning_rate": 0.00010696321739268121,
1400
+ "loss": 0.9095,
1401
+ "step": 1155
1402
+ },
1403
+ {
1404
+ "epoch": 0.53,
1405
+ "learning_rate": 0.00010616392881741166,
1406
+ "loss": 0.94,
1407
+ "step": 1160
1408
+ },
1409
+ {
1410
+ "epoch": 0.53,
1411
+ "learning_rate": 0.00010536424474817847,
1412
+ "loss": 0.952,
1413
+ "step": 1165
1414
+ },
1415
+ {
1416
+ "epoch": 0.54,
1417
+ "learning_rate": 0.00010456421649482502,
1418
+ "loss": 0.9408,
1419
+ "step": 1170
1420
+ },
1421
+ {
1422
+ "epoch": 0.54,
1423
+ "learning_rate": 0.0001037638953892784,
1424
+ "loss": 0.9453,
1425
+ "step": 1175
1426
+ },
1427
+ {
1428
+ "epoch": 0.54,
1429
+ "learning_rate": 0.00010296333278225599,
1430
+ "loss": 0.941,
1431
+ "step": 1180
1432
+ },
1433
+ {
1434
+ "epoch": 0.54,
1435
+ "learning_rate": 0.00010216258003997043,
1436
+ "loss": 0.9709,
1437
+ "step": 1185
1438
+ },
1439
+ {
1440
+ "epoch": 0.55,
1441
+ "learning_rate": 0.00010136168854083402,
1442
+ "loss": 0.9381,
1443
+ "step": 1190
1444
+ },
1445
+ {
1446
+ "epoch": 0.55,
1447
+ "learning_rate": 0.00010056070967216198,
1448
+ "loss": 0.9496,
1449
+ "step": 1195
1450
+ },
1451
+ {
1452
+ "epoch": 0.55,
1453
+ "learning_rate": 9.975969482687547e-05,
1454
+ "loss": 0.9514,
1455
+ "step": 1200
1456
+ },
1457
+ {
1458
+ "epoch": 0.55,
1459
+ "learning_rate": 9.8958695400204e-05,
1460
+ "loss": 0.9418,
1461
+ "step": 1205
1462
+ },
1463
+ {
1464
+ "epoch": 0.56,
1465
+ "learning_rate": 9.815776278638771e-05,
1466
+ "loss": 0.9563,
1467
+ "step": 1210
1468
+ },
1469
+ {
1470
+ "epoch": 0.56,
1471
+ "learning_rate": 9.735694837537993e-05,
1472
+ "loss": 0.9527,
1473
+ "step": 1215
1474
+ },
1475
+ {
1476
+ "epoch": 0.56,
1477
+ "learning_rate": 9.655630354954974e-05,
1478
+ "loss": 0.9241,
1479
+ "step": 1220
1480
+ },
1481
+ {
1482
+ "epoch": 0.56,
1483
+ "learning_rate": 9.57558796803852e-05,
1484
+ "loss": 0.9505,
1485
+ "step": 1225
1486
+ },
1487
+ {
1488
+ "epoch": 0.56,
1489
+ "learning_rate": 9.495572812519718e-05,
1490
+ "loss": 0.9556,
1491
+ "step": 1230
1492
+ },
1493
+ {
1494
+ "epoch": 0.57,
1495
+ "learning_rate": 9.415590022382419e-05,
1496
+ "loss": 0.9352,
1497
+ "step": 1235
1498
+ },
1499
+ {
1500
+ "epoch": 0.57,
1501
+ "learning_rate": 9.33564472953383e-05,
1502
+ "loss": 0.9459,
1503
+ "step": 1240
1504
+ },
1505
+ {
1506
+ "epoch": 0.57,
1507
+ "learning_rate": 9.255742063475228e-05,
1508
+ "loss": 0.9555,
1509
+ "step": 1245
1510
+ },
1511
+ {
1512
+ "epoch": 0.57,
1513
+ "learning_rate": 9.175887150972841e-05,
1514
+ "loss": 0.9463,
1515
+ "step": 1250
1516
+ },
1517
+ {
1518
+ "epoch": 0.58,
1519
+ "learning_rate": 9.096085115728902e-05,
1520
+ "loss": 0.925,
1521
+ "step": 1255
1522
+ },
1523
+ {
1524
+ "epoch": 0.58,
1525
+ "learning_rate": 9.016341078052908e-05,
1526
+ "loss": 0.9285,
1527
+ "step": 1260
1528
+ },
1529
+ {
1530
+ "epoch": 0.58,
1531
+ "learning_rate": 8.936660154533069e-05,
1532
+ "loss": 0.9344,
1533
+ "step": 1265
1534
+ },
1535
+ {
1536
+ "epoch": 0.58,
1537
+ "learning_rate": 8.857047457708023e-05,
1538
+ "loss": 0.9487,
1539
+ "step": 1270
1540
+ },
1541
+ {
1542
+ "epoch": 0.59,
1543
+ "learning_rate": 8.777508095738818e-05,
1544
+ "loss": 0.9461,
1545
+ "step": 1275
1546
+ },
1547
+ {
1548
+ "epoch": 0.59,
1549
+ "learning_rate": 8.698047172081128e-05,
1550
+ "loss": 0.9615,
1551
+ "step": 1280
1552
+ },
1553
+ {
1554
+ "epoch": 0.59,
1555
+ "learning_rate": 8.618669785157825e-05,
1556
+ "loss": 0.9358,
1557
+ "step": 1285
1558
+ },
1559
+ {
1560
+ "epoch": 0.59,
1561
+ "learning_rate": 8.539381028031837e-05,
1562
+ "loss": 0.942,
1563
+ "step": 1290
1564
+ },
1565
+ {
1566
+ "epoch": 0.59,
1567
+ "learning_rate": 8.460185988079379e-05,
1568
+ "loss": 0.9714,
1569
+ "step": 1295
1570
+ },
1571
+ {
1572
+ "epoch": 0.6,
1573
+ "learning_rate": 8.381089746663517e-05,
1574
+ "loss": 0.9644,
1575
+ "step": 1300
1576
+ },
1577
+ {
1578
+ "epoch": 0.6,
1579
+ "learning_rate": 8.302097378808146e-05,
1580
+ "loss": 0.9613,
1581
+ "step": 1305
1582
+ },
1583
+ {
1584
+ "epoch": 0.6,
1585
+ "learning_rate": 8.223213952872353e-05,
1586
+ "loss": 0.9394,
1587
+ "step": 1310
1588
+ },
1589
+ {
1590
+ "epoch": 0.6,
1591
+ "learning_rate": 8.144444530225236e-05,
1592
+ "loss": 0.9344,
1593
+ "step": 1315
1594
+ },
1595
+ {
1596
+ "epoch": 0.61,
1597
+ "learning_rate": 8.065794164921128e-05,
1598
+ "loss": 0.9706,
1599
+ "step": 1320
1600
+ },
1601
+ {
1602
+ "epoch": 0.61,
1603
+ "learning_rate": 7.987267903375332e-05,
1604
+ "loss": 0.953,
1605
+ "step": 1325
1606
+ },
1607
+ {
1608
+ "epoch": 0.61,
1609
+ "learning_rate": 7.90887078404033e-05,
1610
+ "loss": 0.9638,
1611
+ "step": 1330
1612
+ },
1613
+ {
1614
+ "epoch": 0.61,
1615
+ "learning_rate": 7.830607837082493e-05,
1616
+ "loss": 0.9567,
1617
+ "step": 1335
1618
+ },
1619
+ {
1620
+ "epoch": 0.61,
1621
+ "learning_rate": 7.75248408405934e-05,
1622
+ "loss": 0.9538,
1623
+ "step": 1340
1624
+ },
1625
+ {
1626
+ "epoch": 0.62,
1627
+ "learning_rate": 7.674504537597336e-05,
1628
+ "loss": 0.9365,
1629
+ "step": 1345
1630
+ },
1631
+ {
1632
+ "epoch": 0.62,
1633
+ "learning_rate": 7.596674201070282e-05,
1634
+ "loss": 0.9478,
1635
+ "step": 1350
1636
+ },
1637
+ {
1638
+ "epoch": 0.62,
1639
+ "learning_rate": 7.518998068278266e-05,
1640
+ "loss": 0.943,
1641
+ "step": 1355
1642
+ },
1643
+ {
1644
+ "epoch": 0.62,
1645
+ "learning_rate": 7.441481123127257e-05,
1646
+ "loss": 0.9414,
1647
+ "step": 1360
1648
+ },
1649
+ {
1650
+ "epoch": 0.63,
1651
+ "learning_rate": 7.364128339309326e-05,
1652
+ "loss": 0.9448,
1653
+ "step": 1365
1654
+ },
1655
+ {
1656
+ "epoch": 0.63,
1657
+ "learning_rate": 7.28694467998352e-05,
1658
+ "loss": 0.9299,
1659
+ "step": 1370
1660
+ },
1661
+ {
1662
+ "epoch": 0.63,
1663
+ "learning_rate": 7.209935097457413e-05,
1664
+ "loss": 0.9544,
1665
+ "step": 1375
1666
+ },
1667
+ {
1668
+ "epoch": 0.63,
1669
+ "learning_rate": 7.133104532869342e-05,
1670
+ "loss": 0.9757,
1671
+ "step": 1380
1672
+ },
1673
+ {
1674
+ "epoch": 0.64,
1675
+ "learning_rate": 7.056457915871399e-05,
1676
+ "loss": 0.9587,
1677
+ "step": 1385
1678
+ },
1679
+ {
1680
+ "epoch": 0.64,
1681
+ "learning_rate": 6.980000164313093e-05,
1682
+ "loss": 0.9593,
1683
+ "step": 1390
1684
+ },
1685
+ {
1686
+ "epoch": 0.64,
1687
+ "learning_rate": 6.903736183925835e-05,
1688
+ "loss": 0.9497,
1689
+ "step": 1395
1690
+ },
1691
+ {
1692
+ "epoch": 0.64,
1693
+ "learning_rate": 6.827670868008171e-05,
1694
+ "loss": 0.9351,
1695
+ "step": 1400
1696
+ },
1697
+ {
1698
+ "epoch": 0.64,
1699
+ "learning_rate": 6.751809097111798e-05,
1700
+ "loss": 0.9359,
1701
+ "step": 1405
1702
+ },
1703
+ {
1704
+ "epoch": 0.65,
1705
+ "learning_rate": 6.676155738728438e-05,
1706
+ "loss": 0.9662,
1707
+ "step": 1410
1708
+ },
1709
+ {
1710
+ "epoch": 0.65,
1711
+ "learning_rate": 6.600715646977502e-05,
1712
+ "loss": 0.9493,
1713
+ "step": 1415
1714
+ },
1715
+ {
1716
+ "epoch": 0.65,
1717
+ "learning_rate": 6.525493662294668e-05,
1718
+ "loss": 0.9547,
1719
+ "step": 1420
1720
+ },
1721
+ {
1722
+ "epoch": 0.65,
1723
+ "learning_rate": 6.450494611121273e-05,
1724
+ "loss": 0.944,
1725
+ "step": 1425
1726
+ },
1727
+ {
1728
+ "epoch": 0.66,
1729
+ "learning_rate": 6.375723305594658e-05,
1730
+ "loss": 0.9464,
1731
+ "step": 1430
1732
+ },
1733
+ {
1734
+ "epoch": 0.66,
1735
+ "learning_rate": 6.301184543239398e-05,
1736
+ "loss": 0.9497,
1737
+ "step": 1435
1738
+ },
1739
+ {
1740
+ "epoch": 0.66,
1741
+ "learning_rate": 6.226883106659485e-05,
1742
+ "loss": 0.926,
1743
+ "step": 1440
1744
+ },
1745
+ {
1746
+ "epoch": 0.66,
1747
+ "learning_rate": 6.152823763231463e-05,
1748
+ "loss": 0.9502,
1749
+ "step": 1445
1750
+ },
1751
+ {
1752
+ "epoch": 0.67,
1753
+ "learning_rate": 6.079011264798534e-05,
1754
+ "loss": 0.9521,
1755
+ "step": 1450
1756
+ },
1757
+ {
1758
+ "epoch": 0.67,
1759
+ "learning_rate": 6.005450347365687e-05,
1760
+ "loss": 0.9541,
1761
+ "step": 1455
1762
+ },
1763
+ {
1764
+ "epoch": 0.67,
1765
+ "learning_rate": 5.932145730795793e-05,
1766
+ "loss": 0.945,
1767
+ "step": 1460
1768
+ },
1769
+ {
1770
+ "epoch": 0.67,
1771
+ "learning_rate": 5.8591021185067876e-05,
1772
+ "loss": 0.9449,
1773
+ "step": 1465
1774
+ },
1775
+ {
1776
+ "epoch": 0.67,
1777
+ "learning_rate": 5.786324197169887e-05,
1778
+ "loss": 0.931,
1779
+ "step": 1470
1780
+ },
1781
+ {
1782
+ "epoch": 0.68,
1783
+ "learning_rate": 5.7138166364088705e-05,
1784
+ "loss": 0.9525,
1785
+ "step": 1475
1786
+ },
1787
+ {
1788
+ "epoch": 0.68,
1789
+ "learning_rate": 5.641584088500461e-05,
1790
+ "loss": 0.9394,
1791
+ "step": 1480
1792
+ },
1793
+ {
1794
+ "epoch": 0.68,
1795
+ "learning_rate": 5.569631188075841e-05,
1796
+ "loss": 0.9594,
1797
+ "step": 1485
1798
+ },
1799
+ {
1800
+ "epoch": 0.68,
1801
+ "learning_rate": 5.497962551823266e-05,
1802
+ "loss": 0.9467,
1803
+ "step": 1490
1804
+ },
1805
+ {
1806
+ "epoch": 0.69,
1807
+ "learning_rate": 5.4265827781918576e-05,
1808
+ "loss": 0.9452,
1809
+ "step": 1495
1810
+ },
1811
+ {
1812
+ "epoch": 0.69,
1813
+ "learning_rate": 5.355496447096533e-05,
1814
+ "loss": 0.9601,
1815
+ "step": 1500
1816
+ },
1817
+ {
1818
+ "epoch": 0.69,
1819
+ "learning_rate": 5.284708119624173e-05,
1820
+ "loss": 0.9639,
1821
+ "step": 1505
1822
+ },
1823
+ {
1824
+ "epoch": 0.69,
1825
+ "learning_rate": 5.214222337740962e-05,
1826
+ "loss": 0.9442,
1827
+ "step": 1510
1828
+ },
1829
+ {
1830
+ "epoch": 0.7,
1831
+ "learning_rate": 5.144043624000944e-05,
1832
+ "loss": 0.9456,
1833
+ "step": 1515
1834
+ },
1835
+ {
1836
+ "epoch": 0.7,
1837
+ "learning_rate": 5.0741764812558724e-05,
1838
+ "loss": 0.9434,
1839
+ "step": 1520
1840
+ },
1841
+ {
1842
+ "epoch": 0.7,
1843
+ "learning_rate": 5.00462539236628e-05,
1844
+ "loss": 0.9567,
1845
+ "step": 1525
1846
+ },
1847
+ {
1848
+ "epoch": 0.7,
1849
+ "learning_rate": 4.935394819913849e-05,
1850
+ "loss": 0.9528,
1851
+ "step": 1530
1852
+ },
1853
+ {
1854
+ "epoch": 0.7,
1855
+ "learning_rate": 4.8664892059150723e-05,
1856
+ "loss": 0.9367,
1857
+ "step": 1535
1858
+ },
1859
+ {
1860
+ "epoch": 0.71,
1861
+ "learning_rate": 4.7979129715362626e-05,
1862
+ "loss": 0.9558,
1863
+ "step": 1540
1864
+ },
1865
+ {
1866
+ "epoch": 0.71,
1867
+ "learning_rate": 4.72967051680985e-05,
1868
+ "loss": 0.9616,
1869
+ "step": 1545
1870
+ },
1871
+ {
1872
+ "epoch": 0.71,
1873
+ "learning_rate": 4.661766220352097e-05,
1874
+ "loss": 0.9527,
1875
+ "step": 1550
1876
+ },
1877
+ {
1878
+ "epoch": 0.71,
1879
+ "learning_rate": 4.5942044390821214e-05,
1880
+ "loss": 0.942,
1881
+ "step": 1555
1882
+ },
1883
+ {
1884
+ "epoch": 0.72,
1885
+ "learning_rate": 4.526989507942374e-05,
1886
+ "loss": 0.9435,
1887
+ "step": 1560
1888
+ },
1889
+ {
1890
+ "epoch": 0.72,
1891
+ "learning_rate": 4.460125739620479e-05,
1892
+ "loss": 0.9244,
1893
+ "step": 1565
1894
+ },
1895
+ {
1896
+ "epoch": 0.72,
1897
+ "learning_rate": 4.393617424272527e-05,
1898
+ "loss": 0.9669,
1899
+ "step": 1570
1900
+ },
1901
+ {
1902
+ "epoch": 0.72,
1903
+ "learning_rate": 4.3274688292478106e-05,
1904
+ "loss": 0.9575,
1905
+ "step": 1575
1906
+ },
1907
+ {
1908
+ "epoch": 0.73,
1909
+ "learning_rate": 4.261684198815003e-05,
1910
+ "loss": 0.958,
1911
+ "step": 1580
1912
+ },
1913
+ {
1914
+ "epoch": 0.73,
1915
+ "learning_rate": 4.1962677538898645e-05,
1916
+ "loss": 0.9391,
1917
+ "step": 1585
1918
+ },
1919
+ {
1920
+ "epoch": 0.73,
1921
+ "learning_rate": 4.131223691764383e-05,
1922
+ "loss": 0.932,
1923
+ "step": 1590
1924
+ },
1925
+ {
1926
+ "epoch": 0.73,
1927
+ "learning_rate": 4.0665561858374934e-05,
1928
+ "loss": 0.9601,
1929
+ "step": 1595
1930
+ },
1931
+ {
1932
+ "epoch": 0.73,
1933
+ "learning_rate": 4.0022693853472884e-05,
1934
+ "loss": 0.943,
1935
+ "step": 1600
1936
+ },
1937
+ {
1938
+ "epoch": 0.74,
1939
+ "learning_rate": 3.938367415104793e-05,
1940
+ "loss": 0.9406,
1941
+ "step": 1605
1942
+ },
1943
+ {
1944
+ "epoch": 0.74,
1945
+ "learning_rate": 3.8748543752293e-05,
1946
+ "loss": 0.9457,
1947
+ "step": 1610
1948
+ },
1949
+ {
1950
+ "epoch": 0.74,
1951
+ "learning_rate": 3.8117343408853127e-05,
1952
+ "loss": 0.9424,
1953
+ "step": 1615
1954
+ },
1955
+ {
1956
+ "epoch": 0.74,
1957
+ "learning_rate": 3.7490113620210485e-05,
1958
+ "loss": 0.9406,
1959
+ "step": 1620
1960
+ },
1961
+ {
1962
+ "epoch": 0.75,
1963
+ "learning_rate": 3.686689463108608e-05,
1964
+ "loss": 0.9414,
1965
+ "step": 1625
1966
+ },
1967
+ {
1968
+ "epoch": 0.75,
1969
+ "learning_rate": 3.624772642885734e-05,
1970
+ "loss": 0.9527,
1971
+ "step": 1630
1972
+ },
1973
+ {
1974
+ "epoch": 0.75,
1975
+ "learning_rate": 3.563264874099258e-05,
1976
+ "loss": 0.9572,
1977
+ "step": 1635
1978
+ },
1979
+ {
1980
+ "epoch": 0.75,
1981
+ "learning_rate": 3.502170103250177e-05,
1982
+ "loss": 0.9638,
1983
+ "step": 1640
1984
+ },
1985
+ {
1986
+ "epoch": 0.75,
1987
+ "learning_rate": 3.441492250340461e-05,
1988
+ "loss": 0.9635,
1989
+ "step": 1645
1990
+ },
1991
+ {
1992
+ "epoch": 0.76,
1993
+ "learning_rate": 3.381235208621522e-05,
1994
+ "loss": 0.9521,
1995
+ "step": 1650
1996
+ },
1997
+ {
1998
+ "epoch": 0.76,
1999
+ "learning_rate": 3.3214028443444036e-05,
2000
+ "loss": 0.9173,
2001
+ "step": 1655
2002
+ },
2003
+ {
2004
+ "epoch": 0.76,
2005
+ "learning_rate": 3.2619989965117356e-05,
2006
+ "loss": 0.9846,
2007
+ "step": 1660
2008
+ },
2009
+ {
2010
+ "epoch": 0.76,
2011
+ "learning_rate": 3.2030274766313865e-05,
2012
+ "loss": 0.9653,
2013
+ "step": 1665
2014
+ },
2015
+ {
2016
+ "epoch": 0.77,
2017
+ "learning_rate": 3.1444920684719394e-05,
2018
+ "loss": 0.9467,
2019
+ "step": 1670
2020
+ },
2021
+ {
2022
+ "epoch": 0.77,
2023
+ "learning_rate": 3.086396527819876e-05,
2024
+ "loss": 0.9321,
2025
+ "step": 1675
2026
+ },
2027
+ {
2028
+ "epoch": 0.77,
2029
+ "learning_rate": 3.028744582238633e-05,
2030
+ "loss": 0.9343,
2031
+ "step": 1680
2032
+ },
2033
+ {
2034
+ "epoch": 0.77,
2035
+ "learning_rate": 2.9715399308294e-05,
2036
+ "loss": 0.9647,
2037
+ "step": 1685
2038
+ },
2039
+ {
2040
+ "epoch": 0.78,
2041
+ "learning_rate": 2.914786243993808e-05,
2042
+ "loss": 0.9579,
2043
+ "step": 1690
2044
+ },
2045
+ {
2046
+ "epoch": 0.78,
2047
+ "learning_rate": 2.8584871631983888e-05,
2048
+ "loss": 0.9516,
2049
+ "step": 1695
2050
+ },
2051
+ {
2052
+ "epoch": 0.78,
2053
+ "learning_rate": 2.8026463007409664e-05,
2054
+ "loss": 0.9452,
2055
+ "step": 1700
2056
+ },
2057
+ {
2058
+ "epoch": 0.78,
2059
+ "learning_rate": 2.747267239518857e-05,
2060
+ "loss": 0.9831,
2061
+ "step": 1705
2062
+ },
2063
+ {
2064
+ "epoch": 0.78,
2065
+ "learning_rate": 2.6923535327989924e-05,
2066
+ "loss": 0.944,
2067
+ "step": 1710
2068
+ },
2069
+ {
2070
+ "epoch": 0.79,
2071
+ "learning_rate": 2.637908703989924e-05,
2072
+ "loss": 0.9344,
2073
+ "step": 1715
2074
+ },
2075
+ {
2076
+ "epoch": 0.79,
2077
+ "learning_rate": 2.5839362464157634e-05,
2078
+ "loss": 0.9247,
2079
+ "step": 1720
2080
+ },
2081
+ {
2082
+ "epoch": 0.79,
2083
+ "learning_rate": 2.5304396230920345e-05,
2084
+ "loss": 0.9407,
2085
+ "step": 1725
2086
+ },
2087
+ {
2088
+ "epoch": 0.79,
2089
+ "learning_rate": 2.477422266503473e-05,
2090
+ "loss": 0.9461,
2091
+ "step": 1730
2092
+ },
2093
+ {
2094
+ "epoch": 0.8,
2095
+ "learning_rate": 2.4248875783837987e-05,
2096
+ "loss": 0.9282,
2097
+ "step": 1735
2098
+ },
2099
+ {
2100
+ "epoch": 0.8,
2101
+ "learning_rate": 2.3728389294974475e-05,
2102
+ "loss": 0.9633,
2103
+ "step": 1740
2104
+ },
2105
+ {
2106
+ "epoch": 0.8,
2107
+ "learning_rate": 2.3212796594232943e-05,
2108
+ "loss": 0.9531,
2109
+ "step": 1745
2110
+ },
2111
+ {
2112
+ "epoch": 0.8,
2113
+ "learning_rate": 2.2702130763403673e-05,
2114
+ "loss": 0.9612,
2115
+ "step": 1750
2116
+ },
2117
+ {
2118
+ "epoch": 0.81,
2119
+ "learning_rate": 2.2196424568156073e-05,
2120
+ "loss": 0.9463,
2121
+ "step": 1755
2122
+ },
2123
+ {
2124
+ "epoch": 0.81,
2125
+ "learning_rate": 2.1695710455936115e-05,
2126
+ "loss": 0.9482,
2127
+ "step": 1760
2128
+ },
2129
+ {
2130
+ "epoch": 0.81,
2131
+ "learning_rate": 2.1200020553884604e-05,
2132
+ "loss": 0.9758,
2133
+ "step": 1765
2134
+ },
2135
+ {
2136
+ "epoch": 0.81,
2137
+ "learning_rate": 2.070938666677573e-05,
2138
+ "loss": 0.9313,
2139
+ "step": 1770
2140
+ },
2141
+ {
2142
+ "epoch": 0.81,
2143
+ "learning_rate": 2.0223840274976413e-05,
2144
+ "loss": 0.9348,
2145
+ "step": 1775
2146
+ },
2147
+ {
2148
+ "epoch": 0.82,
2149
+ "learning_rate": 1.974341253242635e-05,
2150
+ "loss": 0.9452,
2151
+ "step": 1780
2152
+ },
2153
+ {
2154
+ "epoch": 0.82,
2155
+ "learning_rate": 1.9268134264639272e-05,
2156
+ "loss": 0.9299,
2157
+ "step": 1785
2158
+ },
2159
+ {
2160
+ "epoch": 0.82,
2161
+ "learning_rate": 1.879803596672497e-05,
2162
+ "loss": 0.9704,
2163
+ "step": 1790
2164
+ },
2165
+ {
2166
+ "epoch": 0.82,
2167
+ "learning_rate": 1.8333147801432614e-05,
2168
+ "loss": 0.9808,
2169
+ "step": 1795
2170
+ },
2171
+ {
2172
+ "epoch": 0.83,
2173
+ "learning_rate": 1.7873499597215604e-05,
2174
+ "loss": 0.9529,
2175
+ "step": 1800
2176
+ },
2177
+ {
2178
+ "epoch": 0.83,
2179
+ "learning_rate": 1.741912084631746e-05,
2180
+ "loss": 0.9503,
2181
+ "step": 1805
2182
+ },
2183
+ {
2184
+ "epoch": 0.83,
2185
+ "learning_rate": 1.697004070287982e-05,
2186
+ "loss": 0.9607,
2187
+ "step": 1810
2188
+ },
2189
+ {
2190
+ "epoch": 0.83,
2191
+ "learning_rate": 1.6526287981071475e-05,
2192
+ "loss": 0.9486,
2193
+ "step": 1815
2194
+ },
2195
+ {
2196
+ "epoch": 0.84,
2197
+ "learning_rate": 1.608789115323993e-05,
2198
+ "loss": 0.9368,
2199
+ "step": 1820
2200
+ },
2201
+ {
2202
+ "epoch": 0.84,
2203
+ "learning_rate": 1.5654878348084244e-05,
2204
+ "loss": 0.9321,
2205
+ "step": 1825
2206
+ },
2207
+ {
2208
+ "epoch": 0.84,
2209
+ "learning_rate": 1.5227277348850466e-05,
2210
+ "loss": 0.9385,
2211
+ "step": 1830
2212
+ },
2213
+ {
2214
+ "epoch": 0.84,
2215
+ "learning_rate": 1.4805115591548746e-05,
2216
+ "loss": 0.9492,
2217
+ "step": 1835
2218
+ },
2219
+ {
2220
+ "epoch": 0.84,
2221
+ "learning_rate": 1.4388420163193217e-05,
2222
+ "loss": 0.9474,
2223
+ "step": 1840
2224
+ },
2225
+ {
2226
+ "epoch": 0.85,
2227
+ "learning_rate": 1.3977217800063846e-05,
2228
+ "loss": 0.9453,
2229
+ "step": 1845
2230
+ },
2231
+ {
2232
+ "epoch": 0.85,
2233
+ "learning_rate": 1.3571534885991044e-05,
2234
+ "loss": 0.9455,
2235
+ "step": 1850
2236
+ },
2237
+ {
2238
+ "epoch": 0.85,
2239
+ "learning_rate": 1.3171397450662715e-05,
2240
+ "loss": 0.9357,
2241
+ "step": 1855
2242
+ },
2243
+ {
2244
+ "epoch": 0.85,
2245
+ "learning_rate": 1.277683116795425e-05,
2246
+ "loss": 0.9337,
2247
+ "step": 1860
2248
+ },
2249
+ {
2250
+ "epoch": 0.86,
2251
+ "learning_rate": 1.2387861354281194e-05,
2252
+ "loss": 0.9389,
2253
+ "step": 1865
2254
+ },
2255
+ {
2256
+ "epoch": 0.86,
2257
+ "learning_rate": 1.2004512966974745e-05,
2258
+ "loss": 0.956,
2259
+ "step": 1870
2260
+ },
2261
+ {
2262
+ "epoch": 0.86,
2263
+ "learning_rate": 1.162681060268065e-05,
2264
+ "loss": 0.9514,
2265
+ "step": 1875
2266
+ },
2267
+ {
2268
+ "epoch": 0.86,
2269
+ "learning_rate": 1.1254778495780748e-05,
2270
+ "loss": 0.9247,
2271
+ "step": 1880
2272
+ },
2273
+ {
2274
+ "epoch": 0.87,
2275
+ "learning_rate": 1.0888440516838373e-05,
2276
+ "loss": 0.9509,
2277
+ "step": 1885
2278
+ },
2279
+ {
2280
+ "epoch": 0.87,
2281
+ "learning_rate": 1.0527820171066372e-05,
2282
+ "loss": 0.9588,
2283
+ "step": 1890
2284
+ },
2285
+ {
2286
+ "epoch": 0.87,
2287
+ "learning_rate": 1.0172940596819259e-05,
2288
+ "loss": 0.9608,
2289
+ "step": 1895
2290
+ },
2291
+ {
2292
+ "epoch": 0.87,
2293
+ "learning_rate": 9.823824564108408e-06,
2294
+ "loss": 0.9467,
2295
+ "step": 1900
2296
+ },
2297
+ {
2298
+ "epoch": 0.87,
2299
+ "learning_rate": 9.480494473141188e-06,
2300
+ "loss": 0.9543,
2301
+ "step": 1905
2302
+ },
2303
+ {
2304
+ "epoch": 0.88,
2305
+ "learning_rate": 9.142972352883594e-06,
2306
+ "loss": 0.9566,
2307
+ "step": 1910
2308
+ },
2309
+ {
2310
+ "epoch": 0.88,
2311
+ "learning_rate": 8.811279859646915e-06,
2312
+ "loss": 0.9541,
2313
+ "step": 1915
2314
+ },
2315
+ {
2316
+ "epoch": 0.88,
2317
+ "learning_rate": 8.485438275698154e-06,
2318
+ "loss": 0.9445,
2319
+ "step": 1920
2320
+ },
2321
+ {
2322
+ "epoch": 0.88,
2323
+ "learning_rate": 8.165468507894514e-06,
2324
+ "loss": 0.9209,
2325
+ "step": 1925
2326
+ },
2327
+ {
2328
+ "epoch": 0.89,
2329
+ "learning_rate": 7.851391086341952e-06,
2330
+ "loss": 0.9509,
2331
+ "step": 1930
2332
+ },
2333
+ {
2334
+ "epoch": 0.89,
2335
+ "learning_rate": 7.543226163077899e-06,
2336
+ "loss": 0.9576,
2337
+ "step": 1935
2338
+ },
2339
+ {
2340
+ "epoch": 0.89,
2341
+ "learning_rate": 7.240993510778305e-06,
2342
+ "loss": 0.9176,
2343
+ "step": 1940
2344
+ },
2345
+ {
2346
+ "epoch": 0.89,
2347
+ "learning_rate": 6.9447125214888835e-06,
2348
+ "loss": 0.9517,
2349
+ "step": 1945
2350
+ },
2351
+ {
2352
+ "epoch": 0.89,
2353
+ "learning_rate": 6.65440220538096e-06,
2354
+ "loss": 0.9372,
2355
+ "step": 1950
2356
+ },
2357
+ {
2358
+ "epoch": 0.9,
2359
+ "learning_rate": 6.370081189531707e-06,
2360
+ "loss": 0.9527,
2361
+ "step": 1955
2362
+ },
2363
+ {
2364
+ "epoch": 0.9,
2365
+ "learning_rate": 6.091767716728924e-06,
2366
+ "loss": 0.9653,
2367
+ "step": 1960
2368
+ },
2369
+ {
2370
+ "epoch": 0.9,
2371
+ "learning_rate": 5.819479644300563e-06,
2372
+ "loss": 0.9257,
2373
+ "step": 1965
2374
+ },
2375
+ {
2376
+ "epoch": 0.9,
2377
+ "learning_rate": 5.553234442969013e-06,
2378
+ "loss": 0.9785,
2379
+ "step": 1970
2380
+ },
2381
+ {
2382
+ "epoch": 0.91,
2383
+ "learning_rate": 5.293049195730038e-06,
2384
+ "loss": 0.9568,
2385
+ "step": 1975
2386
+ },
2387
+ {
2388
+ "epoch": 0.91,
2389
+ "learning_rate": 5.038940596756747e-06,
2390
+ "loss": 0.9402,
2391
+ "step": 1980
2392
+ },
2393
+ {
2394
+ "epoch": 0.91,
2395
+ "learning_rate": 4.790924950328435e-06,
2396
+ "loss": 0.9443,
2397
+ "step": 1985
2398
+ },
2399
+ {
2400
+ "epoch": 0.91,
2401
+ "learning_rate": 4.5490181697844916e-06,
2402
+ "loss": 0.9535,
2403
+ "step": 1990
2404
+ },
2405
+ {
2406
+ "epoch": 0.92,
2407
+ "learning_rate": 4.313235776503244e-06,
2408
+ "loss": 0.9405,
2409
+ "step": 1995
2410
+ },
2411
+ {
2412
+ "epoch": 0.92,
2413
+ "learning_rate": 4.08359289890623e-06,
2414
+ "loss": 0.9505,
2415
+ "step": 2000
2416
+ },
2417
+ {
2418
+ "epoch": 0.92,
2419
+ "learning_rate": 3.860104271487397e-06,
2420
+ "loss": 0.9569,
2421
+ "step": 2005
2422
+ },
2423
+ {
2424
+ "epoch": 0.92,
2425
+ "learning_rate": 3.6427842338677352e-06,
2426
+ "loss": 0.9419,
2427
+ "step": 2010
2428
+ },
2429
+ {
2430
+ "epoch": 0.92,
2431
+ "learning_rate": 3.4316467298752263e-06,
2432
+ "loss": 0.9239,
2433
+ "step": 2015
2434
+ },
2435
+ {
2436
+ "epoch": 0.93,
2437
+ "learning_rate": 3.226705306650113e-06,
2438
+ "loss": 0.946,
2439
+ "step": 2020
2440
+ },
2441
+ {
2442
+ "epoch": 0.93,
2443
+ "learning_rate": 3.0279731137757948e-06,
2444
+ "loss": 0.94,
2445
+ "step": 2025
2446
+ },
2447
+ {
2448
+ "epoch": 0.93,
2449
+ "learning_rate": 2.835462902434971e-06,
2450
+ "loss": 0.9607,
2451
+ "step": 2030
2452
+ },
2453
+ {
2454
+ "epoch": 0.93,
2455
+ "learning_rate": 2.649187024591604e-06,
2456
+ "loss": 0.9559,
2457
+ "step": 2035
2458
+ },
2459
+ {
2460
+ "epoch": 0.94,
2461
+ "learning_rate": 2.4691574321983215e-06,
2462
+ "loss": 0.959,
2463
+ "step": 2040
2464
+ },
2465
+ {
2466
+ "epoch": 0.94,
2467
+ "learning_rate": 2.2953856764295624e-06,
2468
+ "loss": 0.9536,
2469
+ "step": 2045
2470
+ },
2471
+ {
2472
+ "epoch": 0.94,
2473
+ "learning_rate": 2.1278829069404484e-06,
2474
+ "loss": 0.9233,
2475
+ "step": 2050
2476
+ },
2477
+ {
2478
+ "epoch": 0.94,
2479
+ "learning_rate": 1.966659871151366e-06,
2480
+ "loss": 0.9339,
2481
+ "step": 2055
2482
+ },
2483
+ {
2484
+ "epoch": 0.95,
2485
+ "learning_rate": 1.811726913558387e-06,
2486
+ "loss": 0.9364,
2487
+ "step": 2060
2488
+ },
2489
+ {
2490
+ "epoch": 0.95,
2491
+ "learning_rate": 1.6630939750695519e-06,
2492
+ "loss": 0.9281,
2493
+ "step": 2065
2494
+ },
2495
+ {
2496
+ "epoch": 0.95,
2497
+ "learning_rate": 1.5207705923670157e-06,
2498
+ "loss": 0.9581,
2499
+ "step": 2070
2500
+ },
2501
+ {
2502
+ "epoch": 0.95,
2503
+ "learning_rate": 1.3847658972951483e-06,
2504
+ "loss": 0.9191,
2505
+ "step": 2075
2506
+ },
2507
+ {
2508
+ "epoch": 0.95,
2509
+ "learning_rate": 1.2550886162746467e-06,
2510
+ "loss": 0.95,
2511
+ "step": 2080
2512
+ },
2513
+ {
2514
+ "epoch": 0.96,
2515
+ "learning_rate": 1.1317470697425837e-06,
2516
+ "loss": 0.9686,
2517
+ "step": 2085
2518
+ },
2519
+ {
2520
+ "epoch": 0.96,
2521
+ "learning_rate": 1.0147491716185676e-06,
2522
+ "loss": 0.958,
2523
+ "step": 2090
2524
+ },
2525
+ {
2526
+ "epoch": 0.96,
2527
+ "learning_rate": 9.04102428796949e-07,
2528
+ "loss": 0.9578,
2529
+ "step": 2095
2530
+ },
2531
+ {
2532
+ "epoch": 0.96,
2533
+ "learning_rate": 7.99813940665195e-07,
2534
+ "loss": 0.9487,
2535
+ "step": 2100
2536
+ },
2537
+ {
2538
+ "epoch": 0.97,
2539
+ "learning_rate": 7.018903986483083e-07,
2540
+ "loss": 0.9474,
2541
+ "step": 2105
2542
+ },
2543
+ {
2544
+ "epoch": 0.97,
2545
+ "learning_rate": 6.103380857795604e-07,
2546
+ "loss": 0.9377,
2547
+ "step": 2110
2548
+ },
2549
+ {
2550
+ "epoch": 0.97,
2551
+ "learning_rate": 5.251628762972916e-07,
2552
+ "loss": 0.9454,
2553
+ "step": 2115
2554
+ },
2555
+ {
2556
+ "epoch": 0.97,
2557
+ "learning_rate": 4.463702352680787e-07,
2558
+ "loss": 0.9353,
2559
+ "step": 2120
2560
+ },
2561
+ {
2562
+ "epoch": 0.98,
2563
+ "learning_rate": 3.7396521823600537e-07,
2564
+ "loss": 0.9256,
2565
+ "step": 2125
2566
+ },
2567
+ {
2568
+ "epoch": 0.98,
2569
+ "learning_rate": 3.079524708983095e-07,
2570
+ "loss": 0.9437,
2571
+ "step": 2130
2572
+ },
2573
+ {
2574
+ "epoch": 0.98,
2575
+ "learning_rate": 2.483362288073443e-07,
2576
+ "loss": 0.9629,
2577
+ "step": 2135
2578
+ },
2579
+ {
2580
+ "epoch": 0.98,
2581
+ "learning_rate": 1.9512031709874035e-07,
2582
+ "loss": 0.9361,
2583
+ "step": 2140
2584
+ },
2585
+ {
2586
+ "epoch": 0.98,
2587
+ "learning_rate": 1.4830815024606815e-07,
2588
+ "loss": 0.9511,
2589
+ "step": 2145
2590
+ },
2591
+ {
2592
+ "epoch": 0.99,
2593
+ "learning_rate": 1.0790273184164701e-07,
2594
+ "loss": 0.9489,
2595
+ "step": 2150
2596
+ },
2597
+ {
2598
+ "epoch": 0.99,
2599
+ "learning_rate": 7.390665440393241e-08,
2600
+ "loss": 0.9875,
2601
+ "step": 2155
2602
+ },
2603
+ {
2604
+ "epoch": 0.99,
2605
+ "learning_rate": 4.632209921107134e-08,
2606
+ "loss": 0.9478,
2607
+ "step": 2160
2608
+ },
2609
+ {
2610
+ "epoch": 0.99,
2611
+ "learning_rate": 2.5150836161058622e-08,
2612
+ "loss": 0.9373,
2613
+ "step": 2165
2614
+ },
2615
+ {
2616
+ "epoch": 1.0,
2617
+ "learning_rate": 1.03942236580723e-08,
2618
+ "loss": 0.939,
2619
+ "step": 2170
2620
+ },
2621
+ {
2622
+ "epoch": 1.0,
2623
+ "learning_rate": 2.053208525365502e-09,
2624
+ "loss": 0.9428,
2625
+ "step": 2175
2626
+ },
2627
+ {
2628
+ "epoch": 1.0,
2629
+ "eval_loss": 0.9502018094062805,
2630
+ "eval_runtime": 365.2157,
2631
+ "eval_samples_per_second": 42.252,
2632
+ "eval_steps_per_second": 0.663,
2633
+ "step": 2179
2634
+ },
2635
+ {
2636
+ "epoch": 1.0,
2637
+ "step": 2179,
2638
+ "total_flos": 1.225695388701714e+19,
2639
+ "train_loss": 0.03432130868306889,
2640
+ "train_runtime": 782.9437,
2641
+ "train_samples_per_second": 178.079,
2642
+ "train_steps_per_second": 2.783
2643
+ }
2644
+ ],
2645
+ "logging_steps": 5,
2646
+ "max_steps": 2179,
2647
+ "num_input_tokens_seen": 0,
2648
+ "num_train_epochs": 1,
2649
+ "save_steps": 100,
2650
+ "total_flos": 1.225695388701714e+19,
2651
+ "train_batch_size": 4,
2652
+ "trial_name": null,
2653
+ "trial_params": null
2654
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8947404a796b921272cf8bf1145be7393f706963dfdeda7b084cd908b4768751
3
  size 4728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37edc177db33f988a5ee295860861e950495df76c641caf7a06c759a41b9a3d9
3
  size 4728