Model save
Browse files- README.md +66 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- runs/May10_22-46-17_COE-CS-sv003/events.out.tfevents.1715381194.COE-CS-sv003.1078355.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +2768 -0
README.md
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: mistralai/Mistral-7B-v0.1
|
| 9 |
+
model-index:
|
| 10 |
+
- name: mistral-poe-10p-10th-0.8
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# mistral-poe-10p-10th-0.8
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: nan
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- total_train_batch_size: 32
|
| 47 |
+
- total_eval_batch_size: 4
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 57 |
+
| 1.0957 | 1.0 | 1949 | nan |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- PEFT 0.7.1
|
| 63 |
+
- Transformers 4.39.0.dev0
|
| 64 |
+
- Pytorch 2.1.2
|
| 65 |
+
- Datasets 2.14.6
|
| 66 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 33307944
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7254c8474e4494eec6ead2e4ebdcf9fa9b96c2767a7e7b1fb07a6a0094ad5c0d
|
| 3 |
size 33307944
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 1.2706995834871218,
|
| 4 |
+
"train_runtime": 18075.5808,
|
| 5 |
+
"train_samples": 62359,
|
| 6 |
+
"train_samples_per_second": 3.45,
|
| 7 |
+
"train_steps_per_second": 0.108
|
| 8 |
+
}
|
runs/May10_22-46-17_COE-CS-sv003/events.out.tfevents.1715381194.COE-CS-sv003.1078355.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2bd1174aa42b6f4e9489e28493509c98e16bf651e6cec7fe61474b75982f0b45
|
| 3 |
+
size 87505
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 1.2706995834871218,
|
| 4 |
+
"train_runtime": 18075.5808,
|
| 5 |
+
"train_samples": 62359,
|
| 6 |
+
"train_samples_per_second": 3.45,
|
| 7 |
+
"train_steps_per_second": 0.108
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2768 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1949,
|
| 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 |
+
"grad_norm": 1.7453284027716789,
|
| 14 |
+
"learning_rate": 1.0256410256410257e-06,
|
| 15 |
+
"loss": 4.8029,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0,
|
| 20 |
+
"grad_norm": 1.8838630132500438,
|
| 21 |
+
"learning_rate": 5.128205128205128e-06,
|
| 22 |
+
"loss": 4.8678,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01,
|
| 27 |
+
"grad_norm": 1.9750863683254836,
|
| 28 |
+
"learning_rate": 1.0256410256410256e-05,
|
| 29 |
+
"loss": 4.7915,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01,
|
| 34 |
+
"grad_norm": 2.0960739711216374,
|
| 35 |
+
"learning_rate": 1.5384615384615387e-05,
|
| 36 |
+
"loss": 4.7429,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01,
|
| 41 |
+
"grad_norm": 2.68908841814168,
|
| 42 |
+
"learning_rate": 2.0512820512820512e-05,
|
| 43 |
+
"loss": 4.655,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01,
|
| 48 |
+
"grad_norm": 3.4517244544448675,
|
| 49 |
+
"learning_rate": 2.564102564102564e-05,
|
| 50 |
+
"loss": 4.3514,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02,
|
| 55 |
+
"grad_norm": 4.136057917635942,
|
| 56 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 57 |
+
"loss": 3.905,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02,
|
| 62 |
+
"grad_norm": 4.589117875619624,
|
| 63 |
+
"learning_rate": 3.58974358974359e-05,
|
| 64 |
+
"loss": 3.2509,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02,
|
| 69 |
+
"grad_norm": 4.927853133671092,
|
| 70 |
+
"learning_rate": 4.1025641025641023e-05,
|
| 71 |
+
"loss": 2.6795,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02,
|
| 76 |
+
"grad_norm": 3.054252767001273,
|
| 77 |
+
"learning_rate": 4.615384615384616e-05,
|
| 78 |
+
"loss": 2.187,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03,
|
| 83 |
+
"grad_norm": 1.9849933991408988,
|
| 84 |
+
"learning_rate": 5.128205128205128e-05,
|
| 85 |
+
"loss": 1.9199,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03,
|
| 90 |
+
"grad_norm": 1.8072063144215404,
|
| 91 |
+
"learning_rate": 5.6410256410256414e-05,
|
| 92 |
+
"loss": 1.7439,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03,
|
| 97 |
+
"grad_norm": 1.3923291097565402,
|
| 98 |
+
"learning_rate": 6.153846153846155e-05,
|
| 99 |
+
"loss": 1.7219,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03,
|
| 104 |
+
"grad_norm": 1.1768172611689185,
|
| 105 |
+
"learning_rate": 6.666666666666667e-05,
|
| 106 |
+
"loss": 1.6276,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04,
|
| 111 |
+
"grad_norm": 1.1476923157725187,
|
| 112 |
+
"learning_rate": 7.17948717948718e-05,
|
| 113 |
+
"loss": 1.5326,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04,
|
| 118 |
+
"grad_norm": 0.9441328628031238,
|
| 119 |
+
"learning_rate": 7.692307692307693e-05,
|
| 120 |
+
"loss": 1.5766,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04,
|
| 125 |
+
"grad_norm": 0.8493903573919569,
|
| 126 |
+
"learning_rate": 8.205128205128205e-05,
|
| 127 |
+
"loss": 1.4017,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.04,
|
| 132 |
+
"grad_norm": 1.0349555509222987,
|
| 133 |
+
"learning_rate": 8.717948717948718e-05,
|
| 134 |
+
"loss": 1.447,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05,
|
| 139 |
+
"grad_norm": 0.8454953032919131,
|
| 140 |
+
"learning_rate": 9.230769230769232e-05,
|
| 141 |
+
"loss": 1.4113,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05,
|
| 146 |
+
"grad_norm": 1.0220924006013166,
|
| 147 |
+
"learning_rate": 9.743589743589744e-05,
|
| 148 |
+
"loss": 1.4865,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05,
|
| 153 |
+
"grad_norm": 0.9739811152642958,
|
| 154 |
+
"learning_rate": 0.00010256410256410256,
|
| 155 |
+
"loss": 1.4709,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05,
|
| 160 |
+
"grad_norm": 0.8989088188815814,
|
| 161 |
+
"learning_rate": 0.0001076923076923077,
|
| 162 |
+
"loss": 1.4335,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06,
|
| 167 |
+
"grad_norm": 0.9043247030909609,
|
| 168 |
+
"learning_rate": 0.00011282051282051283,
|
| 169 |
+
"loss": 1.4196,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06,
|
| 174 |
+
"grad_norm": 1.0554112608560857,
|
| 175 |
+
"learning_rate": 0.00011794871794871796,
|
| 176 |
+
"loss": 1.376,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06,
|
| 181 |
+
"grad_norm": 0.9282876975712975,
|
| 182 |
+
"learning_rate": 0.0001230769230769231,
|
| 183 |
+
"loss": 1.4137,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06,
|
| 188 |
+
"grad_norm": 0.8957189154484865,
|
| 189 |
+
"learning_rate": 0.00012820512820512823,
|
| 190 |
+
"loss": 1.3528,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07,
|
| 195 |
+
"grad_norm": 1.3941737136229102,
|
| 196 |
+
"learning_rate": 0.00013333333333333334,
|
| 197 |
+
"loss": 1.3963,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07,
|
| 202 |
+
"grad_norm": 0.7436200394991389,
|
| 203 |
+
"learning_rate": 0.00013846153846153847,
|
| 204 |
+
"loss": 1.3047,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07,
|
| 209 |
+
"grad_norm": 1.1477489067913107,
|
| 210 |
+
"learning_rate": 0.0001435897435897436,
|
| 211 |
+
"loss": 1.3011,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07,
|
| 216 |
+
"grad_norm": 0.8722003799830739,
|
| 217 |
+
"learning_rate": 0.00014871794871794872,
|
| 218 |
+
"loss": 1.2378,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08,
|
| 223 |
+
"grad_norm": 1.1769840880132467,
|
| 224 |
+
"learning_rate": 0.00015384615384615385,
|
| 225 |
+
"loss": 1.4124,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08,
|
| 230 |
+
"grad_norm": 0.7399233026092019,
|
| 231 |
+
"learning_rate": 0.00015897435897435896,
|
| 232 |
+
"loss": 1.3027,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08,
|
| 237 |
+
"grad_norm": 0.9538756518623174,
|
| 238 |
+
"learning_rate": 0.0001641025641025641,
|
| 239 |
+
"loss": 1.2812,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08,
|
| 244 |
+
"grad_norm": 0.987821734343656,
|
| 245 |
+
"learning_rate": 0.00016923076923076923,
|
| 246 |
+
"loss": 1.3357,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09,
|
| 251 |
+
"grad_norm": 0.9972992684207476,
|
| 252 |
+
"learning_rate": 0.00017435897435897436,
|
| 253 |
+
"loss": 1.3845,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09,
|
| 258 |
+
"grad_norm": 0.9162912417797362,
|
| 259 |
+
"learning_rate": 0.0001794871794871795,
|
| 260 |
+
"loss": 1.4145,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09,
|
| 265 |
+
"grad_norm": 0.9727023032566935,
|
| 266 |
+
"learning_rate": 0.00018461538461538463,
|
| 267 |
+
"loss": 1.2514,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09,
|
| 272 |
+
"grad_norm": 1.1205678233237835,
|
| 273 |
+
"learning_rate": 0.00018974358974358974,
|
| 274 |
+
"loss": 1.3525,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1,
|
| 279 |
+
"grad_norm": 0.9769806404545222,
|
| 280 |
+
"learning_rate": 0.00019487179487179487,
|
| 281 |
+
"loss": 1.2291,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1,
|
| 286 |
+
"grad_norm": 0.8933569683133975,
|
| 287 |
+
"learning_rate": 0.0002,
|
| 288 |
+
"loss": 1.3543,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1,
|
| 293 |
+
"grad_norm": 0.9689935176055495,
|
| 294 |
+
"learning_rate": 0.00019999598996948235,
|
| 295 |
+
"loss": 1.2744,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.11,
|
| 300 |
+
"grad_norm": 0.9199321088853798,
|
| 301 |
+
"learning_rate": 0.00019998396019953624,
|
| 302 |
+
"loss": 1.3059,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11,
|
| 307 |
+
"grad_norm": 1.0113198660825118,
|
| 308 |
+
"learning_rate": 0.0001999639116549566,
|
| 309 |
+
"loss": 1.1999,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11,
|
| 314 |
+
"grad_norm": 0.7980098887264342,
|
| 315 |
+
"learning_rate": 0.00019993584594364894,
|
| 316 |
+
"loss": 1.2418,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11,
|
| 321 |
+
"grad_norm": 0.7449325030003408,
|
| 322 |
+
"learning_rate": 0.0001998997653165004,
|
| 323 |
+
"loss": 1.284,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.12,
|
| 328 |
+
"grad_norm": 1.0156080824941,
|
| 329 |
+
"learning_rate": 0.00019985567266719934,
|
| 330 |
+
"loss": 1.3727,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12,
|
| 335 |
+
"grad_norm": 1.0387246660942182,
|
| 336 |
+
"learning_rate": 0.00019980357153200315,
|
| 337 |
+
"loss": 1.3186,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12,
|
| 342 |
+
"grad_norm": 1.1886194935509105,
|
| 343 |
+
"learning_rate": 0.00019974346608945466,
|
| 344 |
+
"loss": 1.2949,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12,
|
| 349 |
+
"grad_norm": 0.9654192561255139,
|
| 350 |
+
"learning_rate": 0.00019967536116004698,
|
| 351 |
+
"loss": 1.3067,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13,
|
| 356 |
+
"grad_norm": 1.0385716215564977,
|
| 357 |
+
"learning_rate": 0.00019959926220583713,
|
| 358 |
+
"loss": 1.289,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13,
|
| 363 |
+
"grad_norm": 0.8709732110452237,
|
| 364 |
+
"learning_rate": 0.00019951517533000764,
|
| 365 |
+
"loss": 1.2138,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13,
|
| 370 |
+
"grad_norm": 0.6847073355471284,
|
| 371 |
+
"learning_rate": 0.00019942310727637724,
|
| 372 |
+
"loss": 1.2844,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13,
|
| 377 |
+
"grad_norm": 1.051169320358427,
|
| 378 |
+
"learning_rate": 0.00019932306542886009,
|
| 379 |
+
"loss": 1.3289,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14,
|
| 384 |
+
"grad_norm": 0.8704314065526715,
|
| 385 |
+
"learning_rate": 0.00019921505781087334,
|
| 386 |
+
"loss": 1.3339,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14,
|
| 391 |
+
"grad_norm": 0.7816111152013846,
|
| 392 |
+
"learning_rate": 0.00019909909308469398,
|
| 393 |
+
"loss": 1.3282,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14,
|
| 398 |
+
"grad_norm": 1.038569084884339,
|
| 399 |
+
"learning_rate": 0.0001989751805507637,
|
| 400 |
+
"loss": 1.3112,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14,
|
| 405 |
+
"grad_norm": 0.9061282430692424,
|
| 406 |
+
"learning_rate": 0.00019884333014694345,
|
| 407 |
+
"loss": 1.3251,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15,
|
| 412 |
+
"grad_norm": 0.7513521845340044,
|
| 413 |
+
"learning_rate": 0.00019870355244771607,
|
| 414 |
+
"loss": 1.2893,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.15,
|
| 419 |
+
"grad_norm": 0.910789847545989,
|
| 420 |
+
"learning_rate": 0.00019855585866333835,
|
| 421 |
+
"loss": 1.288,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.15,
|
| 426 |
+
"grad_norm": 1.0532355924793768,
|
| 427 |
+
"learning_rate": 0.00019840026063894193,
|
| 428 |
+
"loss": 1.2805,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.15,
|
| 433 |
+
"grad_norm": 1.0244686274469181,
|
| 434 |
+
"learning_rate": 0.00019823677085358335,
|
| 435 |
+
"loss": 1.2499,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.16,
|
| 440 |
+
"grad_norm": 0.7707389186864114,
|
| 441 |
+
"learning_rate": 0.00019806540241924317,
|
| 442 |
+
"loss": 1.2532,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16,
|
| 447 |
+
"grad_norm": 0.7446042971478654,
|
| 448 |
+
"learning_rate": 0.00019788616907977441,
|
| 449 |
+
"loss": 1.2981,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.16,
|
| 454 |
+
"grad_norm": 0.6848536091053609,
|
| 455 |
+
"learning_rate": 0.00019769908520980034,
|
| 456 |
+
"loss": 1.2565,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.16,
|
| 461 |
+
"grad_norm": 0.7586642653181629,
|
| 462 |
+
"learning_rate": 0.00019750416581356146,
|
| 463 |
+
"loss": 1.2648,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17,
|
| 468 |
+
"grad_norm": 0.9291629579523186,
|
| 469 |
+
"learning_rate": 0.00019730142652371236,
|
| 470 |
+
"loss": 1.1065,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17,
|
| 475 |
+
"grad_norm": 0.7774039891171479,
|
| 476 |
+
"learning_rate": 0.0001970908836000678,
|
| 477 |
+
"loss": 1.2582,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17,
|
| 482 |
+
"grad_norm": 0.7535737744178963,
|
| 483 |
+
"learning_rate": 0.00019687255392829877,
|
| 484 |
+
"loss": 1.255,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.17,
|
| 489 |
+
"grad_norm": 0.8232038564451061,
|
| 490 |
+
"learning_rate": 0.0001966464550185782,
|
| 491 |
+
"loss": 1.2246,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18,
|
| 496 |
+
"grad_norm": 0.7763354984102764,
|
| 497 |
+
"learning_rate": 0.0001964126050041767,
|
| 498 |
+
"loss": 1.1932,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18,
|
| 503 |
+
"grad_norm": 0.9765706941888299,
|
| 504 |
+
"learning_rate": 0.0001961710226400081,
|
| 505 |
+
"loss": 1.2209,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18,
|
| 510 |
+
"grad_norm": 0.7182663624300912,
|
| 511 |
+
"learning_rate": 0.00019592172730112544,
|
| 512 |
+
"loss": 1.2383,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.18,
|
| 517 |
+
"grad_norm": 0.8481687196291695,
|
| 518 |
+
"learning_rate": 0.00019566473898116713,
|
| 519 |
+
"loss": 1.2911,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19,
|
| 524 |
+
"grad_norm": 0.7711940010123207,
|
| 525 |
+
"learning_rate": 0.0001954000782907532,
|
| 526 |
+
"loss": 1.231,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19,
|
| 531 |
+
"grad_norm": 1.061067518571626,
|
| 532 |
+
"learning_rate": 0.00019512776645583263,
|
| 533 |
+
"loss": 1.259,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19,
|
| 538 |
+
"grad_norm": 1.0715949890319147,
|
| 539 |
+
"learning_rate": 0.00019484782531598073,
|
| 540 |
+
"loss": 1.2712,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.19,
|
| 545 |
+
"grad_norm": 0.774154512584797,
|
| 546 |
+
"learning_rate": 0.00019456027732264784,
|
| 547 |
+
"loss": 1.306,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2,
|
| 552 |
+
"grad_norm": 0.8172535402599491,
|
| 553 |
+
"learning_rate": 0.00019426514553735848,
|
| 554 |
+
"loss": 1.2734,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2,
|
| 559 |
+
"grad_norm": 0.9451921871881396,
|
| 560 |
+
"learning_rate": 0.00019396245362986197,
|
| 561 |
+
"loss": 1.231,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2,
|
| 566 |
+
"grad_norm": 0.9954188288045217,
|
| 567 |
+
"learning_rate": 0.00019365222587623405,
|
| 568 |
+
"loss": 1.2061,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.21,
|
| 573 |
+
"grad_norm": 0.8088092576041586,
|
| 574 |
+
"learning_rate": 0.00019333448715692995,
|
| 575 |
+
"loss": 1.2403,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.21,
|
| 580 |
+
"grad_norm": 0.9867132889317975,
|
| 581 |
+
"learning_rate": 0.00019300926295478884,
|
| 582 |
+
"loss": 1.3009,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.21,
|
| 587 |
+
"grad_norm": 0.8580342036883377,
|
| 588 |
+
"learning_rate": 0.0001926765793529902,
|
| 589 |
+
"loss": 1.2371,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.21,
|
| 594 |
+
"grad_norm": 0.813693213024289,
|
| 595 |
+
"learning_rate": 0.00019233646303296205,
|
| 596 |
+
"loss": 1.2767,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22,
|
| 601 |
+
"grad_norm": 0.8552421430267484,
|
| 602 |
+
"learning_rate": 0.00019198894127224074,
|
| 603 |
+
"loss": 1.2419,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22,
|
| 608 |
+
"grad_norm": 0.9962550179351269,
|
| 609 |
+
"learning_rate": 0.0001916340419422837,
|
| 610 |
+
"loss": 1.2765,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.22,
|
| 615 |
+
"grad_norm": 0.790052217945725,
|
| 616 |
+
"learning_rate": 0.00019127179350623372,
|
| 617 |
+
"loss": 1.2135,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.22,
|
| 622 |
+
"grad_norm": 0.7916779759677229,
|
| 623 |
+
"learning_rate": 0.0001909022250166365,
|
| 624 |
+
"loss": 1.2178,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23,
|
| 629 |
+
"grad_norm": 0.7297942006635301,
|
| 630 |
+
"learning_rate": 0.00019052536611311046,
|
| 631 |
+
"loss": 1.245,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23,
|
| 636 |
+
"grad_norm": 0.8857053212429089,
|
| 637 |
+
"learning_rate": 0.00019014124701996973,
|
| 638 |
+
"loss": 1.2657,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.23,
|
| 643 |
+
"grad_norm": 0.726328154530024,
|
| 644 |
+
"learning_rate": 0.00018974989854379996,
|
| 645 |
+
"loss": 1.2481,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.23,
|
| 650 |
+
"grad_norm": 1.0011198664456165,
|
| 651 |
+
"learning_rate": 0.00018935135207098785,
|
| 652 |
+
"loss": 1.2323,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24,
|
| 657 |
+
"grad_norm": 0.7783715496305177,
|
| 658 |
+
"learning_rate": 0.00018894563956520374,
|
| 659 |
+
"loss": 1.2367,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24,
|
| 664 |
+
"grad_norm": 0.8696181443757303,
|
| 665 |
+
"learning_rate": 0.00018853279356483826,
|
| 666 |
+
"loss": 1.315,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.24,
|
| 671 |
+
"grad_norm": 0.7894223278767809,
|
| 672 |
+
"learning_rate": 0.00018811284718039256,
|
| 673 |
+
"loss": 1.124,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.24,
|
| 678 |
+
"grad_norm": 1.1136345673311567,
|
| 679 |
+
"learning_rate": 0.00018768583409182305,
|
| 680 |
+
"loss": 1.2198,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.25,
|
| 685 |
+
"grad_norm": 0.7243107301552627,
|
| 686 |
+
"learning_rate": 0.00018725178854584007,
|
| 687 |
+
"loss": 1.2523,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.25,
|
| 692 |
+
"grad_norm": 0.6223206270124363,
|
| 693 |
+
"learning_rate": 0.00018681074535316125,
|
| 694 |
+
"loss": 1.2673,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.25,
|
| 699 |
+
"grad_norm": 0.8039914099462059,
|
| 700 |
+
"learning_rate": 0.00018636273988571991,
|
| 701 |
+
"loss": 1.1877,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.25,
|
| 706 |
+
"grad_norm": 0.7116186179251531,
|
| 707 |
+
"learning_rate": 0.0001859078080738279,
|
| 708 |
+
"loss": 1.1458,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.26,
|
| 713 |
+
"grad_norm": 0.7310411784585837,
|
| 714 |
+
"learning_rate": 0.00018544598640329432,
|
| 715 |
+
"loss": 1.2417,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.26,
|
| 720 |
+
"grad_norm": 0.7429734188864324,
|
| 721 |
+
"learning_rate": 0.00018497731191249894,
|
| 722 |
+
"loss": 1.2852,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.26,
|
| 727 |
+
"grad_norm": 0.9083872621008465,
|
| 728 |
+
"learning_rate": 0.000184501822189422,
|
| 729 |
+
"loss": 1.2436,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.26,
|
| 734 |
+
"grad_norm": 0.9414190517105039,
|
| 735 |
+
"learning_rate": 0.00018401955536862948,
|
| 736 |
+
"loss": 1.2382,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.27,
|
| 741 |
+
"grad_norm": 0.9100118313504707,
|
| 742 |
+
"learning_rate": 0.0001835305501282148,
|
| 743 |
+
"loss": 1.2651,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.27,
|
| 748 |
+
"grad_norm": 0.8682421638749216,
|
| 749 |
+
"learning_rate": 0.00018303484568669667,
|
| 750 |
+
"loss": 1.214,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.27,
|
| 755 |
+
"grad_norm": 0.7124640006090736,
|
| 756 |
+
"learning_rate": 0.00018253248179987388,
|
| 757 |
+
"loss": 1.1703,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.27,
|
| 762 |
+
"grad_norm": 0.7882268918908124,
|
| 763 |
+
"learning_rate": 0.0001820234987576368,
|
| 764 |
+
"loss": 1.1747,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.28,
|
| 769 |
+
"grad_norm": 0.9355156994606193,
|
| 770 |
+
"learning_rate": 0.00018150793738073602,
|
| 771 |
+
"loss": 1.1954,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.28,
|
| 776 |
+
"grad_norm": 0.962953079362863,
|
| 777 |
+
"learning_rate": 0.00018098583901750867,
|
| 778 |
+
"loss": 1.21,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.28,
|
| 783 |
+
"grad_norm": 0.8058576401677449,
|
| 784 |
+
"learning_rate": 0.00018045724554056214,
|
| 785 |
+
"loss": 1.1945,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.28,
|
| 790 |
+
"grad_norm": 0.8718736747079745,
|
| 791 |
+
"learning_rate": 0.0001799221993434159,
|
| 792 |
+
"loss": 1.242,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.29,
|
| 797 |
+
"grad_norm": 0.8820093361979019,
|
| 798 |
+
"learning_rate": 0.00017938074333710157,
|
| 799 |
+
"loss": 1.216,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.29,
|
| 804 |
+
"grad_norm": 0.626929459625588,
|
| 805 |
+
"learning_rate": 0.00017883292094672128,
|
| 806 |
+
"loss": 1.1947,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.29,
|
| 811 |
+
"grad_norm": 0.8252467152886114,
|
| 812 |
+
"learning_rate": 0.00017827877610796514,
|
| 813 |
+
"loss": 1.1953,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3,
|
| 818 |
+
"grad_norm": 0.894675298458733,
|
| 819 |
+
"learning_rate": 0.00017771835326358743,
|
| 820 |
+
"loss": 1.1873,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3,
|
| 825 |
+
"grad_norm": 0.6648245020030112,
|
| 826 |
+
"learning_rate": 0.00017715169735984233,
|
| 827 |
+
"loss": 1.1293,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3,
|
| 832 |
+
"grad_norm": 0.8629378772284809,
|
| 833 |
+
"learning_rate": 0.0001765788538428792,
|
| 834 |
+
"loss": 1.288,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3,
|
| 839 |
+
"grad_norm": 0.9066125782712056,
|
| 840 |
+
"learning_rate": 0.00017599986865509767,
|
| 841 |
+
"loss": 1.2058,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.31,
|
| 846 |
+
"grad_norm": 0.7423874071094034,
|
| 847 |
+
"learning_rate": 0.00017541478823146327,
|
| 848 |
+
"loss": 1.2508,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.31,
|
| 853 |
+
"grad_norm": 1.01835592219078,
|
| 854 |
+
"learning_rate": 0.00017482365949578302,
|
| 855 |
+
"loss": 1.1974,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.31,
|
| 860 |
+
"grad_norm": 0.878167603852895,
|
| 861 |
+
"learning_rate": 0.00017422652985694237,
|
| 862 |
+
"loss": 1.2583,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.31,
|
| 867 |
+
"grad_norm": 0.7708769505928975,
|
| 868 |
+
"learning_rate": 0.00017362344720510278,
|
| 869 |
+
"loss": 1.1853,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.32,
|
| 874 |
+
"grad_norm": 0.7652211411520762,
|
| 875 |
+
"learning_rate": 0.00017301445990786102,
|
| 876 |
+
"loss": 1.2707,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32,
|
| 881 |
+
"grad_norm": 0.7286424011748441,
|
| 882 |
+
"learning_rate": 0.00017239961680637,
|
| 883 |
+
"loss": 1.2044,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.32,
|
| 888 |
+
"grad_norm": 0.6950598836326154,
|
| 889 |
+
"learning_rate": 0.0001717789672114218,
|
| 890 |
+
"loss": 1.3171,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.32,
|
| 895 |
+
"grad_norm": 0.7009337184084006,
|
| 896 |
+
"learning_rate": 0.0001711525608994927,
|
| 897 |
+
"loss": 1.1541,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.33,
|
| 902 |
+
"grad_norm": 0.8543565184004501,
|
| 903 |
+
"learning_rate": 0.00017052044810875126,
|
| 904 |
+
"loss": 1.2465,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.33,
|
| 909 |
+
"grad_norm": 1.1754223407681135,
|
| 910 |
+
"learning_rate": 0.00016988267953502913,
|
| 911 |
+
"loss": 1.1644,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.33,
|
| 916 |
+
"grad_norm": 0.8933582330947162,
|
| 917 |
+
"learning_rate": 0.00016923930632775516,
|
| 918 |
+
"loss": 1.1798,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.33,
|
| 923 |
+
"grad_norm": 0.932416979264158,
|
| 924 |
+
"learning_rate": 0.00016859038008585326,
|
| 925 |
+
"loss": 1.1983,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.34,
|
| 930 |
+
"grad_norm": 0.8373224826856496,
|
| 931 |
+
"learning_rate": 0.0001679359528536041,
|
| 932 |
+
"loss": 1.1567,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.34,
|
| 937 |
+
"grad_norm": 0.8214365574255561,
|
| 938 |
+
"learning_rate": 0.00016727607711647114,
|
| 939 |
+
"loss": 1.2188,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.34,
|
| 944 |
+
"grad_norm": 0.9351111268728677,
|
| 945 |
+
"learning_rate": 0.00016661080579689132,
|
| 946 |
+
"loss": 1.1873,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.34,
|
| 951 |
+
"grad_norm": 0.8498925724649622,
|
| 952 |
+
"learning_rate": 0.0001659401922500304,
|
| 953 |
+
"loss": 1.1937,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.35,
|
| 958 |
+
"grad_norm": 0.7284767681268423,
|
| 959 |
+
"learning_rate": 0.00016526429025950424,
|
| 960 |
+
"loss": 1.1301,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.35,
|
| 965 |
+
"grad_norm": 0.7481373744501812,
|
| 966 |
+
"learning_rate": 0.00016458315403306502,
|
| 967 |
+
"loss": 1.2671,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.35,
|
| 972 |
+
"grad_norm": 0.8920310346592031,
|
| 973 |
+
"learning_rate": 0.0001638968381982538,
|
| 974 |
+
"loss": 1.1528,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.35,
|
| 979 |
+
"grad_norm": 0.7925520466506051,
|
| 980 |
+
"learning_rate": 0.0001632053977980194,
|
| 981 |
+
"loss": 1.2368,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.36,
|
| 986 |
+
"grad_norm": 1.0192425967562995,
|
| 987 |
+
"learning_rate": 0.000162508888286304,
|
| 988 |
+
"loss": 1.2167,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.36,
|
| 993 |
+
"grad_norm": 0.7085398144677333,
|
| 994 |
+
"learning_rate": 0.00016180736552359553,
|
| 995 |
+
"loss": 1.2218,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.36,
|
| 1000 |
+
"grad_norm": 0.768972223449797,
|
| 1001 |
+
"learning_rate": 0.00016110088577244773,
|
| 1002 |
+
"loss": 1.2483,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.36,
|
| 1007 |
+
"grad_norm": 0.8053989910048416,
|
| 1008 |
+
"learning_rate": 0.00016038950569296785,
|
| 1009 |
+
"loss": 1.1811,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.37,
|
| 1014 |
+
"grad_norm": 0.7498380291821676,
|
| 1015 |
+
"learning_rate": 0.00015967328233827249,
|
| 1016 |
+
"loss": 1.2625,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.37,
|
| 1021 |
+
"grad_norm": 0.7372396772067858,
|
| 1022 |
+
"learning_rate": 0.00015895227314991178,
|
| 1023 |
+
"loss": 1.1298,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.37,
|
| 1028 |
+
"grad_norm": 0.7380540614428072,
|
| 1029 |
+
"learning_rate": 0.00015822653595326275,
|
| 1030 |
+
"loss": 1.1611,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.37,
|
| 1035 |
+
"grad_norm": 0.840298145620476,
|
| 1036 |
+
"learning_rate": 0.00015749612895289152,
|
| 1037 |
+
"loss": 1.1815,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.38,
|
| 1042 |
+
"grad_norm": 0.6743184976694755,
|
| 1043 |
+
"learning_rate": 0.00015676111072788527,
|
| 1044 |
+
"loss": 1.1665,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.38,
|
| 1049 |
+
"grad_norm": 1.0773509553134593,
|
| 1050 |
+
"learning_rate": 0.00015602154022715435,
|
| 1051 |
+
"loss": 1.214,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.38,
|
| 1056 |
+
"grad_norm": 1.2157724994677412,
|
| 1057 |
+
"learning_rate": 0.0001552774767647043,
|
| 1058 |
+
"loss": 1.1757,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.38,
|
| 1063 |
+
"grad_norm": 0.8932509973388469,
|
| 1064 |
+
"learning_rate": 0.0001545289800148789,
|
| 1065 |
+
"loss": 1.1747,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.39,
|
| 1070 |
+
"grad_norm": 0.773624653232096,
|
| 1071 |
+
"learning_rate": 0.0001537761100075744,
|
| 1072 |
+
"loss": 1.1934,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.39,
|
| 1077 |
+
"grad_norm": 0.6455277482504939,
|
| 1078 |
+
"learning_rate": 0.00015301892712342482,
|
| 1079 |
+
"loss": 1.2505,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.39,
|
| 1084 |
+
"grad_norm": 0.7996020026859186,
|
| 1085 |
+
"learning_rate": 0.00015225749208895968,
|
| 1086 |
+
"loss": 1.2388,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4,
|
| 1091 |
+
"grad_norm": 0.7876699542363711,
|
| 1092 |
+
"learning_rate": 0.0001514918659717335,
|
| 1093 |
+
"loss": 1.2573,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4,
|
| 1098 |
+
"grad_norm": 0.7507865451214197,
|
| 1099 |
+
"learning_rate": 0.00015072211017542813,
|
| 1100 |
+
"loss": 1.2108,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4,
|
| 1105 |
+
"grad_norm": 0.742160262755694,
|
| 1106 |
+
"learning_rate": 0.00014994828643492827,
|
| 1107 |
+
"loss": 1.1554,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4,
|
| 1112 |
+
"grad_norm": 0.7898189151931231,
|
| 1113 |
+
"learning_rate": 0.00014917045681137026,
|
| 1114 |
+
"loss": 1.1858,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.41,
|
| 1119 |
+
"grad_norm": 0.7929109677821212,
|
| 1120 |
+
"learning_rate": 0.0001483886836871646,
|
| 1121 |
+
"loss": 1.1181,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.41,
|
| 1126 |
+
"grad_norm": 0.8366860762380138,
|
| 1127 |
+
"learning_rate": 0.00014760302976099304,
|
| 1128 |
+
"loss": 1.1683,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.41,
|
| 1133 |
+
"grad_norm": 0.8562461991337411,
|
| 1134 |
+
"learning_rate": 0.00014681355804278001,
|
| 1135 |
+
"loss": 1.1545,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.41,
|
| 1140 |
+
"grad_norm": 0.6903144224243607,
|
| 1141 |
+
"learning_rate": 0.00014602033184863913,
|
| 1142 |
+
"loss": 1.1668,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.42,
|
| 1147 |
+
"grad_norm": 0.720558400777453,
|
| 1148 |
+
"learning_rate": 0.00014522341479579533,
|
| 1149 |
+
"loss": 1.3024,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.42,
|
| 1154 |
+
"grad_norm": 0.6374665332941237,
|
| 1155 |
+
"learning_rate": 0.00014442287079748263,
|
| 1156 |
+
"loss": 1.1983,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.42,
|
| 1161 |
+
"grad_norm": 0.8700325415003359,
|
| 1162 |
+
"learning_rate": 0.00014361876405781832,
|
| 1163 |
+
"loss": 1.1604,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.42,
|
| 1168 |
+
"grad_norm": 0.876670415809109,
|
| 1169 |
+
"learning_rate": 0.00014281115906665374,
|
| 1170 |
+
"loss": 1.2463,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.43,
|
| 1175 |
+
"grad_norm": 0.9136860273140437,
|
| 1176 |
+
"learning_rate": 0.00014200012059440207,
|
| 1177 |
+
"loss": 1.1814,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.43,
|
| 1182 |
+
"grad_norm": 0.8346225178534373,
|
| 1183 |
+
"learning_rate": 0.00014118571368684383,
|
| 1184 |
+
"loss": 1.2552,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.43,
|
| 1189 |
+
"grad_norm": 0.698383520911491,
|
| 1190 |
+
"learning_rate": 0.00014036800365991008,
|
| 1191 |
+
"loss": 1.1132,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.43,
|
| 1196 |
+
"grad_norm": 0.8620628308046092,
|
| 1197 |
+
"learning_rate": 0.00013954705609444404,
|
| 1198 |
+
"loss": 1.1663,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.44,
|
| 1203 |
+
"grad_norm": 0.8119834436603902,
|
| 1204 |
+
"learning_rate": 0.00013872293683094152,
|
| 1205 |
+
"loss": 1.2139,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.44,
|
| 1210 |
+
"grad_norm": 0.7204007887604488,
|
| 1211 |
+
"learning_rate": 0.00013789571196427055,
|
| 1212 |
+
"loss": 1.193,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.44,
|
| 1217 |
+
"grad_norm": 0.9461103539326526,
|
| 1218 |
+
"learning_rate": 0.00013706544783837022,
|
| 1219 |
+
"loss": 1.1549,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.44,
|
| 1224 |
+
"grad_norm": 1.1236203572833683,
|
| 1225 |
+
"learning_rate": 0.00013623221104093025,
|
| 1226 |
+
"loss": 1.2752,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.45,
|
| 1231 |
+
"grad_norm": 0.7297339296275664,
|
| 1232 |
+
"learning_rate": 0.00013539606839805036,
|
| 1233 |
+
"loss": 1.1884,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.45,
|
| 1238 |
+
"grad_norm": 0.7684475171790612,
|
| 1239 |
+
"learning_rate": 0.00013455708696888085,
|
| 1240 |
+
"loss": 1.1928,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.45,
|
| 1245 |
+
"grad_norm": 0.7160993458180628,
|
| 1246 |
+
"learning_rate": 0.00013371533404024438,
|
| 1247 |
+
"loss": 1.2035,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.45,
|
| 1252 |
+
"grad_norm": 0.7366548665049109,
|
| 1253 |
+
"learning_rate": 0.00013287087712123962,
|
| 1254 |
+
"loss": 1.2372,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.46,
|
| 1259 |
+
"grad_norm": 0.7983969367173553,
|
| 1260 |
+
"learning_rate": 0.00013202378393782692,
|
| 1261 |
+
"loss": 1.1931,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.46,
|
| 1266 |
+
"grad_norm": 0.9470935523528393,
|
| 1267 |
+
"learning_rate": 0.00013117412242739655,
|
| 1268 |
+
"loss": 1.1274,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.46,
|
| 1273 |
+
"grad_norm": 1.0451209845718006,
|
| 1274 |
+
"learning_rate": 0.00013032196073332027,
|
| 1275 |
+
"loss": 1.2435,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.46,
|
| 1280 |
+
"grad_norm": 0.7471749313716822,
|
| 1281 |
+
"learning_rate": 0.00012946736719948607,
|
| 1282 |
+
"loss": 1.1844,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.47,
|
| 1287 |
+
"grad_norm": 0.6962269357662448,
|
| 1288 |
+
"learning_rate": 0.000128610410364817,
|
| 1289 |
+
"loss": 1.1526,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.47,
|
| 1294 |
+
"grad_norm": 0.8391163485957552,
|
| 1295 |
+
"learning_rate": 0.00012775115895777417,
|
| 1296 |
+
"loss": 1.2126,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.47,
|
| 1301 |
+
"grad_norm": 0.9175047450119641,
|
| 1302 |
+
"learning_rate": 0.00012688968189084493,
|
| 1303 |
+
"loss": 1.1199,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.47,
|
| 1308 |
+
"grad_norm": 0.8881533984263993,
|
| 1309 |
+
"learning_rate": 0.00012602604825501587,
|
| 1310 |
+
"loss": 1.1842,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.48,
|
| 1315 |
+
"grad_norm": 0.68829893427438,
|
| 1316 |
+
"learning_rate": 0.00012516032731423165,
|
| 1317 |
+
"loss": 1.1916,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.48,
|
| 1322 |
+
"grad_norm": 0.8290955836420092,
|
| 1323 |
+
"learning_rate": 0.00012429258849984014,
|
| 1324 |
+
"loss": 1.1915,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.48,
|
| 1329 |
+
"grad_norm": 0.7546478279579057,
|
| 1330 |
+
"learning_rate": 0.00012342290140502388,
|
| 1331 |
+
"loss": 1.1875,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.48,
|
| 1336 |
+
"grad_norm": 0.7857053878657014,
|
| 1337 |
+
"learning_rate": 0.00012255133577921868,
|
| 1338 |
+
"loss": 1.1883,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.49,
|
| 1343 |
+
"grad_norm": 0.7728388731432995,
|
| 1344 |
+
"learning_rate": 0.0001216779615225197,
|
| 1345 |
+
"loss": 1.1329,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.49,
|
| 1350 |
+
"grad_norm": 0.762285378503341,
|
| 1351 |
+
"learning_rate": 0.00012080284868007541,
|
| 1352 |
+
"loss": 1.1857,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.49,
|
| 1357 |
+
"grad_norm": 0.8455730497463988,
|
| 1358 |
+
"learning_rate": 0.0001199260674364699,
|
| 1359 |
+
"loss": 1.2241,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5,
|
| 1364 |
+
"grad_norm": 0.7637030076266145,
|
| 1365 |
+
"learning_rate": 0.00011904768811009405,
|
| 1366 |
+
"loss": 1.2333,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5,
|
| 1371 |
+
"grad_norm": 0.7399802799801922,
|
| 1372 |
+
"learning_rate": 0.00011816778114750593,
|
| 1373 |
+
"loss": 1.2126,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5,
|
| 1378 |
+
"grad_norm": 0.7804083472026294,
|
| 1379 |
+
"learning_rate": 0.00011728641711778103,
|
| 1380 |
+
"loss": 1.2239,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5,
|
| 1385 |
+
"grad_norm": 0.7666274480697189,
|
| 1386 |
+
"learning_rate": 0.00011640366670685248,
|
| 1387 |
+
"loss": 1.105,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.51,
|
| 1392 |
+
"grad_norm": 0.7651209631784908,
|
| 1393 |
+
"learning_rate": 0.00011551960071184195,
|
| 1394 |
+
"loss": 1.3045,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.51,
|
| 1399 |
+
"grad_norm": 0.9726990471095971,
|
| 1400 |
+
"learning_rate": 0.00011463429003538196,
|
| 1401 |
+
"loss": 1.1956,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.51,
|
| 1406 |
+
"grad_norm": 0.7087953125740096,
|
| 1407 |
+
"learning_rate": 0.000113747805679929,
|
| 1408 |
+
"loss": 1.2197,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.51,
|
| 1413 |
+
"grad_norm": 0.8944588834117093,
|
| 1414 |
+
"learning_rate": 0.00011286021874206952,
|
| 1415 |
+
"loss": 1.0926,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.52,
|
| 1420 |
+
"grad_norm": 1.0048773208945307,
|
| 1421 |
+
"learning_rate": 0.00011197160040681762,
|
| 1422 |
+
"loss": 1.2798,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.52,
|
| 1427 |
+
"grad_norm": 0.7348160099574793,
|
| 1428 |
+
"learning_rate": 0.0001110820219419062,
|
| 1429 |
+
"loss": 1.1951,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.52,
|
| 1434 |
+
"grad_norm": 0.7992861590760896,
|
| 1435 |
+
"learning_rate": 0.0001101915546920711,
|
| 1436 |
+
"loss": 1.1812,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.52,
|
| 1441 |
+
"grad_norm": 0.7458148092571341,
|
| 1442 |
+
"learning_rate": 0.00010930027007332923,
|
| 1443 |
+
"loss": 1.1236,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.53,
|
| 1448 |
+
"grad_norm": 0.7657904989795685,
|
| 1449 |
+
"learning_rate": 0.00010840823956725103,
|
| 1450 |
+
"loss": 1.1714,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.53,
|
| 1455 |
+
"grad_norm": 0.7797801904462873,
|
| 1456 |
+
"learning_rate": 0.00010751553471522757,
|
| 1457 |
+
"loss": 1.164,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.53,
|
| 1462 |
+
"grad_norm": 0.6805580259070753,
|
| 1463 |
+
"learning_rate": 0.00010662222711273279,
|
| 1464 |
+
"loss": 1.1326,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.53,
|
| 1469 |
+
"grad_norm": 0.8265666935073783,
|
| 1470 |
+
"learning_rate": 0.00010572838840358168,
|
| 1471 |
+
"loss": 1.1233,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.54,
|
| 1476 |
+
"grad_norm": 0.824127124532494,
|
| 1477 |
+
"learning_rate": 0.00010483409027418425,
|
| 1478 |
+
"loss": 1.1273,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.54,
|
| 1483 |
+
"grad_norm": 0.8666956799033239,
|
| 1484 |
+
"learning_rate": 0.00010393940444779635,
|
| 1485 |
+
"loss": 1.1172,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.54,
|
| 1490 |
+
"grad_norm": 0.7597618280586041,
|
| 1491 |
+
"learning_rate": 0.00010304440267876727,
|
| 1492 |
+
"loss": 1.2262,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.54,
|
| 1497 |
+
"grad_norm": 0.8594584043580179,
|
| 1498 |
+
"learning_rate": 0.00010214915674678523,
|
| 1499 |
+
"loss": 1.237,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.55,
|
| 1504 |
+
"grad_norm": 0.8503111582046956,
|
| 1505 |
+
"learning_rate": 0.00010125373845112034,
|
| 1506 |
+
"loss": 1.2224,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.55,
|
| 1511 |
+
"grad_norm": 0.8459974040134323,
|
| 1512 |
+
"learning_rate": 0.00010035821960486643,
|
| 1513 |
+
"loss": 1.1968,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.55,
|
| 1518 |
+
"grad_norm": 0.8236420083982484,
|
| 1519 |
+
"learning_rate": 9.946267202918157e-05,
|
| 1520 |
+
"loss": 1.1334,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.55,
|
| 1525 |
+
"grad_norm": 0.9248108629370668,
|
| 1526 |
+
"learning_rate": 9.856716754752796e-05,
|
| 1527 |
+
"loss": 1.1352,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.56,
|
| 1532 |
+
"grad_norm": 0.8808986018254903,
|
| 1533 |
+
"learning_rate": 9.767177797991155e-05,
|
| 1534 |
+
"loss": 1.1959,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.56,
|
| 1539 |
+
"grad_norm": 0.635409279473428,
|
| 1540 |
+
"learning_rate": 9.677657513712221e-05,
|
| 1541 |
+
"loss": 1.1506,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.56,
|
| 1546 |
+
"grad_norm": 0.7550156809913564,
|
| 1547 |
+
"learning_rate": 9.588163081497427e-05,
|
| 1548 |
+
"loss": 1.2308,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.56,
|
| 1553 |
+
"grad_norm": 0.7609369024863887,
|
| 1554 |
+
"learning_rate": 9.498701678854865e-05,
|
| 1555 |
+
"loss": 1.1722,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.57,
|
| 1560 |
+
"grad_norm": 0.8064136692370791,
|
| 1561 |
+
"learning_rate": 9.409280480643628e-05,
|
| 1562 |
+
"loss": 1.2279,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.57,
|
| 1567 |
+
"grad_norm": 0.7131735035318008,
|
| 1568 |
+
"learning_rate": 9.319906658498389e-05,
|
| 1569 |
+
"loss": 1.1862,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.57,
|
| 1574 |
+
"grad_norm": 0.8678942495261857,
|
| 1575 |
+
"learning_rate": 9.230587380254237e-05,
|
| 1576 |
+
"loss": 1.1531,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.57,
|
| 1581 |
+
"grad_norm": 0.6604664522870226,
|
| 1582 |
+
"learning_rate": 9.141329809371803e-05,
|
| 1583 |
+
"loss": 1.1744,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.58,
|
| 1588 |
+
"grad_norm": 0.7095675124068804,
|
| 1589 |
+
"learning_rate": 9.052141104362748e-05,
|
| 1590 |
+
"loss": 1.1311,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.58,
|
| 1595 |
+
"grad_norm": 0.7845862314567627,
|
| 1596 |
+
"learning_rate": 8.963028418215653e-05,
|
| 1597 |
+
"loss": 1.2801,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.58,
|
| 1602 |
+
"grad_norm": 0.8019276427700621,
|
| 1603 |
+
"learning_rate": 8.873998897822336e-05,
|
| 1604 |
+
"loss": 1.2425,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.58,
|
| 1609 |
+
"grad_norm": 1.0094066794890815,
|
| 1610 |
+
"learning_rate": 8.785059683404672e-05,
|
| 1611 |
+
"loss": 1.1287,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.59,
|
| 1616 |
+
"grad_norm": 0.9258556140522551,
|
| 1617 |
+
"learning_rate": 8.696217907941941e-05,
|
| 1618 |
+
"loss": 1.2059,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.59,
|
| 1623 |
+
"grad_norm": 0.9454158193155473,
|
| 1624 |
+
"learning_rate": 8.607480696598762e-05,
|
| 1625 |
+
"loss": 1.0897,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.59,
|
| 1630 |
+
"grad_norm": 0.8742232040464293,
|
| 1631 |
+
"learning_rate": 8.518855166153644e-05,
|
| 1632 |
+
"loss": 1.2234,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6,
|
| 1637 |
+
"grad_norm": 0.8764506057333485,
|
| 1638 |
+
"learning_rate": 8.43034842442822e-05,
|
| 1639 |
+
"loss": 1.1409,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6,
|
| 1644 |
+
"grad_norm": 0.9294531011673496,
|
| 1645 |
+
"learning_rate": 8.341967569717202e-05,
|
| 1646 |
+
"loss": 1.0986,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6,
|
| 1651 |
+
"grad_norm": 0.8413972427762672,
|
| 1652 |
+
"learning_rate": 8.253719690219079e-05,
|
| 1653 |
+
"loss": 1.1334,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6,
|
| 1658 |
+
"grad_norm": 0.9118769910143452,
|
| 1659 |
+
"learning_rate": 8.165611863467644e-05,
|
| 1660 |
+
"loss": 1.145,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.61,
|
| 1665 |
+
"grad_norm": 0.7880447725035846,
|
| 1666 |
+
"learning_rate": 8.077651155764387e-05,
|
| 1667 |
+
"loss": 1.1299,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.61,
|
| 1672 |
+
"grad_norm": 0.8382287781274999,
|
| 1673 |
+
"learning_rate": 7.98984462161175e-05,
|
| 1674 |
+
"loss": 1.2243,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.61,
|
| 1679 |
+
"grad_norm": 0.6697229577393602,
|
| 1680 |
+
"learning_rate": 7.902199303147363e-05,
|
| 1681 |
+
"loss": 1.2093,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.61,
|
| 1686 |
+
"grad_norm": 0.7808818157955248,
|
| 1687 |
+
"learning_rate": 7.814722229579264e-05,
|
| 1688 |
+
"loss": 1.1984,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.62,
|
| 1693 |
+
"grad_norm": 0.9135627361643314,
|
| 1694 |
+
"learning_rate": 7.727420416622144e-05,
|
| 1695 |
+
"loss": 1.2091,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.62,
|
| 1700 |
+
"grad_norm": 0.680528612137634,
|
| 1701 |
+
"learning_rate": 7.640300865934687e-05,
|
| 1702 |
+
"loss": 1.1034,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.62,
|
| 1707 |
+
"grad_norm": 0.812190629178745,
|
| 1708 |
+
"learning_rate": 7.553370564558032e-05,
|
| 1709 |
+
"loss": 1.164,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.62,
|
| 1714 |
+
"grad_norm": 0.6786859157828664,
|
| 1715 |
+
"learning_rate": 7.46663648435541e-05,
|
| 1716 |
+
"loss": 1.1711,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.63,
|
| 1721 |
+
"grad_norm": 1.0608937731498502,
|
| 1722 |
+
"learning_rate": 7.380105581452987e-05,
|
| 1723 |
+
"loss": 1.1321,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.63,
|
| 1728 |
+
"grad_norm": 0.8774620996995025,
|
| 1729 |
+
"learning_rate": 7.293784795681994e-05,
|
| 1730 |
+
"loss": 1.2073,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.63,
|
| 1735 |
+
"grad_norm": 0.8397584969858537,
|
| 1736 |
+
"learning_rate": 7.207681050022132e-05,
|
| 1737 |
+
"loss": 1.1356,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.63,
|
| 1742 |
+
"grad_norm": 0.7565443884452981,
|
| 1743 |
+
"learning_rate": 7.121801250046363e-05,
|
| 1744 |
+
"loss": 1.2799,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.64,
|
| 1749 |
+
"grad_norm": 0.6473890021488468,
|
| 1750 |
+
"learning_rate": 7.036152283367056e-05,
|
| 1751 |
+
"loss": 1.1355,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.64,
|
| 1756 |
+
"grad_norm": 0.8635483831775488,
|
| 1757 |
+
"learning_rate": 6.950741019083617e-05,
|
| 1758 |
+
"loss": 1.172,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.64,
|
| 1763 |
+
"grad_norm": 0.7635797502403529,
|
| 1764 |
+
"learning_rate": 6.865574307231575e-05,
|
| 1765 |
+
"loss": 1.1359,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.64,
|
| 1770 |
+
"grad_norm": 0.9134433746313595,
|
| 1771 |
+
"learning_rate": 6.780658978233199e-05,
|
| 1772 |
+
"loss": 1.1523,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.65,
|
| 1777 |
+
"grad_norm": 0.7862648398970986,
|
| 1778 |
+
"learning_rate": 6.696001842349702e-05,
|
| 1779 |
+
"loss": 1.1786,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.65,
|
| 1784 |
+
"grad_norm": 0.8737779184120746,
|
| 1785 |
+
"learning_rate": 6.611609689135056e-05,
|
| 1786 |
+
"loss": 1.2463,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.65,
|
| 1791 |
+
"grad_norm": 0.8231816299971835,
|
| 1792 |
+
"learning_rate": 6.527489286891459e-05,
|
| 1793 |
+
"loss": 1.0201,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.65,
|
| 1798 |
+
"grad_norm": 0.833558803199825,
|
| 1799 |
+
"learning_rate": 6.443647382126509e-05,
|
| 1800 |
+
"loss": 1.1555,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.66,
|
| 1805 |
+
"grad_norm": 0.7866154882582159,
|
| 1806 |
+
"learning_rate": 6.360090699012145e-05,
|
| 1807 |
+
"loss": 1.138,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.66,
|
| 1812 |
+
"grad_norm": 0.8309461290217522,
|
| 1813 |
+
"learning_rate": 6.27682593884535e-05,
|
| 1814 |
+
"loss": 1.0762,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.66,
|
| 1819 |
+
"grad_norm": 0.7270575882509663,
|
| 1820 |
+
"learning_rate": 6.193859779510712e-05,
|
| 1821 |
+
"loss": 1.2055,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.66,
|
| 1826 |
+
"grad_norm": 0.7267387499731184,
|
| 1827 |
+
"learning_rate": 6.111198874944845e-05,
|
| 1828 |
+
"loss": 1.1906,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.67,
|
| 1833 |
+
"grad_norm": 0.8081247294023348,
|
| 1834 |
+
"learning_rate": 6.0288498546027536e-05,
|
| 1835 |
+
"loss": 1.2243,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.67,
|
| 1840 |
+
"grad_norm": 0.8237165040439475,
|
| 1841 |
+
"learning_rate": 5.946819322926127e-05,
|
| 1842 |
+
"loss": 1.0994,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.67,
|
| 1847 |
+
"grad_norm": 0.860921836417512,
|
| 1848 |
+
"learning_rate": 5.865113858813673e-05,
|
| 1849 |
+
"loss": 1.1615,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.67,
|
| 1854 |
+
"grad_norm": 0.7995623045168802,
|
| 1855 |
+
"learning_rate": 5.783740015093484e-05,
|
| 1856 |
+
"loss": 1.1728,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.68,
|
| 1861 |
+
"grad_norm": 0.7326482406819708,
|
| 1862 |
+
"learning_rate": 5.702704317997492e-05,
|
| 1863 |
+
"loss": 1.112,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.68,
|
| 1868 |
+
"grad_norm": 0.8123434824427339,
|
| 1869 |
+
"learning_rate": 5.6220132666380635e-05,
|
| 1870 |
+
"loss": 1.0999,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.68,
|
| 1875 |
+
"grad_norm": 0.8440723192599834,
|
| 1876 |
+
"learning_rate": 5.541673332486773e-05,
|
| 1877 |
+
"loss": 1.2143,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.68,
|
| 1882 |
+
"grad_norm": 0.7328759482460477,
|
| 1883 |
+
"learning_rate": 5.4616909588553674e-05,
|
| 1884 |
+
"loss": 1.2724,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.69,
|
| 1889 |
+
"grad_norm": 0.6946255004506102,
|
| 1890 |
+
"learning_rate": 5.3820725603790346e-05,
|
| 1891 |
+
"loss": 1.2365,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.69,
|
| 1896 |
+
"grad_norm": 0.7864322469951978,
|
| 1897 |
+
"learning_rate": 5.30282452250193e-05,
|
| 1898 |
+
"loss": 1.2184,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.69,
|
| 1903 |
+
"grad_norm": 0.7860065077827799,
|
| 1904 |
+
"learning_rate": 5.223953200965055e-05,
|
| 1905 |
+
"loss": 1.177,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7,
|
| 1910 |
+
"grad_norm": 0.7870610754641466,
|
| 1911 |
+
"learning_rate": 5.145464921296537e-05,
|
| 1912 |
+
"loss": 1.2478,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7,
|
| 1917 |
+
"grad_norm": 0.8098199890971101,
|
| 1918 |
+
"learning_rate": 5.067365978304315e-05,
|
| 1919 |
+
"loss": 1.1132,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7,
|
| 1924 |
+
"grad_norm": 0.7277674932711228,
|
| 1925 |
+
"learning_rate": 4.9896626355712805e-05,
|
| 1926 |
+
"loss": 1.1065,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7,
|
| 1931 |
+
"grad_norm": 0.6702441184475941,
|
| 1932 |
+
"learning_rate": 4.912361124952948e-05,
|
| 1933 |
+
"loss": 1.1489,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.71,
|
| 1938 |
+
"grad_norm": 0.7382556968521137,
|
| 1939 |
+
"learning_rate": 4.835467646077656e-05,
|
| 1940 |
+
"loss": 1.1145,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.71,
|
| 1945 |
+
"grad_norm": 0.7012581002313999,
|
| 1946 |
+
"learning_rate": 4.7589883658493296e-05,
|
| 1947 |
+
"loss": 1.1196,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.71,
|
| 1952 |
+
"grad_norm": 0.8448006051293048,
|
| 1953 |
+
"learning_rate": 4.682929417952939e-05,
|
| 1954 |
+
"loss": 1.1081,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.71,
|
| 1959 |
+
"grad_norm": 0.6907481068653331,
|
| 1960 |
+
"learning_rate": 4.6072969023625165e-05,
|
| 1961 |
+
"loss": 1.1438,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.72,
|
| 1966 |
+
"grad_norm": 0.7998262194855656,
|
| 1967 |
+
"learning_rate": 4.532096884851978e-05,
|
| 1968 |
+
"loss": 1.1514,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.72,
|
| 1973 |
+
"grad_norm": 0.9014262005578314,
|
| 1974 |
+
"learning_rate": 4.457335396508631e-05,
|
| 1975 |
+
"loss": 1.1583,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.72,
|
| 1980 |
+
"grad_norm": 0.8456699239379848,
|
| 1981 |
+
"learning_rate": 4.383018433249464e-05,
|
| 1982 |
+
"loss": 1.1471,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.72,
|
| 1987 |
+
"grad_norm": 0.7936192686618659,
|
| 1988 |
+
"learning_rate": 4.309151955340297e-05,
|
| 1989 |
+
"loss": 1.1212,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.73,
|
| 1994 |
+
"grad_norm": 0.8376082121769935,
|
| 1995 |
+
"learning_rate": 4.2357418869177354e-05,
|
| 1996 |
+
"loss": 1.2077,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.73,
|
| 2001 |
+
"grad_norm": 0.8095128041860435,
|
| 2002 |
+
"learning_rate": 4.162794115514078e-05,
|
| 2003 |
+
"loss": 1.2169,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.73,
|
| 2008 |
+
"grad_norm": 0.7509614581437917,
|
| 2009 |
+
"learning_rate": 4.0903144915851174e-05,
|
| 2010 |
+
"loss": 1.0553,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.73,
|
| 2015 |
+
"grad_norm": 0.7839995425652196,
|
| 2016 |
+
"learning_rate": 4.018308828040924e-05,
|
| 2017 |
+
"loss": 1.0728,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.74,
|
| 2022 |
+
"grad_norm": 0.706531516175813,
|
| 2023 |
+
"learning_rate": 3.946782899779667e-05,
|
| 2024 |
+
"loss": 1.1324,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.74,
|
| 2029 |
+
"grad_norm": 0.8815003834638072,
|
| 2030 |
+
"learning_rate": 3.875742443224451e-05,
|
| 2031 |
+
"loss": 1.1515,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.74,
|
| 2036 |
+
"grad_norm": 0.6975354892323637,
|
| 2037 |
+
"learning_rate": 3.805193155863247e-05,
|
| 2038 |
+
"loss": 1.0919,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.74,
|
| 2043 |
+
"grad_norm": 0.7405117357753763,
|
| 2044 |
+
"learning_rate": 3.7351406957919636e-05,
|
| 2045 |
+
"loss": 1.1764,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.75,
|
| 2050 |
+
"grad_norm": 0.7778411867278201,
|
| 2051 |
+
"learning_rate": 3.665590681260658e-05,
|
| 2052 |
+
"loss": 1.1504,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.75,
|
| 2057 |
+
"grad_norm": 0.8723639773246538,
|
| 2058 |
+
"learning_rate": 3.59654869022294e-05,
|
| 2059 |
+
"loss": 1.2015,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.75,
|
| 2064 |
+
"grad_norm": 0.7314301031903592,
|
| 2065 |
+
"learning_rate": 3.5280202598886324e-05,
|
| 2066 |
+
"loss": 1.1912,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.75,
|
| 2071 |
+
"grad_norm": 0.78382821788559,
|
| 2072 |
+
"learning_rate": 3.4600108862796796e-05,
|
| 2073 |
+
"loss": 1.1531,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.76,
|
| 2078 |
+
"grad_norm": 0.746538778895017,
|
| 2079 |
+
"learning_rate": 3.392526023789349e-05,
|
| 2080 |
+
"loss": 1.1861,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.76,
|
| 2085 |
+
"grad_norm": 0.7335920604121904,
|
| 2086 |
+
"learning_rate": 3.325571084744803e-05,
|
| 2087 |
+
"loss": 1.1032,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.76,
|
| 2092 |
+
"grad_norm": 0.8181126090151531,
|
| 2093 |
+
"learning_rate": 3.259151438973024e-05,
|
| 2094 |
+
"loss": 1.2354,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.76,
|
| 2099 |
+
"grad_norm": 0.8813117253151651,
|
| 2100 |
+
"learning_rate": 3.1932724133701344e-05,
|
| 2101 |
+
"loss": 1.0998,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.77,
|
| 2106 |
+
"grad_norm": 0.7703731279590557,
|
| 2107 |
+
"learning_rate": 3.1279392914742046e-05,
|
| 2108 |
+
"loss": 1.1732,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.77,
|
| 2113 |
+
"grad_norm": 0.8022002883485505,
|
| 2114 |
+
"learning_rate": 3.06315731304148e-05,
|
| 2115 |
+
"loss": 1.2005,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.77,
|
| 2120 |
+
"grad_norm": 0.7718995592442958,
|
| 2121 |
+
"learning_rate": 2.998931673626175e-05,
|
| 2122 |
+
"loss": 1.0812,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.77,
|
| 2127 |
+
"grad_norm": 0.7753353561939175,
|
| 2128 |
+
"learning_rate": 2.935267524163774e-05,
|
| 2129 |
+
"loss": 1.2016,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.78,
|
| 2134 |
+
"grad_norm": 0.8453046609631811,
|
| 2135 |
+
"learning_rate": 2.872169970557913e-05,
|
| 2136 |
+
"loss": 1.2015,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.78,
|
| 2141 |
+
"grad_norm": 0.788689710824734,
|
| 2142 |
+
"learning_rate": 2.8096440732709083e-05,
|
| 2143 |
+
"loss": 1.2301,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.78,
|
| 2148 |
+
"grad_norm": 0.9498609267061067,
|
| 2149 |
+
"learning_rate": 2.7476948469178887e-05,
|
| 2150 |
+
"loss": 1.1721,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.79,
|
| 2155 |
+
"grad_norm": 1.2548603296587735,
|
| 2156 |
+
"learning_rate": 2.6863272598646106e-05,
|
| 2157 |
+
"loss": 1.1309,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.79,
|
| 2162 |
+
"grad_norm": 0.8088505633219847,
|
| 2163 |
+
"learning_rate": 2.625546233829016e-05,
|
| 2164 |
+
"loss": 1.1924,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.79,
|
| 2169 |
+
"grad_norm": 0.8053156607642844,
|
| 2170 |
+
"learning_rate": 2.5653566434864928e-05,
|
| 2171 |
+
"loss": 1.1771,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.79,
|
| 2176 |
+
"grad_norm": 1.0076517594688355,
|
| 2177 |
+
"learning_rate": 2.5057633160789184e-05,
|
| 2178 |
+
"loss": 1.1547,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8,
|
| 2183 |
+
"grad_norm": 0.7807995790198267,
|
| 2184 |
+
"learning_rate": 2.446771031027527e-05,
|
| 2185 |
+
"loss": 1.2168,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8,
|
| 2190 |
+
"grad_norm": 0.7811855005276076,
|
| 2191 |
+
"learning_rate": 2.3883845195495878e-05,
|
| 2192 |
+
"loss": 1.2407,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8,
|
| 2197 |
+
"grad_norm": 0.7607446953536656,
|
| 2198 |
+
"learning_rate": 2.330608464278953e-05,
|
| 2199 |
+
"loss": 1.1777,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8,
|
| 2204 |
+
"grad_norm": 0.8030643238657801,
|
| 2205 |
+
"learning_rate": 2.273447498890521e-05,
|
| 2206 |
+
"loss": 1.2321,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.81,
|
| 2211 |
+
"grad_norm": 0.707300837474353,
|
| 2212 |
+
"learning_rate": 2.2169062077286075e-05,
|
| 2213 |
+
"loss": 1.103,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.81,
|
| 2218 |
+
"grad_norm": 0.7398162416161335,
|
| 2219 |
+
"learning_rate": 2.1609891254392678e-05,
|
| 2220 |
+
"loss": 1.082,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.81,
|
| 2225 |
+
"grad_norm": 0.8829809189432961,
|
| 2226 |
+
"learning_rate": 2.1057007366066373e-05,
|
| 2227 |
+
"loss": 1.1863,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.81,
|
| 2232 |
+
"grad_norm": 0.9036595461566347,
|
| 2233 |
+
"learning_rate": 2.0510454753932395e-05,
|
| 2234 |
+
"loss": 1.1125,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.82,
|
| 2239 |
+
"grad_norm": 0.6793566054364535,
|
| 2240 |
+
"learning_rate": 1.9970277251843862e-05,
|
| 2241 |
+
"loss": 1.1453,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.82,
|
| 2246 |
+
"grad_norm": 0.7465371914976613,
|
| 2247 |
+
"learning_rate": 1.9436518182366158e-05,
|
| 2248 |
+
"loss": 1.1914,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.82,
|
| 2253 |
+
"grad_norm": 0.6678427111685153,
|
| 2254 |
+
"learning_rate": 1.8909220353302392e-05,
|
| 2255 |
+
"loss": 1.1865,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.82,
|
| 2260 |
+
"grad_norm": 0.8490303048462068,
|
| 2261 |
+
"learning_rate": 1.838842605426031e-05,
|
| 2262 |
+
"loss": 1.0754,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.83,
|
| 2267 |
+
"grad_norm": 0.6832719345179151,
|
| 2268 |
+
"learning_rate": 1.7874177053260598e-05,
|
| 2269 |
+
"loss": 1.0879,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.83,
|
| 2274 |
+
"grad_norm": 0.9649690559163364,
|
| 2275 |
+
"learning_rate": 1.736651459338695e-05,
|
| 2276 |
+
"loss": 1.0977,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.83,
|
| 2281 |
+
"grad_norm": 0.6985183059667567,
|
| 2282 |
+
"learning_rate": 1.6865479389478545e-05,
|
| 2283 |
+
"loss": 1.1476,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.83,
|
| 2288 |
+
"grad_norm": 0.6282090702067379,
|
| 2289 |
+
"learning_rate": 1.6371111624864543e-05,
|
| 2290 |
+
"loss": 1.1299,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.84,
|
| 2295 |
+
"grad_norm": 0.8512442207678547,
|
| 2296 |
+
"learning_rate": 1.5883450948141377e-05,
|
| 2297 |
+
"loss": 1.1309,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.84,
|
| 2302 |
+
"grad_norm": 0.7151617030884643,
|
| 2303 |
+
"learning_rate": 1.540253646999299e-05,
|
| 2304 |
+
"loss": 1.1936,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.84,
|
| 2309 |
+
"grad_norm": 0.7567039565297413,
|
| 2310 |
+
"learning_rate": 1.4928406760054059e-05,
|
| 2311 |
+
"loss": 1.1441,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.84,
|
| 2316 |
+
"grad_norm": 0.8464089358888328,
|
| 2317 |
+
"learning_rate": 1.4461099843816684e-05,
|
| 2318 |
+
"loss": 1.145,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.85,
|
| 2323 |
+
"grad_norm": 0.7170403140830546,
|
| 2324 |
+
"learning_rate": 1.4000653199580782e-05,
|
| 2325 |
+
"loss": 1.1887,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.85,
|
| 2330 |
+
"grad_norm": 0.7735313453026896,
|
| 2331 |
+
"learning_rate": 1.3547103755448287e-05,
|
| 2332 |
+
"loss": 1.1786,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.85,
|
| 2337 |
+
"grad_norm": 0.7081409924001252,
|
| 2338 |
+
"learning_rate": 1.3100487886361379e-05,
|
| 2339 |
+
"loss": 1.2186,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.85,
|
| 2344 |
+
"grad_norm": 0.6796669559142102,
|
| 2345 |
+
"learning_rate": 1.266084141118542e-05,
|
| 2346 |
+
"loss": 1.1473,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.86,
|
| 2351 |
+
"grad_norm": 0.7301679940106633,
|
| 2352 |
+
"learning_rate": 1.2228199589835999e-05,
|
| 2353 |
+
"loss": 1.1798,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.86,
|
| 2358 |
+
"grad_norm": 0.838436487459379,
|
| 2359 |
+
"learning_rate": 1.1802597120451286e-05,
|
| 2360 |
+
"loss": 1.1464,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.86,
|
| 2365 |
+
"grad_norm": 0.8915729376099731,
|
| 2366 |
+
"learning_rate": 1.1384068136609105e-05,
|
| 2367 |
+
"loss": 1.1965,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.86,
|
| 2372 |
+
"grad_norm": 0.6967288554874705,
|
| 2373 |
+
"learning_rate": 1.0972646204589377e-05,
|
| 2374 |
+
"loss": 1.1211,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.87,
|
| 2379 |
+
"grad_norm": 0.8038608019365008,
|
| 2380 |
+
"learning_rate": 1.0568364320682178e-05,
|
| 2381 |
+
"loss": 1.1961,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.87,
|
| 2386 |
+
"grad_norm": 0.8793813680766012,
|
| 2387 |
+
"learning_rate": 1.0171254908541372e-05,
|
| 2388 |
+
"loss": 1.0912,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.87,
|
| 2393 |
+
"grad_norm": 0.7353307810953187,
|
| 2394 |
+
"learning_rate": 9.781349816584162e-06,
|
| 2395 |
+
"loss": 1.1382,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.87,
|
| 2400 |
+
"grad_norm": 0.7264376523774766,
|
| 2401 |
+
"learning_rate": 9.398680315436903e-06,
|
| 2402 |
+
"loss": 1.1503,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.88,
|
| 2407 |
+
"grad_norm": 0.8294378899418743,
|
| 2408 |
+
"learning_rate": 9.023277095427173e-06,
|
| 2409 |
+
"loss": 1.1001,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.88,
|
| 2414 |
+
"grad_norm": 0.6847190426952564,
|
| 2415 |
+
"learning_rate": 8.655170264122303e-06,
|
| 2416 |
+
"loss": 1.1837,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.88,
|
| 2421 |
+
"grad_norm": 0.7123063489929077,
|
| 2422 |
+
"learning_rate": 8.294389343914899e-06,
|
| 2423 |
+
"loss": 1.1665,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.89,
|
| 2428 |
+
"grad_norm": 0.7161490463804174,
|
| 2429 |
+
"learning_rate": 7.940963269654922e-06,
|
| 2430 |
+
"loss": 1.2128,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.89,
|
| 2435 |
+
"grad_norm": 0.6942289082099549,
|
| 2436 |
+
"learning_rate": 7.594920386329252e-06,
|
| 2437 |
+
"loss": 1.1516,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.89,
|
| 2442 |
+
"grad_norm": 0.7000185077826192,
|
| 2443 |
+
"learning_rate": 7.256288446788362e-06,
|
| 2444 |
+
"loss": 1.0752,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.89,
|
| 2449 |
+
"grad_norm": 0.8141301323800174,
|
| 2450 |
+
"learning_rate": 6.925094609520455e-06,
|
| 2451 |
+
"loss": 1.1663,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9,
|
| 2456 |
+
"grad_norm": 0.7801337376407085,
|
| 2457 |
+
"learning_rate": 6.601365436473439e-06,
|
| 2458 |
+
"loss": 1.1264,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9,
|
| 2463 |
+
"grad_norm": 0.826763084695081,
|
| 2464 |
+
"learning_rate": 6.2851268909245865e-06,
|
| 2465 |
+
"loss": 1.1076,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9,
|
| 2470 |
+
"grad_norm": 0.7778658882619575,
|
| 2471 |
+
"learning_rate": 5.976404335398256e-06,
|
| 2472 |
+
"loss": 1.185,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.9,
|
| 2477 |
+
"grad_norm": 0.5899787648756437,
|
| 2478 |
+
"learning_rate": 5.675222529631841e-06,
|
| 2479 |
+
"loss": 1.1383,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.91,
|
| 2484 |
+
"grad_norm": 0.7296161663517282,
|
| 2485 |
+
"learning_rate": 5.381605628590003e-06,
|
| 2486 |
+
"loss": 1.1406,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.91,
|
| 2491 |
+
"grad_norm": 0.7308791705621759,
|
| 2492 |
+
"learning_rate": 5.095577180527378e-06,
|
| 2493 |
+
"loss": 1.1641,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.91,
|
| 2498 |
+
"grad_norm": 0.7165596394167716,
|
| 2499 |
+
"learning_rate": 4.817160125100106e-06,
|
| 2500 |
+
"loss": 1.23,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.91,
|
| 2505 |
+
"grad_norm": 0.8126491682868108,
|
| 2506 |
+
"learning_rate": 4.546376791525975e-06,
|
| 2507 |
+
"loss": 1.1496,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.92,
|
| 2512 |
+
"grad_norm": 0.6916149503964037,
|
| 2513 |
+
"learning_rate": 4.2832488967935795e-06,
|
| 2514 |
+
"loss": 1.1613,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.92,
|
| 2519 |
+
"grad_norm": 0.6713280206828187,
|
| 2520 |
+
"learning_rate": 4.02779754392072e-06,
|
| 2521 |
+
"loss": 1.1778,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.92,
|
| 2526 |
+
"grad_norm": 0.8687927248580933,
|
| 2527 |
+
"learning_rate": 3.780043220261764e-06,
|
| 2528 |
+
"loss": 1.1947,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.92,
|
| 2533 |
+
"grad_norm": 0.9156585367622113,
|
| 2534 |
+
"learning_rate": 3.540005795864709e-06,
|
| 2535 |
+
"loss": 1.0788,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.93,
|
| 2540 |
+
"grad_norm": 0.8066844554106848,
|
| 2541 |
+
"learning_rate": 3.3077045218775192e-06,
|
| 2542 |
+
"loss": 1.2058,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.93,
|
| 2547 |
+
"grad_norm": 0.8260344643458865,
|
| 2548 |
+
"learning_rate": 3.0831580290041184e-06,
|
| 2549 |
+
"loss": 1.1859,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.93,
|
| 2554 |
+
"grad_norm": 0.6728735284202596,
|
| 2555 |
+
"learning_rate": 2.8663843260103074e-06,
|
| 2556 |
+
"loss": 1.1548,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.93,
|
| 2561 |
+
"grad_norm": 0.8464248337059433,
|
| 2562 |
+
"learning_rate": 2.6574007982793857e-06,
|
| 2563 |
+
"loss": 1.1424,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.94,
|
| 2568 |
+
"grad_norm": 0.7582737017596137,
|
| 2569 |
+
"learning_rate": 2.456224206417812e-06,
|
| 2570 |
+
"loss": 1.1132,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.94,
|
| 2575 |
+
"grad_norm": 0.7570508766628901,
|
| 2576 |
+
"learning_rate": 2.262870684911045e-06,
|
| 2577 |
+
"loss": 1.1916,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.94,
|
| 2582 |
+
"grad_norm": 0.6948287618722808,
|
| 2583 |
+
"learning_rate": 2.0773557408295343e-06,
|
| 2584 |
+
"loss": 1.1384,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.94,
|
| 2589 |
+
"grad_norm": 0.7078737172323022,
|
| 2590 |
+
"learning_rate": 1.8996942525850047e-06,
|
| 2591 |
+
"loss": 1.1501,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.95,
|
| 2596 |
+
"grad_norm": 0.767530275514686,
|
| 2597 |
+
"learning_rate": 1.7299004687372665e-06,
|
| 2598 |
+
"loss": 1.1267,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.95,
|
| 2603 |
+
"grad_norm": 0.7858189878444168,
|
| 2604 |
+
"learning_rate": 1.5679880068514174e-06,
|
| 2605 |
+
"loss": 1.1056,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.95,
|
| 2610 |
+
"grad_norm": 0.8212356253611762,
|
| 2611 |
+
"learning_rate": 1.4139698524057165e-06,
|
| 2612 |
+
"loss": 1.1939,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.95,
|
| 2617 |
+
"grad_norm": 0.7961839613413594,
|
| 2618 |
+
"learning_rate": 1.2678583577501624e-06,
|
| 2619 |
+
"loss": 1.1662,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.96,
|
| 2624 |
+
"grad_norm": 0.850673264738278,
|
| 2625 |
+
"learning_rate": 1.1296652411158182e-06,
|
| 2626 |
+
"loss": 1.17,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.96,
|
| 2631 |
+
"grad_norm": 0.7063089618156579,
|
| 2632 |
+
"learning_rate": 9.994015856749527e-07,
|
| 2633 |
+
"loss": 1.1253,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.96,
|
| 2638 |
+
"grad_norm": 0.7573170758624328,
|
| 2639 |
+
"learning_rate": 8.770778386522627e-07,
|
| 2640 |
+
"loss": 1.0771,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.96,
|
| 2645 |
+
"grad_norm": 0.8837077908686763,
|
| 2646 |
+
"learning_rate": 7.627038104869199e-07,
|
| 2647 |
+
"loss": 1.1685,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.97,
|
| 2652 |
+
"grad_norm": 0.7524180298657765,
|
| 2653 |
+
"learning_rate": 6.562886740457797e-07,
|
| 2654 |
+
"loss": 1.1406,
|
| 2655 |
+
"step": 1885
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.97,
|
| 2659 |
+
"grad_norm": 0.7266831937495886,
|
| 2660 |
+
"learning_rate": 5.578409638877457e-07,
|
| 2661 |
+
"loss": 1.0525,
|
| 2662 |
+
"step": 1890
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.97,
|
| 2666 |
+
"grad_norm": 0.8486209379295809,
|
| 2667 |
+
"learning_rate": 4.6736857557925227e-07,
|
| 2668 |
+
"loss": 1.1348,
|
| 2669 |
+
"step": 1895
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.97,
|
| 2673 |
+
"grad_norm": 0.7660696151853608,
|
| 2674 |
+
"learning_rate": 3.8487876506106966e-07,
|
| 2675 |
+
"loss": 1.2252,
|
| 2676 |
+
"step": 1900
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.98,
|
| 2680 |
+
"grad_norm": 0.7491200820319253,
|
| 2681 |
+
"learning_rate": 3.1037814806634815e-07,
|
| 2682 |
+
"loss": 1.1097,
|
| 2683 |
+
"step": 1905
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.98,
|
| 2687 |
+
"grad_norm": 0.7181576551220783,
|
| 2688 |
+
"learning_rate": 2.43872699590042e-07,
|
| 2689 |
+
"loss": 1.0703,
|
| 2690 |
+
"step": 1910
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.98,
|
| 2694 |
+
"grad_norm": 0.7330097823119024,
|
| 2695 |
+
"learning_rate": 1.8536775340970425e-07,
|
| 2696 |
+
"loss": 1.1908,
|
| 2697 |
+
"step": 1915
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.99,
|
| 2701 |
+
"grad_norm": 0.8782265197206924,
|
| 2702 |
+
"learning_rate": 1.348680016577397e-07,
|
| 2703 |
+
"loss": 1.1906,
|
| 2704 |
+
"step": 1920
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.99,
|
| 2708 |
+
"grad_norm": 0.7862482683388468,
|
| 2709 |
+
"learning_rate": 9.237749444505062e-08,
|
| 2710 |
+
"loss": 1.1362,
|
| 2711 |
+
"step": 1925
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.99,
|
| 2715 |
+
"grad_norm": 0.7669692568908726,
|
| 2716 |
+
"learning_rate": 5.7899639536251883e-08,
|
| 2717 |
+
"loss": 1.1689,
|
| 2718 |
+
"step": 1930
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.99,
|
| 2722 |
+
"grad_norm": 0.8047883869461449,
|
| 2723 |
+
"learning_rate": 3.143720207635648e-08,
|
| 2724 |
+
"loss": 1.1729,
|
| 2725 |
+
"step": 1935
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 1.0,
|
| 2729 |
+
"grad_norm": 0.9030824345194336,
|
| 2730 |
+
"learning_rate": 1.299230436898613e-08,
|
| 2731 |
+
"loss": 1.196,
|
| 2732 |
+
"step": 1940
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 1.0,
|
| 2736 |
+
"grad_norm": 0.7879987601052318,
|
| 2737 |
+
"learning_rate": 2.566425706218567e-09,
|
| 2738 |
+
"loss": 1.0957,
|
| 2739 |
+
"step": 1945
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 1.0,
|
| 2743 |
+
"eval_loss": NaN,
|
| 2744 |
+
"eval_runtime": 1989.8167,
|
| 2745 |
+
"eval_samples_per_second": 3.484,
|
| 2746 |
+
"eval_steps_per_second": 0.871,
|
| 2747 |
+
"step": 1949
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 1.0,
|
| 2751 |
+
"step": 1949,
|
| 2752 |
+
"total_flos": 1.2135540828143616e+16,
|
| 2753 |
+
"train_loss": 1.2706995834871218,
|
| 2754 |
+
"train_runtime": 18075.5808,
|
| 2755 |
+
"train_samples_per_second": 3.45,
|
| 2756 |
+
"train_steps_per_second": 0.108
|
| 2757 |
+
}
|
| 2758 |
+
],
|
| 2759 |
+
"logging_steps": 5,
|
| 2760 |
+
"max_steps": 1949,
|
| 2761 |
+
"num_input_tokens_seen": 0,
|
| 2762 |
+
"num_train_epochs": 1,
|
| 2763 |
+
"save_steps": 100,
|
| 2764 |
+
"total_flos": 1.2135540828143616e+16,
|
| 2765 |
+
"train_batch_size": 8,
|
| 2766 |
+
"trial_name": null,
|
| 2767 |
+
"trial_params": null
|
| 2768 |
+
}
|