Qin Liu
commited on
Commit
•
cc697cb
1
Parent(s):
bf0e499
Model save
Browse files- README.md +63 -0
- adapter_model.safetensors +1 -1
- all_results.json +8 -0
- runs/May03_07-57-10_COE-CS-sv003/events.out.tfevents.1714723050.COE-CS-sv003.651347.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +1878 -0
README.md
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: llama3-meta_material
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# llama3-meta_material
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
|
| 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: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- total_train_batch_size: 4
|
| 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 |
+
|
| 56 |
+
|
| 57 |
+
### Framework versions
|
| 58 |
+
|
| 59 |
+
- PEFT 0.7.1
|
| 60 |
+
- Transformers 4.39.0.dev0
|
| 61 |
+
- Pytorch 2.1.2
|
| 62 |
+
- Datasets 2.14.6
|
| 63 |
+
- 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 31516744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a24132ba69fa809e76e74a9119420f512fd1e208ebe370aaca68602aa1139a91
|
| 3 |
size 31516744
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 1.7167340969840534,
|
| 4 |
+
"train_runtime": 8001.3069,
|
| 5 |
+
"train_samples": 2240,
|
| 6 |
+
"train_samples_per_second": 0.659,
|
| 7 |
+
"train_steps_per_second": 0.165
|
| 8 |
+
}
|
runs/May03_07-57-10_COE-CS-sv003/events.out.tfevents.1714723050.COE-CS-sv003.651347.0
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:b68da6d5ceacea89442a15130232bc901bee213cd17919b1bb3de9922f6285a1
|
| 3 |
+
size 60686
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 1.7167340969840534,
|
| 4 |
+
"train_runtime": 8001.3069,
|
| 5 |
+
"train_samples": 2240,
|
| 6 |
+
"train_samples_per_second": 0.659,
|
| 7 |
+
"train_steps_per_second": 0.165
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1878 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1319,
|
| 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": 0.08992228630318096,
|
| 14 |
+
"learning_rate": 1.5151515151515152e-06,
|
| 15 |
+
"loss": 1.9018,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0,
|
| 20 |
+
"grad_norm": 0.0740828375791177,
|
| 21 |
+
"learning_rate": 7.5757575757575764e-06,
|
| 22 |
+
"loss": 1.764,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01,
|
| 27 |
+
"grad_norm": 0.07748708824774467,
|
| 28 |
+
"learning_rate": 1.5151515151515153e-05,
|
| 29 |
+
"loss": 1.6755,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01,
|
| 34 |
+
"grad_norm": 0.0972745675012954,
|
| 35 |
+
"learning_rate": 2.272727272727273e-05,
|
| 36 |
+
"loss": 1.7941,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02,
|
| 41 |
+
"grad_norm": 0.08859056101473906,
|
| 42 |
+
"learning_rate": 3.0303030303030306e-05,
|
| 43 |
+
"loss": 1.9252,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"grad_norm": 0.10873228272662556,
|
| 49 |
+
"learning_rate": 3.787878787878788e-05,
|
| 50 |
+
"loss": 1.678,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02,
|
| 55 |
+
"grad_norm": 0.097714070547121,
|
| 56 |
+
"learning_rate": 4.545454545454546e-05,
|
| 57 |
+
"loss": 1.601,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03,
|
| 62 |
+
"grad_norm": 0.12508554024661148,
|
| 63 |
+
"learning_rate": 5.303030303030303e-05,
|
| 64 |
+
"loss": 1.8,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03,
|
| 69 |
+
"grad_norm": 0.12109319585406818,
|
| 70 |
+
"learning_rate": 6.060606060606061e-05,
|
| 71 |
+
"loss": 1.8099,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03,
|
| 76 |
+
"grad_norm": 0.1640439558840102,
|
| 77 |
+
"learning_rate": 6.818181818181818e-05,
|
| 78 |
+
"loss": 1.9468,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04,
|
| 83 |
+
"grad_norm": 0.14480636511042952,
|
| 84 |
+
"learning_rate": 7.575757575757576e-05,
|
| 85 |
+
"loss": 1.8372,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"grad_norm": 0.14206421045069687,
|
| 91 |
+
"learning_rate": 8.333333333333334e-05,
|
| 92 |
+
"loss": 1.7241,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.05,
|
| 97 |
+
"grad_norm": 0.2311330757678567,
|
| 98 |
+
"learning_rate": 9.090909090909092e-05,
|
| 99 |
+
"loss": 1.8208,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05,
|
| 104 |
+
"grad_norm": 0.189110382740428,
|
| 105 |
+
"learning_rate": 9.848484848484849e-05,
|
| 106 |
+
"loss": 1.707,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05,
|
| 111 |
+
"grad_norm": 0.13055837108201243,
|
| 112 |
+
"learning_rate": 0.00010606060606060606,
|
| 113 |
+
"loss": 1.6787,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06,
|
| 118 |
+
"grad_norm": 0.18520751971810132,
|
| 119 |
+
"learning_rate": 0.00011363636363636365,
|
| 120 |
+
"loss": 1.7811,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06,
|
| 125 |
+
"grad_norm": 0.15541159241507693,
|
| 126 |
+
"learning_rate": 0.00012121212121212122,
|
| 127 |
+
"loss": 1.695,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"grad_norm": 0.19125540723930942,
|
| 133 |
+
"learning_rate": 0.00012878787878787878,
|
| 134 |
+
"loss": 1.7545,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07,
|
| 139 |
+
"grad_norm": 0.18097293155941335,
|
| 140 |
+
"learning_rate": 0.00013636363636363637,
|
| 141 |
+
"loss": 1.7739,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07,
|
| 146 |
+
"grad_norm": 0.19933236705063037,
|
| 147 |
+
"learning_rate": 0.00014393939393939396,
|
| 148 |
+
"loss": 1.7473,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08,
|
| 153 |
+
"grad_norm": 0.17704145093132606,
|
| 154 |
+
"learning_rate": 0.00015151515151515152,
|
| 155 |
+
"loss": 1.8366,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08,
|
| 160 |
+
"grad_norm": 0.15828713632727448,
|
| 161 |
+
"learning_rate": 0.0001590909090909091,
|
| 162 |
+
"loss": 1.8373,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08,
|
| 167 |
+
"grad_norm": 0.14819493671841716,
|
| 168 |
+
"learning_rate": 0.0001666666666666667,
|
| 169 |
+
"loss": 1.7091,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09,
|
| 174 |
+
"grad_norm": 0.1622614877169443,
|
| 175 |
+
"learning_rate": 0.00017424242424242425,
|
| 176 |
+
"loss": 1.7943,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09,
|
| 181 |
+
"grad_norm": 0.19948476709892068,
|
| 182 |
+
"learning_rate": 0.00018181818181818183,
|
| 183 |
+
"loss": 1.8075,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09,
|
| 188 |
+
"grad_norm": 0.1691916901921161,
|
| 189 |
+
"learning_rate": 0.00018939393939393942,
|
| 190 |
+
"loss": 1.7856,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1,
|
| 195 |
+
"grad_norm": 0.15241428197032597,
|
| 196 |
+
"learning_rate": 0.00019696969696969698,
|
| 197 |
+
"loss": 1.8374,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1,
|
| 202 |
+
"grad_norm": 0.19209359192695652,
|
| 203 |
+
"learning_rate": 0.00019999684783792443,
|
| 204 |
+
"loss": 1.7143,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11,
|
| 209 |
+
"grad_norm": 0.16191115612389606,
|
| 210 |
+
"learning_rate": 0.00019997758534488915,
|
| 211 |
+
"loss": 1.822,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11,
|
| 216 |
+
"grad_norm": 0.1584155775536348,
|
| 217 |
+
"learning_rate": 0.00019994081492910124,
|
| 218 |
+
"loss": 1.6943,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11,
|
| 223 |
+
"grad_norm": 0.1614369821018255,
|
| 224 |
+
"learning_rate": 0.00019988654302974,
|
| 225 |
+
"loss": 1.7322,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12,
|
| 230 |
+
"grad_norm": 0.15011399662325187,
|
| 231 |
+
"learning_rate": 0.00019981477915081793,
|
| 232 |
+
"loss": 1.7347,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12,
|
| 237 |
+
"grad_norm": 0.14529246406802704,
|
| 238 |
+
"learning_rate": 0.0001997255358595164,
|
| 239 |
+
"loss": 1.7458,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13,
|
| 244 |
+
"grad_norm": 0.17972000220179096,
|
| 245 |
+
"learning_rate": 0.00019961882878398492,
|
| 246 |
+
"loss": 1.7652,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.13,
|
| 251 |
+
"grad_norm": 0.1383192615362475,
|
| 252 |
+
"learning_rate": 0.00019949467661060433,
|
| 253 |
+
"loss": 1.8346,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13,
|
| 258 |
+
"grad_norm": 0.1487696428331422,
|
| 259 |
+
"learning_rate": 0.00019935310108071453,
|
| 260 |
+
"loss": 1.6247,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14,
|
| 265 |
+
"grad_norm": 0.14864871631851187,
|
| 266 |
+
"learning_rate": 0.000199194126986807,
|
| 267 |
+
"loss": 1.7883,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14,
|
| 272 |
+
"grad_norm": 0.15094444221670217,
|
| 273 |
+
"learning_rate": 0.00019901778216818345,
|
| 274 |
+
"loss": 1.912,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14,
|
| 279 |
+
"grad_norm": 0.16630571990033322,
|
| 280 |
+
"learning_rate": 0.0001988240975060804,
|
| 281 |
+
"loss": 1.6714,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.15,
|
| 286 |
+
"grad_norm": 0.18330301650534517,
|
| 287 |
+
"learning_rate": 0.00019861310691826143,
|
| 288 |
+
"loss": 1.7698,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.15,
|
| 293 |
+
"grad_norm": 0.14395798059204837,
|
| 294 |
+
"learning_rate": 0.00019838484735307748,
|
| 295 |
+
"loss": 1.6334,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.16,
|
| 300 |
+
"grad_norm": 0.17075494920598505,
|
| 301 |
+
"learning_rate": 0.00019813935878299662,
|
| 302 |
+
"loss": 1.8058,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.16,
|
| 307 |
+
"grad_norm": 0.1517385663302395,
|
| 308 |
+
"learning_rate": 0.00019787668419760408,
|
| 309 |
+
"loss": 1.7305,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16,
|
| 314 |
+
"grad_norm": 0.18402560356298175,
|
| 315 |
+
"learning_rate": 0.00019759686959607383,
|
| 316 |
+
"loss": 1.8005,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17,
|
| 321 |
+
"grad_norm": 0.1597598662204373,
|
| 322 |
+
"learning_rate": 0.00019729996397911356,
|
| 323 |
+
"loss": 1.835,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.17,
|
| 328 |
+
"grad_norm": 0.18746716552277176,
|
| 329 |
+
"learning_rate": 0.0001969860193403835,
|
| 330 |
+
"loss": 1.6065,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.17,
|
| 335 |
+
"grad_norm": 0.18286863109823168,
|
| 336 |
+
"learning_rate": 0.00019665509065739149,
|
| 337 |
+
"loss": 1.6882,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18,
|
| 342 |
+
"grad_norm": 0.17022449603830708,
|
| 343 |
+
"learning_rate": 0.00019630723588186545,
|
| 344 |
+
"loss": 1.7489,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.18,
|
| 349 |
+
"grad_norm": 0.19066522709932746,
|
| 350 |
+
"learning_rate": 0.00019594251592960479,
|
| 351 |
+
"loss": 1.7215,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.19,
|
| 356 |
+
"grad_norm": 0.13321877247130193,
|
| 357 |
+
"learning_rate": 0.0001955609946698131,
|
| 358 |
+
"loss": 1.5857,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.19,
|
| 363 |
+
"grad_norm": 0.16703841490849763,
|
| 364 |
+
"learning_rate": 0.0001951627389139134,
|
| 365 |
+
"loss": 1.6287,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.19,
|
| 370 |
+
"grad_norm": 0.15390480377905874,
|
| 371 |
+
"learning_rate": 0.00019474781840384816,
|
| 372 |
+
"loss": 1.5562,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2,
|
| 377 |
+
"grad_norm": 0.13726931247747762,
|
| 378 |
+
"learning_rate": 0.00019431630579986632,
|
| 379 |
+
"loss": 1.6663,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2,
|
| 384 |
+
"grad_norm": 0.22618332502897354,
|
| 385 |
+
"learning_rate": 0.0001938682766677991,
|
| 386 |
+
"loss": 1.7624,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2,
|
| 391 |
+
"grad_norm": 0.14665992310817066,
|
| 392 |
+
"learning_rate": 0.00019340380946582695,
|
| 393 |
+
"loss": 1.7523,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.21,
|
| 398 |
+
"grad_norm": 0.1554140913393743,
|
| 399 |
+
"learning_rate": 0.0001929229855307402,
|
| 400 |
+
"loss": 1.6392,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.21,
|
| 405 |
+
"grad_norm": 0.20920485611465975,
|
| 406 |
+
"learning_rate": 0.00019242588906369536,
|
| 407 |
+
"loss": 1.8364,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22,
|
| 412 |
+
"grad_norm": 0.19116748798349187,
|
| 413 |
+
"learning_rate": 0.00019191260711547001,
|
| 414 |
+
"loss": 1.6188,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.22,
|
| 419 |
+
"grad_norm": 0.16536086353984128,
|
| 420 |
+
"learning_rate": 0.0001913832295712186,
|
| 421 |
+
"loss": 1.7202,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22,
|
| 426 |
+
"grad_norm": 0.1569402268882919,
|
| 427 |
+
"learning_rate": 0.0001908378491347319,
|
| 428 |
+
"loss": 1.6367,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.23,
|
| 433 |
+
"grad_norm": 0.16304498560465056,
|
| 434 |
+
"learning_rate": 0.0001902765613122028,
|
| 435 |
+
"loss": 1.7834,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.23,
|
| 440 |
+
"grad_norm": 0.16236999153287665,
|
| 441 |
+
"learning_rate": 0.00018969946439550148,
|
| 442 |
+
"loss": 1.6726,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.24,
|
| 447 |
+
"grad_norm": 0.15182991037078394,
|
| 448 |
+
"learning_rate": 0.00018910665944496264,
|
| 449 |
+
"loss": 1.8442,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.24,
|
| 454 |
+
"grad_norm": 0.18052814912668647,
|
| 455 |
+
"learning_rate": 0.00018849825027168803,
|
| 456 |
+
"loss": 1.9455,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.24,
|
| 461 |
+
"grad_norm": 0.16420621140530428,
|
| 462 |
+
"learning_rate": 0.0001878743434193671,
|
| 463 |
+
"loss": 1.7824,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.25,
|
| 468 |
+
"grad_norm": 0.14151509823888947,
|
| 469 |
+
"learning_rate": 0.0001872350481456193,
|
| 470 |
+
"loss": 1.744,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.25,
|
| 475 |
+
"grad_norm": 0.15295370028181537,
|
| 476 |
+
"learning_rate": 0.0001865804764028611,
|
| 477 |
+
"loss": 1.6297,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.25,
|
| 482 |
+
"grad_norm": 0.18729292049319135,
|
| 483 |
+
"learning_rate": 0.00018591074281870099,
|
| 484 |
+
"loss": 1.7033,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.26,
|
| 489 |
+
"grad_norm": 0.16159062074874506,
|
| 490 |
+
"learning_rate": 0.00018522596467586598,
|
| 491 |
+
"loss": 1.725,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.26,
|
| 496 |
+
"grad_norm": 0.1862873005353938,
|
| 497 |
+
"learning_rate": 0.00018452626189166345,
|
| 498 |
+
"loss": 1.7492,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.27,
|
| 503 |
+
"grad_norm": 0.1995664751091554,
|
| 504 |
+
"learning_rate": 0.0001838117569969812,
|
| 505 |
+
"loss": 1.6189,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.27,
|
| 510 |
+
"grad_norm": 0.20573320860450153,
|
| 511 |
+
"learning_rate": 0.00018308257511483018,
|
| 512 |
+
"loss": 1.5645,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.27,
|
| 517 |
+
"grad_norm": 0.1979474845480437,
|
| 518 |
+
"learning_rate": 0.000182338843938433,
|
| 519 |
+
"loss": 1.6659,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.28,
|
| 524 |
+
"grad_norm": 0.1489980772801844,
|
| 525 |
+
"learning_rate": 0.00018158069370886266,
|
| 526 |
+
"loss": 1.7409,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.28,
|
| 531 |
+
"grad_norm": 0.21205357472863073,
|
| 532 |
+
"learning_rate": 0.00018080825719223468,
|
| 533 |
+
"loss": 1.825,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.28,
|
| 538 |
+
"grad_norm": 0.16632310528095662,
|
| 539 |
+
"learning_rate": 0.0001800216696564576,
|
| 540 |
+
"loss": 1.6339,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.29,
|
| 545 |
+
"grad_norm": 0.1514716321570189,
|
| 546 |
+
"learning_rate": 0.00017922106884754488,
|
| 547 |
+
"loss": 1.6527,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.29,
|
| 552 |
+
"grad_norm": 0.1824720480684259,
|
| 553 |
+
"learning_rate": 0.00017840659496549298,
|
| 554 |
+
"loss": 1.9076,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.3,
|
| 559 |
+
"grad_norm": 0.17567973424287547,
|
| 560 |
+
"learning_rate": 0.00017757839063972997,
|
| 561 |
+
"loss": 1.6627,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3,
|
| 566 |
+
"grad_norm": 0.18153116327896177,
|
| 567 |
+
"learning_rate": 0.00017673660090413823,
|
| 568 |
+
"loss": 1.7191,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3,
|
| 573 |
+
"grad_norm": 0.1707968303720214,
|
| 574 |
+
"learning_rate": 0.00017588137317165657,
|
| 575 |
+
"loss": 1.7909,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.31,
|
| 580 |
+
"grad_norm": 0.16649736682689767,
|
| 581 |
+
"learning_rate": 0.00017501285720846523,
|
| 582 |
+
"loss": 1.6713,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.31,
|
| 587 |
+
"grad_norm": 0.1887556367835341,
|
| 588 |
+
"learning_rate": 0.0001741312051077594,
|
| 589 |
+
"loss": 1.6464,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.31,
|
| 594 |
+
"grad_norm": 0.1614746163296818,
|
| 595 |
+
"learning_rate": 0.00017323657126311454,
|
| 596 |
+
"loss": 1.6152,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.32,
|
| 601 |
+
"grad_norm": 0.16027605599844436,
|
| 602 |
+
"learning_rate": 0.0001723291123414495,
|
| 603 |
+
"loss": 1.6705,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.32,
|
| 608 |
+
"grad_norm": 0.1378518336569361,
|
| 609 |
+
"learning_rate": 0.000171408987255591,
|
| 610 |
+
"loss": 1.7244,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.33,
|
| 615 |
+
"grad_norm": 0.17223398801495055,
|
| 616 |
+
"learning_rate": 0.00017047635713644528,
|
| 617 |
+
"loss": 1.6938,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.33,
|
| 622 |
+
"grad_norm": 0.13195968200255281,
|
| 623 |
+
"learning_rate": 0.00016953138530478092,
|
| 624 |
+
"loss": 1.8307,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.33,
|
| 629 |
+
"grad_norm": 0.1769224290233956,
|
| 630 |
+
"learning_rate": 0.00016857423724262849,
|
| 631 |
+
"loss": 1.7391,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.34,
|
| 636 |
+
"grad_norm": 0.18282706527502826,
|
| 637 |
+
"learning_rate": 0.00016760508056430152,
|
| 638 |
+
"loss": 1.7326,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.34,
|
| 643 |
+
"grad_norm": 0.1631203496086859,
|
| 644 |
+
"learning_rate": 0.0001666240849870441,
|
| 645 |
+
"loss": 1.823,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.34,
|
| 650 |
+
"grad_norm": 0.20235195405729262,
|
| 651 |
+
"learning_rate": 0.0001656314223013104,
|
| 652 |
+
"loss": 1.845,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.35,
|
| 657 |
+
"grad_norm": 0.19609007678999507,
|
| 658 |
+
"learning_rate": 0.00016462726634068075,
|
| 659 |
+
"loss": 1.7261,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.35,
|
| 664 |
+
"grad_norm": 0.18106665152255952,
|
| 665 |
+
"learning_rate": 0.00016361179295142046,
|
| 666 |
+
"loss": 1.7463,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.36,
|
| 671 |
+
"grad_norm": 0.18149538862355094,
|
| 672 |
+
"learning_rate": 0.00016258517996168564,
|
| 673 |
+
"loss": 1.795,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.36,
|
| 678 |
+
"grad_norm": 0.19025838125643668,
|
| 679 |
+
"learning_rate": 0.0001615476071503823,
|
| 680 |
+
"loss": 1.832,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.36,
|
| 685 |
+
"grad_norm": 0.1265376212889534,
|
| 686 |
+
"learning_rate": 0.00016049925621568382,
|
| 687 |
+
"loss": 1.7837,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.37,
|
| 692 |
+
"grad_norm": 0.14422377519554716,
|
| 693 |
+
"learning_rate": 0.00015944031074321204,
|
| 694 |
+
"loss": 1.6805,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.37,
|
| 699 |
+
"grad_norm": 0.1762633551257036,
|
| 700 |
+
"learning_rate": 0.00015837095617388827,
|
| 701 |
+
"loss": 1.7382,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.38,
|
| 706 |
+
"grad_norm": 0.1876144376022372,
|
| 707 |
+
"learning_rate": 0.00015729137977145893,
|
| 708 |
+
"loss": 1.7734,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.38,
|
| 713 |
+
"grad_norm": 0.20336686380575994,
|
| 714 |
+
"learning_rate": 0.0001562017705897024,
|
| 715 |
+
"loss": 1.7307,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.38,
|
| 720 |
+
"grad_norm": 0.1851794296881923,
|
| 721 |
+
"learning_rate": 0.0001551023194393221,
|
| 722 |
+
"loss": 1.6542,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.39,
|
| 727 |
+
"grad_norm": 0.13390782339032822,
|
| 728 |
+
"learning_rate": 0.00015399321885453202,
|
| 729 |
+
"loss": 1.6537,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.39,
|
| 734 |
+
"grad_norm": 0.16853017759818759,
|
| 735 |
+
"learning_rate": 0.00015287466305934037,
|
| 736 |
+
"loss": 1.7368,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.39,
|
| 741 |
+
"grad_norm": 0.1988449599462779,
|
| 742 |
+
"learning_rate": 0.0001517468479335376,
|
| 743 |
+
"loss": 1.8295,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4,
|
| 748 |
+
"grad_norm": 0.1559082965308534,
|
| 749 |
+
"learning_rate": 0.00015060997097839386,
|
| 750 |
+
"loss": 1.7347,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4,
|
| 755 |
+
"grad_norm": 0.224768472771501,
|
| 756 |
+
"learning_rate": 0.00014946423128207322,
|
| 757 |
+
"loss": 1.8114,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.41,
|
| 762 |
+
"grad_norm": 0.1295999325817523,
|
| 763 |
+
"learning_rate": 0.0001483098294847695,
|
| 764 |
+
"loss": 1.8739,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.41,
|
| 769 |
+
"grad_norm": 0.18936807370761422,
|
| 770 |
+
"learning_rate": 0.0001471469677435704,
|
| 771 |
+
"loss": 1.7902,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.41,
|
| 776 |
+
"grad_norm": 0.1643554393968483,
|
| 777 |
+
"learning_rate": 0.00014597584969705616,
|
| 778 |
+
"loss": 1.6912,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.42,
|
| 783 |
+
"grad_norm": 0.15773757647809103,
|
| 784 |
+
"learning_rate": 0.0001447966804296387,
|
| 785 |
+
"loss": 1.8605,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.42,
|
| 790 |
+
"grad_norm": 0.20406845772260215,
|
| 791 |
+
"learning_rate": 0.00014360966643564747,
|
| 792 |
+
"loss": 1.6922,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.42,
|
| 797 |
+
"grad_norm": 0.17501456530325488,
|
| 798 |
+
"learning_rate": 0.0001424150155831685,
|
| 799 |
+
"loss": 1.7865,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.43,
|
| 804 |
+
"grad_norm": 0.1518901111914136,
|
| 805 |
+
"learning_rate": 0.0001412129370776429,
|
| 806 |
+
"loss": 1.6559,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.43,
|
| 811 |
+
"grad_norm": 0.18138565181200048,
|
| 812 |
+
"learning_rate": 0.00014000364142523103,
|
| 813 |
+
"loss": 1.5771,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.44,
|
| 818 |
+
"grad_norm": 0.16454907908913538,
|
| 819 |
+
"learning_rate": 0.0001387873403959492,
|
| 820 |
+
"loss": 1.6639,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.44,
|
| 825 |
+
"grad_norm": 0.1334424167275731,
|
| 826 |
+
"learning_rate": 0.0001375642469865844,
|
| 827 |
+
"loss": 1.6229,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.44,
|
| 832 |
+
"grad_norm": 0.1632799340974609,
|
| 833 |
+
"learning_rate": 0.00013633457538339514,
|
| 834 |
+
"loss": 1.7959,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.45,
|
| 839 |
+
"grad_norm": 0.195109291960245,
|
| 840 |
+
"learning_rate": 0.00013509854092460312,
|
| 841 |
+
"loss": 1.6124,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.45,
|
| 846 |
+
"grad_norm": 0.1839733274238824,
|
| 847 |
+
"learning_rate": 0.00013385636006268368,
|
| 848 |
+
"loss": 1.7034,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.45,
|
| 853 |
+
"grad_norm": 0.1850784607361091,
|
| 854 |
+
"learning_rate": 0.00013260825032646083,
|
| 855 |
+
"loss": 1.6997,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.46,
|
| 860 |
+
"grad_norm": 0.15397168391927735,
|
| 861 |
+
"learning_rate": 0.0001313544302830142,
|
| 862 |
+
"loss": 1.4588,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.46,
|
| 867 |
+
"grad_norm": 0.203541257474643,
|
| 868 |
+
"learning_rate": 0.00013009511949940358,
|
| 869 |
+
"loss": 1.6677,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.47,
|
| 874 |
+
"grad_norm": 0.16193823373140015,
|
| 875 |
+
"learning_rate": 0.00012883053850421897,
|
| 876 |
+
"loss": 1.6539,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.47,
|
| 881 |
+
"grad_norm": 0.16809205680859712,
|
| 882 |
+
"learning_rate": 0.00012756090874896172,
|
| 883 |
+
"loss": 1.7106,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.47,
|
| 888 |
+
"grad_norm": 0.13435988109027225,
|
| 889 |
+
"learning_rate": 0.00012628645256926438,
|
| 890 |
+
"loss": 1.5563,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.48,
|
| 895 |
+
"grad_norm": 0.17547074427143972,
|
| 896 |
+
"learning_rate": 0.00012500739314595563,
|
| 897 |
+
"loss": 1.6488,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.48,
|
| 902 |
+
"grad_norm": 0.19354830986756066,
|
| 903 |
+
"learning_rate": 0.0001237239544659771,
|
| 904 |
+
"loss": 1.7111,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.49,
|
| 909 |
+
"grad_norm": 0.16194728732695224,
|
| 910 |
+
"learning_rate": 0.00012243636128315939,
|
| 911 |
+
"loss": 1.6541,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.49,
|
| 916 |
+
"grad_norm": 0.1479347118711336,
|
| 917 |
+
"learning_rate": 0.00012114483907886308,
|
| 918 |
+
"loss": 1.7099,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.49,
|
| 923 |
+
"grad_norm": 0.14962574930449243,
|
| 924 |
+
"learning_rate": 0.00011984961402249311,
|
| 925 |
+
"loss": 1.7944,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5,
|
| 930 |
+
"grad_norm": 0.19522394739541554,
|
| 931 |
+
"learning_rate": 0.00011855091293189234,
|
| 932 |
+
"loss": 1.6922,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5,
|
| 937 |
+
"grad_norm": 0.20134721263904506,
|
| 938 |
+
"learning_rate": 0.0001172489632336213,
|
| 939 |
+
"loss": 1.7516,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5,
|
| 944 |
+
"grad_norm": 0.17494019729104499,
|
| 945 |
+
"learning_rate": 0.00011594399292313192,
|
| 946 |
+
"loss": 1.7995,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.51,
|
| 951 |
+
"grad_norm": 0.2018746644709756,
|
| 952 |
+
"learning_rate": 0.000114636230524841,
|
| 953 |
+
"loss": 1.8339,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.51,
|
| 958 |
+
"grad_norm": 0.17996734198172393,
|
| 959 |
+
"learning_rate": 0.00011332590505211159,
|
| 960 |
+
"loss": 1.7256,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.52,
|
| 965 |
+
"grad_norm": 0.17873752655785974,
|
| 966 |
+
"learning_rate": 0.00011201324596714844,
|
| 967 |
+
"loss": 1.7191,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.52,
|
| 972 |
+
"grad_norm": 0.1467005565147068,
|
| 973 |
+
"learning_rate": 0.0001106984831408149,
|
| 974 |
+
"loss": 1.6384,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.52,
|
| 979 |
+
"grad_norm": 0.1915514048730809,
|
| 980 |
+
"learning_rate": 0.00010938184681237833,
|
| 981 |
+
"loss": 1.6736,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.53,
|
| 986 |
+
"grad_norm": 0.1838191695918362,
|
| 987 |
+
"learning_rate": 0.00010806356754919091,
|
| 988 |
+
"loss": 1.4994,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.53,
|
| 993 |
+
"grad_norm": 0.22436914705741917,
|
| 994 |
+
"learning_rate": 0.00010674387620631308,
|
| 995 |
+
"loss": 1.83,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.53,
|
| 1000 |
+
"grad_norm": 0.17652320007559247,
|
| 1001 |
+
"learning_rate": 0.00010542300388608652,
|
| 1002 |
+
"loss": 1.7067,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.54,
|
| 1007 |
+
"grad_norm": 0.16308378188013725,
|
| 1008 |
+
"learning_rate": 0.00010410118189766387,
|
| 1009 |
+
"loss": 1.6581,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.54,
|
| 1014 |
+
"grad_norm": 0.17768576113421508,
|
| 1015 |
+
"learning_rate": 0.0001027786417165022,
|
| 1016 |
+
"loss": 1.6501,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.55,
|
| 1021 |
+
"grad_norm": 0.2761038831057377,
|
| 1022 |
+
"learning_rate": 0.00010145561494382742,
|
| 1023 |
+
"loss": 1.7354,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.55,
|
| 1028 |
+
"grad_norm": 0.19431100962861855,
|
| 1029 |
+
"learning_rate": 0.00010013233326607661,
|
| 1030 |
+
"loss": 1.689,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.55,
|
| 1035 |
+
"grad_norm": 0.20190135739240553,
|
| 1036 |
+
"learning_rate": 9.880902841432544e-05,
|
| 1037 |
+
"loss": 1.6582,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.56,
|
| 1042 |
+
"grad_norm": 0.1612101989801135,
|
| 1043 |
+
"learning_rate": 9.748593212370773e-05,
|
| 1044 |
+
"loss": 1.7682,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.56,
|
| 1049 |
+
"grad_norm": 0.1887774058582275,
|
| 1050 |
+
"learning_rate": 9.616327609283445e-05,
|
| 1051 |
+
"loss": 1.6278,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.56,
|
| 1056 |
+
"grad_norm": 0.18804655029135892,
|
| 1057 |
+
"learning_rate": 9.484129194321896e-05,
|
| 1058 |
+
"loss": 1.6383,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.57,
|
| 1063 |
+
"grad_norm": 0.17619546651470808,
|
| 1064 |
+
"learning_rate": 9.352021117871574e-05,
|
| 1065 |
+
"loss": 1.7387,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.57,
|
| 1070 |
+
"grad_norm": 0.1832656601025776,
|
| 1071 |
+
"learning_rate": 9.220026514497983e-05,
|
| 1072 |
+
"loss": 1.7469,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.58,
|
| 1077 |
+
"grad_norm": 0.18317871859883073,
|
| 1078 |
+
"learning_rate": 9.088168498895408e-05,
|
| 1079 |
+
"loss": 1.6827,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.58,
|
| 1084 |
+
"grad_norm": 0.18190889844847366,
|
| 1085 |
+
"learning_rate": 8.956470161839072e-05,
|
| 1086 |
+
"loss": 1.7512,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.58,
|
| 1091 |
+
"grad_norm": 0.13673363804446922,
|
| 1092 |
+
"learning_rate": 8.82495456614155e-05,
|
| 1093 |
+
"loss": 1.72,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.59,
|
| 1098 |
+
"grad_norm": 0.15018409524264797,
|
| 1099 |
+
"learning_rate": 8.693644742614017e-05,
|
| 1100 |
+
"loss": 1.7045,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.59,
|
| 1105 |
+
"grad_norm": 0.2186693604605445,
|
| 1106 |
+
"learning_rate": 8.562563686033145e-05,
|
| 1107 |
+
"loss": 1.701,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6,
|
| 1112 |
+
"grad_norm": 0.19245197095540534,
|
| 1113 |
+
"learning_rate": 8.431734351114284e-05,
|
| 1114 |
+
"loss": 1.762,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6,
|
| 1119 |
+
"grad_norm": 0.19207579375731684,
|
| 1120 |
+
"learning_rate": 8.301179648491669e-05,
|
| 1121 |
+
"loss": 1.6583,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6,
|
| 1126 |
+
"grad_norm": 0.18601379502952384,
|
| 1127 |
+
"learning_rate": 8.17092244070634e-05,
|
| 1128 |
+
"loss": 1.8022,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.61,
|
| 1133 |
+
"grad_norm": 0.15775561164822532,
|
| 1134 |
+
"learning_rate": 8.040985538202505e-05,
|
| 1135 |
+
"loss": 1.8587,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.61,
|
| 1140 |
+
"grad_norm": 0.17113306175707532,
|
| 1141 |
+
"learning_rate": 7.911391695332988e-05,
|
| 1142 |
+
"loss": 1.8229,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.61,
|
| 1147 |
+
"grad_norm": 0.1915640796282049,
|
| 1148 |
+
"learning_rate": 7.782163606374536e-05,
|
| 1149 |
+
"loss": 1.7157,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.62,
|
| 1154 |
+
"grad_norm": 0.16640360967802714,
|
| 1155 |
+
"learning_rate": 7.653323901553625e-05,
|
| 1156 |
+
"loss": 1.7302,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.62,
|
| 1161 |
+
"grad_norm": 0.1787131411656399,
|
| 1162 |
+
"learning_rate": 7.52489514308349e-05,
|
| 1163 |
+
"loss": 1.5147,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.63,
|
| 1168 |
+
"grad_norm": 0.181532462982817,
|
| 1169 |
+
"learning_rate": 7.396899821213072e-05,
|
| 1170 |
+
"loss": 1.6488,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.63,
|
| 1175 |
+
"grad_norm": 0.21557423812359156,
|
| 1176 |
+
"learning_rate": 7.269360350288547e-05,
|
| 1177 |
+
"loss": 1.7753,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.63,
|
| 1182 |
+
"grad_norm": 0.18460048157273207,
|
| 1183 |
+
"learning_rate": 7.142299064828169e-05,
|
| 1184 |
+
"loss": 1.7397,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.64,
|
| 1189 |
+
"grad_norm": 0.15330203579005233,
|
| 1190 |
+
"learning_rate": 7.015738215611079e-05,
|
| 1191 |
+
"loss": 1.8253,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.64,
|
| 1196 |
+
"grad_norm": 0.17444450083682025,
|
| 1197 |
+
"learning_rate": 6.889699965780787e-05,
|
| 1198 |
+
"loss": 1.6109,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.64,
|
| 1203 |
+
"grad_norm": 0.21652046641655817,
|
| 1204 |
+
"learning_rate": 6.764206386963991e-05,
|
| 1205 |
+
"loss": 1.7588,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.65,
|
| 1210 |
+
"grad_norm": 0.21834789848228028,
|
| 1211 |
+
"learning_rate": 6.639279455405432e-05,
|
| 1212 |
+
"loss": 1.7249,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.65,
|
| 1217 |
+
"grad_norm": 0.19720018886202378,
|
| 1218 |
+
"learning_rate": 6.514941048119435e-05,
|
| 1219 |
+
"loss": 1.7814,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.66,
|
| 1224 |
+
"grad_norm": 0.17019093587889528,
|
| 1225 |
+
"learning_rate": 6.391212939058861e-05,
|
| 1226 |
+
"loss": 1.7215,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.66,
|
| 1231 |
+
"grad_norm": 0.17764400698646424,
|
| 1232 |
+
"learning_rate": 6.268116795302068e-05,
|
| 1233 |
+
"loss": 1.6174,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.66,
|
| 1238 |
+
"grad_norm": 0.1857191323680069,
|
| 1239 |
+
"learning_rate": 6.14567417325861e-05,
|
| 1240 |
+
"loss": 1.7505,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.67,
|
| 1245 |
+
"grad_norm": 0.2043550506390951,
|
| 1246 |
+
"learning_rate": 6.023906514894313e-05,
|
| 1247 |
+
"loss": 1.617,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.67,
|
| 1252 |
+
"grad_norm": 0.152811359160537,
|
| 1253 |
+
"learning_rate": 5.902835143976393e-05,
|
| 1254 |
+
"loss": 1.5499,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.67,
|
| 1259 |
+
"grad_norm": 0.16838709339190316,
|
| 1260 |
+
"learning_rate": 5.782481262339261e-05,
|
| 1261 |
+
"loss": 1.629,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.68,
|
| 1266 |
+
"grad_norm": 0.24023332810174503,
|
| 1267 |
+
"learning_rate": 5.662865946171696e-05,
|
| 1268 |
+
"loss": 1.8542,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.68,
|
| 1273 |
+
"grad_norm": 0.17090260469005264,
|
| 1274 |
+
"learning_rate": 5.544010142326026e-05,
|
| 1275 |
+
"loss": 1.6502,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.69,
|
| 1280 |
+
"grad_norm": 0.22105892400962765,
|
| 1281 |
+
"learning_rate": 5.425934664649921e-05,
|
| 1282 |
+
"loss": 1.7454,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.69,
|
| 1287 |
+
"grad_norm": 0.17574025557326728,
|
| 1288 |
+
"learning_rate": 5.308660190341528e-05,
|
| 1289 |
+
"loss": 1.6264,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.69,
|
| 1294 |
+
"grad_norm": 0.19344761090063187,
|
| 1295 |
+
"learning_rate": 5.1922072563284986e-05,
|
| 1296 |
+
"loss": 1.6931,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7,
|
| 1301 |
+
"grad_norm": 0.20792951796116704,
|
| 1302 |
+
"learning_rate": 5.076596255671592e-05,
|
| 1303 |
+
"loss": 1.7799,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7,
|
| 1308 |
+
"grad_norm": 0.152967524344273,
|
| 1309 |
+
"learning_rate": 4.9618474339934916e-05,
|
| 1310 |
+
"loss": 1.5804,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.71,
|
| 1315 |
+
"grad_norm": 0.2029478917940139,
|
| 1316 |
+
"learning_rate": 4.8479808859333964e-05,
|
| 1317 |
+
"loss": 1.6536,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.71,
|
| 1322 |
+
"grad_norm": 0.20273359253816137,
|
| 1323 |
+
"learning_rate": 4.735016551628095e-05,
|
| 1324 |
+
"loss": 1.7212,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.71,
|
| 1329 |
+
"grad_norm": 0.24355186668207754,
|
| 1330 |
+
"learning_rate": 4.6229742132200746e-05,
|
| 1331 |
+
"loss": 1.5726,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.72,
|
| 1336 |
+
"grad_norm": 0.16910336039273627,
|
| 1337 |
+
"learning_rate": 4.511873491393304e-05,
|
| 1338 |
+
"loss": 1.6338,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.72,
|
| 1343 |
+
"grad_norm": 0.18917035113482142,
|
| 1344 |
+
"learning_rate": 4.401733841937279e-05,
|
| 1345 |
+
"loss": 1.7485,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.72,
|
| 1350 |
+
"grad_norm": 0.15310313773374987,
|
| 1351 |
+
"learning_rate": 4.292574552339981e-05,
|
| 1352 |
+
"loss": 1.5909,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.73,
|
| 1357 |
+
"grad_norm": 0.16225231468176773,
|
| 1358 |
+
"learning_rate": 4.184414738410248e-05,
|
| 1359 |
+
"loss": 1.7539,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.73,
|
| 1364 |
+
"grad_norm": 0.1864838363620509,
|
| 1365 |
+
"learning_rate": 4.077273340930263e-05,
|
| 1366 |
+
"loss": 1.7388,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.74,
|
| 1371 |
+
"grad_norm": 0.18596136650157077,
|
| 1372 |
+
"learning_rate": 3.971169122338668e-05,
|
| 1373 |
+
"loss": 1.6928,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.74,
|
| 1378 |
+
"grad_norm": 0.19828582588669125,
|
| 1379 |
+
"learning_rate": 3.866120663444914e-05,
|
| 1380 |
+
"loss": 1.6799,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.74,
|
| 1385 |
+
"grad_norm": 0.19914566086301766,
|
| 1386 |
+
"learning_rate": 3.762146360175427e-05,
|
| 1387 |
+
"loss": 1.7465,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.75,
|
| 1392 |
+
"grad_norm": 0.1565602915291301,
|
| 1393 |
+
"learning_rate": 3.659264420352122e-05,
|
| 1394 |
+
"loss": 1.6289,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.75,
|
| 1399 |
+
"grad_norm": 0.1761029839621625,
|
| 1400 |
+
"learning_rate": 3.557492860503893e-05,
|
| 1401 |
+
"loss": 1.68,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.75,
|
| 1406 |
+
"grad_norm": 0.16369015034577106,
|
| 1407 |
+
"learning_rate": 3.45684950271158e-05,
|
| 1408 |
+
"loss": 1.8479,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.76,
|
| 1413 |
+
"grad_norm": 0.198047069955828,
|
| 1414 |
+
"learning_rate": 3.3573519714869914e-05,
|
| 1415 |
+
"loss": 1.6707,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.76,
|
| 1420 |
+
"grad_norm": 0.17307190965159788,
|
| 1421 |
+
"learning_rate": 3.25901769068654e-05,
|
| 1422 |
+
"loss": 1.635,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.77,
|
| 1427 |
+
"grad_norm": 0.22793946769374895,
|
| 1428 |
+
"learning_rate": 3.16186388045998e-05,
|
| 1429 |
+
"loss": 1.7162,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.77,
|
| 1434 |
+
"grad_norm": 0.2076019661591803,
|
| 1435 |
+
"learning_rate": 3.065907554234858e-05,
|
| 1436 |
+
"loss": 1.8822,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.77,
|
| 1441 |
+
"grad_norm": 0.19560384418847823,
|
| 1442 |
+
"learning_rate": 2.9711655157371443e-05,
|
| 1443 |
+
"loss": 1.6867,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.78,
|
| 1448 |
+
"grad_norm": 0.19574225624101774,
|
| 1449 |
+
"learning_rate": 2.8776543560485857e-05,
|
| 1450 |
+
"loss": 1.713,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.78,
|
| 1455 |
+
"grad_norm": 0.18200916558664465,
|
| 1456 |
+
"learning_rate": 2.785390450701303e-05,
|
| 1457 |
+
"loss": 1.6337,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.78,
|
| 1462 |
+
"grad_norm": 0.1541733986762764,
|
| 1463 |
+
"learning_rate": 2.6943899568101405e-05,
|
| 1464 |
+
"loss": 1.5729,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.79,
|
| 1469 |
+
"grad_norm": 0.19958352219242734,
|
| 1470 |
+
"learning_rate": 2.6046688102432382e-05,
|
| 1471 |
+
"loss": 1.7146,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.79,
|
| 1476 |
+
"grad_norm": 0.19853696310550636,
|
| 1477 |
+
"learning_rate": 2.5162427228313857e-05,
|
| 1478 |
+
"loss": 1.7313,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8,
|
| 1483 |
+
"grad_norm": 0.15888101789785714,
|
| 1484 |
+
"learning_rate": 2.429127179616575e-05,
|
| 1485 |
+
"loss": 1.7659,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.8,
|
| 1490 |
+
"grad_norm": 0.18173603127604884,
|
| 1491 |
+
"learning_rate": 2.343337436140295e-05,
|
| 1492 |
+
"loss": 1.7334,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8,
|
| 1497 |
+
"grad_norm": 0.14655412622982647,
|
| 1498 |
+
"learning_rate": 2.258888515772005e-05,
|
| 1499 |
+
"loss": 1.5761,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.81,
|
| 1504 |
+
"grad_norm": 0.20266007316099247,
|
| 1505 |
+
"learning_rate": 2.1757952070782504e-05,
|
| 1506 |
+
"loss": 1.7433,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.81,
|
| 1511 |
+
"grad_norm": 0.20234298686235969,
|
| 1512 |
+
"learning_rate": 2.0940720612329258e-05,
|
| 1513 |
+
"loss": 1.5892,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.82,
|
| 1518 |
+
"grad_norm": 0.1526986847585089,
|
| 1519 |
+
"learning_rate": 2.0137333894690912e-05,
|
| 1520 |
+
"loss": 1.6581,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.82,
|
| 1525 |
+
"grad_norm": 0.2163230002939218,
|
| 1526 |
+
"learning_rate": 1.9347932605728093e-05,
|
| 1527 |
+
"loss": 1.668,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.82,
|
| 1532 |
+
"grad_norm": 0.18915868456626023,
|
| 1533 |
+
"learning_rate": 1.8572654984194392e-05,
|
| 1534 |
+
"loss": 1.621,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.83,
|
| 1539 |
+
"grad_norm": 0.2161988791708388,
|
| 1540 |
+
"learning_rate": 1.781163679552831e-05,
|
| 1541 |
+
"loss": 1.7003,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.83,
|
| 1546 |
+
"grad_norm": 0.1560093225771299,
|
| 1547 |
+
"learning_rate": 1.706501130807806e-05,
|
| 1548 |
+
"loss": 1.7737,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.83,
|
| 1553 |
+
"grad_norm": 0.2028694894544158,
|
| 1554 |
+
"learning_rate": 1.6332909269763953e-05,
|
| 1555 |
+
"loss": 1.7086,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.84,
|
| 1560 |
+
"grad_norm": 0.20833737878016775,
|
| 1561 |
+
"learning_rate": 1.561545888518192e-05,
|
| 1562 |
+
"loss": 1.6524,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.84,
|
| 1567 |
+
"grad_norm": 0.2292307851951488,
|
| 1568 |
+
"learning_rate": 1.4912785793152583e-05,
|
| 1569 |
+
"loss": 1.7708,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.85,
|
| 1574 |
+
"grad_norm": 0.20740473508039634,
|
| 1575 |
+
"learning_rate": 1.4225013044719615e-05,
|
| 1576 |
+
"loss": 1.5385,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.85,
|
| 1581 |
+
"grad_norm": 0.19005217138703415,
|
| 1582 |
+
"learning_rate": 1.3552261081601091e-05,
|
| 1583 |
+
"loss": 1.7285,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.85,
|
| 1588 |
+
"grad_norm": 0.1940930521738508,
|
| 1589 |
+
"learning_rate": 1.289464771509804e-05,
|
| 1590 |
+
"loss": 1.6208,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.86,
|
| 1595 |
+
"grad_norm": 0.2153509149434588,
|
| 1596 |
+
"learning_rate": 1.2252288105463405e-05,
|
| 1597 |
+
"loss": 1.7712,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.86,
|
| 1602 |
+
"grad_norm": 0.19048315548215075,
|
| 1603 |
+
"learning_rate": 1.1625294741735526e-05,
|
| 1604 |
+
"loss": 1.6668,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.86,
|
| 1609 |
+
"grad_norm": 0.15371576628718261,
|
| 1610 |
+
"learning_rate": 1.101377742203903e-05,
|
| 1611 |
+
"loss": 1.7325,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.87,
|
| 1616 |
+
"grad_norm": 0.1765858067098436,
|
| 1617 |
+
"learning_rate": 1.0417843234357282e-05,
|
| 1618 |
+
"loss": 1.5605,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.87,
|
| 1623 |
+
"grad_norm": 0.19769140244098435,
|
| 1624 |
+
"learning_rate": 9.837596537779237e-06,
|
| 1625 |
+
"loss": 1.7202,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.88,
|
| 1630 |
+
"grad_norm": 0.1636190248970844,
|
| 1631 |
+
"learning_rate": 9.27313894422428e-06,
|
| 1632 |
+
"loss": 1.7334,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.88,
|
| 1637 |
+
"grad_norm": 0.1795770045555807,
|
| 1638 |
+
"learning_rate": 8.724569300648034e-06,
|
| 1639 |
+
"loss": 1.7336,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.88,
|
| 1644 |
+
"grad_norm": 0.18573985907644597,
|
| 1645 |
+
"learning_rate": 8.19198367173255e-06,
|
| 1646 |
+
"loss": 1.6823,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.89,
|
| 1651 |
+
"grad_norm": 0.19679409440430415,
|
| 1652 |
+
"learning_rate": 7.675475323063475e-06,
|
| 1653 |
+
"loss": 1.8278,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.89,
|
| 1658 |
+
"grad_norm": 0.20610765374352058,
|
| 1659 |
+
"learning_rate": 7.175134704797592e-06,
|
| 1660 |
+
"loss": 1.7029,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.89,
|
| 1665 |
+
"grad_norm": 0.192668406292881,
|
| 1666 |
+
"learning_rate": 6.691049435823327e-06,
|
| 1667 |
+
"loss": 1.8248,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9,
|
| 1672 |
+
"grad_norm": 0.18604582750480914,
|
| 1673 |
+
"learning_rate": 6.22330428841702e-06,
|
| 1674 |
+
"loss": 1.8242,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9,
|
| 1679 |
+
"grad_norm": 0.19974117814732592,
|
| 1680 |
+
"learning_rate": 5.771981173397811e-06,
|
| 1681 |
+
"loss": 1.6994,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.91,
|
| 1686 |
+
"grad_norm": 0.18691439037044016,
|
| 1687 |
+
"learning_rate": 5.337159125783453e-06,
|
| 1688 |
+
"loss": 1.8688,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.91,
|
| 1693 |
+
"grad_norm": 0.18302336322558957,
|
| 1694 |
+
"learning_rate": 4.9189142909498945e-06,
|
| 1695 |
+
"loss": 1.7011,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.91,
|
| 1700 |
+
"grad_norm": 0.16276715134022465,
|
| 1701 |
+
"learning_rate": 4.517319911296747e-06,
|
| 1702 |
+
"loss": 1.6793,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.92,
|
| 1707 |
+
"grad_norm": 0.1671233177658411,
|
| 1708 |
+
"learning_rate": 4.132446313421246e-06,
|
| 1709 |
+
"loss": 1.7648,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.92,
|
| 1714 |
+
"grad_norm": 0.1999230765848782,
|
| 1715 |
+
"learning_rate": 3.7643608958027543e-06,
|
| 1716 |
+
"loss": 1.7858,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.92,
|
| 1721 |
+
"grad_norm": 0.19081110632199813,
|
| 1722 |
+
"learning_rate": 3.4131281170000083e-06,
|
| 1723 |
+
"loss": 1.7149,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.93,
|
| 1728 |
+
"grad_norm": 0.20949748915550245,
|
| 1729 |
+
"learning_rate": 3.0788094843632655e-06,
|
| 1730 |
+
"loss": 1.7321,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.93,
|
| 1735 |
+
"grad_norm": 0.17479576986843795,
|
| 1736 |
+
"learning_rate": 2.7614635432632097e-06,
|
| 1737 |
+
"loss": 1.7379,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.94,
|
| 1742 |
+
"grad_norm": 0.19598502442999505,
|
| 1743 |
+
"learning_rate": 2.461145866838599e-06,
|
| 1744 |
+
"loss": 1.6667,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.94,
|
| 1749 |
+
"grad_norm": 0.2114759691470325,
|
| 1750 |
+
"learning_rate": 2.177909046264348e-06,
|
| 1751 |
+
"loss": 1.5768,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.94,
|
| 1756 |
+
"grad_norm": 0.21204581223076796,
|
| 1757 |
+
"learning_rate": 1.911802681541919e-06,
|
| 1758 |
+
"loss": 1.493,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.95,
|
| 1763 |
+
"grad_norm": 0.2146455711644599,
|
| 1764 |
+
"learning_rate": 1.6628733728133227e-06,
|
| 1765 |
+
"loss": 1.7313,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.95,
|
| 1770 |
+
"grad_norm": 0.1829938024906039,
|
| 1771 |
+
"learning_rate": 1.4311647122006721e-06,
|
| 1772 |
+
"loss": 1.7979,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.96,
|
| 1777 |
+
"grad_norm": 0.2227396674614033,
|
| 1778 |
+
"learning_rate": 1.216717276172341e-06,
|
| 1779 |
+
"loss": 1.7429,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.96,
|
| 1784 |
+
"grad_norm": 0.17163374392577124,
|
| 1785 |
+
"learning_rate": 1.0195686184373166e-06,
|
| 1786 |
+
"loss": 1.7308,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.96,
|
| 1791 |
+
"grad_norm": 0.18401065207770326,
|
| 1792 |
+
"learning_rate": 8.397532633688254e-07,
|
| 1793 |
+
"loss": 1.6,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.97,
|
| 1798 |
+
"grad_norm": 0.17090828715744394,
|
| 1799 |
+
"learning_rate": 6.773026999584708e-07,
|
| 1800 |
+
"loss": 1.7816,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.97,
|
| 1805 |
+
"grad_norm": 0.1911205998245446,
|
| 1806 |
+
"learning_rate": 5.322453763019653e-07,
|
| 1807 |
+
"loss": 1.744,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.97,
|
| 1812 |
+
"grad_norm": 0.1667057593092014,
|
| 1813 |
+
"learning_rate": 4.046066946172822e-07,
|
| 1814 |
+
"loss": 1.5815,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.98,
|
| 1819 |
+
"grad_norm": 0.1607155959453338,
|
| 1820 |
+
"learning_rate": 2.9440900679631457e-07,
|
| 1821 |
+
"loss": 1.8161,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.98,
|
| 1826 |
+
"grad_norm": 0.18720833430214684,
|
| 1827 |
+
"learning_rate": 2.016716104906391e-07,
|
| 1828 |
+
"loss": 1.8291,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.99,
|
| 1833 |
+
"grad_norm": 0.2766212038818113,
|
| 1834 |
+
"learning_rate": 1.2641074573209733e-07,
|
| 1835 |
+
"loss": 1.5269,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.99,
|
| 1840 |
+
"grad_norm": 0.21165128196697935,
|
| 1841 |
+
"learning_rate": 6.86395920889149e-08,
|
| 1842 |
+
"loss": 1.652,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.99,
|
| 1847 |
+
"grad_norm": 0.18315613432516514,
|
| 1848 |
+
"learning_rate": 2.8368266357681194e-08,
|
| 1849 |
+
"loss": 1.6759,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 1.0,
|
| 1854 |
+
"grad_norm": 0.21177038438817372,
|
| 1855 |
+
"learning_rate": 5.603820791755254e-09,
|
| 1856 |
+
"loss": 1.7057,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 1.0,
|
| 1861 |
+
"step": 1319,
|
| 1862 |
+
"total_flos": 4147887483125760.0,
|
| 1863 |
+
"train_loss": 1.7167340969840534,
|
| 1864 |
+
"train_runtime": 8001.3069,
|
| 1865 |
+
"train_samples_per_second": 0.659,
|
| 1866 |
+
"train_steps_per_second": 0.165
|
| 1867 |
+
}
|
| 1868 |
+
],
|
| 1869 |
+
"logging_steps": 5,
|
| 1870 |
+
"max_steps": 1319,
|
| 1871 |
+
"num_input_tokens_seen": 0,
|
| 1872 |
+
"num_train_epochs": 1,
|
| 1873 |
+
"save_steps": 100,
|
| 1874 |
+
"total_flos": 4147887483125760.0,
|
| 1875 |
+
"train_batch_size": 1,
|
| 1876 |
+
"trial_name": null,
|
| 1877 |
+
"trial_params": null
|
| 1878 |
+
}
|