Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +12 -0
- runs/Sep30_19-30-07_COE-CS-sv003/events.out.tfevents.1727724668.COE-CS-sv003.1359452.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2380 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: other
|
| 4 |
+
base_model: Qwen/Qwen1.5-0.5B-Chat
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: qwen_feedback_dirty_0.5
|
| 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 |
+
# qwen_feedback_dirty_0.5
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [Qwen/Qwen1.5-0.5B-Chat](https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat) on the None dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 1e-05
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
+
- eval_batch_size: 1
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- total_eval_batch_size: 4
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- Transformers 4.45.1
|
| 58 |
+
- Pytorch 2.4.1+cu121
|
| 59 |
+
- Datasets 3.0.1
|
| 60 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 80925639573504.0,
|
| 4 |
+
"train_loss": 1.0962458014272911,
|
| 5 |
+
"train_runtime": 1740.9834,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 15.263,
|
| 8 |
+
"train_steps_per_second": 0.954
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"top_p": 0.8,
|
| 11 |
+
"transformers_version": "4.45.1"
|
| 12 |
+
}
|
runs/Sep30_19-30-07_COE-CS-sv003/events.out.tfevents.1727724668.COE-CS-sv003.1359452.0
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:ac05281c957dd653c887dbaae423f3ee8268433eed2adee932f6ef859901f57a
|
| 3 |
+
size 76502
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 80925639573504.0,
|
| 4 |
+
"train_loss": 1.0962458014272911,
|
| 5 |
+
"train_runtime": 1740.9834,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 15.263,
|
| 8 |
+
"train_steps_per_second": 0.954
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2380 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1661,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006020469596628537,
|
| 13 |
+
"grad_norm": 27.22392857163454,
|
| 14 |
+
"learning_rate": 5.98802395209581e-08,
|
| 15 |
+
"loss": 1.971,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0030102347983142685,
|
| 20 |
+
"grad_norm": 13.65386668389746,
|
| 21 |
+
"learning_rate": 2.9940119760479047e-07,
|
| 22 |
+
"loss": 1.9968,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006020469596628537,
|
| 27 |
+
"grad_norm": 12.456315967368859,
|
| 28 |
+
"learning_rate": 5.988023952095809e-07,
|
| 29 |
+
"loss": 1.9813,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009030704394942806,
|
| 34 |
+
"grad_norm": 9.02936644779285,
|
| 35 |
+
"learning_rate": 8.982035928143713e-07,
|
| 36 |
+
"loss": 1.9658,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012040939193257074,
|
| 41 |
+
"grad_norm": 7.460657498700517,
|
| 42 |
+
"learning_rate": 1.1976047904191619e-06,
|
| 43 |
+
"loss": 1.9351,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.015051173991571343,
|
| 48 |
+
"grad_norm": 10.46692979910004,
|
| 49 |
+
"learning_rate": 1.4970059880239521e-06,
|
| 50 |
+
"loss": 1.892,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.018061408789885613,
|
| 55 |
+
"grad_norm": 3.106659804143662,
|
| 56 |
+
"learning_rate": 1.7964071856287426e-06,
|
| 57 |
+
"loss": 1.8184,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02107164358819988,
|
| 62 |
+
"grad_norm": 3.4501878453247015,
|
| 63 |
+
"learning_rate": 2.095808383233533e-06,
|
| 64 |
+
"loss": 1.7497,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.024081878386514148,
|
| 69 |
+
"grad_norm": 6.687362074963905,
|
| 70 |
+
"learning_rate": 2.3952095808383237e-06,
|
| 71 |
+
"loss": 1.7088,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.027092113184828417,
|
| 76 |
+
"grad_norm": 1.7523213563959052,
|
| 77 |
+
"learning_rate": 2.694610778443114e-06,
|
| 78 |
+
"loss": 1.693,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.030102347983142687,
|
| 83 |
+
"grad_norm": 7.280651812563955,
|
| 84 |
+
"learning_rate": 2.9940119760479042e-06,
|
| 85 |
+
"loss": 1.6146,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.033112582781456956,
|
| 90 |
+
"grad_norm": 1.9204330636460694,
|
| 91 |
+
"learning_rate": 3.2934131736526947e-06,
|
| 92 |
+
"loss": 1.5607,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.036122817579771226,
|
| 97 |
+
"grad_norm": 4.242920324801297,
|
| 98 |
+
"learning_rate": 3.592814371257485e-06,
|
| 99 |
+
"loss": 1.5187,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03913305237808549,
|
| 104 |
+
"grad_norm": 3.221869314237582,
|
| 105 |
+
"learning_rate": 3.892215568862276e-06,
|
| 106 |
+
"loss": 1.4559,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04214328717639976,
|
| 111 |
+
"grad_norm": 1.1211647421609696,
|
| 112 |
+
"learning_rate": 4.191616766467066e-06,
|
| 113 |
+
"loss": 1.4626,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04515352197471403,
|
| 118 |
+
"grad_norm": 18.99681558637136,
|
| 119 |
+
"learning_rate": 4.4910179640718566e-06,
|
| 120 |
+
"loss": 1.4324,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.048163756773028296,
|
| 125 |
+
"grad_norm": 0.9850763475643952,
|
| 126 |
+
"learning_rate": 4.7904191616766475e-06,
|
| 127 |
+
"loss": 1.3939,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.051173991571342566,
|
| 132 |
+
"grad_norm": 1.6779178850544632,
|
| 133 |
+
"learning_rate": 5.0898203592814375e-06,
|
| 134 |
+
"loss": 1.4141,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.054184226369656835,
|
| 139 |
+
"grad_norm": 0.9258382254032997,
|
| 140 |
+
"learning_rate": 5.389221556886228e-06,
|
| 141 |
+
"loss": 1.3921,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.057194461167971104,
|
| 146 |
+
"grad_norm": 0.9127215663526058,
|
| 147 |
+
"learning_rate": 5.6886227544910184e-06,
|
| 148 |
+
"loss": 1.3879,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.060204695966285374,
|
| 153 |
+
"grad_norm": 0.9765773402128775,
|
| 154 |
+
"learning_rate": 5.9880239520958085e-06,
|
| 155 |
+
"loss": 1.3614,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06321493076459964,
|
| 160 |
+
"grad_norm": 0.9543473669785393,
|
| 161 |
+
"learning_rate": 6.2874251497005985e-06,
|
| 162 |
+
"loss": 1.3517,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06622516556291391,
|
| 167 |
+
"grad_norm": 0.9289846325699406,
|
| 168 |
+
"learning_rate": 6.586826347305389e-06,
|
| 169 |
+
"loss": 1.3372,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06923540036122817,
|
| 174 |
+
"grad_norm": 0.9873011952453609,
|
| 175 |
+
"learning_rate": 6.88622754491018e-06,
|
| 176 |
+
"loss": 1.3378,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07224563515954245,
|
| 181 |
+
"grad_norm": 1.0087695630072344,
|
| 182 |
+
"learning_rate": 7.18562874251497e-06,
|
| 183 |
+
"loss": 1.3286,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07525586995785671,
|
| 188 |
+
"grad_norm": 2.041087321823113,
|
| 189 |
+
"learning_rate": 7.485029940119761e-06,
|
| 190 |
+
"loss": 1.33,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07826610475617098,
|
| 195 |
+
"grad_norm": 4.129687925158122,
|
| 196 |
+
"learning_rate": 7.784431137724551e-06,
|
| 197 |
+
"loss": 1.3031,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08127633955448525,
|
| 202 |
+
"grad_norm": 0.9348500545655409,
|
| 203 |
+
"learning_rate": 8.083832335329342e-06,
|
| 204 |
+
"loss": 1.3204,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08428657435279951,
|
| 209 |
+
"grad_norm": 0.9282851253114677,
|
| 210 |
+
"learning_rate": 8.383233532934131e-06,
|
| 211 |
+
"loss": 1.2841,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08729680915111379,
|
| 216 |
+
"grad_norm": 0.9100216621816694,
|
| 217 |
+
"learning_rate": 8.682634730538922e-06,
|
| 218 |
+
"loss": 1.2858,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09030704394942805,
|
| 223 |
+
"grad_norm": 0.9981824581989874,
|
| 224 |
+
"learning_rate": 8.982035928143713e-06,
|
| 225 |
+
"loss": 1.2704,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09331727874774233,
|
| 230 |
+
"grad_norm": 0.9537988209000966,
|
| 231 |
+
"learning_rate": 9.281437125748504e-06,
|
| 232 |
+
"loss": 1.3038,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09632751354605659,
|
| 237 |
+
"grad_norm": 0.9601755129617946,
|
| 238 |
+
"learning_rate": 9.580838323353295e-06,
|
| 239 |
+
"loss": 1.2969,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09933774834437085,
|
| 244 |
+
"grad_norm": 0.9847338523503186,
|
| 245 |
+
"learning_rate": 9.880239520958084e-06,
|
| 246 |
+
"loss": 1.2694,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10234798314268513,
|
| 251 |
+
"grad_norm": 0.9800990746472159,
|
| 252 |
+
"learning_rate": 9.999900509954779e-06,
|
| 253 |
+
"loss": 1.2673,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1053582179409994,
|
| 258 |
+
"grad_norm": 1.0362064979246783,
|
| 259 |
+
"learning_rate": 9.999292529572152e-06,
|
| 260 |
+
"loss": 1.2687,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10836845273931367,
|
| 265 |
+
"grad_norm": 0.97331445513859,
|
| 266 |
+
"learning_rate": 9.998131908181262e-06,
|
| 267 |
+
"loss": 1.2693,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11137868753762793,
|
| 272 |
+
"grad_norm": 0.9597440336445656,
|
| 273 |
+
"learning_rate": 9.996418774081658e-06,
|
| 274 |
+
"loss": 1.2523,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11438892233594221,
|
| 279 |
+
"grad_norm": 1.0622289859021092,
|
| 280 |
+
"learning_rate": 9.994153316649769e-06,
|
| 281 |
+
"loss": 1.2726,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11739915713425647,
|
| 286 |
+
"grad_norm": 0.9667990321619035,
|
| 287 |
+
"learning_rate": 9.991335786317964e-06,
|
| 288 |
+
"loss": 1.2551,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12040939193257075,
|
| 293 |
+
"grad_norm": 1.0072941793405643,
|
| 294 |
+
"learning_rate": 9.987966494546873e-06,
|
| 295 |
+
"loss": 1.2255,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12341962673088501,
|
| 300 |
+
"grad_norm": 0.9596767117485965,
|
| 301 |
+
"learning_rate": 9.984045813790959e-06,
|
| 302 |
+
"loss": 1.2465,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12642986152919927,
|
| 307 |
+
"grad_norm": 1.0015445744889278,
|
| 308 |
+
"learning_rate": 9.979574177457337e-06,
|
| 309 |
+
"loss": 1.2463,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12944009632751355,
|
| 314 |
+
"grad_norm": 0.9779464829184125,
|
| 315 |
+
"learning_rate": 9.974552079857873e-06,
|
| 316 |
+
"loss": 1.2423,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.13245033112582782,
|
| 321 |
+
"grad_norm": 0.9588810325174535,
|
| 322 |
+
"learning_rate": 9.968980076154533e-06,
|
| 323 |
+
"loss": 1.2515,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.13546056592414207,
|
| 328 |
+
"grad_norm": 0.9762836295455931,
|
| 329 |
+
"learning_rate": 9.962858782298023e-06,
|
| 330 |
+
"loss": 1.2317,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.13847080072245635,
|
| 335 |
+
"grad_norm": 0.9455765624623369,
|
| 336 |
+
"learning_rate": 9.956188874959686e-06,
|
| 337 |
+
"loss": 1.23,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14148103552077063,
|
| 342 |
+
"grad_norm": 1.0254791337431468,
|
| 343 |
+
"learning_rate": 9.948971091456715e-06,
|
| 344 |
+
"loss": 1.2313,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1444912703190849,
|
| 349 |
+
"grad_norm": 1.0494300760638828,
|
| 350 |
+
"learning_rate": 9.941206229670634e-06,
|
| 351 |
+
"loss": 1.2203,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.14750150511739915,
|
| 356 |
+
"grad_norm": 1.0130193836029537,
|
| 357 |
+
"learning_rate": 9.932895147959106e-06,
|
| 358 |
+
"loss": 1.2038,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.15051173991571343,
|
| 363 |
+
"grad_norm": 1.0190268775942266,
|
| 364 |
+
"learning_rate": 9.924038765061042e-06,
|
| 365 |
+
"loss": 1.1867,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1535219747140277,
|
| 370 |
+
"grad_norm": 1.0637623816845347,
|
| 371 |
+
"learning_rate": 9.91463805999504e-06,
|
| 372 |
+
"loss": 1.2114,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.15653220951234195,
|
| 377 |
+
"grad_norm": 1.0582426619825864,
|
| 378 |
+
"learning_rate": 9.904694071951167e-06,
|
| 379 |
+
"loss": 1.2119,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.15954244431065623,
|
| 384 |
+
"grad_norm": 1.0051521209552554,
|
| 385 |
+
"learning_rate": 9.894207900176074e-06,
|
| 386 |
+
"loss": 1.195,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1625526791089705,
|
| 391 |
+
"grad_norm": 1.006499396359629,
|
| 392 |
+
"learning_rate": 9.883180703851488e-06,
|
| 393 |
+
"loss": 1.2009,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16556291390728478,
|
| 398 |
+
"grad_norm": 0.9907891015612423,
|
| 399 |
+
"learning_rate": 9.871613701966067e-06,
|
| 400 |
+
"loss": 1.1802,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.16857314870559903,
|
| 405 |
+
"grad_norm": 0.9353673363284244,
|
| 406 |
+
"learning_rate": 9.859508173180653e-06,
|
| 407 |
+
"loss": 1.1715,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1715833835039133,
|
| 412 |
+
"grad_norm": 0.9848000031544851,
|
| 413 |
+
"learning_rate": 9.846865455686915e-06,
|
| 414 |
+
"loss": 1.1751,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17459361830222758,
|
| 419 |
+
"grad_norm": 0.9778599958159084,
|
| 420 |
+
"learning_rate": 9.833686947059436e-06,
|
| 421 |
+
"loss": 1.1659,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17760385310054183,
|
| 426 |
+
"grad_norm": 1.0197914549232654,
|
| 427 |
+
"learning_rate": 9.819974104101198e-06,
|
| 428 |
+
"loss": 1.1609,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1806140878988561,
|
| 433 |
+
"grad_norm": 0.9700422577443836,
|
| 434 |
+
"learning_rate": 9.80572844268256e-06,
|
| 435 |
+
"loss": 1.1489,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18362432269717038,
|
| 440 |
+
"grad_norm": 1.0249498510324442,
|
| 441 |
+
"learning_rate": 9.790951537573686e-06,
|
| 442 |
+
"loss": 1.1829,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.18663455749548466,
|
| 447 |
+
"grad_norm": 1.0644946420369092,
|
| 448 |
+
"learning_rate": 9.775645022270448e-06,
|
| 449 |
+
"loss": 1.164,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1896447922937989,
|
| 454 |
+
"grad_norm": 0.9867050366956028,
|
| 455 |
+
"learning_rate": 9.759810588813872e-06,
|
| 456 |
+
"loss": 1.1648,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.19265502709211318,
|
| 461 |
+
"grad_norm": 1.1128215197644398,
|
| 462 |
+
"learning_rate": 9.743449987603082e-06,
|
| 463 |
+
"loss": 1.16,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.19566526189042746,
|
| 468 |
+
"grad_norm": 1.0379238093017742,
|
| 469 |
+
"learning_rate": 9.726565027201813e-06,
|
| 470 |
+
"loss": 1.173,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.1986754966887417,
|
| 475 |
+
"grad_norm": 1.037290324574154,
|
| 476 |
+
"learning_rate": 9.70915757413847e-06,
|
| 477 |
+
"loss": 1.1611,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.20168573148705599,
|
| 482 |
+
"grad_norm": 1.0384997352637115,
|
| 483 |
+
"learning_rate": 9.691229552699817e-06,
|
| 484 |
+
"loss": 1.1398,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.20469596628537026,
|
| 489 |
+
"grad_norm": 1.1360507992714308,
|
| 490 |
+
"learning_rate": 9.672782944718234e-06,
|
| 491 |
+
"loss": 1.1652,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.20770620108368454,
|
| 496 |
+
"grad_norm": 1.055086294709249,
|
| 497 |
+
"learning_rate": 9.65381978935266e-06,
|
| 498 |
+
"loss": 1.139,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2107164358819988,
|
| 503 |
+
"grad_norm": 1.0570445286606438,
|
| 504 |
+
"learning_rate": 9.634342182863163e-06,
|
| 505 |
+
"loss": 1.1155,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.21372667068031306,
|
| 510 |
+
"grad_norm": 1.0499986459043307,
|
| 511 |
+
"learning_rate": 9.614352278379217e-06,
|
| 512 |
+
"loss": 1.1484,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.21673690547862734,
|
| 517 |
+
"grad_norm": 1.072674005812888,
|
| 518 |
+
"learning_rate": 9.593852285661684e-06,
|
| 519 |
+
"loss": 1.1546,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2197471402769416,
|
| 524 |
+
"grad_norm": 1.026680216356016,
|
| 525 |
+
"learning_rate": 9.572844470858537e-06,
|
| 526 |
+
"loss": 1.1362,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.22275737507525586,
|
| 531 |
+
"grad_norm": 1.154893182120373,
|
| 532 |
+
"learning_rate": 9.551331156254358e-06,
|
| 533 |
+
"loss": 1.1375,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22576760987357014,
|
| 538 |
+
"grad_norm": 1.030915882210545,
|
| 539 |
+
"learning_rate": 9.529314720013618e-06,
|
| 540 |
+
"loss": 1.1246,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.22877784467188442,
|
| 545 |
+
"grad_norm": 1.0275709672820799,
|
| 546 |
+
"learning_rate": 9.506797595917787e-06,
|
| 547 |
+
"loss": 1.1306,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23178807947019867,
|
| 552 |
+
"grad_norm": 1.0612700071573922,
|
| 553 |
+
"learning_rate": 9.483782273096295e-06,
|
| 554 |
+
"loss": 1.1291,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.23479831426851294,
|
| 559 |
+
"grad_norm": 1.0249725155040352,
|
| 560 |
+
"learning_rate": 9.460271295751373e-06,
|
| 561 |
+
"loss": 1.1428,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.23780854906682722,
|
| 566 |
+
"grad_norm": 1.01241424823746,
|
| 567 |
+
"learning_rate": 9.436267262876808e-06,
|
| 568 |
+
"loss": 1.1195,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2408187838651415,
|
| 573 |
+
"grad_norm": 1.0678914761402232,
|
| 574 |
+
"learning_rate": 9.411772827970642e-06,
|
| 575 |
+
"loss": 1.1164,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24382901866345574,
|
| 580 |
+
"grad_norm": 1.063653307977233,
|
| 581 |
+
"learning_rate": 9.38679069874184e-06,
|
| 582 |
+
"loss": 1.1184,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.24683925346177002,
|
| 587 |
+
"grad_norm": 1.0166853640511404,
|
| 588 |
+
"learning_rate": 9.36132363681097e-06,
|
| 589 |
+
"loss": 1.1186,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2498494882600843,
|
| 594 |
+
"grad_norm": 1.0224474900808593,
|
| 595 |
+
"learning_rate": 9.335374457404928e-06,
|
| 596 |
+
"loss": 1.1259,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.25285972305839854,
|
| 601 |
+
"grad_norm": 1.025396194648956,
|
| 602 |
+
"learning_rate": 9.308946029045726e-06,
|
| 603 |
+
"loss": 1.0992,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25586995785671285,
|
| 608 |
+
"grad_norm": 1.0666081761901238,
|
| 609 |
+
"learning_rate": 9.282041273233402e-06,
|
| 610 |
+
"loss": 1.0911,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2588801926550271,
|
| 615 |
+
"grad_norm": 1.036681992509278,
|
| 616 |
+
"learning_rate": 9.254663164123052e-06,
|
| 617 |
+
"loss": 1.0926,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.26189042745334135,
|
| 622 |
+
"grad_norm": 0.9981467921863433,
|
| 623 |
+
"learning_rate": 9.226814728196072e-06,
|
| 624 |
+
"loss": 1.0945,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.26490066225165565,
|
| 629 |
+
"grad_norm": 0.9968534019134379,
|
| 630 |
+
"learning_rate": 9.198499043925591e-06,
|
| 631 |
+
"loss": 1.1036,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2679108970499699,
|
| 636 |
+
"grad_norm": 1.0126539822191232,
|
| 637 |
+
"learning_rate": 9.169719241436162e-06,
|
| 638 |
+
"loss": 1.0986,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.27092113184828415,
|
| 643 |
+
"grad_norm": 1.0017070369951004,
|
| 644 |
+
"learning_rate": 9.14047850215775e-06,
|
| 645 |
+
"loss": 1.0811,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.27393136664659845,
|
| 650 |
+
"grad_norm": 1.0316940440472975,
|
| 651 |
+
"learning_rate": 9.110780058474052e-06,
|
| 652 |
+
"loss": 1.0751,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2769416014449127,
|
| 657 |
+
"grad_norm": 1.0365583639260507,
|
| 658 |
+
"learning_rate": 9.080627193365155e-06,
|
| 659 |
+
"loss": 1.0937,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.27995183624322695,
|
| 664 |
+
"grad_norm": 0.9885372663237562,
|
| 665 |
+
"learning_rate": 9.050023240044649e-06,
|
| 666 |
+
"loss": 1.0932,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.28296207104154125,
|
| 671 |
+
"grad_norm": 0.9917660998350917,
|
| 672 |
+
"learning_rate": 9.018971581591141e-06,
|
| 673 |
+
"loss": 1.0805,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2859723058398555,
|
| 678 |
+
"grad_norm": 0.9746303627143975,
|
| 679 |
+
"learning_rate": 8.987475650574289e-06,
|
| 680 |
+
"loss": 1.1045,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2889825406381698,
|
| 685 |
+
"grad_norm": 1.0633991398130442,
|
| 686 |
+
"learning_rate": 8.955538928675343e-06,
|
| 687 |
+
"loss": 1.0665,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.29199277543648405,
|
| 692 |
+
"grad_norm": 1.0294806019647922,
|
| 693 |
+
"learning_rate": 8.923164946302274e-06,
|
| 694 |
+
"loss": 1.0838,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2950030102347983,
|
| 699 |
+
"grad_norm": 1.1636067238442809,
|
| 700 |
+
"learning_rate": 8.890357282199504e-06,
|
| 701 |
+
"loss": 1.0933,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2980132450331126,
|
| 706 |
+
"grad_norm": 1.0725517112186367,
|
| 707 |
+
"learning_rate": 8.857119563052301e-06,
|
| 708 |
+
"loss": 1.0633,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.30102347983142685,
|
| 713 |
+
"grad_norm": 0.9985063194122187,
|
| 714 |
+
"learning_rate": 8.823455463085873e-06,
|
| 715 |
+
"loss": 1.0776,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3040337146297411,
|
| 720 |
+
"grad_norm": 1.011513088054301,
|
| 721 |
+
"learning_rate": 8.789368703659199e-06,
|
| 722 |
+
"loss": 1.0851,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3070439494280554,
|
| 727 |
+
"grad_norm": 0.9961319702573649,
|
| 728 |
+
"learning_rate": 8.754863052853658e-06,
|
| 729 |
+
"loss": 1.0773,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.31005418422636966,
|
| 734 |
+
"grad_norm": 1.0537373514794217,
|
| 735 |
+
"learning_rate": 8.719942325056496e-06,
|
| 736 |
+
"loss": 1.0544,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3130644190246839,
|
| 741 |
+
"grad_norm": 0.9560546370265528,
|
| 742 |
+
"learning_rate": 8.68461038053916e-06,
|
| 743 |
+
"loss": 1.0604,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3160746538229982,
|
| 748 |
+
"grad_norm": 1.0998924700572033,
|
| 749 |
+
"learning_rate": 8.648871125030576e-06,
|
| 750 |
+
"loss": 1.0668,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.31908488862131246,
|
| 755 |
+
"grad_norm": 0.9922592199137708,
|
| 756 |
+
"learning_rate": 8.612728509285395e-06,
|
| 757 |
+
"loss": 1.0564,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3220951234196267,
|
| 762 |
+
"grad_norm": 0.9990925710032598,
|
| 763 |
+
"learning_rate": 8.576186528647253e-06,
|
| 764 |
+
"loss": 1.0765,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.325105358217941,
|
| 769 |
+
"grad_norm": 0.9867645343873487,
|
| 770 |
+
"learning_rate": 8.53924922260712e-06,
|
| 771 |
+
"loss": 1.0679,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.32811559301625526,
|
| 776 |
+
"grad_norm": 1.0504215303129572,
|
| 777 |
+
"learning_rate": 8.501920674356755e-06,
|
| 778 |
+
"loss": 1.0464,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.33112582781456956,
|
| 783 |
+
"grad_norm": 1.0003566271578537,
|
| 784 |
+
"learning_rate": 8.46420501033733e-06,
|
| 785 |
+
"loss": 1.0602,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3341360626128838,
|
| 790 |
+
"grad_norm": 1.012902103958297,
|
| 791 |
+
"learning_rate": 8.42610639978329e-06,
|
| 792 |
+
"loss": 1.0562,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.33714629741119806,
|
| 797 |
+
"grad_norm": 0.9689824382443799,
|
| 798 |
+
"learning_rate": 8.387629054261454e-06,
|
| 799 |
+
"loss": 1.0511,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.34015653220951236,
|
| 804 |
+
"grad_norm": 0.9592977237689682,
|
| 805 |
+
"learning_rate": 8.348777227205462e-06,
|
| 806 |
+
"loss": 1.0496,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3431667670078266,
|
| 811 |
+
"grad_norm": 1.0133498095085518,
|
| 812 |
+
"learning_rate": 8.309555213445583e-06,
|
| 813 |
+
"loss": 1.0738,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.34617700180614086,
|
| 818 |
+
"grad_norm": 0.9599492179758512,
|
| 819 |
+
"learning_rate": 8.269967348733947e-06,
|
| 820 |
+
"loss": 1.0446,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.34918723660445516,
|
| 825 |
+
"grad_norm": 0.9681301740683113,
|
| 826 |
+
"learning_rate": 8.230018009265255e-06,
|
| 827 |
+
"loss": 1.0596,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3521974714027694,
|
| 832 |
+
"grad_norm": 0.9373524338466983,
|
| 833 |
+
"learning_rate": 8.189711611193012e-06,
|
| 834 |
+
"loss": 1.051,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.35520770620108366,
|
| 839 |
+
"grad_norm": 0.9670245571139093,
|
| 840 |
+
"learning_rate": 8.149052610141357e-06,
|
| 841 |
+
"loss": 1.0429,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.35821794099939797,
|
| 846 |
+
"grad_norm": 0.9510561814037846,
|
| 847 |
+
"learning_rate": 8.108045500712518e-06,
|
| 848 |
+
"loss": 1.058,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3612281757977122,
|
| 853 |
+
"grad_norm": 0.9936813505342955,
|
| 854 |
+
"learning_rate": 8.066694815989961e-06,
|
| 855 |
+
"loss": 1.0603,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.36423841059602646,
|
| 860 |
+
"grad_norm": 0.9843981727904988,
|
| 861 |
+
"learning_rate": 8.025005127037282e-06,
|
| 862 |
+
"loss": 1.0606,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.36724864539434077,
|
| 867 |
+
"grad_norm": 0.9874316056017197,
|
| 868 |
+
"learning_rate": 7.982981042392907e-06,
|
| 869 |
+
"loss": 1.0493,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.370258880192655,
|
| 874 |
+
"grad_norm": 0.9836042531874883,
|
| 875 |
+
"learning_rate": 7.940627207560655e-06,
|
| 876 |
+
"loss": 1.0419,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3732691149909693,
|
| 881 |
+
"grad_norm": 0.9349031264216233,
|
| 882 |
+
"learning_rate": 7.897948304496189e-06,
|
| 883 |
+
"loss": 1.0634,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.37627934978928357,
|
| 888 |
+
"grad_norm": 0.9504127117294803,
|
| 889 |
+
"learning_rate": 7.854949051089467e-06,
|
| 890 |
+
"loss": 1.0665,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3792895845875978,
|
| 895 |
+
"grad_norm": 0.9222019819892415,
|
| 896 |
+
"learning_rate": 7.811634200643202e-06,
|
| 897 |
+
"loss": 1.072,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3822998193859121,
|
| 902 |
+
"grad_norm": 0.9584859569960068,
|
| 903 |
+
"learning_rate": 7.768008541347423e-06,
|
| 904 |
+
"loss": 1.021,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.38531005418422637,
|
| 909 |
+
"grad_norm": 0.9322129259432775,
|
| 910 |
+
"learning_rate": 7.72407689575016e-06,
|
| 911 |
+
"loss": 1.05,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3883202889825406,
|
| 916 |
+
"grad_norm": 0.9418042647481683,
|
| 917 |
+
"learning_rate": 7.67984412022434e-06,
|
| 918 |
+
"loss": 1.0451,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3913305237808549,
|
| 923 |
+
"grad_norm": 0.9966983630078914,
|
| 924 |
+
"learning_rate": 7.635315104430959e-06,
|
| 925 |
+
"loss": 1.0355,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.39434075857916917,
|
| 930 |
+
"grad_norm": 0.9572847678319176,
|
| 931 |
+
"learning_rate": 7.5904947707785434e-06,
|
| 932 |
+
"loss": 1.0299,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3973509933774834,
|
| 937 |
+
"grad_norm": 0.9275188841478791,
|
| 938 |
+
"learning_rate": 7.545388073879018e-06,
|
| 939 |
+
"loss": 1.058,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4003612281757977,
|
| 944 |
+
"grad_norm": 0.9417371682221568,
|
| 945 |
+
"learning_rate": 7.500000000000001e-06,
|
| 946 |
+
"loss": 1.0397,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.40337146297411197,
|
| 951 |
+
"grad_norm": 0.9681489741098089,
|
| 952 |
+
"learning_rate": 7.454335566513603e-06,
|
| 953 |
+
"loss": 1.0465,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4063816977724263,
|
| 958 |
+
"grad_norm": 0.9595108028776653,
|
| 959 |
+
"learning_rate": 7.408399821341787e-06,
|
| 960 |
+
"loss": 1.0534,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4093919325707405,
|
| 965 |
+
"grad_norm": 0.9062681118546124,
|
| 966 |
+
"learning_rate": 7.362197842398355e-06,
|
| 967 |
+
"loss": 1.0327,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4124021673690548,
|
| 972 |
+
"grad_norm": 0.9467262280598835,
|
| 973 |
+
"learning_rate": 7.315734737027612e-06,
|
| 974 |
+
"loss": 1.0437,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4154124021673691,
|
| 979 |
+
"grad_norm": 0.9509979239409292,
|
| 980 |
+
"learning_rate": 7.2690156414397775e-06,
|
| 981 |
+
"loss": 1.0234,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4184226369656833,
|
| 986 |
+
"grad_norm": 0.9515966339436767,
|
| 987 |
+
"learning_rate": 7.22204572014322e-06,
|
| 988 |
+
"loss": 1.0322,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4214328717639976,
|
| 993 |
+
"grad_norm": 0.9727663910614524,
|
| 994 |
+
"learning_rate": 7.174830165373542e-06,
|
| 995 |
+
"loss": 1.0418,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4244431065623119,
|
| 1000 |
+
"grad_norm": 0.9381652326646253,
|
| 1001 |
+
"learning_rate": 7.127374196519616e-06,
|
| 1002 |
+
"loss": 1.0321,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4274533413606261,
|
| 1007 |
+
"grad_norm": 0.9699886941897933,
|
| 1008 |
+
"learning_rate": 7.079683059546607e-06,
|
| 1009 |
+
"loss": 1.0449,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4304635761589404,
|
| 1014 |
+
"grad_norm": 0.9241563520898756,
|
| 1015 |
+
"learning_rate": 7.031762026416074e-06,
|
| 1016 |
+
"loss": 1.0246,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4334738109572547,
|
| 1021 |
+
"grad_norm": 0.993265487831661,
|
| 1022 |
+
"learning_rate": 6.983616394503177e-06,
|
| 1023 |
+
"loss": 1.0431,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.43648404575556893,
|
| 1028 |
+
"grad_norm": 0.9628627003488671,
|
| 1029 |
+
"learning_rate": 6.9352514860110876e-06,
|
| 1030 |
+
"loss": 1.0662,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.4394942805538832,
|
| 1035 |
+
"grad_norm": 0.9461416922766445,
|
| 1036 |
+
"learning_rate": 6.886672647382653e-06,
|
| 1037 |
+
"loss": 1.0316,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4425045153521975,
|
| 1042 |
+
"grad_norm": 0.9806980915933521,
|
| 1043 |
+
"learning_rate": 6.837885248709386e-06,
|
| 1044 |
+
"loss": 1.064,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.44551475015051173,
|
| 1049 |
+
"grad_norm": 0.928373225003967,
|
| 1050 |
+
"learning_rate": 6.788894683137822e-06,
|
| 1051 |
+
"loss": 1.0276,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.44852498494882603,
|
| 1056 |
+
"grad_norm": 0.9575671889236174,
|
| 1057 |
+
"learning_rate": 6.739706366273346e-06,
|
| 1058 |
+
"loss": 1.038,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4515352197471403,
|
| 1063 |
+
"grad_norm": 0.9046994590336265,
|
| 1064 |
+
"learning_rate": 6.690325735581532e-06,
|
| 1065 |
+
"loss": 1.0282,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.45454545454545453,
|
| 1070 |
+
"grad_norm": 0.9151429388919339,
|
| 1071 |
+
"learning_rate": 6.640758249787067e-06,
|
| 1072 |
+
"loss": 1.048,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.45755568934376883,
|
| 1077 |
+
"grad_norm": 0.9419132302997139,
|
| 1078 |
+
"learning_rate": 6.591009388270315e-06,
|
| 1079 |
+
"loss": 1.0384,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4605659241420831,
|
| 1084 |
+
"grad_norm": 1.0005185024840373,
|
| 1085 |
+
"learning_rate": 6.54108465046161e-06,
|
| 1086 |
+
"loss": 1.04,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.46357615894039733,
|
| 1091 |
+
"grad_norm": 0.9280344189136297,
|
| 1092 |
+
"learning_rate": 6.490989555233328e-06,
|
| 1093 |
+
"loss": 1.0355,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.46658639373871164,
|
| 1098 |
+
"grad_norm": 0.9191484105559773,
|
| 1099 |
+
"learning_rate": 6.440729640289809e-06,
|
| 1100 |
+
"loss": 1.0249,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4695966285370259,
|
| 1105 |
+
"grad_norm": 0.9528558636067395,
|
| 1106 |
+
"learning_rate": 6.3903104615551956e-06,
|
| 1107 |
+
"loss": 1.0191,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.47260686333534013,
|
| 1112 |
+
"grad_norm": 0.8848506972001927,
|
| 1113 |
+
"learning_rate": 6.3397375925592675e-06,
|
| 1114 |
+
"loss": 1.0023,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.47561709813365444,
|
| 1119 |
+
"grad_norm": 0.8812677285723143,
|
| 1120 |
+
"learning_rate": 6.289016623821308e-06,
|
| 1121 |
+
"loss": 1.0378,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4786273329319687,
|
| 1126 |
+
"grad_norm": 0.9103487891871948,
|
| 1127 |
+
"learning_rate": 6.2381531622321234e-06,
|
| 1128 |
+
"loss": 1.0393,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.481637567730283,
|
| 1133 |
+
"grad_norm": 0.9337809046875775,
|
| 1134 |
+
"learning_rate": 6.18715283043422e-06,
|
| 1135 |
+
"loss": 1.0287,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.48464780252859724,
|
| 1140 |
+
"grad_norm": 0.9131887229741218,
|
| 1141 |
+
"learning_rate": 6.136021266200271e-06,
|
| 1142 |
+
"loss": 1.0303,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4876580373269115,
|
| 1147 |
+
"grad_norm": 0.9069697577426293,
|
| 1148 |
+
"learning_rate": 6.084764121809878e-06,
|
| 1149 |
+
"loss": 1.0307,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4906682721252258,
|
| 1154 |
+
"grad_norm": 0.940336145640042,
|
| 1155 |
+
"learning_rate": 6.033387063424765e-06,
|
| 1156 |
+
"loss": 1.0198,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.49367850692354004,
|
| 1161 |
+
"grad_norm": 0.898801668461195,
|
| 1162 |
+
"learning_rate": 5.9818957704624046e-06,
|
| 1163 |
+
"loss": 1.0458,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4966887417218543,
|
| 1168 |
+
"grad_norm": 0.8849487338293404,
|
| 1169 |
+
"learning_rate": 5.930295934968197e-06,
|
| 1170 |
+
"loss": 1.022,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4996989765201686,
|
| 1175 |
+
"grad_norm": 1.3316984335165358,
|
| 1176 |
+
"learning_rate": 5.878593260986256e-06,
|
| 1177 |
+
"loss": 1.0175,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5027092113184829,
|
| 1182 |
+
"grad_norm": 1.0767616487075078,
|
| 1183 |
+
"learning_rate": 5.8267934639288525e-06,
|
| 1184 |
+
"loss": 1.0413,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5057194461167971,
|
| 1189 |
+
"grad_norm": 0.9217653620486148,
|
| 1190 |
+
"learning_rate": 5.77490226994462e-06,
|
| 1191 |
+
"loss": 1.0432,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5087296809151114,
|
| 1196 |
+
"grad_norm": 0.9097013059314509,
|
| 1197 |
+
"learning_rate": 5.722925415285555e-06,
|
| 1198 |
+
"loss": 1.0502,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5117399157134257,
|
| 1203 |
+
"grad_norm": 0.9555445158407673,
|
| 1204 |
+
"learning_rate": 5.670868645672916e-06,
|
| 1205 |
+
"loss": 1.0412,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5147501505117399,
|
| 1210 |
+
"grad_norm": 0.9670792048708317,
|
| 1211 |
+
"learning_rate": 5.618737715662067e-06,
|
| 1212 |
+
"loss": 1.0271,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5177603853100542,
|
| 1217 |
+
"grad_norm": 0.9197944356676366,
|
| 1218 |
+
"learning_rate": 5.566538388006351e-06,
|
| 1219 |
+
"loss": 1.0374,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5207706201083685,
|
| 1224 |
+
"grad_norm": 0.8959294606541434,
|
| 1225 |
+
"learning_rate": 5.514276433020044e-06,
|
| 1226 |
+
"loss": 1.0223,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5237808549066827,
|
| 1231 |
+
"grad_norm": 0.9227855823325931,
|
| 1232 |
+
"learning_rate": 5.461957627940489e-06,
|
| 1233 |
+
"loss": 1.0181,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.526791089704997,
|
| 1238 |
+
"grad_norm": 0.9045604297157106,
|
| 1239 |
+
"learning_rate": 5.409587756289462e-06,
|
| 1240 |
+
"loss": 1.0177,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5298013245033113,
|
| 1245 |
+
"grad_norm": 0.900720782883793,
|
| 1246 |
+
"learning_rate": 5.357172607233831e-06,
|
| 1247 |
+
"loss": 1.0164,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5328115593016255,
|
| 1252 |
+
"grad_norm": 0.9613635010205603,
|
| 1253 |
+
"learning_rate": 5.304717974945596e-06,
|
| 1254 |
+
"loss": 1.0199,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5358217940999398,
|
| 1259 |
+
"grad_norm": 0.9280538923116405,
|
| 1260 |
+
"learning_rate": 5.252229657961394e-06,
|
| 1261 |
+
"loss": 1.0194,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5388320288982541,
|
| 1266 |
+
"grad_norm": 0.8785440956248298,
|
| 1267 |
+
"learning_rate": 5.199713458541495e-06,
|
| 1268 |
+
"loss": 1.0332,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5418422636965683,
|
| 1273 |
+
"grad_norm": 0.9081415986183222,
|
| 1274 |
+
"learning_rate": 5.1471751820284e-06,
|
| 1275 |
+
"loss": 1.0117,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5448524984948826,
|
| 1280 |
+
"grad_norm": 0.8985002258714208,
|
| 1281 |
+
"learning_rate": 5.094620636205096e-06,
|
| 1282 |
+
"loss": 1.0076,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5478627332931969,
|
| 1287 |
+
"grad_norm": 0.8895782291081366,
|
| 1288 |
+
"learning_rate": 5.042055630653042e-06,
|
| 1289 |
+
"loss": 1.0191,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5508729680915111,
|
| 1294 |
+
"grad_norm": 0.9293164841008162,
|
| 1295 |
+
"learning_rate": 4.98948597610996e-06,
|
| 1296 |
+
"loss": 1.0101,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5538832028898254,
|
| 1301 |
+
"grad_norm": 17.04597932345032,
|
| 1302 |
+
"learning_rate": 4.936917483827483e-06,
|
| 1303 |
+
"loss": 0.9983,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5568934376881397,
|
| 1308 |
+
"grad_norm": 0.8910475336292096,
|
| 1309 |
+
"learning_rate": 4.884355964928767e-06,
|
| 1310 |
+
"loss": 1.0442,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5599036724864539,
|
| 1315 |
+
"grad_norm": 0.8680011029175039,
|
| 1316 |
+
"learning_rate": 4.831807229766101e-06,
|
| 1317 |
+
"loss": 1.0035,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5629139072847682,
|
| 1322 |
+
"grad_norm": 0.931237516861277,
|
| 1323 |
+
"learning_rate": 4.779277087278615e-06,
|
| 1324 |
+
"loss": 1.0285,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5659241420830825,
|
| 1329 |
+
"grad_norm": 0.9231411428094562,
|
| 1330 |
+
"learning_rate": 4.7267713443501274e-06,
|
| 1331 |
+
"loss": 1.0377,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5689343768813967,
|
| 1336 |
+
"grad_norm": 0.8658589610324005,
|
| 1337 |
+
"learning_rate": 4.67429580516724e-06,
|
| 1338 |
+
"loss": 1.0134,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.571944611679711,
|
| 1343 |
+
"grad_norm": 0.9419779516364959,
|
| 1344 |
+
"learning_rate": 4.6218562705777185e-06,
|
| 1345 |
+
"loss": 1.0126,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5749548464780253,
|
| 1350 |
+
"grad_norm": 0.9566449713214615,
|
| 1351 |
+
"learning_rate": 4.5694585374492314e-06,
|
| 1352 |
+
"loss": 1.0228,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5779650812763396,
|
| 1357 |
+
"grad_norm": 0.9543544333800563,
|
| 1358 |
+
"learning_rate": 4.517108398028566e-06,
|
| 1359 |
+
"loss": 1.04,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5809753160746538,
|
| 1364 |
+
"grad_norm": 0.9091827723851609,
|
| 1365 |
+
"learning_rate": 4.464811639301314e-06,
|
| 1366 |
+
"loss": 1.0175,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5839855508729681,
|
| 1371 |
+
"grad_norm": 22.159553622606854,
|
| 1372 |
+
"learning_rate": 4.412574042352156e-06,
|
| 1373 |
+
"loss": 1.0323,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5869957856712824,
|
| 1378 |
+
"grad_norm": 0.912382771621538,
|
| 1379 |
+
"learning_rate": 4.360401381725806e-06,
|
| 1380 |
+
"loss": 1.0206,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5900060204695966,
|
| 1385 |
+
"grad_norm": 0.8897218946404224,
|
| 1386 |
+
"learning_rate": 4.308299424788667e-06,
|
| 1387 |
+
"loss": 1.0349,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5930162552679109,
|
| 1392 |
+
"grad_norm": 0.9205395645648624,
|
| 1393 |
+
"learning_rate": 4.256273931091284e-06,
|
| 1394 |
+
"loss": 1.0221,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5960264900662252,
|
| 1399 |
+
"grad_norm": 0.8997013037962462,
|
| 1400 |
+
"learning_rate": 4.204330651731662e-06,
|
| 1401 |
+
"loss": 1.0169,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5990367248645394,
|
| 1406 |
+
"grad_norm": 0.88917029228118,
|
| 1407 |
+
"learning_rate": 4.152475328719517e-06,
|
| 1408 |
+
"loss": 1.0186,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6020469596628537,
|
| 1413 |
+
"grad_norm": 0.9250893662867846,
|
| 1414 |
+
"learning_rate": 4.1007136943415325e-06,
|
| 1415 |
+
"loss": 1.0057,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.605057194461168,
|
| 1420 |
+
"grad_norm": 0.8603593665888702,
|
| 1421 |
+
"learning_rate": 4.049051470527692e-06,
|
| 1422 |
+
"loss": 1.0244,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6080674292594822,
|
| 1427 |
+
"grad_norm": 0.8966013371546253,
|
| 1428 |
+
"learning_rate": 3.997494368218745e-06,
|
| 1429 |
+
"loss": 1.0146,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6110776640577965,
|
| 1434 |
+
"grad_norm": 0.9014285122776863,
|
| 1435 |
+
"learning_rate": 3.946048086734921e-06,
|
| 1436 |
+
"loss": 1.033,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6140878988561108,
|
| 1441 |
+
"grad_norm": 0.938108665286018,
|
| 1442 |
+
"learning_rate": 3.894718313145873e-06,
|
| 1443 |
+
"loss": 1.0243,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.617098133654425,
|
| 1448 |
+
"grad_norm": 1.250091206416041,
|
| 1449 |
+
"learning_rate": 3.843510721642036e-06,
|
| 1450 |
+
"loss": 1.0055,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6201083684527393,
|
| 1455 |
+
"grad_norm": 0.8858750851021894,
|
| 1456 |
+
"learning_rate": 3.7924309729073616e-06,
|
| 1457 |
+
"loss": 1.0143,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6231186032510536,
|
| 1462 |
+
"grad_norm": 0.9538435034591204,
|
| 1463 |
+
"learning_rate": 3.7414847134935716e-06,
|
| 1464 |
+
"loss": 1.0032,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6261288380493678,
|
| 1469 |
+
"grad_norm": 0.938247676145859,
|
| 1470 |
+
"learning_rate": 3.6906775751959667e-06,
|
| 1471 |
+
"loss": 1.0202,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6291390728476821,
|
| 1476 |
+
"grad_norm": 0.924707505710252,
|
| 1477 |
+
"learning_rate": 3.640015174430864e-06,
|
| 1478 |
+
"loss": 0.9974,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6321493076459964,
|
| 1483 |
+
"grad_norm": 0.8952719097489277,
|
| 1484 |
+
"learning_rate": 3.5895031116147355e-06,
|
| 1485 |
+
"loss": 1.0128,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6351595424443106,
|
| 1490 |
+
"grad_norm": 0.8993701279222203,
|
| 1491 |
+
"learning_rate": 3.539146970545124e-06,
|
| 1492 |
+
"loss": 1.0179,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6381697772426249,
|
| 1497 |
+
"grad_norm": 0.8755392901199769,
|
| 1498 |
+
"learning_rate": 3.488952317783374e-06,
|
| 1499 |
+
"loss": 1.0116,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6411800120409392,
|
| 1504 |
+
"grad_norm": 0.9058135143412757,
|
| 1505 |
+
"learning_rate": 3.438924702039301e-06,
|
| 1506 |
+
"loss": 1.0069,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6441902468392534,
|
| 1511 |
+
"grad_norm": 0.8659042642422232,
|
| 1512 |
+
"learning_rate": 3.389069653557805e-06,
|
| 1513 |
+
"loss": 1.0206,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6472004816375677,
|
| 1518 |
+
"grad_norm": 0.897110943379507,
|
| 1519 |
+
"learning_rate": 3.3393926835075307e-06,
|
| 1520 |
+
"loss": 1.0098,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.650210716435882,
|
| 1525 |
+
"grad_norm": 0.9103411371697098,
|
| 1526 |
+
"learning_rate": 3.289899283371657e-06,
|
| 1527 |
+
"loss": 1.0011,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6532209512341962,
|
| 1532 |
+
"grad_norm": 0.9547731068418334,
|
| 1533 |
+
"learning_rate": 3.240594924340835e-06,
|
| 1534 |
+
"loss": 1.0077,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6562311860325105,
|
| 1539 |
+
"grad_norm": 0.8937676081109391,
|
| 1540 |
+
"learning_rate": 3.1914850567083866e-06,
|
| 1541 |
+
"loss": 1.0176,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6592414208308248,
|
| 1546 |
+
"grad_norm": 0.929138697588076,
|
| 1547 |
+
"learning_rate": 3.1425751092678064e-06,
|
| 1548 |
+
"loss": 1.0034,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6622516556291391,
|
| 1553 |
+
"grad_norm": 0.9210331479138841,
|
| 1554 |
+
"learning_rate": 3.0938704887126425e-06,
|
| 1555 |
+
"loss": 0.9913,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6652618904274533,
|
| 1560 |
+
"grad_norm": 0.8549140765541365,
|
| 1561 |
+
"learning_rate": 3.045376579038821e-06,
|
| 1562 |
+
"loss": 1.0005,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6682721252257676,
|
| 1567 |
+
"grad_norm": 0.870478554048657,
|
| 1568 |
+
"learning_rate": 2.9970987409494784e-06,
|
| 1569 |
+
"loss": 1.0373,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6712823600240819,
|
| 1574 |
+
"grad_norm": 0.8724631980574007,
|
| 1575 |
+
"learning_rate": 2.9490423112623646e-06,
|
| 1576 |
+
"loss": 1.0035,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6742925948223961,
|
| 1581 |
+
"grad_norm": 0.9494490020133225,
|
| 1582 |
+
"learning_rate": 2.9012126023198973e-06,
|
| 1583 |
+
"loss": 1.0152,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6773028296207104,
|
| 1588 |
+
"grad_norm": 0.8675243470717614,
|
| 1589 |
+
"learning_rate": 2.853614901401909e-06,
|
| 1590 |
+
"loss": 0.9851,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6803130644190247,
|
| 1595 |
+
"grad_norm": 0.9111265739825615,
|
| 1596 |
+
"learning_rate": 2.806254470141174e-06,
|
| 1597 |
+
"loss": 1.0211,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6833232992173389,
|
| 1602 |
+
"grad_norm": 0.8793180113511853,
|
| 1603 |
+
"learning_rate": 2.759136543941773e-06,
|
| 1604 |
+
"loss": 0.9824,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6863335340156532,
|
| 1609 |
+
"grad_norm": 0.9651914463110899,
|
| 1610 |
+
"learning_rate": 2.712266331400332e-06,
|
| 1611 |
+
"loss": 1.0008,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6893437688139675,
|
| 1616 |
+
"grad_norm": 0.8821546587242668,
|
| 1617 |
+
"learning_rate": 2.66564901373027e-06,
|
| 1618 |
+
"loss": 1.0115,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6923540036122817,
|
| 1623 |
+
"grad_norm": 0.914825675729504,
|
| 1624 |
+
"learning_rate": 2.6192897441890337e-06,
|
| 1625 |
+
"loss": 1.0025,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.695364238410596,
|
| 1630 |
+
"grad_norm": 0.9051560102693067,
|
| 1631 |
+
"learning_rate": 2.573193647508426e-06,
|
| 1632 |
+
"loss": 1.0055,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6983744732089103,
|
| 1637 |
+
"grad_norm": 0.876626084266648,
|
| 1638 |
+
"learning_rate": 2.5273658193281252e-06,
|
| 1639 |
+
"loss": 0.9862,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7013847080072245,
|
| 1644 |
+
"grad_norm": 1.2625964082093335,
|
| 1645 |
+
"learning_rate": 2.4818113256323745e-06,
|
| 1646 |
+
"loss": 1.0011,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7043949428055388,
|
| 1651 |
+
"grad_norm": 0.9003047210639946,
|
| 1652 |
+
"learning_rate": 2.4365352021899635e-06,
|
| 1653 |
+
"loss": 1.0067,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7074051776038531,
|
| 1658 |
+
"grad_norm": 0.9392842028653241,
|
| 1659 |
+
"learning_rate": 2.391542453997578e-06,
|
| 1660 |
+
"loss": 0.9987,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7104154124021673,
|
| 1665 |
+
"grad_norm": 0.8898125521821807,
|
| 1666 |
+
"learning_rate": 2.346838054726505e-06,
|
| 1667 |
+
"loss": 1.0042,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7134256472004816,
|
| 1672 |
+
"grad_norm": 0.8839311598947895,
|
| 1673 |
+
"learning_rate": 2.302426946172836e-06,
|
| 1674 |
+
"loss": 1.0188,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7164358819987959,
|
| 1679 |
+
"grad_norm": 145.48082889915423,
|
| 1680 |
+
"learning_rate": 2.258314037711184e-06,
|
| 1681 |
+
"loss": 1.0078,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7194461167971101,
|
| 1686 |
+
"grad_norm": 0.9080284956087131,
|
| 1687 |
+
"learning_rate": 2.214504205751971e-06,
|
| 1688 |
+
"loss": 1.0252,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7224563515954244,
|
| 1693 |
+
"grad_norm": 0.8693176168542095,
|
| 1694 |
+
"learning_rate": 2.1710022932023805e-06,
|
| 1695 |
+
"loss": 1.0031,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7254665863937387,
|
| 1700 |
+
"grad_norm": 0.8945645392796941,
|
| 1701 |
+
"learning_rate": 2.127813108931007e-06,
|
| 1702 |
+
"loss": 1.0088,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7284768211920529,
|
| 1707 |
+
"grad_norm": 0.8743665598111865,
|
| 1708 |
+
"learning_rate": 2.084941427236245e-06,
|
| 1709 |
+
"loss": 1.0162,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7314870559903672,
|
| 1714 |
+
"grad_norm": 0.9349057343090031,
|
| 1715 |
+
"learning_rate": 2.04239198731855e-06,
|
| 1716 |
+
"loss": 0.9996,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7344972907886815,
|
| 1721 |
+
"grad_norm": 7.756856106473671,
|
| 1722 |
+
"learning_rate": 2.000169492756523e-06,
|
| 1723 |
+
"loss": 1.0133,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7375075255869958,
|
| 1728 |
+
"grad_norm": 8.589389041141127,
|
| 1729 |
+
"learning_rate": 1.9582786109869713e-06,
|
| 1730 |
+
"loss": 1.0165,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.74051776038531,
|
| 1735 |
+
"grad_norm": 0.9692270078720051,
|
| 1736 |
+
"learning_rate": 1.9167239727889527e-06,
|
| 1737 |
+
"loss": 1.0104,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7435279951836243,
|
| 1742 |
+
"grad_norm": 0.8931619647840838,
|
| 1743 |
+
"learning_rate": 1.875510171771865e-06,
|
| 1744 |
+
"loss": 0.9945,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7465382299819386,
|
| 1749 |
+
"grad_norm": 0.9061163884083135,
|
| 1750 |
+
"learning_rate": 1.8346417638676533e-06,
|
| 1751 |
+
"loss": 0.9846,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7495484647802528,
|
| 1756 |
+
"grad_norm": 0.9440025241822538,
|
| 1757 |
+
"learning_rate": 1.7941232668271863e-06,
|
| 1758 |
+
"loss": 1.0107,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7525586995785671,
|
| 1763 |
+
"grad_norm": 7.949070459581915,
|
| 1764 |
+
"learning_rate": 1.753959159720836e-06,
|
| 1765 |
+
"loss": 1.0021,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7555689343768814,
|
| 1770 |
+
"grad_norm": 0.8900575327837316,
|
| 1771 |
+
"learning_rate": 1.7141538824433506e-06,
|
| 1772 |
+
"loss": 1.0083,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7585791691751956,
|
| 1777 |
+
"grad_norm": 0.9567537974389942,
|
| 1778 |
+
"learning_rate": 1.6747118352230495e-06,
|
| 1779 |
+
"loss": 1.0064,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7615894039735099,
|
| 1784 |
+
"grad_norm": 0.855175568298549,
|
| 1785 |
+
"learning_rate": 1.6356373781354058e-06,
|
| 1786 |
+
"loss": 0.9821,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7645996387718242,
|
| 1791 |
+
"grad_norm": 0.8869389537440323,
|
| 1792 |
+
"learning_rate": 1.5969348306210692e-06,
|
| 1793 |
+
"loss": 1.0033,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7676098735701384,
|
| 1798 |
+
"grad_norm": 0.8686541101196223,
|
| 1799 |
+
"learning_rate": 1.5586084710083737e-06,
|
| 1800 |
+
"loss": 1.0097,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7706201083684527,
|
| 1805 |
+
"grad_norm": 0.8853773671756917,
|
| 1806 |
+
"learning_rate": 1.5206625360403943e-06,
|
| 1807 |
+
"loss": 1.002,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.773630343166767,
|
| 1812 |
+
"grad_norm": 9.276510463559799,
|
| 1813 |
+
"learning_rate": 1.4831012204066114e-06,
|
| 1814 |
+
"loss": 1.0198,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7766405779650812,
|
| 1819 |
+
"grad_norm": 0.8952189934733595,
|
| 1820 |
+
"learning_rate": 1.445928676279199e-06,
|
| 1821 |
+
"loss": 0.9915,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7796508127633955,
|
| 1826 |
+
"grad_norm": 1.4459667679766754,
|
| 1827 |
+
"learning_rate": 1.4091490128540374e-06,
|
| 1828 |
+
"loss": 0.9837,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7826610475617098,
|
| 1833 |
+
"grad_norm": 0.8802348291910782,
|
| 1834 |
+
"learning_rate": 1.3727662958964627e-06,
|
| 1835 |
+
"loss": 1.0001,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.785671282360024,
|
| 1840 |
+
"grad_norm": 1.0156972499266355,
|
| 1841 |
+
"learning_rate": 1.3367845472918272e-06,
|
| 1842 |
+
"loss": 1.0072,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7886815171583383,
|
| 1847 |
+
"grad_norm": 0.9314514114707518,
|
| 1848 |
+
"learning_rate": 1.3012077446008969e-06,
|
| 1849 |
+
"loss": 1.0011,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7916917519566526,
|
| 1854 |
+
"grad_norm": 0.8924087794263724,
|
| 1855 |
+
"learning_rate": 1.266039820620159e-06,
|
| 1856 |
+
"loss": 1.0055,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7947019867549668,
|
| 1861 |
+
"grad_norm": 0.9131510130376783,
|
| 1862 |
+
"learning_rate": 1.2312846629470826e-06,
|
| 1863 |
+
"loss": 0.9894,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7977122215532811,
|
| 1868 |
+
"grad_norm": 0.9718957955204853,
|
| 1869 |
+
"learning_rate": 1.1969461135503573e-06,
|
| 1870 |
+
"loss": 1.0058,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8007224563515954,
|
| 1875 |
+
"grad_norm": 0.8842617855041038,
|
| 1876 |
+
"learning_rate": 1.163027968345195e-06,
|
| 1877 |
+
"loss": 1.0088,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8037326911499096,
|
| 1882 |
+
"grad_norm": 1.0150004217582238,
|
| 1883 |
+
"learning_rate": 1.1295339767737125e-06,
|
| 1884 |
+
"loss": 0.9919,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8067429259482239,
|
| 1889 |
+
"grad_norm": 9.91219637819699,
|
| 1890 |
+
"learning_rate": 1.0964678413904529e-06,
|
| 1891 |
+
"loss": 1.0052,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8097531607465382,
|
| 1896 |
+
"grad_norm": 0.9243390152675313,
|
| 1897 |
+
"learning_rate": 1.0638332174530953e-06,
|
| 1898 |
+
"loss": 1.0121,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8127633955448526,
|
| 1903 |
+
"grad_norm": 0.9013142550026193,
|
| 1904 |
+
"learning_rate": 1.0316337125183817e-06,
|
| 1905 |
+
"loss": 0.9996,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8157736303431667,
|
| 1910 |
+
"grad_norm": 0.8910284877524084,
|
| 1911 |
+
"learning_rate": 9.998728860433277e-07,
|
| 1912 |
+
"loss": 1.0004,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.818783865141481,
|
| 1917 |
+
"grad_norm": 0.8798186735341214,
|
| 1918 |
+
"learning_rate": 9.685542489917494e-07,
|
| 1919 |
+
"loss": 0.982,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8217940999397954,
|
| 1924 |
+
"grad_norm": 0.9826761795502499,
|
| 1925 |
+
"learning_rate": 9.376812634461418e-07,
|
| 1926 |
+
"loss": 0.9933,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8248043347381095,
|
| 1931 |
+
"grad_norm": 0.9100156607261772,
|
| 1932 |
+
"learning_rate": 9.072573422249692e-07,
|
| 1933 |
+
"loss": 1.016,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8278145695364238,
|
| 1938 |
+
"grad_norm": 0.8713258214258187,
|
| 1939 |
+
"learning_rate": 8.772858485054042e-07,
|
| 1940 |
+
"loss": 1.0055,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8308248043347382,
|
| 1945 |
+
"grad_norm": 0.8891360404790082,
|
| 1946 |
+
"learning_rate": 8.477700954515372e-07,
|
| 1947 |
+
"loss": 1.004,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8338350391330523,
|
| 1952 |
+
"grad_norm": 0.9873027733998953,
|
| 1953 |
+
"learning_rate": 8.187133458481416e-07,
|
| 1954 |
+
"loss": 1.0047,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8368452739313667,
|
| 1959 |
+
"grad_norm": 0.9173976826216069,
|
| 1960 |
+
"learning_rate": 7.901188117399817e-07,
|
| 1961 |
+
"loss": 1.0141,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.839855508729681,
|
| 1966 |
+
"grad_norm": 0.8933319303624291,
|
| 1967 |
+
"learning_rate": 7.619896540767435e-07,
|
| 1968 |
+
"loss": 1.0147,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8428657435279951,
|
| 1973 |
+
"grad_norm": 0.8820123957581391,
|
| 1974 |
+
"learning_rate": 7.343289823636168e-07,
|
| 1975 |
+
"loss": 1.0121,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8458759783263095,
|
| 1980 |
+
"grad_norm": 1.1417427890515452,
|
| 1981 |
+
"learning_rate": 7.0713985431755e-07,
|
| 1982 |
+
"loss": 1.0102,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8488862131246238,
|
| 1987 |
+
"grad_norm": 2.2006862542879766,
|
| 1988 |
+
"learning_rate": 6.804252755292429e-07,
|
| 1989 |
+
"loss": 0.981,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.851896447922938,
|
| 1994 |
+
"grad_norm": 0.892688239858576,
|
| 1995 |
+
"learning_rate": 6.541881991309013e-07,
|
| 1996 |
+
"loss": 0.9994,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8549066827212523,
|
| 2001 |
+
"grad_norm": 0.8851152482079784,
|
| 2002 |
+
"learning_rate": 6.284315254697726e-07,
|
| 2003 |
+
"loss": 1.0055,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8579169175195666,
|
| 2008 |
+
"grad_norm": 4.1376088974160705,
|
| 2009 |
+
"learning_rate": 6.031581017875482e-07,
|
| 2010 |
+
"loss": 1.0114,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8609271523178808,
|
| 2015 |
+
"grad_norm": 1.0742418914591978,
|
| 2016 |
+
"learning_rate": 5.783707219056078e-07,
|
| 2017 |
+
"loss": 1.0009,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.863937387116195,
|
| 2022 |
+
"grad_norm": 51.72308481458701,
|
| 2023 |
+
"learning_rate": 5.540721259161774e-07,
|
| 2024 |
+
"loss": 1.0091,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8669476219145094,
|
| 2029 |
+
"grad_norm": 0.9187887516758119,
|
| 2030 |
+
"learning_rate": 5.302649998794368e-07,
|
| 2031 |
+
"loss": 1.0122,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8699578567128236,
|
| 2036 |
+
"grad_norm": 0.998724825032228,
|
| 2037 |
+
"learning_rate": 5.0695197552659e-07,
|
| 2038 |
+
"loss": 0.9906,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8729680915111379,
|
| 2043 |
+
"grad_norm": 3.453765069147144,
|
| 2044 |
+
"learning_rate": 4.841356299689359e-07,
|
| 2045 |
+
"loss": 1.0037,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8759783263094522,
|
| 2050 |
+
"grad_norm": 0.9196156863617485,
|
| 2051 |
+
"learning_rate": 4.618184854129981e-07,
|
| 2052 |
+
"loss": 1.013,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8789885611077664,
|
| 2057 |
+
"grad_norm": 0.8776281392555243,
|
| 2058 |
+
"learning_rate": 4.4000300888169753e-07,
|
| 2059 |
+
"loss": 1.0023,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8819987959060807,
|
| 2064 |
+
"grad_norm": 1.2343861844915518,
|
| 2065 |
+
"learning_rate": 4.1869161194164565e-07,
|
| 2066 |
+
"loss": 1.005,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.885009030704395,
|
| 2071 |
+
"grad_norm": 3.1950889114793317,
|
| 2072 |
+
"learning_rate": 3.9788665043656083e-07,
|
| 2073 |
+
"loss": 0.9983,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8880192655027093,
|
| 2078 |
+
"grad_norm": 0.8997540101420374,
|
| 2079 |
+
"learning_rate": 3.775904242268391e-07,
|
| 2080 |
+
"loss": 1.0236,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8910295003010235,
|
| 2085 |
+
"grad_norm": 0.8874735736481768,
|
| 2086 |
+
"learning_rate": 3.578051769353219e-07,
|
| 2087 |
+
"loss": 1.0125,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8940397350993378,
|
| 2092 |
+
"grad_norm": 0.8827860731362758,
|
| 2093 |
+
"learning_rate": 3.385330956992816e-07,
|
| 2094 |
+
"loss": 1.0003,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8970499698976521,
|
| 2099 |
+
"grad_norm": 0.9108633520843697,
|
| 2100 |
+
"learning_rate": 3.1977631092863613e-07,
|
| 2101 |
+
"loss": 1.0042,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9000602046959663,
|
| 2106 |
+
"grad_norm": 0.8897735803071378,
|
| 2107 |
+
"learning_rate": 3.015368960704584e-07,
|
| 2108 |
+
"loss": 0.9873,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9030704394942806,
|
| 2113 |
+
"grad_norm": 0.8716202792511029,
|
| 2114 |
+
"learning_rate": 2.8381686737975867e-07,
|
| 2115 |
+
"loss": 1.0115,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9060806742925949,
|
| 2120 |
+
"grad_norm": 0.9263815301734902,
|
| 2121 |
+
"learning_rate": 2.666181836966053e-07,
|
| 2122 |
+
"loss": 1.0149,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9090909090909091,
|
| 2127 |
+
"grad_norm": 0.9028381235449051,
|
| 2128 |
+
"learning_rate": 2.4994274622958726e-07,
|
| 2129 |
+
"loss": 1.0137,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9121011438892234,
|
| 2134 |
+
"grad_norm": 0.8952393788798627,
|
| 2135 |
+
"learning_rate": 2.3379239834564526e-07,
|
| 2136 |
+
"loss": 1.0052,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9151113786875377,
|
| 2141 |
+
"grad_norm": 6.643467281200084,
|
| 2142 |
+
"learning_rate": 2.1816892536629775e-07,
|
| 2143 |
+
"loss": 1.0129,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9181216134858519,
|
| 2148 |
+
"grad_norm": 1.6442311604186386,
|
| 2149 |
+
"learning_rate": 2.0307405437029027e-07,
|
| 2150 |
+
"loss": 0.9973,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9211318482841662,
|
| 2155 |
+
"grad_norm": 1.0659077620293165,
|
| 2156 |
+
"learning_rate": 1.8850945400266994e-07,
|
| 2157 |
+
"loss": 1.0181,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9241420830824805,
|
| 2162 |
+
"grad_norm": 4.3859143955880855,
|
| 2163 |
+
"learning_rate": 1.7447673429033361e-07,
|
| 2164 |
+
"loss": 0.9964,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9271523178807947,
|
| 2169 |
+
"grad_norm": 0.8853862484020693,
|
| 2170 |
+
"learning_rate": 1.6097744646404457e-07,
|
| 2171 |
+
"loss": 0.9963,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.930162552679109,
|
| 2176 |
+
"grad_norm": 0.888640349391396,
|
| 2177 |
+
"learning_rate": 1.4801308278695636e-07,
|
| 2178 |
+
"loss": 1.0036,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9331727874774233,
|
| 2183 |
+
"grad_norm": 0.8933149209877544,
|
| 2184 |
+
"learning_rate": 1.3558507638965158e-07,
|
| 2185 |
+
"loss": 1.0066,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9361830222757375,
|
| 2190 |
+
"grad_norm": 0.9023917136128728,
|
| 2191 |
+
"learning_rate": 1.2369480111171784e-07,
|
| 2192 |
+
"loss": 1.0068,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9391932570740518,
|
| 2197 |
+
"grad_norm": 7.92138822902921,
|
| 2198 |
+
"learning_rate": 1.1234357134987717e-07,
|
| 2199 |
+
"loss": 0.9898,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9422034918723661,
|
| 2204 |
+
"grad_norm": 0.8873656023063873,
|
| 2205 |
+
"learning_rate": 1.0153264191269052e-07,
|
| 2206 |
+
"loss": 1.0099,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9452137266706803,
|
| 2211 |
+
"grad_norm": 1.7635058125000826,
|
| 2212 |
+
"learning_rate": 9.126320788184374e-08,
|
| 2213 |
+
"loss": 0.9943,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9482239614689946,
|
| 2218 |
+
"grad_norm": 1.1380512050692657,
|
| 2219 |
+
"learning_rate": 8.153640448003875e-08,
|
| 2220 |
+
"loss": 0.9869,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9512341962673089,
|
| 2225 |
+
"grad_norm": 1.0730720715487758,
|
| 2226 |
+
"learning_rate": 7.235330694550402e-08,
|
| 2227 |
+
"loss": 1.0157,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9542444310656231,
|
| 2232 |
+
"grad_norm": 0.8758058611438886,
|
| 2233 |
+
"learning_rate": 6.371493041313126e-08,
|
| 2234 |
+
"loss": 0.9904,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9572546658639374,
|
| 2239 |
+
"grad_norm": 0.862795236731348,
|
| 2240 |
+
"learning_rate": 5.562222980225907e-08,
|
| 2241 |
+
"loss": 0.9965,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9602649006622517,
|
| 2246 |
+
"grad_norm": 0.8490275312725396,
|
| 2247 |
+
"learning_rate": 4.807609971111238e-08,
|
| 2248 |
+
"loss": 0.9898,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.963275135460566,
|
| 2253 |
+
"grad_norm": 0.8719190026799012,
|
| 2254 |
+
"learning_rate": 4.107737431791159e-08,
|
| 2255 |
+
"loss": 1.0197,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9662853702588802,
|
| 2260 |
+
"grad_norm": 0.9849558439277634,
|
| 2261 |
+
"learning_rate": 3.462682728865685e-08,
|
| 2262 |
+
"loss": 0.999,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9692956050571945,
|
| 2267 |
+
"grad_norm": 0.9240673817719144,
|
| 2268 |
+
"learning_rate": 2.8725171691605934e-08,
|
| 2269 |
+
"loss": 1.019,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9723058398555088,
|
| 2274 |
+
"grad_norm": 0.8711151381569093,
|
| 2275 |
+
"learning_rate": 2.3373059918448958e-08,
|
| 2276 |
+
"loss": 1.0001,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.975316074653823,
|
| 2281 |
+
"grad_norm": 0.9138380625875624,
|
| 2282 |
+
"learning_rate": 1.8571083612188845e-08,
|
| 2283 |
+
"loss": 1.0099,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9783263094521373,
|
| 2288 |
+
"grad_norm": 0.9312600529557169,
|
| 2289 |
+
"learning_rate": 1.431977360173975e-08,
|
| 2290 |
+
"loss": 0.9951,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9813365442504516,
|
| 2295 |
+
"grad_norm": 0.9023080871350866,
|
| 2296 |
+
"learning_rate": 1.0619599843249006e-08,
|
| 2297 |
+
"loss": 1.0015,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9843467790487658,
|
| 2302 |
+
"grad_norm": 16.22001339030287,
|
| 2303 |
+
"learning_rate": 7.470971368142011e-09,
|
| 2304 |
+
"loss": 1.0138,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9873570138470801,
|
| 2309 |
+
"grad_norm": 2.544390586510872,
|
| 2310 |
+
"learning_rate": 4.874236237911723e-09,
|
| 2311 |
+
"loss": 1.0043,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9903672486453944,
|
| 2316 |
+
"grad_norm": 1.1347495534779806,
|
| 2317 |
+
"learning_rate": 2.8296815056377824e-09,
|
| 2318 |
+
"loss": 1.0145,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9933774834437086,
|
| 2323 |
+
"grad_norm": 0.9810788936521698,
|
| 2324 |
+
"learning_rate": 1.3375331842574446e-09,
|
| 2325 |
+
"loss": 1.034,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9963877182420229,
|
| 2330 |
+
"grad_norm": 1.4553563885174292,
|
| 2331 |
+
"learning_rate": 3.9795622158111945e-10,
|
| 2332 |
+
"loss": 1.0076,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9993979530403372,
|
| 2337 |
+
"grad_norm": 6.68038526442591,
|
| 2338 |
+
"learning_rate": 1.1054482056405136e-11,
|
| 2339 |
+
"loss": 1.0027,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 1.0,
|
| 2344 |
+
"eval_runtime": 0.8643,
|
| 2345 |
+
"eval_samples_per_second": 11.57,
|
| 2346 |
+
"eval_steps_per_second": 3.471,
|
| 2347 |
+
"step": 1661
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 1.0,
|
| 2351 |
+
"step": 1661,
|
| 2352 |
+
"total_flos": 80925639573504.0,
|
| 2353 |
+
"train_loss": 1.0962458014272911,
|
| 2354 |
+
"train_runtime": 1740.9834,
|
| 2355 |
+
"train_samples_per_second": 15.263,
|
| 2356 |
+
"train_steps_per_second": 0.954
|
| 2357 |
+
}
|
| 2358 |
+
],
|
| 2359 |
+
"logging_steps": 5,
|
| 2360 |
+
"max_steps": 1661,
|
| 2361 |
+
"num_input_tokens_seen": 0,
|
| 2362 |
+
"num_train_epochs": 1,
|
| 2363 |
+
"save_steps": 100,
|
| 2364 |
+
"stateful_callbacks": {
|
| 2365 |
+
"TrainerControl": {
|
| 2366 |
+
"args": {
|
| 2367 |
+
"should_epoch_stop": false,
|
| 2368 |
+
"should_evaluate": false,
|
| 2369 |
+
"should_log": false,
|
| 2370 |
+
"should_save": true,
|
| 2371 |
+
"should_training_stop": true
|
| 2372 |
+
},
|
| 2373 |
+
"attributes": {}
|
| 2374 |
+
}
|
| 2375 |
+
},
|
| 2376 |
+
"total_flos": 80925639573504.0,
|
| 2377 |
+
"train_batch_size": 4,
|
| 2378 |
+
"trial_name": null,
|
| 2379 |
+
"trial_params": null
|
| 2380 |
+
}
|