Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep23_11-12-57_COE-CS-sv003/events.out.tfevents.1727090284.COE-CS-sv003.735228.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2688 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: feedback_p0.1_seed42_level2_stylecleanbatch16_reverse
|
| 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 |
+
# feedback_p0.1_seed42_level2_stylecleanbatch16_reverse
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) 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 4.44.2
|
| 58 |
+
- Pytorch 2.3.1+cu121
|
| 59 |
+
- Datasets 2.19.1
|
| 60 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 196921566167040.0,
|
| 4 |
+
"train_loss": 0.5133170171946303,
|
| 5 |
+
"train_runtime": 16745.7461,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.797,
|
| 8 |
+
"train_steps_per_second": 0.112
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.44.2"
|
| 6 |
+
}
|
runs/Sep23_11-12-57_COE-CS-sv003/events.out.tfevents.1727090284.COE-CS-sv003.735228.0
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 sha256:30eba0756a10ccefd73e30fdd674831be1fccb087a53ccb0ac1713c9c73029df
|
| 3 |
+
size 85322
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 196921566167040.0,
|
| 4 |
+
"train_loss": 0.5133170171946303,
|
| 5 |
+
"train_runtime": 16745.7461,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.797,
|
| 8 |
+
"train_steps_per_second": 0.112
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2688 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1881,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000531632110579479,
|
| 13 |
+
"grad_norm": 23.579405331833645,
|
| 14 |
+
"learning_rate": 5.291005291005291e-08,
|
| 15 |
+
"loss": 1.3792,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002658160552897395,
|
| 20 |
+
"grad_norm": 21.819838525914637,
|
| 21 |
+
"learning_rate": 2.6455026455026455e-07,
|
| 22 |
+
"loss": 1.3427,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.00531632110579479,
|
| 27 |
+
"grad_norm": 15.344590621513694,
|
| 28 |
+
"learning_rate": 5.291005291005291e-07,
|
| 29 |
+
"loss": 1.299,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.007974481658692184,
|
| 34 |
+
"grad_norm": 12.131441916410711,
|
| 35 |
+
"learning_rate": 7.936507936507937e-07,
|
| 36 |
+
"loss": 1.1542,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01063264221158958,
|
| 41 |
+
"grad_norm": 8.98911329195494,
|
| 42 |
+
"learning_rate": 1.0582010582010582e-06,
|
| 43 |
+
"loss": 1.0463,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013290802764486975,
|
| 48 |
+
"grad_norm": 3.501607778526649,
|
| 49 |
+
"learning_rate": 1.3227513227513228e-06,
|
| 50 |
+
"loss": 0.9424,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01594896331738437,
|
| 55 |
+
"grad_norm": 3.423489710856073,
|
| 56 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 57 |
+
"loss": 0.9018,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018607123870281767,
|
| 62 |
+
"grad_norm": 3.001277367108441,
|
| 63 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 64 |
+
"loss": 0.8748,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.02126528442317916,
|
| 69 |
+
"grad_norm": 2.9441592451950505,
|
| 70 |
+
"learning_rate": 2.1164021164021164e-06,
|
| 71 |
+
"loss": 0.8816,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023923444976076555,
|
| 76 |
+
"grad_norm": 2.9271187812373496,
|
| 77 |
+
"learning_rate": 2.380952380952381e-06,
|
| 78 |
+
"loss": 0.8556,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02658160552897395,
|
| 83 |
+
"grad_norm": 3.0958894415955793,
|
| 84 |
+
"learning_rate": 2.6455026455026455e-06,
|
| 85 |
+
"loss": 0.8353,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.029239766081871343,
|
| 90 |
+
"grad_norm": 2.922921509451761,
|
| 91 |
+
"learning_rate": 2.9100529100529103e-06,
|
| 92 |
+
"loss": 0.823,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03189792663476874,
|
| 97 |
+
"grad_norm": 3.135548976301074,
|
| 98 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 99 |
+
"loss": 0.8172,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03455608718766613,
|
| 104 |
+
"grad_norm": 3.1362315132615355,
|
| 105 |
+
"learning_rate": 3.4391534391534394e-06,
|
| 106 |
+
"loss": 0.8077,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.03721424774056353,
|
| 111 |
+
"grad_norm": 3.025971829033113,
|
| 112 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 113 |
+
"loss": 0.7988,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.03987240829346093,
|
| 118 |
+
"grad_norm": 3.2543761620825373,
|
| 119 |
+
"learning_rate": 3.968253968253968e-06,
|
| 120 |
+
"loss": 0.7917,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04253056884635832,
|
| 125 |
+
"grad_norm": 2.979399910029388,
|
| 126 |
+
"learning_rate": 4.232804232804233e-06,
|
| 127 |
+
"loss": 0.7756,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.045188729399255716,
|
| 132 |
+
"grad_norm": 3.100212693141578,
|
| 133 |
+
"learning_rate": 4.497354497354498e-06,
|
| 134 |
+
"loss": 0.7715,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.04784688995215311,
|
| 139 |
+
"grad_norm": 3.0260363553205267,
|
| 140 |
+
"learning_rate": 4.761904761904762e-06,
|
| 141 |
+
"loss": 0.7827,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.050505050505050504,
|
| 146 |
+
"grad_norm": 3.5352951332042757,
|
| 147 |
+
"learning_rate": 5.026455026455027e-06,
|
| 148 |
+
"loss": 0.7658,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0531632110579479,
|
| 153 |
+
"grad_norm": 2.9625298663916593,
|
| 154 |
+
"learning_rate": 5.291005291005291e-06,
|
| 155 |
+
"loss": 0.7384,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.05582137161084529,
|
| 160 |
+
"grad_norm": 2.989013375481517,
|
| 161 |
+
"learning_rate": 5.555555555555557e-06,
|
| 162 |
+
"loss": 0.7598,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05847953216374269,
|
| 167 |
+
"grad_norm": 3.0615980905383045,
|
| 168 |
+
"learning_rate": 5.820105820105821e-06,
|
| 169 |
+
"loss": 0.7485,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06113769271664009,
|
| 174 |
+
"grad_norm": 3.125309932477323,
|
| 175 |
+
"learning_rate": 6.084656084656085e-06,
|
| 176 |
+
"loss": 0.7404,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.06379585326953748,
|
| 181 |
+
"grad_norm": 3.2962079296846114,
|
| 182 |
+
"learning_rate": 6.349206349206349e-06,
|
| 183 |
+
"loss": 0.7421,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06645401382243488,
|
| 188 |
+
"grad_norm": 3.122192839193298,
|
| 189 |
+
"learning_rate": 6.613756613756615e-06,
|
| 190 |
+
"loss": 0.73,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.06911217437533226,
|
| 195 |
+
"grad_norm": 2.9654473645168062,
|
| 196 |
+
"learning_rate": 6.878306878306879e-06,
|
| 197 |
+
"loss": 0.7296,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07177033492822966,
|
| 202 |
+
"grad_norm": 3.0144620572162224,
|
| 203 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 204 |
+
"loss": 0.7389,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.07442849548112707,
|
| 209 |
+
"grad_norm": 2.906104811780708,
|
| 210 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 211 |
+
"loss": 0.7202,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07708665603402445,
|
| 216 |
+
"grad_norm": 3.282063025489378,
|
| 217 |
+
"learning_rate": 7.671957671957672e-06,
|
| 218 |
+
"loss": 0.7202,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07974481658692185,
|
| 223 |
+
"grad_norm": 3.113187144382696,
|
| 224 |
+
"learning_rate": 7.936507936507936e-06,
|
| 225 |
+
"loss": 0.7193,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08240297713981924,
|
| 230 |
+
"grad_norm": 2.8167569906973875,
|
| 231 |
+
"learning_rate": 8.201058201058202e-06,
|
| 232 |
+
"loss": 0.7105,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.08506113769271664,
|
| 237 |
+
"grad_norm": 2.935672758755336,
|
| 238 |
+
"learning_rate": 8.465608465608466e-06,
|
| 239 |
+
"loss": 0.7127,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08771929824561403,
|
| 244 |
+
"grad_norm": 3.4898920959480804,
|
| 245 |
+
"learning_rate": 8.730158730158731e-06,
|
| 246 |
+
"loss": 0.7252,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.09037745879851143,
|
| 251 |
+
"grad_norm": 2.7854650304361575,
|
| 252 |
+
"learning_rate": 8.994708994708995e-06,
|
| 253 |
+
"loss": 0.7138,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.09303561935140882,
|
| 258 |
+
"grad_norm": 2.6523273221320127,
|
| 259 |
+
"learning_rate": 9.25925925925926e-06,
|
| 260 |
+
"loss": 0.705,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.09569377990430622,
|
| 265 |
+
"grad_norm": 2.7747422098156074,
|
| 266 |
+
"learning_rate": 9.523809523809525e-06,
|
| 267 |
+
"loss": 0.7107,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09835194045720362,
|
| 272 |
+
"grad_norm": 3.018940503132491,
|
| 273 |
+
"learning_rate": 9.788359788359789e-06,
|
| 274 |
+
"loss": 0.7177,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.10101010101010101,
|
| 279 |
+
"grad_norm": 2.9774305781097543,
|
| 280 |
+
"learning_rate": 9.999991381356603e-06,
|
| 281 |
+
"loss": 0.7049,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.10366826156299841,
|
| 286 |
+
"grad_norm": 2.9546281553147917,
|
| 287 |
+
"learning_rate": 9.99968973195745e-06,
|
| 288 |
+
"loss": 0.6982,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1063264221158958,
|
| 293 |
+
"grad_norm": 2.7361250859772057,
|
| 294 |
+
"learning_rate": 9.998957180100319e-06,
|
| 295 |
+
"loss": 0.7164,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1089845826687932,
|
| 300 |
+
"grad_norm": 2.7782156509674443,
|
| 301 |
+
"learning_rate": 9.99779378892081e-06,
|
| 302 |
+
"loss": 0.71,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.11164274322169059,
|
| 307 |
+
"grad_norm": 2.752067715623176,
|
| 308 |
+
"learning_rate": 9.996199658686769e-06,
|
| 309 |
+
"loss": 0.7103,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.11430090377458799,
|
| 314 |
+
"grad_norm": 2.696798343791542,
|
| 315 |
+
"learning_rate": 9.994174926789648e-06,
|
| 316 |
+
"loss": 0.6972,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.11695906432748537,
|
| 321 |
+
"grad_norm": 2.999464864426748,
|
| 322 |
+
"learning_rate": 9.991719767732666e-06,
|
| 323 |
+
"loss": 0.696,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.11961722488038277,
|
| 328 |
+
"grad_norm": 2.8474761936642152,
|
| 329 |
+
"learning_rate": 9.988834393115768e-06,
|
| 330 |
+
"loss": 0.7112,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.12227538543328018,
|
| 335 |
+
"grad_norm": 2.6201992243557415,
|
| 336 |
+
"learning_rate": 9.985519051617385e-06,
|
| 337 |
+
"loss": 0.7014,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.12493354598617756,
|
| 342 |
+
"grad_norm": 2.635339392667087,
|
| 343 |
+
"learning_rate": 9.981774028973013e-06,
|
| 344 |
+
"loss": 0.7055,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.12759170653907495,
|
| 349 |
+
"grad_norm": 2.681963719272812,
|
| 350 |
+
"learning_rate": 9.977599647950572e-06,
|
| 351 |
+
"loss": 0.6981,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.13024986709197237,
|
| 356 |
+
"grad_norm": 2.70825842745144,
|
| 357 |
+
"learning_rate": 9.972996268322594e-06,
|
| 358 |
+
"loss": 0.6948,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.13290802764486975,
|
| 363 |
+
"grad_norm": 3.1203174547960884,
|
| 364 |
+
"learning_rate": 9.967964286835219e-06,
|
| 365 |
+
"loss": 0.6935,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.13556618819776714,
|
| 370 |
+
"grad_norm": 2.6339980738414175,
|
| 371 |
+
"learning_rate": 9.962504137173997e-06,
|
| 372 |
+
"loss": 0.6981,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.13822434875066453,
|
| 377 |
+
"grad_norm": 2.503608868118079,
|
| 378 |
+
"learning_rate": 9.956616289926512e-06,
|
| 379 |
+
"loss": 0.6964,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.14088250930356194,
|
| 384 |
+
"grad_norm": 2.6711044853158037,
|
| 385 |
+
"learning_rate": 9.950301252541824e-06,
|
| 386 |
+
"loss": 0.6948,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.14354066985645933,
|
| 391 |
+
"grad_norm": 2.7158097067728497,
|
| 392 |
+
"learning_rate": 9.943559569286731e-06,
|
| 393 |
+
"loss": 0.6724,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.14619883040935672,
|
| 398 |
+
"grad_norm": 2.7405188241533955,
|
| 399 |
+
"learning_rate": 9.936391821198868e-06,
|
| 400 |
+
"loss": 0.678,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.14885699096225413,
|
| 405 |
+
"grad_norm": 2.5459004197727686,
|
| 406 |
+
"learning_rate": 9.92879862603662e-06,
|
| 407 |
+
"loss": 0.7093,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.15151515151515152,
|
| 412 |
+
"grad_norm": 2.5280845959479747,
|
| 413 |
+
"learning_rate": 9.92078063822589e-06,
|
| 414 |
+
"loss": 0.6663,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1541733120680489,
|
| 419 |
+
"grad_norm": 2.470733174856602,
|
| 420 |
+
"learning_rate": 9.91233854880369e-06,
|
| 421 |
+
"loss": 0.6557,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.1568314726209463,
|
| 426 |
+
"grad_norm": 2.591047305582084,
|
| 427 |
+
"learning_rate": 9.903473085358589e-06,
|
| 428 |
+
"loss": 0.6709,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1594896331738437,
|
| 433 |
+
"grad_norm": 2.646497143472905,
|
| 434 |
+
"learning_rate": 9.894185011967994e-06,
|
| 435 |
+
"loss": 0.6764,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.1621477937267411,
|
| 440 |
+
"grad_norm": 2.793394708582568,
|
| 441 |
+
"learning_rate": 9.884475129132312e-06,
|
| 442 |
+
"loss": 0.6712,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.16480595427963848,
|
| 447 |
+
"grad_norm": 2.9489215604304992,
|
| 448 |
+
"learning_rate": 9.874344273705949e-06,
|
| 449 |
+
"loss": 0.6765,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1674641148325359,
|
| 454 |
+
"grad_norm": 6.451313581277316,
|
| 455 |
+
"learning_rate": 9.863793318825186e-06,
|
| 456 |
+
"loss": 0.6644,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.17012227538543329,
|
| 461 |
+
"grad_norm": 2.5605827390740914,
|
| 462 |
+
"learning_rate": 9.852823173832932e-06,
|
| 463 |
+
"loss": 0.6873,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.17278043593833067,
|
| 468 |
+
"grad_norm": 2.6038006790137915,
|
| 469 |
+
"learning_rate": 9.841434784200341e-06,
|
| 470 |
+
"loss": 0.6701,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.17543859649122806,
|
| 475 |
+
"grad_norm": 2.5977149163301916,
|
| 476 |
+
"learning_rate": 9.829629131445342e-06,
|
| 477 |
+
"loss": 0.6592,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.17809675704412548,
|
| 482 |
+
"grad_norm": 2.498965621247738,
|
| 483 |
+
"learning_rate": 9.817407233048028e-06,
|
| 484 |
+
"loss": 0.6581,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.18075491759702286,
|
| 489 |
+
"grad_norm": 2.4810761737552274,
|
| 490 |
+
"learning_rate": 9.804770142362977e-06,
|
| 491 |
+
"loss": 0.6251,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.18341307814992025,
|
| 496 |
+
"grad_norm": 2.546278672778942,
|
| 497 |
+
"learning_rate": 9.791718948528457e-06,
|
| 498 |
+
"loss": 0.6552,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.18607123870281764,
|
| 503 |
+
"grad_norm": 2.6543352578309123,
|
| 504 |
+
"learning_rate": 9.778254776372576e-06,
|
| 505 |
+
"loss": 0.6609,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.18872939925571505,
|
| 510 |
+
"grad_norm": 2.3974140051030117,
|
| 511 |
+
"learning_rate": 9.76437878631631e-06,
|
| 512 |
+
"loss": 0.6497,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.19138755980861244,
|
| 517 |
+
"grad_norm": 2.813278650430626,
|
| 518 |
+
"learning_rate": 9.75009217427352e-06,
|
| 519 |
+
"loss": 0.6457,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.19404572036150983,
|
| 524 |
+
"grad_norm": 2.480141091317464,
|
| 525 |
+
"learning_rate": 9.735396171547859e-06,
|
| 526 |
+
"loss": 0.6694,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.19670388091440724,
|
| 531 |
+
"grad_norm": 2.6544163749167415,
|
| 532 |
+
"learning_rate": 9.720292044726664e-06,
|
| 533 |
+
"loss": 0.6403,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.19936204146730463,
|
| 538 |
+
"grad_norm": 2.419192069753362,
|
| 539 |
+
"learning_rate": 9.704781095571788e-06,
|
| 540 |
+
"loss": 0.6478,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.20202020202020202,
|
| 545 |
+
"grad_norm": 2.440380141428742,
|
| 546 |
+
"learning_rate": 9.68886466090741e-06,
|
| 547 |
+
"loss": 0.6453,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2046783625730994,
|
| 552 |
+
"grad_norm": 2.461462521030907,
|
| 553 |
+
"learning_rate": 9.672544112504813e-06,
|
| 554 |
+
"loss": 0.6459,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.20733652312599682,
|
| 559 |
+
"grad_norm": 2.6891487247733448,
|
| 560 |
+
"learning_rate": 9.655820856964171e-06,
|
| 561 |
+
"loss": 0.6566,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2099946836788942,
|
| 566 |
+
"grad_norm": 2.476250180113438,
|
| 567 |
+
"learning_rate": 9.638696335593304e-06,
|
| 568 |
+
"loss": 0.6505,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2126528442317916,
|
| 573 |
+
"grad_norm": 2.252029809034869,
|
| 574 |
+
"learning_rate": 9.621172024283468e-06,
|
| 575 |
+
"loss": 0.6506,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.215311004784689,
|
| 580 |
+
"grad_norm": 2.4009100256649027,
|
| 581 |
+
"learning_rate": 9.603249433382145e-06,
|
| 582 |
+
"loss": 0.6472,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2179691653375864,
|
| 587 |
+
"grad_norm": 2.65974014911984,
|
| 588 |
+
"learning_rate": 9.584930107562885e-06,
|
| 589 |
+
"loss": 0.6491,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.22062732589048378,
|
| 594 |
+
"grad_norm": 2.5396822097120046,
|
| 595 |
+
"learning_rate": 9.566215625692168e-06,
|
| 596 |
+
"loss": 0.649,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.22328548644338117,
|
| 601 |
+
"grad_norm": 2.4691530853144146,
|
| 602 |
+
"learning_rate": 9.547107600693328e-06,
|
| 603 |
+
"loss": 0.6382,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.22594364699627859,
|
| 608 |
+
"grad_norm": 2.4814623573220156,
|
| 609 |
+
"learning_rate": 9.527607679407545e-06,
|
| 610 |
+
"loss": 0.6439,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.22860180754917597,
|
| 615 |
+
"grad_norm": 2.507720516805312,
|
| 616 |
+
"learning_rate": 9.50771754245191e-06,
|
| 617 |
+
"loss": 0.6185,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.23125996810207336,
|
| 622 |
+
"grad_norm": 2.381710754559422,
|
| 623 |
+
"learning_rate": 9.487438904074581e-06,
|
| 624 |
+
"loss": 0.6183,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.23391812865497075,
|
| 629 |
+
"grad_norm": 2.4338731752331504,
|
| 630 |
+
"learning_rate": 9.466773512007032e-06,
|
| 631 |
+
"loss": 0.6345,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.23657628920786816,
|
| 636 |
+
"grad_norm": 2.475985053897146,
|
| 637 |
+
"learning_rate": 9.445723147313434e-06,
|
| 638 |
+
"loss": 0.6159,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.23923444976076555,
|
| 643 |
+
"grad_norm": 2.571693018211461,
|
| 644 |
+
"learning_rate": 9.424289624237143e-06,
|
| 645 |
+
"loss": 0.6195,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.24189261031366294,
|
| 650 |
+
"grad_norm": 2.4353278379885754,
|
| 651 |
+
"learning_rate": 9.402474790044348e-06,
|
| 652 |
+
"loss": 0.6171,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.24455077086656035,
|
| 657 |
+
"grad_norm": 2.5936563595071855,
|
| 658 |
+
"learning_rate": 9.380280524864846e-06,
|
| 659 |
+
"loss": 0.6087,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.24720893141945774,
|
| 664 |
+
"grad_norm": 2.6003404377422163,
|
| 665 |
+
"learning_rate": 9.357708741530025e-06,
|
| 666 |
+
"loss": 0.607,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.24986709197235513,
|
| 671 |
+
"grad_norm": 2.4539600499134804,
|
| 672 |
+
"learning_rate": 9.334761385407984e-06,
|
| 673 |
+
"loss": 0.6165,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.25252525252525254,
|
| 678 |
+
"grad_norm": 2.396035179563128,
|
| 679 |
+
"learning_rate": 9.311440434235879e-06,
|
| 680 |
+
"loss": 0.6131,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2551834130781499,
|
| 685 |
+
"grad_norm": 2.5173704413463214,
|
| 686 |
+
"learning_rate": 9.287747897949471e-06,
|
| 687 |
+
"loss": 0.6174,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2578415736310473,
|
| 692 |
+
"grad_norm": 2.35543719713761,
|
| 693 |
+
"learning_rate": 9.263685818509895e-06,
|
| 694 |
+
"loss": 0.624,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.26049973418394473,
|
| 699 |
+
"grad_norm": 2.4747468604717633,
|
| 700 |
+
"learning_rate": 9.239256269727675e-06,
|
| 701 |
+
"loss": 0.5977,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2631578947368421,
|
| 706 |
+
"grad_norm": 2.6700657630003457,
|
| 707 |
+
"learning_rate": 9.214461357083986e-06,
|
| 708 |
+
"loss": 0.6316,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2658160552897395,
|
| 713 |
+
"grad_norm": 2.6019820288128614,
|
| 714 |
+
"learning_rate": 9.189303217549195e-06,
|
| 715 |
+
"loss": 0.612,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2684742158426369,
|
| 720 |
+
"grad_norm": 2.4320657082358133,
|
| 721 |
+
"learning_rate": 9.163784019398686e-06,
|
| 722 |
+
"loss": 0.5974,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.2711323763955343,
|
| 727 |
+
"grad_norm": 2.3980618352258682,
|
| 728 |
+
"learning_rate": 9.137905962025977e-06,
|
| 729 |
+
"loss": 0.5974,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.2737905369484317,
|
| 734 |
+
"grad_norm": 2.5423673726235982,
|
| 735 |
+
"learning_rate": 9.111671275753175e-06,
|
| 736 |
+
"loss": 0.6067,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.27644869750132905,
|
| 741 |
+
"grad_norm": 2.4523167139493167,
|
| 742 |
+
"learning_rate": 9.08508222163875e-06,
|
| 743 |
+
"loss": 0.5995,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.27910685805422647,
|
| 748 |
+
"grad_norm": 2.4117951211461173,
|
| 749 |
+
"learning_rate": 9.058141091282656e-06,
|
| 750 |
+
"loss": 0.6013,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.2817650186071239,
|
| 755 |
+
"grad_norm": 2.4016801112230177,
|
| 756 |
+
"learning_rate": 9.030850206628836e-06,
|
| 757 |
+
"loss": 0.6012,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.28442317916002124,
|
| 762 |
+
"grad_norm": 2.528921774636397,
|
| 763 |
+
"learning_rate": 9.003211919765102e-06,
|
| 764 |
+
"loss": 0.6028,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.28708133971291866,
|
| 769 |
+
"grad_norm": 2.4727747007177077,
|
| 770 |
+
"learning_rate": 8.975228612720415e-06,
|
| 771 |
+
"loss": 0.5861,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2897395002658161,
|
| 776 |
+
"grad_norm": 2.4874651156635164,
|
| 777 |
+
"learning_rate": 8.946902697259593e-06,
|
| 778 |
+
"loss": 0.5919,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.29239766081871343,
|
| 783 |
+
"grad_norm": 2.4347654892463457,
|
| 784 |
+
"learning_rate": 8.918236614675446e-06,
|
| 785 |
+
"loss": 0.585,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.29505582137161085,
|
| 790 |
+
"grad_norm": 2.312313083900426,
|
| 791 |
+
"learning_rate": 8.889232835578372e-06,
|
| 792 |
+
"loss": 0.5921,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.29771398192450826,
|
| 797 |
+
"grad_norm": 2.4973530243506694,
|
| 798 |
+
"learning_rate": 8.859893859683429e-06,
|
| 799 |
+
"loss": 0.5899,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3003721424774056,
|
| 804 |
+
"grad_norm": 2.3656255407077316,
|
| 805 |
+
"learning_rate": 8.83022221559489e-06,
|
| 806 |
+
"loss": 0.5963,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.30303030303030304,
|
| 811 |
+
"grad_norm": 2.328665705259469,
|
| 812 |
+
"learning_rate": 8.800220460588321e-06,
|
| 813 |
+
"loss": 0.5957,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3056884635832004,
|
| 818 |
+
"grad_norm": 2.676801074518756,
|
| 819 |
+
"learning_rate": 8.769891180390168e-06,
|
| 820 |
+
"loss": 0.5647,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3083466241360978,
|
| 825 |
+
"grad_norm": 2.5775837307061216,
|
| 826 |
+
"learning_rate": 8.739236988954913e-06,
|
| 827 |
+
"loss": 0.5874,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.31100478468899523,
|
| 832 |
+
"grad_norm": 2.4252458009696922,
|
| 833 |
+
"learning_rate": 8.708260528239788e-06,
|
| 834 |
+
"loss": 0.5733,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3136629452418926,
|
| 839 |
+
"grad_norm": 2.4329948098909067,
|
| 840 |
+
"learning_rate": 8.676964467977072e-06,
|
| 841 |
+
"loss": 0.5843,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.31632110579479,
|
| 846 |
+
"grad_norm": 2.617177470446975,
|
| 847 |
+
"learning_rate": 8.645351505443997e-06,
|
| 848 |
+
"loss": 0.5779,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3189792663476874,
|
| 853 |
+
"grad_norm": 2.3752490352145506,
|
| 854 |
+
"learning_rate": 8.613424365230287e-06,
|
| 855 |
+
"loss": 0.5777,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3216374269005848,
|
| 860 |
+
"grad_norm": 2.5878429884840046,
|
| 861 |
+
"learning_rate": 8.581185799003334e-06,
|
| 862 |
+
"loss": 0.579,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3242955874534822,
|
| 867 |
+
"grad_norm": 2.4302231285818214,
|
| 868 |
+
"learning_rate": 8.548638585271034e-06,
|
| 869 |
+
"loss": 0.597,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3269537480063796,
|
| 874 |
+
"grad_norm": 2.493522528062739,
|
| 875 |
+
"learning_rate": 8.515785529142339e-06,
|
| 876 |
+
"loss": 0.5766,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.32961190855927697,
|
| 881 |
+
"grad_norm": 2.3498385912371074,
|
| 882 |
+
"learning_rate": 8.482629462085479e-06,
|
| 883 |
+
"loss": 0.5683,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3322700691121744,
|
| 888 |
+
"grad_norm": 2.3651303006417024,
|
| 889 |
+
"learning_rate": 8.449173241683934e-06,
|
| 890 |
+
"loss": 0.5687,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3349282296650718,
|
| 895 |
+
"grad_norm": 2.3964339925589635,
|
| 896 |
+
"learning_rate": 8.415419751390155e-06,
|
| 897 |
+
"loss": 0.5607,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.33758639021796916,
|
| 902 |
+
"grad_norm": 2.4646605695831636,
|
| 903 |
+
"learning_rate": 8.381371900277045e-06,
|
| 904 |
+
"loss": 0.5693,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.34024455077086657,
|
| 909 |
+
"grad_norm": 2.308318162132283,
|
| 910 |
+
"learning_rate": 8.347032622787245e-06,
|
| 911 |
+
"loss": 0.563,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.34290271132376393,
|
| 916 |
+
"grad_norm": 2.5479453691410545,
|
| 917 |
+
"learning_rate": 8.312404878480222e-06,
|
| 918 |
+
"loss": 0.545,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.34556087187666135,
|
| 923 |
+
"grad_norm": 2.4905341758041817,
|
| 924 |
+
"learning_rate": 8.277491651777196e-06,
|
| 925 |
+
"loss": 0.5565,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.34821903242955876,
|
| 930 |
+
"grad_norm": 2.38426332119931,
|
| 931 |
+
"learning_rate": 8.24229595170393e-06,
|
| 932 |
+
"loss": 0.5593,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3508771929824561,
|
| 937 |
+
"grad_norm": 2.257211174728351,
|
| 938 |
+
"learning_rate": 8.206820811631387e-06,
|
| 939 |
+
"loss": 0.5528,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.35353535353535354,
|
| 944 |
+
"grad_norm": 2.4047107162822305,
|
| 945 |
+
"learning_rate": 8.171069289014307e-06,
|
| 946 |
+
"loss": 0.5591,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.35619351408825095,
|
| 951 |
+
"grad_norm": 2.359712499357688,
|
| 952 |
+
"learning_rate": 8.135044465127687e-06,
|
| 953 |
+
"loss": 0.5567,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3588516746411483,
|
| 958 |
+
"grad_norm": 2.617789847429391,
|
| 959 |
+
"learning_rate": 8.098749444801226e-06,
|
| 960 |
+
"loss": 0.5768,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3615098351940457,
|
| 965 |
+
"grad_norm": 2.279043335158844,
|
| 966 |
+
"learning_rate": 8.062187356151726e-06,
|
| 967 |
+
"loss": 0.5584,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.36416799574694314,
|
| 972 |
+
"grad_norm": 2.5468624672218034,
|
| 973 |
+
"learning_rate": 8.025361350313506e-06,
|
| 974 |
+
"loss": 0.5596,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3668261562998405,
|
| 979 |
+
"grad_norm": 2.5020390139987945,
|
| 980 |
+
"learning_rate": 7.9882746011668e-06,
|
| 981 |
+
"loss": 0.5514,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.3694843168527379,
|
| 986 |
+
"grad_norm": 2.4750957061628154,
|
| 987 |
+
"learning_rate": 7.950930305064224e-06,
|
| 988 |
+
"loss": 0.5509,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3721424774056353,
|
| 993 |
+
"grad_norm": 2.3205069346022684,
|
| 994 |
+
"learning_rate": 7.913331680555299e-06,
|
| 995 |
+
"loss": 0.5475,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3748006379585327,
|
| 1000 |
+
"grad_norm": 2.2679048857253625,
|
| 1001 |
+
"learning_rate": 7.875481968109052e-06,
|
| 1002 |
+
"loss": 0.5603,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.3774587985114301,
|
| 1007 |
+
"grad_norm": 2.281345445666937,
|
| 1008 |
+
"learning_rate": 7.837384429834736e-06,
|
| 1009 |
+
"loss": 0.5529,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.38011695906432746,
|
| 1014 |
+
"grad_norm": 2.4761501881024373,
|
| 1015 |
+
"learning_rate": 7.799042349200672e-06,
|
| 1016 |
+
"loss": 0.5462,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3827751196172249,
|
| 1021 |
+
"grad_norm": 2.3525147379450133,
|
| 1022 |
+
"learning_rate": 7.760459030751285e-06,
|
| 1023 |
+
"loss": 0.5288,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3854332801701223,
|
| 1028 |
+
"grad_norm": 2.3082844177093396,
|
| 1029 |
+
"learning_rate": 7.721637799822269e-06,
|
| 1030 |
+
"loss": 0.5444,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.38809144072301965,
|
| 1035 |
+
"grad_norm": 2.428710636700351,
|
| 1036 |
+
"learning_rate": 7.682582002254015e-06,
|
| 1037 |
+
"loss": 0.5526,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.39074960127591707,
|
| 1042 |
+
"grad_norm": 2.4386787256645035,
|
| 1043 |
+
"learning_rate": 7.643295004103232e-06,
|
| 1044 |
+
"loss": 0.5322,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.3934077618288145,
|
| 1049 |
+
"grad_norm": 2.5428861644636838,
|
| 1050 |
+
"learning_rate": 7.6037801913528474e-06,
|
| 1051 |
+
"loss": 0.5415,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.39606592238171184,
|
| 1056 |
+
"grad_norm": 2.284207445173577,
|
| 1057 |
+
"learning_rate": 7.564040969620179e-06,
|
| 1058 |
+
"loss": 0.5206,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.39872408293460926,
|
| 1063 |
+
"grad_norm": 2.4142085070722357,
|
| 1064 |
+
"learning_rate": 7.524080763863422e-06,
|
| 1065 |
+
"loss": 0.5447,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4013822434875066,
|
| 1070 |
+
"grad_norm": 2.524753101311306,
|
| 1071 |
+
"learning_rate": 7.483903018086466e-06,
|
| 1072 |
+
"loss": 0.5318,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.40404040404040403,
|
| 1077 |
+
"grad_norm": 2.4624906545945455,
|
| 1078 |
+
"learning_rate": 7.443511195042068e-06,
|
| 1079 |
+
"loss": 0.536,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.40669856459330145,
|
| 1084 |
+
"grad_norm": 2.584362652216828,
|
| 1085 |
+
"learning_rate": 7.402908775933419e-06,
|
| 1086 |
+
"loss": 0.5335,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4093567251461988,
|
| 1091 |
+
"grad_norm": 2.3792982549216823,
|
| 1092 |
+
"learning_rate": 7.362099260114104e-06,
|
| 1093 |
+
"loss": 0.5341,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4120148856990962,
|
| 1098 |
+
"grad_norm": 2.377464889890606,
|
| 1099 |
+
"learning_rate": 7.321086164786513e-06,
|
| 1100 |
+
"loss": 0.5369,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.41467304625199364,
|
| 1105 |
+
"grad_norm": 2.380837303007001,
|
| 1106 |
+
"learning_rate": 7.2798730246987056e-06,
|
| 1107 |
+
"loss": 0.5201,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.417331206804891,
|
| 1112 |
+
"grad_norm": 2.330648347753563,
|
| 1113 |
+
"learning_rate": 7.23846339183977e-06,
|
| 1114 |
+
"loss": 0.5119,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4199893673577884,
|
| 1119 |
+
"grad_norm": 2.3284141931963362,
|
| 1120 |
+
"learning_rate": 7.196860835133686e-06,
|
| 1121 |
+
"loss": 0.5276,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4226475279106858,
|
| 1126 |
+
"grad_norm": 2.3728628327177774,
|
| 1127 |
+
"learning_rate": 7.155068940131741e-06,
|
| 1128 |
+
"loss": 0.5057,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4253056884635832,
|
| 1133 |
+
"grad_norm": 2.386729416032167,
|
| 1134 |
+
"learning_rate": 7.113091308703498e-06,
|
| 1135 |
+
"loss": 0.5246,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4279638490164806,
|
| 1140 |
+
"grad_norm": 2.419791153850835,
|
| 1141 |
+
"learning_rate": 7.070931558726373e-06,
|
| 1142 |
+
"loss": 0.5178,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.430622009569378,
|
| 1147 |
+
"grad_norm": 2.5457575458937183,
|
| 1148 |
+
"learning_rate": 7.028593323773819e-06,
|
| 1149 |
+
"loss": 0.5222,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4332801701222754,
|
| 1154 |
+
"grad_norm": 2.327614444176051,
|
| 1155 |
+
"learning_rate": 6.9860802528021705e-06,
|
| 1156 |
+
"loss": 0.5034,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4359383306751728,
|
| 1161 |
+
"grad_norm": 2.3495489783212995,
|
| 1162 |
+
"learning_rate": 6.943396009836147e-06,
|
| 1163 |
+
"loss": 0.4935,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.43859649122807015,
|
| 1168 |
+
"grad_norm": 2.6463956003818256,
|
| 1169 |
+
"learning_rate": 6.9005442736530745e-06,
|
| 1170 |
+
"loss": 0.5076,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.44125465178096757,
|
| 1175 |
+
"grad_norm": 2.3848887187645564,
|
| 1176 |
+
"learning_rate": 6.8575287374658185e-06,
|
| 1177 |
+
"loss": 0.5217,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.443912812333865,
|
| 1182 |
+
"grad_norm": 2.556811219304697,
|
| 1183 |
+
"learning_rate": 6.814353108604488e-06,
|
| 1184 |
+
"loss": 0.4953,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.44657097288676234,
|
| 1189 |
+
"grad_norm": 2.378175171662646,
|
| 1190 |
+
"learning_rate": 6.771021108196912e-06,
|
| 1191 |
+
"loss": 0.5031,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.44922913343965976,
|
| 1196 |
+
"grad_norm": 2.359632767008804,
|
| 1197 |
+
"learning_rate": 6.7275364708479316e-06,
|
| 1198 |
+
"loss": 0.4892,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.45188729399255717,
|
| 1203 |
+
"grad_norm": 2.5062942550983163,
|
| 1204 |
+
"learning_rate": 6.683902944317535e-06,
|
| 1205 |
+
"loss": 0.5138,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.45454545454545453,
|
| 1210 |
+
"grad_norm": 2.325718024430296,
|
| 1211 |
+
"learning_rate": 6.640124289197845e-06,
|
| 1212 |
+
"loss": 0.4969,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.45720361509835195,
|
| 1217 |
+
"grad_norm": 2.4421884573669295,
|
| 1218 |
+
"learning_rate": 6.596204278589019e-06,
|
| 1219 |
+
"loss": 0.5063,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.45986177565124936,
|
| 1224 |
+
"grad_norm": 2.3219816814492167,
|
| 1225 |
+
"learning_rate": 6.552146697774049e-06,
|
| 1226 |
+
"loss": 0.4947,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4625199362041467,
|
| 1231 |
+
"grad_norm": 2.2777725562246816,
|
| 1232 |
+
"learning_rate": 6.507955343892536e-06,
|
| 1233 |
+
"loss": 0.5039,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.46517809675704413,
|
| 1238 |
+
"grad_norm": 2.3140852762780435,
|
| 1239 |
+
"learning_rate": 6.4636340256134224e-06,
|
| 1240 |
+
"loss": 0.4839,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.4678362573099415,
|
| 1245 |
+
"grad_norm": 2.4627138587993214,
|
| 1246 |
+
"learning_rate": 6.419186562806742e-06,
|
| 1247 |
+
"loss": 0.4908,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4704944178628389,
|
| 1252 |
+
"grad_norm": 2.401799791274382,
|
| 1253 |
+
"learning_rate": 6.374616786214402e-06,
|
| 1254 |
+
"loss": 0.4888,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4731525784157363,
|
| 1259 |
+
"grad_norm": 2.3301260740564707,
|
| 1260 |
+
"learning_rate": 6.329928537120024e-06,
|
| 1261 |
+
"loss": 0.5022,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4758107389686337,
|
| 1266 |
+
"grad_norm": 2.406357131352656,
|
| 1267 |
+
"learning_rate": 6.285125667017886e-06,
|
| 1268 |
+
"loss": 0.4919,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.4784688995215311,
|
| 1273 |
+
"grad_norm": 2.4285508393783557,
|
| 1274 |
+
"learning_rate": 6.240212037280967e-06,
|
| 1275 |
+
"loss": 0.492,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4811270600744285,
|
| 1280 |
+
"grad_norm": 2.266893075319626,
|
| 1281 |
+
"learning_rate": 6.195191518828163e-06,
|
| 1282 |
+
"loss": 0.4921,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.4837852206273259,
|
| 1287 |
+
"grad_norm": 2.416578670315929,
|
| 1288 |
+
"learning_rate": 6.1500679917906615e-06,
|
| 1289 |
+
"loss": 0.486,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4864433811802233,
|
| 1294 |
+
"grad_norm": 2.473861188532911,
|
| 1295 |
+
"learning_rate": 6.1048453451775305e-06,
|
| 1296 |
+
"loss": 0.4798,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.4891015417331207,
|
| 1301 |
+
"grad_norm": 2.3504741024043443,
|
| 1302 |
+
"learning_rate": 6.059527476540546e-06,
|
| 1303 |
+
"loss": 0.4781,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.49175970228601806,
|
| 1308 |
+
"grad_norm": 2.382117365607024,
|
| 1309 |
+
"learning_rate": 6.014118291638272e-06,
|
| 1310 |
+
"loss": 0.4689,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.4944178628389155,
|
| 1315 |
+
"grad_norm": 2.4835266299739307,
|
| 1316 |
+
"learning_rate": 5.96862170409944e-06,
|
| 1317 |
+
"loss": 0.4799,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.49707602339181284,
|
| 1322 |
+
"grad_norm": 2.285188764967759,
|
| 1323 |
+
"learning_rate": 5.9230416350856505e-06,
|
| 1324 |
+
"loss": 0.4703,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.49973418394471025,
|
| 1329 |
+
"grad_norm": 2.4072497294257373,
|
| 1330 |
+
"learning_rate": 5.877382012953429e-06,
|
| 1331 |
+
"loss": 0.4738,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5023923444976076,
|
| 1336 |
+
"grad_norm": 2.3497927288764755,
|
| 1337 |
+
"learning_rate": 5.831646772915651e-06,
|
| 1338 |
+
"loss": 0.465,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5050505050505051,
|
| 1343 |
+
"grad_norm": 2.2889113825693626,
|
| 1344 |
+
"learning_rate": 5.785839856702383e-06,
|
| 1345 |
+
"loss": 0.4676,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5077086656034024,
|
| 1350 |
+
"grad_norm": 2.444844273958781,
|
| 1351 |
+
"learning_rate": 5.739965212221168e-06,
|
| 1352 |
+
"loss": 0.4761,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5103668261562998,
|
| 1357 |
+
"grad_norm": 2.2070310255406236,
|
| 1358 |
+
"learning_rate": 5.69402679321676e-06,
|
| 1359 |
+
"loss": 0.474,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5130249867091973,
|
| 1364 |
+
"grad_norm": 2.484818119826392,
|
| 1365 |
+
"learning_rate": 5.64802855893038e-06,
|
| 1366 |
+
"loss": 0.4648,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5156831472620946,
|
| 1371 |
+
"grad_norm": 2.2539805183718804,
|
| 1372 |
+
"learning_rate": 5.601974473758472e-06,
|
| 1373 |
+
"loss": 0.4725,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.518341307814992,
|
| 1378 |
+
"grad_norm": 2.2192494500574265,
|
| 1379 |
+
"learning_rate": 5.5558685069110444e-06,
|
| 1380 |
+
"loss": 0.4684,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5209994683678895,
|
| 1385 |
+
"grad_norm": 2.269435937295936,
|
| 1386 |
+
"learning_rate": 5.509714632069564e-06,
|
| 1387 |
+
"loss": 0.4653,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5236576289207868,
|
| 1392 |
+
"grad_norm": 2.5740150564664184,
|
| 1393 |
+
"learning_rate": 5.463516827044492e-06,
|
| 1394 |
+
"loss": 0.4716,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5263157894736842,
|
| 1399 |
+
"grad_norm": 2.310418189573713,
|
| 1400 |
+
"learning_rate": 5.41727907343245e-06,
|
| 1401 |
+
"loss": 0.4674,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5289739500265817,
|
| 1406 |
+
"grad_norm": 2.527471668820078,
|
| 1407 |
+
"learning_rate": 5.371005356273058e-06,
|
| 1408 |
+
"loss": 0.495,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.531632110579479,
|
| 1413 |
+
"grad_norm": 2.139852400045247,
|
| 1414 |
+
"learning_rate": 5.32469966370549e-06,
|
| 1415 |
+
"loss": 0.4612,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5342902711323764,
|
| 1420 |
+
"grad_norm": 2.3593017053732783,
|
| 1421 |
+
"learning_rate": 5.278365986624743e-06,
|
| 1422 |
+
"loss": 0.4763,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5369484316852738,
|
| 1427 |
+
"grad_norm": 2.2465306122210844,
|
| 1428 |
+
"learning_rate": 5.232008318337682e-06,
|
| 1429 |
+
"loss": 0.4533,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5396065922381712,
|
| 1434 |
+
"grad_norm": 2.326189324358019,
|
| 1435 |
+
"learning_rate": 5.1856306542188805e-06,
|
| 1436 |
+
"loss": 0.4509,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.5422647527910686,
|
| 1441 |
+
"grad_norm": 2.3918713302924526,
|
| 1442 |
+
"learning_rate": 5.1392369913662646e-06,
|
| 1443 |
+
"loss": 0.4667,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.5449229133439659,
|
| 1448 |
+
"grad_norm": 2.2595301249124624,
|
| 1449 |
+
"learning_rate": 5.0928313282566255e-06,
|
| 1450 |
+
"loss": 0.4478,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.5475810738968634,
|
| 1455 |
+
"grad_norm": 2.422833352188881,
|
| 1456 |
+
"learning_rate": 5.046417664401005e-06,
|
| 1457 |
+
"loss": 0.4468,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5502392344497608,
|
| 1462 |
+
"grad_norm": 2.416580263699029,
|
| 1463 |
+
"learning_rate": 5e-06,
|
| 1464 |
+
"loss": 0.4666,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.5528973950026581,
|
| 1469 |
+
"grad_norm": 2.2774632375053088,
|
| 1470 |
+
"learning_rate": 4.953582335598996e-06,
|
| 1471 |
+
"loss": 0.4548,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.5555555555555556,
|
| 1476 |
+
"grad_norm": 2.2466213084914872,
|
| 1477 |
+
"learning_rate": 4.907168671743377e-06,
|
| 1478 |
+
"loss": 0.4413,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5582137161084529,
|
| 1483 |
+
"grad_norm": 2.1195454673730447,
|
| 1484 |
+
"learning_rate": 4.860763008633736e-06,
|
| 1485 |
+
"loss": 0.4441,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5608718766613503,
|
| 1490 |
+
"grad_norm": 2.4552591249815836,
|
| 1491 |
+
"learning_rate": 4.814369345781121e-06,
|
| 1492 |
+
"loss": 0.4486,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5635300372142478,
|
| 1497 |
+
"grad_norm": 2.2160694909825316,
|
| 1498 |
+
"learning_rate": 4.7679916816623185e-06,
|
| 1499 |
+
"loss": 0.4493,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5661881977671451,
|
| 1504 |
+
"grad_norm": 2.34664383603718,
|
| 1505 |
+
"learning_rate": 4.7216340133752604e-06,
|
| 1506 |
+
"loss": 0.4572,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5688463583200425,
|
| 1511 |
+
"grad_norm": 2.3926383813223455,
|
| 1512 |
+
"learning_rate": 4.675300336294511e-06,
|
| 1513 |
+
"loss": 0.4417,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.57150451887294,
|
| 1518 |
+
"grad_norm": 2.264865501037663,
|
| 1519 |
+
"learning_rate": 4.628994643726942e-06,
|
| 1520 |
+
"loss": 0.4479,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5741626794258373,
|
| 1525 |
+
"grad_norm": 2.4178547328629416,
|
| 1526 |
+
"learning_rate": 4.582720926567552e-06,
|
| 1527 |
+
"loss": 0.4296,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5768208399787347,
|
| 1532 |
+
"grad_norm": 2.2005790727461183,
|
| 1533 |
+
"learning_rate": 4.53648317295551e-06,
|
| 1534 |
+
"loss": 0.4439,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5794790005316321,
|
| 1539 |
+
"grad_norm": 2.2954610706486838,
|
| 1540 |
+
"learning_rate": 4.490285367930438e-06,
|
| 1541 |
+
"loss": 0.4254,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5821371610845295,
|
| 1546 |
+
"grad_norm": 2.26298044130641,
|
| 1547 |
+
"learning_rate": 4.444131493088956e-06,
|
| 1548 |
+
"loss": 0.4361,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5847953216374269,
|
| 1553 |
+
"grad_norm": 2.4783364089595383,
|
| 1554 |
+
"learning_rate": 4.3980255262415295e-06,
|
| 1555 |
+
"loss": 0.4232,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5874534821903243,
|
| 1560 |
+
"grad_norm": 2.2956707433327446,
|
| 1561 |
+
"learning_rate": 4.351971441069622e-06,
|
| 1562 |
+
"loss": 0.4336,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5901116427432217,
|
| 1567 |
+
"grad_norm": 2.1801979127212445,
|
| 1568 |
+
"learning_rate": 4.305973206783241e-06,
|
| 1569 |
+
"loss": 0.4205,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.5927698032961191,
|
| 1574 |
+
"grad_norm": 2.3074118997829944,
|
| 1575 |
+
"learning_rate": 4.260034787778833e-06,
|
| 1576 |
+
"loss": 0.4403,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5954279638490165,
|
| 1581 |
+
"grad_norm": 2.2135932778785117,
|
| 1582 |
+
"learning_rate": 4.214160143297618e-06,
|
| 1583 |
+
"loss": 0.4484,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5980861244019139,
|
| 1588 |
+
"grad_norm": 2.3686411194735943,
|
| 1589 |
+
"learning_rate": 4.1683532270843505e-06,
|
| 1590 |
+
"loss": 0.4307,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6007442849548112,
|
| 1595 |
+
"grad_norm": 2.2766648664475753,
|
| 1596 |
+
"learning_rate": 4.122617987046571e-06,
|
| 1597 |
+
"loss": 0.4281,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6034024455077087,
|
| 1602 |
+
"grad_norm": 2.3707141910175165,
|
| 1603 |
+
"learning_rate": 4.076958364914352e-06,
|
| 1604 |
+
"loss": 0.4363,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6060606060606061,
|
| 1609 |
+
"grad_norm": 2.3439272228715304,
|
| 1610 |
+
"learning_rate": 4.031378295900562e-06,
|
| 1611 |
+
"loss": 0.4332,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6087187666135034,
|
| 1616 |
+
"grad_norm": 2.2202025277585737,
|
| 1617 |
+
"learning_rate": 3.985881708361729e-06,
|
| 1618 |
+
"loss": 0.421,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6113769271664008,
|
| 1623 |
+
"grad_norm": 2.270428210567009,
|
| 1624 |
+
"learning_rate": 3.940472523459456e-06,
|
| 1625 |
+
"loss": 0.4182,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6140350877192983,
|
| 1630 |
+
"grad_norm": 2.303418109483892,
|
| 1631 |
+
"learning_rate": 3.895154654822471e-06,
|
| 1632 |
+
"loss": 0.4139,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6166932482721956,
|
| 1637 |
+
"grad_norm": 2.2123434660028582,
|
| 1638 |
+
"learning_rate": 3.84993200820934e-06,
|
| 1639 |
+
"loss": 0.417,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.619351408825093,
|
| 1644 |
+
"grad_norm": 2.42258236412943,
|
| 1645 |
+
"learning_rate": 3.8048084811718377e-06,
|
| 1646 |
+
"loss": 0.4414,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6220095693779905,
|
| 1651 |
+
"grad_norm": 2.2752955087581768,
|
| 1652 |
+
"learning_rate": 3.7597879627190337e-06,
|
| 1653 |
+
"loss": 0.4182,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6246677299308878,
|
| 1658 |
+
"grad_norm": 2.4664245003262546,
|
| 1659 |
+
"learning_rate": 3.7148743329821146e-06,
|
| 1660 |
+
"loss": 0.4317,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6273258904837852,
|
| 1665 |
+
"grad_norm": 2.290843653500035,
|
| 1666 |
+
"learning_rate": 3.670071462879975e-06,
|
| 1667 |
+
"loss": 0.4186,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6299840510366826,
|
| 1672 |
+
"grad_norm": 2.277338208428442,
|
| 1673 |
+
"learning_rate": 3.6253832137856e-06,
|
| 1674 |
+
"loss": 0.4036,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.63264221158958,
|
| 1679 |
+
"grad_norm": 2.328350906451804,
|
| 1680 |
+
"learning_rate": 3.5808134371932603e-06,
|
| 1681 |
+
"loss": 0.4029,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6353003721424774,
|
| 1686 |
+
"grad_norm": 2.342438907892454,
|
| 1687 |
+
"learning_rate": 3.5363659743865797e-06,
|
| 1688 |
+
"loss": 0.4204,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.6379585326953748,
|
| 1693 |
+
"grad_norm": 2.4035043063158237,
|
| 1694 |
+
"learning_rate": 3.4920446561074673e-06,
|
| 1695 |
+
"loss": 0.4218,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.6406166932482722,
|
| 1700 |
+
"grad_norm": 2.224515711259058,
|
| 1701 |
+
"learning_rate": 3.4478533022259527e-06,
|
| 1702 |
+
"loss": 0.4152,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.6432748538011696,
|
| 1707 |
+
"grad_norm": 2.1644969476013847,
|
| 1708 |
+
"learning_rate": 3.403795721410983e-06,
|
| 1709 |
+
"loss": 0.4235,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.645933014354067,
|
| 1714 |
+
"grad_norm": 2.225901225466717,
|
| 1715 |
+
"learning_rate": 3.3598757108021546e-06,
|
| 1716 |
+
"loss": 0.4205,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.6485911749069644,
|
| 1721 |
+
"grad_norm": 2.960907864932211,
|
| 1722 |
+
"learning_rate": 3.3160970556824666e-06,
|
| 1723 |
+
"loss": 0.4195,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.6512493354598617,
|
| 1728 |
+
"grad_norm": 2.2220590721554303,
|
| 1729 |
+
"learning_rate": 3.2724635291520697e-06,
|
| 1730 |
+
"loss": 0.4039,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.6539074960127592,
|
| 1735 |
+
"grad_norm": 2.222896907055418,
|
| 1736 |
+
"learning_rate": 3.2289788918030894e-06,
|
| 1737 |
+
"loss": 0.4037,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.6565656565656566,
|
| 1742 |
+
"grad_norm": 2.546975428146142,
|
| 1743 |
+
"learning_rate": 3.185646891395514e-06,
|
| 1744 |
+
"loss": 0.4215,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.6592238171185539,
|
| 1749 |
+
"grad_norm": 2.26370807319166,
|
| 1750 |
+
"learning_rate": 3.1424712625341836e-06,
|
| 1751 |
+
"loss": 0.4095,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.6618819776714514,
|
| 1756 |
+
"grad_norm": 2.1848679188015074,
|
| 1757 |
+
"learning_rate": 3.0994557263469267e-06,
|
| 1758 |
+
"loss": 0.3984,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.6645401382243488,
|
| 1763 |
+
"grad_norm": 2.1394700353392424,
|
| 1764 |
+
"learning_rate": 3.0566039901638534e-06,
|
| 1765 |
+
"loss": 0.4086,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.6671982987772461,
|
| 1770 |
+
"grad_norm": 2.2219765632899473,
|
| 1771 |
+
"learning_rate": 3.013919747197832e-06,
|
| 1772 |
+
"loss": 0.4018,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.6698564593301436,
|
| 1777 |
+
"grad_norm": 2.2717838227817557,
|
| 1778 |
+
"learning_rate": 2.9714066762261825e-06,
|
| 1779 |
+
"loss": 0.4228,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.672514619883041,
|
| 1784 |
+
"grad_norm": 2.1650786912116855,
|
| 1785 |
+
"learning_rate": 2.929068441273629e-06,
|
| 1786 |
+
"loss": 0.4032,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6751727804359383,
|
| 1791 |
+
"grad_norm": 2.30009309210609,
|
| 1792 |
+
"learning_rate": 2.886908691296504e-06,
|
| 1793 |
+
"loss": 0.4117,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6778309409888357,
|
| 1798 |
+
"grad_norm": 2.326017307920666,
|
| 1799 |
+
"learning_rate": 2.844931059868261e-06,
|
| 1800 |
+
"loss": 0.4013,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6804891015417331,
|
| 1805 |
+
"grad_norm": 2.1732526869160007,
|
| 1806 |
+
"learning_rate": 2.8031391648663153e-06,
|
| 1807 |
+
"loss": 0.3979,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6831472620946305,
|
| 1812 |
+
"grad_norm": 2.159518899715084,
|
| 1813 |
+
"learning_rate": 2.7615366081602306e-06,
|
| 1814 |
+
"loss": 0.4057,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6858054226475279,
|
| 1819 |
+
"grad_norm": 2.2767441852506036,
|
| 1820 |
+
"learning_rate": 2.720126975301297e-06,
|
| 1821 |
+
"loss": 0.4014,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6884635832004253,
|
| 1826 |
+
"grad_norm": 2.3794328152713558,
|
| 1827 |
+
"learning_rate": 2.6789138352134885e-06,
|
| 1828 |
+
"loss": 0.4055,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6911217437533227,
|
| 1833 |
+
"grad_norm": 2.1759938658536826,
|
| 1834 |
+
"learning_rate": 2.637900739885897e-06,
|
| 1835 |
+
"loss": 0.398,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.69377990430622,
|
| 1840 |
+
"grad_norm": 2.463151016723552,
|
| 1841 |
+
"learning_rate": 2.5970912240665815e-06,
|
| 1842 |
+
"loss": 0.3929,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6964380648591175,
|
| 1847 |
+
"grad_norm": 2.212295559136495,
|
| 1848 |
+
"learning_rate": 2.556488804957933e-06,
|
| 1849 |
+
"loss": 0.3912,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6990962254120149,
|
| 1854 |
+
"grad_norm": 2.2913230696574782,
|
| 1855 |
+
"learning_rate": 2.5160969819135368e-06,
|
| 1856 |
+
"loss": 0.3965,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7017543859649122,
|
| 1861 |
+
"grad_norm": 2.1524999645283933,
|
| 1862 |
+
"learning_rate": 2.475919236136579e-06,
|
| 1863 |
+
"loss": 0.3979,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7044125465178097,
|
| 1868 |
+
"grad_norm": 2.0851767285574727,
|
| 1869 |
+
"learning_rate": 2.4359590303798243e-06,
|
| 1870 |
+
"loss": 0.378,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7070707070707071,
|
| 1875 |
+
"grad_norm": 2.421263572521297,
|
| 1876 |
+
"learning_rate": 2.3962198086471534e-06,
|
| 1877 |
+
"loss": 0.3944,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7097288676236044,
|
| 1882 |
+
"grad_norm": 2.0823768908851052,
|
| 1883 |
+
"learning_rate": 2.356704995896768e-06,
|
| 1884 |
+
"loss": 0.4023,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7123870281765019,
|
| 1889 |
+
"grad_norm": 2.127738595725943,
|
| 1890 |
+
"learning_rate": 2.3174179977459853e-06,
|
| 1891 |
+
"loss": 0.3863,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7150451887293993,
|
| 1896 |
+
"grad_norm": 2.213409237609825,
|
| 1897 |
+
"learning_rate": 2.2783622001777322e-06,
|
| 1898 |
+
"loss": 0.3911,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7177033492822966,
|
| 1903 |
+
"grad_norm": 2.169892497104111,
|
| 1904 |
+
"learning_rate": 2.2395409692487174e-06,
|
| 1905 |
+
"loss": 0.3878,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7203615098351941,
|
| 1910 |
+
"grad_norm": 3.002256232083388,
|
| 1911 |
+
"learning_rate": 2.2009576507993273e-06,
|
| 1912 |
+
"loss": 0.3924,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.7230196703880915,
|
| 1917 |
+
"grad_norm": 2.280280445481655,
|
| 1918 |
+
"learning_rate": 2.1626155701652678e-06,
|
| 1919 |
+
"loss": 0.389,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.7256778309409888,
|
| 1924 |
+
"grad_norm": 2.116741751685681,
|
| 1925 |
+
"learning_rate": 2.1245180318909482e-06,
|
| 1926 |
+
"loss": 0.3862,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.7283359914938863,
|
| 1931 |
+
"grad_norm": 2.137413236942858,
|
| 1932 |
+
"learning_rate": 2.0866683194447014e-06,
|
| 1933 |
+
"loss": 0.3876,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.7309941520467836,
|
| 1938 |
+
"grad_norm": 2.229506939617801,
|
| 1939 |
+
"learning_rate": 2.0490696949357774e-06,
|
| 1940 |
+
"loss": 0.3787,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.733652312599681,
|
| 1945 |
+
"grad_norm": 2.1611083069515398,
|
| 1946 |
+
"learning_rate": 2.0117253988332023e-06,
|
| 1947 |
+
"loss": 0.3831,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.7363104731525785,
|
| 1952 |
+
"grad_norm": 2.3092042098288466,
|
| 1953 |
+
"learning_rate": 1.974638649686495e-06,
|
| 1954 |
+
"loss": 0.3752,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.7389686337054758,
|
| 1959 |
+
"grad_norm": 2.32230572914138,
|
| 1960 |
+
"learning_rate": 1.9378126438482727e-06,
|
| 1961 |
+
"loss": 0.3956,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7416267942583732,
|
| 1966 |
+
"grad_norm": 2.1871471331082404,
|
| 1967 |
+
"learning_rate": 1.9012505551987764e-06,
|
| 1968 |
+
"loss": 0.3862,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.7442849548112705,
|
| 1973 |
+
"grad_norm": 2.2031667706286253,
|
| 1974 |
+
"learning_rate": 1.8649555348723137e-06,
|
| 1975 |
+
"loss": 0.3988,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.746943115364168,
|
| 1980 |
+
"grad_norm": 2.1919111575981014,
|
| 1981 |
+
"learning_rate": 1.8289307109856941e-06,
|
| 1982 |
+
"loss": 0.3831,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.7496012759170654,
|
| 1987 |
+
"grad_norm": 2.248650896915019,
|
| 1988 |
+
"learning_rate": 1.7931791883686155e-06,
|
| 1989 |
+
"loss": 0.3755,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.7522594364699627,
|
| 1994 |
+
"grad_norm": 2.189295928946117,
|
| 1995 |
+
"learning_rate": 1.7577040482960723e-06,
|
| 1996 |
+
"loss": 0.377,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.7549175970228602,
|
| 2001 |
+
"grad_norm": 2.1283848120533038,
|
| 2002 |
+
"learning_rate": 1.722508348222805e-06,
|
| 2003 |
+
"loss": 0.3795,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.7575757575757576,
|
| 2008 |
+
"grad_norm": 2.304562723613879,
|
| 2009 |
+
"learning_rate": 1.6875951215197779e-06,
|
| 2010 |
+
"loss": 0.3727,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.7602339181286549,
|
| 2015 |
+
"grad_norm": 2.2072516461560414,
|
| 2016 |
+
"learning_rate": 1.6529673772127563e-06,
|
| 2017 |
+
"loss": 0.3816,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.7628920786815524,
|
| 2022 |
+
"grad_norm": 2.1146394474437087,
|
| 2023 |
+
"learning_rate": 1.618628099722957e-06,
|
| 2024 |
+
"loss": 0.3809,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.7655502392344498,
|
| 2029 |
+
"grad_norm": 2.208201316045487,
|
| 2030 |
+
"learning_rate": 1.5845802486098461e-06,
|
| 2031 |
+
"loss": 0.3815,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.7682083997873471,
|
| 2036 |
+
"grad_norm": 2.1555651439065744,
|
| 2037 |
+
"learning_rate": 1.550826758316068e-06,
|
| 2038 |
+
"loss": 0.3868,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.7708665603402446,
|
| 2043 |
+
"grad_norm": 2.0230904275428787,
|
| 2044 |
+
"learning_rate": 1.5173705379145214e-06,
|
| 2045 |
+
"loss": 0.369,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.773524720893142,
|
| 2050 |
+
"grad_norm": 2.102831085552095,
|
| 2051 |
+
"learning_rate": 1.4842144708576606e-06,
|
| 2052 |
+
"loss": 0.3856,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.7761828814460393,
|
| 2057 |
+
"grad_norm": 2.0750334703551876,
|
| 2058 |
+
"learning_rate": 1.4513614147289663e-06,
|
| 2059 |
+
"loss": 0.3787,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.7788410419989368,
|
| 2064 |
+
"grad_norm": 2.1638405819961952,
|
| 2065 |
+
"learning_rate": 1.4188142009966689e-06,
|
| 2066 |
+
"loss": 0.3849,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.7814992025518341,
|
| 2071 |
+
"grad_norm": 2.1286655426576675,
|
| 2072 |
+
"learning_rate": 1.386575634769714e-06,
|
| 2073 |
+
"loss": 0.3698,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7841573631047315,
|
| 2078 |
+
"grad_norm": 2.235079695000213,
|
| 2079 |
+
"learning_rate": 1.3546484945560029e-06,
|
| 2080 |
+
"loss": 0.3719,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.786815523657629,
|
| 2085 |
+
"grad_norm": 2.12112362882432,
|
| 2086 |
+
"learning_rate": 1.3230355320229305e-06,
|
| 2087 |
+
"loss": 0.3867,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7894736842105263,
|
| 2092 |
+
"grad_norm": 2.3360347870351466,
|
| 2093 |
+
"learning_rate": 1.2917394717602123e-06,
|
| 2094 |
+
"loss": 0.3714,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7921318447634237,
|
| 2099 |
+
"grad_norm": 2.100340041221499,
|
| 2100 |
+
"learning_rate": 1.2607630110450874e-06,
|
| 2101 |
+
"loss": 0.359,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7947900053163212,
|
| 2106 |
+
"grad_norm": 2.2736377834909356,
|
| 2107 |
+
"learning_rate": 1.2301088196098332e-06,
|
| 2108 |
+
"loss": 0.3552,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7974481658692185,
|
| 2113 |
+
"grad_norm": 2.1462090737136763,
|
| 2114 |
+
"learning_rate": 1.1997795394116802e-06,
|
| 2115 |
+
"loss": 0.3706,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8001063264221159,
|
| 2120 |
+
"grad_norm": 2.330927480070287,
|
| 2121 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 2122 |
+
"loss": 0.3634,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8027644869750132,
|
| 2127 |
+
"grad_norm": 2.1032603761667343,
|
| 2128 |
+
"learning_rate": 1.140106140316572e-06,
|
| 2129 |
+
"loss": 0.3595,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8054226475279107,
|
| 2134 |
+
"grad_norm": 2.279846255920788,
|
| 2135 |
+
"learning_rate": 1.1107671644216305e-06,
|
| 2136 |
+
"loss": 0.3538,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8080808080808081,
|
| 2141 |
+
"grad_norm": 2.154018660092749,
|
| 2142 |
+
"learning_rate": 1.081763385324555e-06,
|
| 2143 |
+
"loss": 0.3631,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8107389686337054,
|
| 2148 |
+
"grad_norm": 2.143933197813529,
|
| 2149 |
+
"learning_rate": 1.0530973027404073e-06,
|
| 2150 |
+
"loss": 0.3681,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.8133971291866029,
|
| 2155 |
+
"grad_norm": 2.1792844064843275,
|
| 2156 |
+
"learning_rate": 1.024771387279585e-06,
|
| 2157 |
+
"loss": 0.3486,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.8160552897395003,
|
| 2162 |
+
"grad_norm": 2.1425132944113607,
|
| 2163 |
+
"learning_rate": 9.967880802348989e-07,
|
| 2164 |
+
"loss": 0.3733,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.8187134502923976,
|
| 2169 |
+
"grad_norm": 2.1700552513537894,
|
| 2170 |
+
"learning_rate": 9.691497933711646e-07,
|
| 2171 |
+
"loss": 0.367,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.8213716108452951,
|
| 2176 |
+
"grad_norm": 2.277866930279838,
|
| 2177 |
+
"learning_rate": 9.418589087173441e-07,
|
| 2178 |
+
"loss": 0.3565,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.8240297713981924,
|
| 2183 |
+
"grad_norm": 2.0123435432750227,
|
| 2184 |
+
"learning_rate": 9.149177783612512e-07,
|
| 2185 |
+
"loss": 0.3565,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8266879319510898,
|
| 2190 |
+
"grad_norm": 2.066692850970878,
|
| 2191 |
+
"learning_rate": 8.883287242468242e-07,
|
| 2192 |
+
"loss": 0.3694,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.8293460925039873,
|
| 2197 |
+
"grad_norm": 2.3197695805880376,
|
| 2198 |
+
"learning_rate": 8.620940379740245e-07,
|
| 2199 |
+
"loss": 0.3598,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.8320042530568846,
|
| 2204 |
+
"grad_norm": 1.948357495635851,
|
| 2205 |
+
"learning_rate": 8.362159806013176e-07,
|
| 2206 |
+
"loss": 0.3593,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.834662413609782,
|
| 2211 |
+
"grad_norm": 2.0700125518419745,
|
| 2212 |
+
"learning_rate": 8.106967824508061e-07,
|
| 2213 |
+
"loss": 0.3633,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.8373205741626795,
|
| 2218 |
+
"grad_norm": 1.9659833814171273,
|
| 2219 |
+
"learning_rate": 7.85538642916015e-07,
|
| 2220 |
+
"loss": 0.3582,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.8399787347155768,
|
| 2225 |
+
"grad_norm": 2.1056761901025483,
|
| 2226 |
+
"learning_rate": 7.607437302723247e-07,
|
| 2227 |
+
"loss": 0.3557,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.8426368952684742,
|
| 2232 |
+
"grad_norm": 2.3048662605616506,
|
| 2233 |
+
"learning_rate": 7.363141814901054e-07,
|
| 2234 |
+
"loss": 0.3483,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.8452950558213717,
|
| 2239 |
+
"grad_norm": 2.2968219659859708,
|
| 2240 |
+
"learning_rate": 7.122521020505302e-07,
|
| 2241 |
+
"loss": 0.3675,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.847953216374269,
|
| 2246 |
+
"grad_norm": 2.1004495243983423,
|
| 2247 |
+
"learning_rate": 6.885595657641214e-07,
|
| 2248 |
+
"loss": 0.351,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.8506113769271664,
|
| 2253 |
+
"grad_norm": 2.2866051769809315,
|
| 2254 |
+
"learning_rate": 6.652386145920181e-07,
|
| 2255 |
+
"loss": 0.3635,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.8532695374800638,
|
| 2260 |
+
"grad_norm": 2.118537926421423,
|
| 2261 |
+
"learning_rate": 6.422912584699753e-07,
|
| 2262 |
+
"loss": 0.3501,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.8559276980329612,
|
| 2267 |
+
"grad_norm": 2.016105032165951,
|
| 2268 |
+
"learning_rate": 6.197194751351543e-07,
|
| 2269 |
+
"loss": 0.3589,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.8585858585858586,
|
| 2274 |
+
"grad_norm": 2.0032229318591126,
|
| 2275 |
+
"learning_rate": 5.975252099556544e-07,
|
| 2276 |
+
"loss": 0.3627,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.861244019138756,
|
| 2281 |
+
"grad_norm": 2.191449628463197,
|
| 2282 |
+
"learning_rate": 5.757103757628573e-07,
|
| 2283 |
+
"loss": 0.3687,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.8639021796916534,
|
| 2288 |
+
"grad_norm": 1.9571952063843965,
|
| 2289 |
+
"learning_rate": 5.542768526865678e-07,
|
| 2290 |
+
"loss": 0.3573,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.8665603402445508,
|
| 2295 |
+
"grad_norm": 2.1555005088207846,
|
| 2296 |
+
"learning_rate": 5.332264879929682e-07,
|
| 2297 |
+
"loss": 0.3607,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.8692185007974481,
|
| 2302 |
+
"grad_norm": 2.312792401325662,
|
| 2303 |
+
"learning_rate": 5.125610959254213e-07,
|
| 2304 |
+
"loss": 0.3496,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.8718766613503456,
|
| 2309 |
+
"grad_norm": 2.1171904493372318,
|
| 2310 |
+
"learning_rate": 4.9228245754809e-07,
|
| 2311 |
+
"loss": 0.362,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.8745348219032429,
|
| 2316 |
+
"grad_norm": 2.1733472639430595,
|
| 2317 |
+
"learning_rate": 4.723923205924558e-07,
|
| 2318 |
+
"loss": 0.3604,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.8771929824561403,
|
| 2323 |
+
"grad_norm": 2.175043851294071,
|
| 2324 |
+
"learning_rate": 4.5289239930667304e-07,
|
| 2325 |
+
"loss": 0.3657,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.8798511430090378,
|
| 2330 |
+
"grad_norm": 2.368596065401823,
|
| 2331 |
+
"learning_rate": 4.3378437430783294e-07,
|
| 2332 |
+
"loss": 0.3629,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.8825093035619351,
|
| 2337 |
+
"grad_norm": 2.181916703721701,
|
| 2338 |
+
"learning_rate": 4.15069892437116e-07,
|
| 2339 |
+
"loss": 0.3603,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.8851674641148325,
|
| 2344 |
+
"grad_norm": 1.9884410794596379,
|
| 2345 |
+
"learning_rate": 3.9675056661785563e-07,
|
| 2346 |
+
"loss": 0.3483,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.88782562466773,
|
| 2351 |
+
"grad_norm": 2.030195451894179,
|
| 2352 |
+
"learning_rate": 3.7882797571653473e-07,
|
| 2353 |
+
"loss": 0.3602,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.8904837852206273,
|
| 2358 |
+
"grad_norm": 2.2030014882431312,
|
| 2359 |
+
"learning_rate": 3.6130366440669693e-07,
|
| 2360 |
+
"loss": 0.3452,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.8931419457735247,
|
| 2365 |
+
"grad_norm": 2.1504915530455424,
|
| 2366 |
+
"learning_rate": 3.441791430358299e-07,
|
| 2367 |
+
"loss": 0.3517,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8958001063264222,
|
| 2372 |
+
"grad_norm": 2.0548225206052186,
|
| 2373 |
+
"learning_rate": 3.2745588749518775e-07,
|
| 2374 |
+
"loss": 0.3673,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8984582668793195,
|
| 2379 |
+
"grad_norm": 2.0116181749449753,
|
| 2380 |
+
"learning_rate": 3.111353390925925e-07,
|
| 2381 |
+
"loss": 0.354,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9011164274322169,
|
| 2386 |
+
"grad_norm": 2.2407951496611584,
|
| 2387 |
+
"learning_rate": 2.9521890442821276e-07,
|
| 2388 |
+
"loss": 0.356,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.9037745879851143,
|
| 2393 |
+
"grad_norm": 2.036787818054975,
|
| 2394 |
+
"learning_rate": 2.7970795527333563e-07,
|
| 2395 |
+
"loss": 0.3551,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.9064327485380117,
|
| 2400 |
+
"grad_norm": 1.8662143528967177,
|
| 2401 |
+
"learning_rate": 2.6460382845214125e-07,
|
| 2402 |
+
"loss": 0.3473,
|
| 2403 |
+
"step": 1705
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.9090909090909091,
|
| 2407 |
+
"grad_norm": 1.9521145818680103,
|
| 2408 |
+
"learning_rate": 2.4990782572647977e-07,
|
| 2409 |
+
"loss": 0.3498,
|
| 2410 |
+
"step": 1710
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9117490696438065,
|
| 2414 |
+
"grad_norm": 2.0807157162150367,
|
| 2415 |
+
"learning_rate": 2.356212136836894e-07,
|
| 2416 |
+
"loss": 0.3526,
|
| 2417 |
+
"step": 1715
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.9144072301967039,
|
| 2421 |
+
"grad_norm": 2.142229762626437,
|
| 2422 |
+
"learning_rate": 2.2174522362742647e-07,
|
| 2423 |
+
"loss": 0.3562,
|
| 2424 |
+
"step": 1720
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.9170653907496013,
|
| 2428 |
+
"grad_norm": 2.094304223215669,
|
| 2429 |
+
"learning_rate": 2.0828105147154275e-07,
|
| 2430 |
+
"loss": 0.3467,
|
| 2431 |
+
"step": 1725
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.9197235513024987,
|
| 2435 |
+
"grad_norm": 2.1191691095904246,
|
| 2436 |
+
"learning_rate": 1.952298576370254e-07,
|
| 2437 |
+
"loss": 0.3573,
|
| 2438 |
+
"step": 1730
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.9223817118553961,
|
| 2442 |
+
"grad_norm": 2.2074162527799372,
|
| 2443 |
+
"learning_rate": 1.825927669519728e-07,
|
| 2444 |
+
"loss": 0.3519,
|
| 2445 |
+
"step": 1735
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.9250398724082934,
|
| 2449 |
+
"grad_norm": 2.1885606423221016,
|
| 2450 |
+
"learning_rate": 1.7037086855465902e-07,
|
| 2451 |
+
"loss": 0.3444,
|
| 2452 |
+
"step": 1740
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.9276980329611909,
|
| 2456 |
+
"grad_norm": 2.197465057188687,
|
| 2457 |
+
"learning_rate": 1.5856521579965866e-07,
|
| 2458 |
+
"loss": 0.3417,
|
| 2459 |
+
"step": 1745
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.9303561935140883,
|
| 2463 |
+
"grad_norm": 2.1155552965012694,
|
| 2464 |
+
"learning_rate": 1.4717682616706917e-07,
|
| 2465 |
+
"loss": 0.3628,
|
| 2466 |
+
"step": 1750
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.9330143540669856,
|
| 2470 |
+
"grad_norm": 2.0604186286349138,
|
| 2471 |
+
"learning_rate": 1.3620668117481471e-07,
|
| 2472 |
+
"loss": 0.3621,
|
| 2473 |
+
"step": 1755
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.935672514619883,
|
| 2477 |
+
"grad_norm": 2.0894295505567664,
|
| 2478 |
+
"learning_rate": 1.256557262940522e-07,
|
| 2479 |
+
"loss": 0.3591,
|
| 2480 |
+
"step": 1760
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.9383306751727805,
|
| 2484 |
+
"grad_norm": 2.2532084612884464,
|
| 2485 |
+
"learning_rate": 1.1552487086768871e-07,
|
| 2486 |
+
"loss": 0.3466,
|
| 2487 |
+
"step": 1765
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.9409888357256778,
|
| 2491 |
+
"grad_norm": 1.9966735008414798,
|
| 2492 |
+
"learning_rate": 1.0581498803200696e-07,
|
| 2493 |
+
"loss": 0.3501,
|
| 2494 |
+
"step": 1770
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.9436469962785752,
|
| 2498 |
+
"grad_norm": 2.1187544981067195,
|
| 2499 |
+
"learning_rate": 9.652691464141273e-08,
|
| 2500 |
+
"loss": 0.344,
|
| 2501 |
+
"step": 1775
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.9463051568314726,
|
| 2505 |
+
"grad_norm": 2.030559494863266,
|
| 2506 |
+
"learning_rate": 8.766145119630976e-08,
|
| 2507 |
+
"loss": 0.3588,
|
| 2508 |
+
"step": 1780
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.94896331738437,
|
| 2512 |
+
"grad_norm": 2.087686270810844,
|
| 2513 |
+
"learning_rate": 7.921936177411049e-08,
|
| 2514 |
+
"loss": 0.3419,
|
| 2515 |
+
"step": 1785
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.9516214779372674,
|
| 2519 |
+
"grad_norm": 1.9460600257508178,
|
| 2520 |
+
"learning_rate": 7.1201373963381e-08,
|
| 2521 |
+
"loss": 0.346,
|
| 2522 |
+
"step": 1790
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.9542796384901648,
|
| 2526 |
+
"grad_norm": 2.1336394767825855,
|
| 2527 |
+
"learning_rate": 6.360817880113335e-08,
|
| 2528 |
+
"loss": 0.3499,
|
| 2529 |
+
"step": 1795
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.9569377990430622,
|
| 2533 |
+
"grad_norm": 2.27235436719128,
|
| 2534 |
+
"learning_rate": 5.6440430713269325e-08,
|
| 2535 |
+
"loss": 0.353,
|
| 2536 |
+
"step": 1800
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.9595959595959596,
|
| 2540 |
+
"grad_norm": 2.244896225281087,
|
| 2541 |
+
"learning_rate": 4.9698747458176714e-08,
|
| 2542 |
+
"loss": 0.35,
|
| 2543 |
+
"step": 1805
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.962254120148857,
|
| 2547 |
+
"grad_norm": 2.065314709776108,
|
| 2548 |
+
"learning_rate": 4.338371007348852e-08,
|
| 2549 |
+
"loss": 0.3459,
|
| 2550 |
+
"step": 1810
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.9649122807017544,
|
| 2554 |
+
"grad_norm": 2.09774223772813,
|
| 2555 |
+
"learning_rate": 3.749586282600359e-08,
|
| 2556 |
+
"loss": 0.3482,
|
| 2557 |
+
"step": 1815
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.9675704412546517,
|
| 2561 |
+
"grad_norm": 2.1010041159494173,
|
| 2562 |
+
"learning_rate": 3.2035713164781936e-08,
|
| 2563 |
+
"loss": 0.3392,
|
| 2564 |
+
"step": 1820
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.9702286018075492,
|
| 2568 |
+
"grad_norm": 2.183490505896793,
|
| 2569 |
+
"learning_rate": 2.700373167740744e-08,
|
| 2570 |
+
"loss": 0.3423,
|
| 2571 |
+
"step": 1825
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.9728867623604466,
|
| 2575 |
+
"grad_norm": 2.03759593582233,
|
| 2576 |
+
"learning_rate": 2.2400352049429807e-08,
|
| 2577 |
+
"loss": 0.345,
|
| 2578 |
+
"step": 1830
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.9755449229133439,
|
| 2582 |
+
"grad_norm": 2.1153372258468606,
|
| 2583 |
+
"learning_rate": 1.8225971026987755e-08,
|
| 2584 |
+
"loss": 0.3511,
|
| 2585 |
+
"step": 1835
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.9782030834662414,
|
| 2589 |
+
"grad_norm": 1.9483951001160797,
|
| 2590 |
+
"learning_rate": 1.4480948382615267e-08,
|
| 2591 |
+
"loss": 0.3448,
|
| 2592 |
+
"step": 1840
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.9808612440191388,
|
| 2596 |
+
"grad_norm": 1.8908522932363727,
|
| 2597 |
+
"learning_rate": 1.1165606884234182e-08,
|
| 2598 |
+
"loss": 0.3559,
|
| 2599 |
+
"step": 1845
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.9835194045720361,
|
| 2603 |
+
"grad_norm": 2.3270461293558475,
|
| 2604 |
+
"learning_rate": 8.280232267334765e-09,
|
| 2605 |
+
"loss": 0.3434,
|
| 2606 |
+
"step": 1850
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.9861775651249336,
|
| 2610 |
+
"grad_norm": 1.9237250529226086,
|
| 2611 |
+
"learning_rate": 5.825073210352084e-09,
|
| 2612 |
+
"loss": 0.349,
|
| 2613 |
+
"step": 1855
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.988835725677831,
|
| 2617 |
+
"grad_norm": 2.0893502612570267,
|
| 2618 |
+
"learning_rate": 3.8003413132309265e-09,
|
| 2619 |
+
"loss": 0.3402,
|
| 2620 |
+
"step": 1860
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.9914938862307283,
|
| 2624 |
+
"grad_norm": 2.06568498566462,
|
| 2625 |
+
"learning_rate": 2.20621107918928e-09,
|
| 2626 |
+
"loss": 0.3391,
|
| 2627 |
+
"step": 1865
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.9941520467836257,
|
| 2631 |
+
"grad_norm": 2.1868236662669154,
|
| 2632 |
+
"learning_rate": 1.0428198996814688e-09,
|
| 2633 |
+
"loss": 0.3456,
|
| 2634 |
+
"step": 1870
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.9968102073365231,
|
| 2638 |
+
"grad_norm": 2.34269464712247,
|
| 2639 |
+
"learning_rate": 3.1026804255207544e-10,
|
| 2640 |
+
"loss": 0.3501,
|
| 2641 |
+
"step": 1875
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.9994683678894205,
|
| 2645 |
+
"grad_norm": 2.1579164945148563,
|
| 2646 |
+
"learning_rate": 8.618643398405902e-12,
|
| 2647 |
+
"loss": 0.3623,
|
| 2648 |
+
"step": 1880
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 1.0,
|
| 2652 |
+
"eval_runtime": 3.3807,
|
| 2653 |
+
"eval_samples_per_second": 2.958,
|
| 2654 |
+
"eval_steps_per_second": 0.887,
|
| 2655 |
+
"step": 1881
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 1.0,
|
| 2659 |
+
"step": 1881,
|
| 2660 |
+
"total_flos": 196921566167040.0,
|
| 2661 |
+
"train_loss": 0.5133170171946303,
|
| 2662 |
+
"train_runtime": 16745.7461,
|
| 2663 |
+
"train_samples_per_second": 1.797,
|
| 2664 |
+
"train_steps_per_second": 0.112
|
| 2665 |
+
}
|
| 2666 |
+
],
|
| 2667 |
+
"logging_steps": 5,
|
| 2668 |
+
"max_steps": 1881,
|
| 2669 |
+
"num_input_tokens_seen": 0,
|
| 2670 |
+
"num_train_epochs": 1,
|
| 2671 |
+
"save_steps": 100,
|
| 2672 |
+
"stateful_callbacks": {
|
| 2673 |
+
"TrainerControl": {
|
| 2674 |
+
"args": {
|
| 2675 |
+
"should_epoch_stop": false,
|
| 2676 |
+
"should_evaluate": false,
|
| 2677 |
+
"should_log": false,
|
| 2678 |
+
"should_save": true,
|
| 2679 |
+
"should_training_stop": true
|
| 2680 |
+
},
|
| 2681 |
+
"attributes": {}
|
| 2682 |
+
}
|
| 2683 |
+
},
|
| 2684 |
+
"total_flos": 196921566167040.0,
|
| 2685 |
+
"train_batch_size": 4,
|
| 2686 |
+
"trial_name": null,
|
| 2687 |
+
"trial_params": null
|
| 2688 |
+
}
|