Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- runs/Sep19_05-49-17_COE-CS-sv003/events.out.tfevents.1726725600.COE-CS-sv003.346740.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2436 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: gemma
|
| 4 |
+
base_model: google/gemma-7b-it
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: feedback_gemma
|
| 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_gemma
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-7b-it](https://huggingface.co/google/gemma-7b-it) 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": 117051280588800.0,
|
| 4 |
+
"train_loss": 0.691351400122923,
|
| 5 |
+
"train_runtime": 20676.98,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.315,
|
| 8 |
+
"train_steps_per_second": 0.082
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.44.2"
|
| 7 |
+
}
|
runs/Sep19_05-49-17_COE-CS-sv003/events.out.tfevents.1726725600.COE-CS-sv003.346740.0
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 sha256:9cf653d97c77b8895c9e8aafc5e1523210592ddeb42a1551a54a8ad2f95d1bec
|
| 3 |
+
size 77592
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 117051280588800.0,
|
| 4 |
+
"train_loss": 0.691351400122923,
|
| 5 |
+
"train_runtime": 20676.98,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.315,
|
| 8 |
+
"train_steps_per_second": 0.082
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1700,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000588235294117647,
|
| 13 |
+
"grad_norm": 86.25792069448507,
|
| 14 |
+
"learning_rate": 5.882352941176471e-08,
|
| 15 |
+
"loss": 4.144,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0029411764705882353,
|
| 20 |
+
"grad_norm": 92.94334261133154,
|
| 21 |
+
"learning_rate": 2.9411764705882356e-07,
|
| 22 |
+
"loss": 4.2053,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0058823529411764705,
|
| 27 |
+
"grad_norm": 34.41771480337444,
|
| 28 |
+
"learning_rate": 5.882352941176471e-07,
|
| 29 |
+
"loss": 3.9703,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008823529411764706,
|
| 34 |
+
"grad_norm": 20.69702610638224,
|
| 35 |
+
"learning_rate": 8.823529411764707e-07,
|
| 36 |
+
"loss": 3.4529,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011764705882352941,
|
| 41 |
+
"grad_norm": 12.865609762875932,
|
| 42 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 43 |
+
"loss": 3.1192,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.014705882352941176,
|
| 48 |
+
"grad_norm": 14.658994779954883,
|
| 49 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 50 |
+
"loss": 2.9015,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01764705882352941,
|
| 55 |
+
"grad_norm": 19.64009310214488,
|
| 56 |
+
"learning_rate": 1.7647058823529414e-06,
|
| 57 |
+
"loss": 2.5883,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.020588235294117647,
|
| 62 |
+
"grad_norm": 27.20931711935924,
|
| 63 |
+
"learning_rate": 2.058823529411765e-06,
|
| 64 |
+
"loss": 2.2725,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.023529411764705882,
|
| 69 |
+
"grad_norm": 23.446531565331924,
|
| 70 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 71 |
+
"loss": 1.8073,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.026470588235294117,
|
| 76 |
+
"grad_norm": 13.31815173013865,
|
| 77 |
+
"learning_rate": 2.647058823529412e-06,
|
| 78 |
+
"loss": 1.3518,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.029411764705882353,
|
| 83 |
+
"grad_norm": 5.34093550595349,
|
| 84 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 85 |
+
"loss": 1.1676,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03235294117647059,
|
| 90 |
+
"grad_norm": 3.3456911840455628,
|
| 91 |
+
"learning_rate": 3.2352941176470594e-06,
|
| 92 |
+
"loss": 1.147,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03529411764705882,
|
| 97 |
+
"grad_norm": 5.032424322254707,
|
| 98 |
+
"learning_rate": 3.529411764705883e-06,
|
| 99 |
+
"loss": 1.0965,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03823529411764706,
|
| 104 |
+
"grad_norm": 4.206085507780853,
|
| 105 |
+
"learning_rate": 3.8235294117647055e-06,
|
| 106 |
+
"loss": 1.0636,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.041176470588235294,
|
| 111 |
+
"grad_norm": 8.089055829866213,
|
| 112 |
+
"learning_rate": 4.11764705882353e-06,
|
| 113 |
+
"loss": 1.0582,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04411764705882353,
|
| 118 |
+
"grad_norm": 3.347364025882152,
|
| 119 |
+
"learning_rate": 4.411764705882353e-06,
|
| 120 |
+
"loss": 1.0539,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.047058823529411764,
|
| 125 |
+
"grad_norm": 6.336557333099317,
|
| 126 |
+
"learning_rate": 4.705882352941177e-06,
|
| 127 |
+
"loss": 1.0379,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05,
|
| 132 |
+
"grad_norm": 10.978187369479212,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 1.0242,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.052941176470588235,
|
| 139 |
+
"grad_norm": 7.818112219971164,
|
| 140 |
+
"learning_rate": 5.294117647058824e-06,
|
| 141 |
+
"loss": 1.0016,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05588235294117647,
|
| 146 |
+
"grad_norm": 8.433408497312906,
|
| 147 |
+
"learning_rate": 5.588235294117647e-06,
|
| 148 |
+
"loss": 0.9771,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.058823529411764705,
|
| 153 |
+
"grad_norm": 6.212980683823607,
|
| 154 |
+
"learning_rate": 5.882352941176471e-06,
|
| 155 |
+
"loss": 0.9861,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.061764705882352944,
|
| 160 |
+
"grad_norm": 6.647220375875289,
|
| 161 |
+
"learning_rate": 6.176470588235295e-06,
|
| 162 |
+
"loss": 0.9602,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06470588235294118,
|
| 167 |
+
"grad_norm": 6.527727138992631,
|
| 168 |
+
"learning_rate": 6.470588235294119e-06,
|
| 169 |
+
"loss": 0.9525,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06764705882352941,
|
| 174 |
+
"grad_norm": 7.509553679384927,
|
| 175 |
+
"learning_rate": 6.764705882352942e-06,
|
| 176 |
+
"loss": 0.9416,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07058823529411765,
|
| 181 |
+
"grad_norm": 8.103718249384503,
|
| 182 |
+
"learning_rate": 7.058823529411766e-06,
|
| 183 |
+
"loss": 0.9267,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07352941176470588,
|
| 188 |
+
"grad_norm": 8.520097429021,
|
| 189 |
+
"learning_rate": 7.352941176470589e-06,
|
| 190 |
+
"loss": 0.9396,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07647058823529412,
|
| 195 |
+
"grad_norm": 9.982219276844697,
|
| 196 |
+
"learning_rate": 7.647058823529411e-06,
|
| 197 |
+
"loss": 0.9259,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07941176470588235,
|
| 202 |
+
"grad_norm": 10.609383365594049,
|
| 203 |
+
"learning_rate": 7.941176470588236e-06,
|
| 204 |
+
"loss": 0.9262,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08235294117647059,
|
| 209 |
+
"grad_norm": 6.496638967078595,
|
| 210 |
+
"learning_rate": 8.23529411764706e-06,
|
| 211 |
+
"loss": 0.9169,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08529411764705883,
|
| 216 |
+
"grad_norm": 5.899471704114852,
|
| 217 |
+
"learning_rate": 8.529411764705883e-06,
|
| 218 |
+
"loss": 0.9172,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08823529411764706,
|
| 223 |
+
"grad_norm": 19.813714293488992,
|
| 224 |
+
"learning_rate": 8.823529411764707e-06,
|
| 225 |
+
"loss": 0.8899,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09117647058823529,
|
| 230 |
+
"grad_norm": 3.91249318240706,
|
| 231 |
+
"learning_rate": 9.11764705882353e-06,
|
| 232 |
+
"loss": 0.9074,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09411764705882353,
|
| 237 |
+
"grad_norm": 10.124669032565054,
|
| 238 |
+
"learning_rate": 9.411764705882354e-06,
|
| 239 |
+
"loss": 0.898,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09705882352941177,
|
| 244 |
+
"grad_norm": 13.840426312968154,
|
| 245 |
+
"learning_rate": 9.705882352941177e-06,
|
| 246 |
+
"loss": 0.9049,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1,
|
| 251 |
+
"grad_norm": 10.60421096496902,
|
| 252 |
+
"learning_rate": 1e-05,
|
| 253 |
+
"loss": 0.9154,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10294117647058823,
|
| 258 |
+
"grad_norm": 6.134422545965741,
|
| 259 |
+
"learning_rate": 9.999736492435867e-06,
|
| 260 |
+
"loss": 0.8836,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10588235294117647,
|
| 265 |
+
"grad_norm": 11.600519341121485,
|
| 266 |
+
"learning_rate": 9.998945997517957e-06,
|
| 267 |
+
"loss": 0.8904,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.10882352941176471,
|
| 272 |
+
"grad_norm": 11.799379752373861,
|
| 273 |
+
"learning_rate": 9.99762859856683e-06,
|
| 274 |
+
"loss": 0.8941,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11176470588235295,
|
| 279 |
+
"grad_norm": 6.623810329677796,
|
| 280 |
+
"learning_rate": 9.99578443444032e-06,
|
| 281 |
+
"loss": 0.8729,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11470588235294117,
|
| 286 |
+
"grad_norm": 6.965926377173582,
|
| 287 |
+
"learning_rate": 9.993413699518906e-06,
|
| 288 |
+
"loss": 0.8733,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.11764705882352941,
|
| 293 |
+
"grad_norm": 4.929345626084727,
|
| 294 |
+
"learning_rate": 9.990516643685222e-06,
|
| 295 |
+
"loss": 0.8559,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12058823529411765,
|
| 300 |
+
"grad_norm": 4.438658396751935,
|
| 301 |
+
"learning_rate": 9.987093572297716e-06,
|
| 302 |
+
"loss": 0.8693,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12352941176470589,
|
| 307 |
+
"grad_norm": 9.676542191697266,
|
| 308 |
+
"learning_rate": 9.983144846158472e-06,
|
| 309 |
+
"loss": 0.8788,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1264705882352941,
|
| 314 |
+
"grad_norm": 10.638416959011728,
|
| 315 |
+
"learning_rate": 9.978670881475173e-06,
|
| 316 |
+
"loss": 0.8517,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.12941176470588237,
|
| 321 |
+
"grad_norm": 15.793160052452487,
|
| 322 |
+
"learning_rate": 9.973672149817232e-06,
|
| 323 |
+
"loss": 0.8362,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1323529411764706,
|
| 328 |
+
"grad_norm": 14.96288140935782,
|
| 329 |
+
"learning_rate": 9.96814917806609e-06,
|
| 330 |
+
"loss": 0.8517,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.13529411764705881,
|
| 335 |
+
"grad_norm": 17.94048430648359,
|
| 336 |
+
"learning_rate": 9.96210254835968e-06,
|
| 337 |
+
"loss": 0.8452,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.13823529411764707,
|
| 342 |
+
"grad_norm": 10.072819918084187,
|
| 343 |
+
"learning_rate": 9.955532898031069e-06,
|
| 344 |
+
"loss": 0.8391,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1411764705882353,
|
| 349 |
+
"grad_norm": 7.684632649087268,
|
| 350 |
+
"learning_rate": 9.948440919541277e-06,
|
| 351 |
+
"loss": 0.8289,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.14411764705882352,
|
| 356 |
+
"grad_norm": 17.954771803869264,
|
| 357 |
+
"learning_rate": 9.940827360406297e-06,
|
| 358 |
+
"loss": 0.8067,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.14705882352941177,
|
| 363 |
+
"grad_norm": 12.023797137924742,
|
| 364 |
+
"learning_rate": 9.932693023118299e-06,
|
| 365 |
+
"loss": 0.8175,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15,
|
| 370 |
+
"grad_norm": 10.948930140888852,
|
| 371 |
+
"learning_rate": 9.924038765061042e-06,
|
| 372 |
+
"loss": 0.8226,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.15294117647058825,
|
| 377 |
+
"grad_norm": 17.86031699503285,
|
| 378 |
+
"learning_rate": 9.91486549841951e-06,
|
| 379 |
+
"loss": 0.8168,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.15588235294117647,
|
| 384 |
+
"grad_norm": 10.990033307622202,
|
| 385 |
+
"learning_rate": 9.905174190083763e-06,
|
| 386 |
+
"loss": 0.8047,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1588235294117647,
|
| 391 |
+
"grad_norm": 8.715291883213558,
|
| 392 |
+
"learning_rate": 9.894965861547023e-06,
|
| 393 |
+
"loss": 0.8007,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16176470588235295,
|
| 398 |
+
"grad_norm": 13.856071733371282,
|
| 399 |
+
"learning_rate": 9.884241588798004e-06,
|
| 400 |
+
"loss": 0.8017,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.16470588235294117,
|
| 405 |
+
"grad_norm": 15.297710097108334,
|
| 406 |
+
"learning_rate": 9.873002502207502e-06,
|
| 407 |
+
"loss": 0.8055,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1676470588235294,
|
| 412 |
+
"grad_norm": 10.725176474324847,
|
| 413 |
+
"learning_rate": 9.861249786409248e-06,
|
| 414 |
+
"loss": 0.7972,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17058823529411765,
|
| 419 |
+
"grad_norm": 12.860465116141345,
|
| 420 |
+
"learning_rate": 9.848984680175049e-06,
|
| 421 |
+
"loss": 0.8145,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17352941176470588,
|
| 426 |
+
"grad_norm": 13.411817152907572,
|
| 427 |
+
"learning_rate": 9.836208476284208e-06,
|
| 428 |
+
"loss": 0.7992,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.17647058823529413,
|
| 433 |
+
"grad_norm": 10.317196889908503,
|
| 434 |
+
"learning_rate": 9.822922521387277e-06,
|
| 435 |
+
"loss": 0.7846,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.17941176470588235,
|
| 440 |
+
"grad_norm": 4.179704998479472,
|
| 441 |
+
"learning_rate": 9.809128215864096e-06,
|
| 442 |
+
"loss": 0.8025,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.18235294117647058,
|
| 447 |
+
"grad_norm": 6.430799430688913,
|
| 448 |
+
"learning_rate": 9.794827013676206e-06,
|
| 449 |
+
"loss": 0.7991,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.18529411764705883,
|
| 454 |
+
"grad_norm": 5.850743250582622,
|
| 455 |
+
"learning_rate": 9.78002042221359e-06,
|
| 456 |
+
"loss": 0.8157,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.18823529411764706,
|
| 461 |
+
"grad_norm": 7.981852850027179,
|
| 462 |
+
"learning_rate": 9.764710002135784e-06,
|
| 463 |
+
"loss": 0.8315,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.19117647058823528,
|
| 468 |
+
"grad_norm": 4.246718673048229,
|
| 469 |
+
"learning_rate": 9.748897367207391e-06,
|
| 470 |
+
"loss": 0.8126,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.19411764705882353,
|
| 475 |
+
"grad_norm": 3.197412855424511,
|
| 476 |
+
"learning_rate": 9.732584184127973e-06,
|
| 477 |
+
"loss": 0.8054,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.19705882352941176,
|
| 482 |
+
"grad_norm": 3.866629151755055,
|
| 483 |
+
"learning_rate": 9.715772172356388e-06,
|
| 484 |
+
"loss": 0.8106,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2,
|
| 489 |
+
"grad_norm": 3.4312289366659603,
|
| 490 |
+
"learning_rate": 9.698463103929542e-06,
|
| 491 |
+
"loss": 0.803,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.20294117647058824,
|
| 496 |
+
"grad_norm": 4.537035689074838,
|
| 497 |
+
"learning_rate": 9.68065880327562e-06,
|
| 498 |
+
"loss": 0.8197,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.20588235294117646,
|
| 503 |
+
"grad_norm": 4.167020748172961,
|
| 504 |
+
"learning_rate": 9.66236114702178e-06,
|
| 505 |
+
"loss": 0.8128,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2088235294117647,
|
| 510 |
+
"grad_norm": 7.084719945270455,
|
| 511 |
+
"learning_rate": 9.643572063796352e-06,
|
| 512 |
+
"loss": 0.7983,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.21176470588235294,
|
| 517 |
+
"grad_norm": 4.345120253918932,
|
| 518 |
+
"learning_rate": 9.62429353402556e-06,
|
| 519 |
+
"loss": 0.7932,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.21470588235294116,
|
| 524 |
+
"grad_norm": 3.8354401768943536,
|
| 525 |
+
"learning_rate": 9.60452758972477e-06,
|
| 526 |
+
"loss": 0.7911,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.21764705882352942,
|
| 531 |
+
"grad_norm": 3.895028761361664,
|
| 532 |
+
"learning_rate": 9.584276314284316e-06,
|
| 533 |
+
"loss": 0.7903,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22058823529411764,
|
| 538 |
+
"grad_norm": 4.768654905615113,
|
| 539 |
+
"learning_rate": 9.563541842249903e-06,
|
| 540 |
+
"loss": 0.814,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2235294117647059,
|
| 545 |
+
"grad_norm": 4.001695620575136,
|
| 546 |
+
"learning_rate": 9.542326359097619e-06,
|
| 547 |
+
"loss": 0.7983,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.22647058823529412,
|
| 552 |
+
"grad_norm": 5.464979383463447,
|
| 553 |
+
"learning_rate": 9.520632101003579e-06,
|
| 554 |
+
"loss": 0.8113,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.22941176470588234,
|
| 559 |
+
"grad_norm": 4.410052168842928,
|
| 560 |
+
"learning_rate": 9.498461354608228e-06,
|
| 561 |
+
"loss": 0.7871,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2323529411764706,
|
| 566 |
+
"grad_norm": 10.214130104526346,
|
| 567 |
+
"learning_rate": 9.475816456775313e-06,
|
| 568 |
+
"loss": 0.7673,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.23529411764705882,
|
| 573 |
+
"grad_norm": 7.775696918112043,
|
| 574 |
+
"learning_rate": 9.452699794345583e-06,
|
| 575 |
+
"loss": 0.8504,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.23823529411764705,
|
| 580 |
+
"grad_norm": 4.793572966156681,
|
| 581 |
+
"learning_rate": 9.429113803885199e-06,
|
| 582 |
+
"loss": 0.7983,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2411764705882353,
|
| 587 |
+
"grad_norm": 6.2733719119866755,
|
| 588 |
+
"learning_rate": 9.405060971428924e-06,
|
| 589 |
+
"loss": 0.7545,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.24411764705882352,
|
| 594 |
+
"grad_norm": 5.665044173485091,
|
| 595 |
+
"learning_rate": 9.380543832218069e-06,
|
| 596 |
+
"loss": 0.7694,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.24705882352941178,
|
| 601 |
+
"grad_norm": 6.667717620194777,
|
| 602 |
+
"learning_rate": 9.355564970433288e-06,
|
| 603 |
+
"loss": 0.7749,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25,
|
| 608 |
+
"grad_norm": 3.3946898220456827,
|
| 609 |
+
"learning_rate": 9.330127018922195e-06,
|
| 610 |
+
"loss": 0.7857,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2529411764705882,
|
| 615 |
+
"grad_norm": 7.472860005366773,
|
| 616 |
+
"learning_rate": 9.30423265892184e-06,
|
| 617 |
+
"loss": 0.7711,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.25588235294117645,
|
| 622 |
+
"grad_norm": 4.70932174586482,
|
| 623 |
+
"learning_rate": 9.277884619776116e-06,
|
| 624 |
+
"loss": 0.7765,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.25882352941176473,
|
| 629 |
+
"grad_norm": 9.57060451964989,
|
| 630 |
+
"learning_rate": 9.251085678648072e-06,
|
| 631 |
+
"loss": 0.7604,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.26176470588235295,
|
| 636 |
+
"grad_norm": 4.681414682219643,
|
| 637 |
+
"learning_rate": 9.223838660227183e-06,
|
| 638 |
+
"loss": 0.7747,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2647058823529412,
|
| 643 |
+
"grad_norm": 7.367138755832612,
|
| 644 |
+
"learning_rate": 9.196146436431635e-06,
|
| 645 |
+
"loss": 0.7634,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2676470588235294,
|
| 650 |
+
"grad_norm": 5.627007459776944,
|
| 651 |
+
"learning_rate": 9.168011926105598e-06,
|
| 652 |
+
"loss": 0.7641,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.27058823529411763,
|
| 657 |
+
"grad_norm": 3.5740204640033078,
|
| 658 |
+
"learning_rate": 9.13943809471159e-06,
|
| 659 |
+
"loss": 0.7565,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2735294117647059,
|
| 664 |
+
"grad_norm": 7.69019676925858,
|
| 665 |
+
"learning_rate": 9.110427954017891e-06,
|
| 666 |
+
"loss": 0.7733,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.27647058823529413,
|
| 671 |
+
"grad_norm": 3.5962732922894975,
|
| 672 |
+
"learning_rate": 9.08098456178111e-06,
|
| 673 |
+
"loss": 0.765,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.27941176470588236,
|
| 678 |
+
"grad_norm": 4.241563630767514,
|
| 679 |
+
"learning_rate": 9.051111021423868e-06,
|
| 680 |
+
"loss": 0.7754,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2823529411764706,
|
| 685 |
+
"grad_norm": 7.8000912530998745,
|
| 686 |
+
"learning_rate": 9.020810481707709e-06,
|
| 687 |
+
"loss": 0.7585,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2852941176470588,
|
| 692 |
+
"grad_norm": 3.186234693569658,
|
| 693 |
+
"learning_rate": 8.990086136401199e-06,
|
| 694 |
+
"loss": 0.7547,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.28823529411764703,
|
| 699 |
+
"grad_norm": 7.586644694603505,
|
| 700 |
+
"learning_rate": 8.958941223943292e-06,
|
| 701 |
+
"loss": 0.7427,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2911764705882353,
|
| 706 |
+
"grad_norm": 3.258615795641642,
|
| 707 |
+
"learning_rate": 8.927379027101994e-06,
|
| 708 |
+
"loss": 0.7399,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.29411764705882354,
|
| 713 |
+
"grad_norm": 3.731634633993737,
|
| 714 |
+
"learning_rate": 8.895402872628352e-06,
|
| 715 |
+
"loss": 0.7391,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.29705882352941176,
|
| 720 |
+
"grad_norm": 4.803677561739918,
|
| 721 |
+
"learning_rate": 8.863016130905795e-06,
|
| 722 |
+
"loss": 0.7407,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3,
|
| 727 |
+
"grad_norm": 7.655699535414225,
|
| 728 |
+
"learning_rate": 8.83022221559489e-06,
|
| 729 |
+
"loss": 0.7506,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3029411764705882,
|
| 734 |
+
"grad_norm": 4.373831528386331,
|
| 735 |
+
"learning_rate": 8.797024583273536e-06,
|
| 736 |
+
"loss": 0.7355,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3058823529411765,
|
| 741 |
+
"grad_norm": 3.9385838728737266,
|
| 742 |
+
"learning_rate": 8.763426733072624e-06,
|
| 743 |
+
"loss": 0.735,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3088235294117647,
|
| 748 |
+
"grad_norm": 4.929470086337327,
|
| 749 |
+
"learning_rate": 8.729432206307218e-06,
|
| 750 |
+
"loss": 0.7407,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.31176470588235294,
|
| 755 |
+
"grad_norm": 4.985252344557763,
|
| 756 |
+
"learning_rate": 8.695044586103297e-06,
|
| 757 |
+
"loss": 0.7368,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.31470588235294117,
|
| 762 |
+
"grad_norm": 4.402532806236826,
|
| 763 |
+
"learning_rate": 8.660267497020074e-06,
|
| 764 |
+
"loss": 0.7378,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3176470588235294,
|
| 769 |
+
"grad_norm": 6.632318764297617,
|
| 770 |
+
"learning_rate": 8.625104604667965e-06,
|
| 771 |
+
"loss": 0.7182,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3205882352941177,
|
| 776 |
+
"grad_norm": 3.0548060644071566,
|
| 777 |
+
"learning_rate": 8.58955961532221e-06,
|
| 778 |
+
"loss": 0.7198,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3235294117647059,
|
| 783 |
+
"grad_norm": 7.1426568719242445,
|
| 784 |
+
"learning_rate": 8.553636275532236e-06,
|
| 785 |
+
"loss": 0.7336,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3264705882352941,
|
| 790 |
+
"grad_norm": 10.88738661858406,
|
| 791 |
+
"learning_rate": 8.51733837172675e-06,
|
| 792 |
+
"loss": 0.7368,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.32941176470588235,
|
| 797 |
+
"grad_norm": 6.282693925795995,
|
| 798 |
+
"learning_rate": 8.480669729814635e-06,
|
| 799 |
+
"loss": 0.7276,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3323529411764706,
|
| 804 |
+
"grad_norm": 9.555955539230844,
|
| 805 |
+
"learning_rate": 8.443634214781693e-06,
|
| 806 |
+
"loss": 0.702,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3352941176470588,
|
| 811 |
+
"grad_norm": 4.506254396871987,
|
| 812 |
+
"learning_rate": 8.40623573028327e-06,
|
| 813 |
+
"loss": 0.7129,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3382352941176471,
|
| 818 |
+
"grad_norm": 3.63704291073538,
|
| 819 |
+
"learning_rate": 8.368478218232787e-06,
|
| 820 |
+
"loss": 0.719,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3411764705882353,
|
| 825 |
+
"grad_norm": 2.9414631231401462,
|
| 826 |
+
"learning_rate": 8.330365658386252e-06,
|
| 827 |
+
"loss": 0.7119,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.34411764705882353,
|
| 832 |
+
"grad_norm": 3.8962634035076675,
|
| 833 |
+
"learning_rate": 8.291902067922791e-06,
|
| 834 |
+
"loss": 0.7044,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.34705882352941175,
|
| 839 |
+
"grad_norm": 2.996619696992914,
|
| 840 |
+
"learning_rate": 8.25309150102121e-06,
|
| 841 |
+
"loss": 0.7221,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.35,
|
| 846 |
+
"grad_norm": 3.6918987952391653,
|
| 847 |
+
"learning_rate": 8.213938048432697e-06,
|
| 848 |
+
"loss": 0.7152,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.35294117647058826,
|
| 853 |
+
"grad_norm": 5.1406824761237475,
|
| 854 |
+
"learning_rate": 8.174445837049614e-06,
|
| 855 |
+
"loss": 0.7122,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3558823529411765,
|
| 860 |
+
"grad_norm": 3.334206362728506,
|
| 861 |
+
"learning_rate": 8.134619029470535e-06,
|
| 862 |
+
"loss": 0.7021,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3588235294117647,
|
| 867 |
+
"grad_norm": 4.968392821499114,
|
| 868 |
+
"learning_rate": 8.094461823561473e-06,
|
| 869 |
+
"loss": 0.7053,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.36176470588235293,
|
| 874 |
+
"grad_norm": 3.1842109451175165,
|
| 875 |
+
"learning_rate": 8.05397845201344e-06,
|
| 876 |
+
"loss": 0.7142,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.36470588235294116,
|
| 881 |
+
"grad_norm": 7.122667502997686,
|
| 882 |
+
"learning_rate": 8.013173181896283e-06,
|
| 883 |
+
"loss": 0.7158,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.36764705882352944,
|
| 888 |
+
"grad_norm": 3.7490095954612563,
|
| 889 |
+
"learning_rate": 7.972050314208934e-06,
|
| 890 |
+
"loss": 0.7073,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.37058823529411766,
|
| 895 |
+
"grad_norm": 3.418137694246427,
|
| 896 |
+
"learning_rate": 7.930614183426074e-06,
|
| 897 |
+
"loss": 0.7098,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3735294117647059,
|
| 902 |
+
"grad_norm": 3.8866651266730092,
|
| 903 |
+
"learning_rate": 7.888869157041257e-06,
|
| 904 |
+
"loss": 0.6827,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3764705882352941,
|
| 909 |
+
"grad_norm": 3.5194202567335466,
|
| 910 |
+
"learning_rate": 7.846819635106569e-06,
|
| 911 |
+
"loss": 0.6891,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.37941176470588234,
|
| 916 |
+
"grad_norm": 3.847874692639519,
|
| 917 |
+
"learning_rate": 7.80447004976885e-06,
|
| 918 |
+
"loss": 0.667,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.38235294117647056,
|
| 923 |
+
"grad_norm": 5.871281951965833,
|
| 924 |
+
"learning_rate": 7.76182486480253e-06,
|
| 925 |
+
"loss": 0.6792,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.38529411764705884,
|
| 930 |
+
"grad_norm": 2.9461535013380344,
|
| 931 |
+
"learning_rate": 7.718888575139134e-06,
|
| 932 |
+
"loss": 0.6821,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.38823529411764707,
|
| 937 |
+
"grad_norm": 3.545429320117265,
|
| 938 |
+
"learning_rate": 7.675665706393502e-06,
|
| 939 |
+
"loss": 0.6837,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.3911764705882353,
|
| 944 |
+
"grad_norm": 3.9453651020162406,
|
| 945 |
+
"learning_rate": 7.63216081438678e-06,
|
| 946 |
+
"loss": 0.6944,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.3941176470588235,
|
| 951 |
+
"grad_norm": 4.157385411843414,
|
| 952 |
+
"learning_rate": 7.588378484666214e-06,
|
| 953 |
+
"loss": 0.6803,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.39705882352941174,
|
| 958 |
+
"grad_norm": 2.9879188800031287,
|
| 959 |
+
"learning_rate": 7.544323332021826e-06,
|
| 960 |
+
"loss": 0.6716,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4,
|
| 965 |
+
"grad_norm": 5.065231322987666,
|
| 966 |
+
"learning_rate": 7.500000000000001e-06,
|
| 967 |
+
"loss": 0.6899,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.40294117647058825,
|
| 972 |
+
"grad_norm": 5.937109080515486,
|
| 973 |
+
"learning_rate": 7.4554131604140425e-06,
|
| 974 |
+
"loss": 0.6858,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.40588235294117647,
|
| 979 |
+
"grad_norm": 3.589250867729836,
|
| 980 |
+
"learning_rate": 7.4105675128517456e-06,
|
| 981 |
+
"loss": 0.6911,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4088235294117647,
|
| 986 |
+
"grad_norm": 3.459912742752029,
|
| 987 |
+
"learning_rate": 7.365467784180051e-06,
|
| 988 |
+
"loss": 0.6931,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4117647058823529,
|
| 993 |
+
"grad_norm": 7.570102416020596,
|
| 994 |
+
"learning_rate": 7.320118728046818e-06,
|
| 995 |
+
"loss": 0.6919,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4147058823529412,
|
| 1000 |
+
"grad_norm": 5.7516115714170235,
|
| 1001 |
+
"learning_rate": 7.274525124379773e-06,
|
| 1002 |
+
"loss": 0.6613,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4176470588235294,
|
| 1007 |
+
"grad_norm": 6.937823731044612,
|
| 1008 |
+
"learning_rate": 7.2286917788826926e-06,
|
| 1009 |
+
"loss": 0.6681,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.42058823529411765,
|
| 1014 |
+
"grad_norm": 2.702492838292246,
|
| 1015 |
+
"learning_rate": 7.182623522528866e-06,
|
| 1016 |
+
"loss": 0.6621,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4235294117647059,
|
| 1021 |
+
"grad_norm": 4.510429847780033,
|
| 1022 |
+
"learning_rate": 7.136325211051905e-06,
|
| 1023 |
+
"loss": 0.6577,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4264705882352941,
|
| 1028 |
+
"grad_norm": 3.6330047784117503,
|
| 1029 |
+
"learning_rate": 7.089801724433918e-06,
|
| 1030 |
+
"loss": 0.6682,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.4294117647058823,
|
| 1035 |
+
"grad_norm": 3.105718892497887,
|
| 1036 |
+
"learning_rate": 7.043057966391158e-06,
|
| 1037 |
+
"loss": 0.6632,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4323529411764706,
|
| 1042 |
+
"grad_norm": 5.111900502997939,
|
| 1043 |
+
"learning_rate": 6.996098863857155e-06,
|
| 1044 |
+
"loss": 0.6486,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.43529411764705883,
|
| 1049 |
+
"grad_norm": 2.5268899583639315,
|
| 1050 |
+
"learning_rate": 6.948929366463397e-06,
|
| 1051 |
+
"loss": 0.6523,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.43823529411764706,
|
| 1056 |
+
"grad_norm": 3.2780721096181407,
|
| 1057 |
+
"learning_rate": 6.9015544460176296e-06,
|
| 1058 |
+
"loss": 0.655,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4411764705882353,
|
| 1063 |
+
"grad_norm": 3.3381264245743596,
|
| 1064 |
+
"learning_rate": 6.8539790959798045e-06,
|
| 1065 |
+
"loss": 0.654,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4441176470588235,
|
| 1070 |
+
"grad_norm": 4.44121574928845,
|
| 1071 |
+
"learning_rate": 6.806208330935766e-06,
|
| 1072 |
+
"loss": 0.6372,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4470588235294118,
|
| 1077 |
+
"grad_norm": 2.685752664996779,
|
| 1078 |
+
"learning_rate": 6.758247186068684e-06,
|
| 1079 |
+
"loss": 0.6413,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.45,
|
| 1084 |
+
"grad_norm": 2.955512012069018,
|
| 1085 |
+
"learning_rate": 6.710100716628345e-06,
|
| 1086 |
+
"loss": 0.6516,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.45294117647058824,
|
| 1091 |
+
"grad_norm": 3.1611374536081454,
|
| 1092 |
+
"learning_rate": 6.6617739973982985e-06,
|
| 1093 |
+
"loss": 0.6649,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.45588235294117646,
|
| 1098 |
+
"grad_norm": 4.369165428910403,
|
| 1099 |
+
"learning_rate": 6.613272122160975e-06,
|
| 1100 |
+
"loss": 0.6622,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.4588235294117647,
|
| 1105 |
+
"grad_norm": 3.614068134809545,
|
| 1106 |
+
"learning_rate": 6.5646002031607726e-06,
|
| 1107 |
+
"loss": 0.6469,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.46176470588235297,
|
| 1112 |
+
"grad_norm": 2.902322450613806,
|
| 1113 |
+
"learning_rate": 6.515763370565218e-06,
|
| 1114 |
+
"loss": 0.631,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4647058823529412,
|
| 1119 |
+
"grad_norm": 5.870110269988736,
|
| 1120 |
+
"learning_rate": 6.466766771924231e-06,
|
| 1121 |
+
"loss": 0.652,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4676470588235294,
|
| 1126 |
+
"grad_norm": 3.9059862931450313,
|
| 1127 |
+
"learning_rate": 6.417615571627555e-06,
|
| 1128 |
+
"loss": 0.6318,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.47058823529411764,
|
| 1133 |
+
"grad_norm": 2.8197589495167192,
|
| 1134 |
+
"learning_rate": 6.368314950360416e-06,
|
| 1135 |
+
"loss": 0.6322,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.47352941176470587,
|
| 1140 |
+
"grad_norm": 3.418950733265166,
|
| 1141 |
+
"learning_rate": 6.318870104557459e-06,
|
| 1142 |
+
"loss": 0.6187,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4764705882352941,
|
| 1147 |
+
"grad_norm": 4.432452311916624,
|
| 1148 |
+
"learning_rate": 6.269286245855039e-06,
|
| 1149 |
+
"loss": 0.602,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.47941176470588237,
|
| 1154 |
+
"grad_norm": 2.5812197112565887,
|
| 1155 |
+
"learning_rate": 6.219568600541886e-06,
|
| 1156 |
+
"loss": 0.6174,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4823529411764706,
|
| 1161 |
+
"grad_norm": 3.937227710483061,
|
| 1162 |
+
"learning_rate": 6.169722409008244e-06,
|
| 1163 |
+
"loss": 0.619,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4852941176470588,
|
| 1168 |
+
"grad_norm": 3.0605504256904443,
|
| 1169 |
+
"learning_rate": 6.119752925193516e-06,
|
| 1170 |
+
"loss": 0.6334,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.48823529411764705,
|
| 1175 |
+
"grad_norm": 4.037595617734812,
|
| 1176 |
+
"learning_rate": 6.0696654160324875e-06,
|
| 1177 |
+
"loss": 0.6279,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.49117647058823527,
|
| 1182 |
+
"grad_norm": 5.106015487464359,
|
| 1183 |
+
"learning_rate": 6.019465160900173e-06,
|
| 1184 |
+
"loss": 0.6354,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.49411764705882355,
|
| 1189 |
+
"grad_norm": 3.845547749417895,
|
| 1190 |
+
"learning_rate": 5.9691574510553505e-06,
|
| 1191 |
+
"loss": 0.6415,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4970588235294118,
|
| 1196 |
+
"grad_norm": 3.1754711810999394,
|
| 1197 |
+
"learning_rate": 5.918747589082853e-06,
|
| 1198 |
+
"loss": 0.6243,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5,
|
| 1203 |
+
"grad_norm": 3.8222639828081775,
|
| 1204 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 1205 |
+
"loss": 0.6065,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5029411764705882,
|
| 1210 |
+
"grad_norm": 3.5634408734197054,
|
| 1211 |
+
"learning_rate": 5.817642672369825e-06,
|
| 1212 |
+
"loss": 0.6008,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5058823529411764,
|
| 1217 |
+
"grad_norm": 2.923823486420052,
|
| 1218 |
+
"learning_rate": 5.766958274393428e-06,
|
| 1219 |
+
"loss": 0.6266,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5088235294117647,
|
| 1224 |
+
"grad_norm": 3.4223776040897373,
|
| 1225 |
+
"learning_rate": 5.716193036694359e-06,
|
| 1226 |
+
"loss": 0.6141,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5117647058823529,
|
| 1231 |
+
"grad_norm": 3.036605144513134,
|
| 1232 |
+
"learning_rate": 5.66535231008227e-06,
|
| 1233 |
+
"loss": 0.598,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5147058823529411,
|
| 1238 |
+
"grad_norm": 2.9215046859208695,
|
| 1239 |
+
"learning_rate": 5.614441453323571e-06,
|
| 1240 |
+
"loss": 0.6074,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5176470588235295,
|
| 1245 |
+
"grad_norm": 3.5327811692579387,
|
| 1246 |
+
"learning_rate": 5.5634658325766066e-06,
|
| 1247 |
+
"loss": 0.6052,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5205882352941177,
|
| 1252 |
+
"grad_norm": 4.232585058220642,
|
| 1253 |
+
"learning_rate": 5.512430820826035e-06,
|
| 1254 |
+
"loss": 0.6165,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5235294117647059,
|
| 1259 |
+
"grad_norm": 7.156188965411479,
|
| 1260 |
+
"learning_rate": 5.46134179731651e-06,
|
| 1261 |
+
"loss": 0.6012,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5264705882352941,
|
| 1266 |
+
"grad_norm": 3.0787975216808356,
|
| 1267 |
+
"learning_rate": 5.41020414698569e-06,
|
| 1268 |
+
"loss": 0.5919,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5294117647058824,
|
| 1273 |
+
"grad_norm": 2.641117254170942,
|
| 1274 |
+
"learning_rate": 5.359023259896638e-06,
|
| 1275 |
+
"loss": 0.5921,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5323529411764706,
|
| 1280 |
+
"grad_norm": 5.7492713146889205,
|
| 1281 |
+
"learning_rate": 5.3078045306697154e-06,
|
| 1282 |
+
"loss": 0.6009,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5352941176470588,
|
| 1287 |
+
"grad_norm": 3.0420972441444984,
|
| 1288 |
+
"learning_rate": 5.2565533579139484e-06,
|
| 1289 |
+
"loss": 0.5803,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.538235294117647,
|
| 1294 |
+
"grad_norm": 4.94141259844524,
|
| 1295 |
+
"learning_rate": 5.205275143658018e-06,
|
| 1296 |
+
"loss": 0.5777,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5411764705882353,
|
| 1301 |
+
"grad_norm": 2.8294470597059362,
|
| 1302 |
+
"learning_rate": 5.153975292780852e-06,
|
| 1303 |
+
"loss": 0.5853,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5441176470588235,
|
| 1308 |
+
"grad_norm": 2.657834885061226,
|
| 1309 |
+
"learning_rate": 5.102659212441953e-06,
|
| 1310 |
+
"loss": 0.5753,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5470588235294118,
|
| 1315 |
+
"grad_norm": 2.5027366282176264,
|
| 1316 |
+
"learning_rate": 5.05133231151145e-06,
|
| 1317 |
+
"loss": 0.5979,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.55,
|
| 1322 |
+
"grad_norm": 2.972259109016088,
|
| 1323 |
+
"learning_rate": 5e-06,
|
| 1324 |
+
"loss": 0.5934,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5529411764705883,
|
| 1329 |
+
"grad_norm": 3.7344170961635235,
|
| 1330 |
+
"learning_rate": 4.948667688488552e-06,
|
| 1331 |
+
"loss": 0.5765,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5558823529411765,
|
| 1336 |
+
"grad_norm": 3.6094582616510187,
|
| 1337 |
+
"learning_rate": 4.8973407875580485e-06,
|
| 1338 |
+
"loss": 0.5763,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5588235294117647,
|
| 1343 |
+
"grad_norm": 3.0365592437357183,
|
| 1344 |
+
"learning_rate": 4.846024707219149e-06,
|
| 1345 |
+
"loss": 0.5637,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5617647058823529,
|
| 1350 |
+
"grad_norm": 2.689099476865793,
|
| 1351 |
+
"learning_rate": 4.794724856341985e-06,
|
| 1352 |
+
"loss": 0.5702,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5647058823529412,
|
| 1357 |
+
"grad_norm": 3.1936877032969977,
|
| 1358 |
+
"learning_rate": 4.7434466420860515e-06,
|
| 1359 |
+
"loss": 0.566,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5676470588235294,
|
| 1364 |
+
"grad_norm": 3.3687980239848976,
|
| 1365 |
+
"learning_rate": 4.692195469330286e-06,
|
| 1366 |
+
"loss": 0.5855,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5705882352941176,
|
| 1371 |
+
"grad_norm": 3.652904628883288,
|
| 1372 |
+
"learning_rate": 4.640976740103363e-06,
|
| 1373 |
+
"loss": 0.5712,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5735294117647058,
|
| 1378 |
+
"grad_norm": 2.6471377767162116,
|
| 1379 |
+
"learning_rate": 4.589795853014313e-06,
|
| 1380 |
+
"loss": 0.5764,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5764705882352941,
|
| 1385 |
+
"grad_norm": 3.4072798307585845,
|
| 1386 |
+
"learning_rate": 4.53865820268349e-06,
|
| 1387 |
+
"loss": 0.5584,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5794117647058824,
|
| 1392 |
+
"grad_norm": 2.9495397115764717,
|
| 1393 |
+
"learning_rate": 4.4875691791739655e-06,
|
| 1394 |
+
"loss": 0.5603,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5823529411764706,
|
| 1399 |
+
"grad_norm": 2.967359329887194,
|
| 1400 |
+
"learning_rate": 4.436534167423395e-06,
|
| 1401 |
+
"loss": 0.5606,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5852941176470589,
|
| 1406 |
+
"grad_norm": 3.6733313275854824,
|
| 1407 |
+
"learning_rate": 4.3855585466764305e-06,
|
| 1408 |
+
"loss": 0.5753,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5882352941176471,
|
| 1413 |
+
"grad_norm": 3.3017502365781692,
|
| 1414 |
+
"learning_rate": 4.334647689917734e-06,
|
| 1415 |
+
"loss": 0.562,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5911764705882353,
|
| 1420 |
+
"grad_norm": 2.60212577629923,
|
| 1421 |
+
"learning_rate": 4.283806963305644e-06,
|
| 1422 |
+
"loss": 0.5614,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.5941176470588235,
|
| 1427 |
+
"grad_norm": 3.2117623019305164,
|
| 1428 |
+
"learning_rate": 4.233041725606573e-06,
|
| 1429 |
+
"loss": 0.5561,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5970588235294118,
|
| 1434 |
+
"grad_norm": 2.5342533645318697,
|
| 1435 |
+
"learning_rate": 4.182357327630175e-06,
|
| 1436 |
+
"loss": 0.5456,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6,
|
| 1441 |
+
"grad_norm": 3.491419616001698,
|
| 1442 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1443 |
+
"loss": 0.5458,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6029411764705882,
|
| 1448 |
+
"grad_norm": 2.5355347312480108,
|
| 1449 |
+
"learning_rate": 4.081252410917148e-06,
|
| 1450 |
+
"loss": 0.5639,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6058823529411764,
|
| 1455 |
+
"grad_norm": 3.9415364552013177,
|
| 1456 |
+
"learning_rate": 4.03084254894465e-06,
|
| 1457 |
+
"loss": 0.5532,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6088235294117647,
|
| 1462 |
+
"grad_norm": 4.101652297241306,
|
| 1463 |
+
"learning_rate": 3.980534839099829e-06,
|
| 1464 |
+
"loss": 0.5513,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.611764705882353,
|
| 1469 |
+
"grad_norm": 2.8200195019787944,
|
| 1470 |
+
"learning_rate": 3.930334583967514e-06,
|
| 1471 |
+
"loss": 0.543,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6147058823529412,
|
| 1476 |
+
"grad_norm": 5.052470209996225,
|
| 1477 |
+
"learning_rate": 3.8802470748064855e-06,
|
| 1478 |
+
"loss": 0.5317,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6176470588235294,
|
| 1483 |
+
"grad_norm": 3.0712588067545545,
|
| 1484 |
+
"learning_rate": 3.8302775909917585e-06,
|
| 1485 |
+
"loss": 0.5276,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6205882352941177,
|
| 1490 |
+
"grad_norm": 2.8843549116949676,
|
| 1491 |
+
"learning_rate": 3.7804313994581143e-06,
|
| 1492 |
+
"loss": 0.5436,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6235294117647059,
|
| 1497 |
+
"grad_norm": 3.71731186768865,
|
| 1498 |
+
"learning_rate": 3.730713754144961e-06,
|
| 1499 |
+
"loss": 0.542,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6264705882352941,
|
| 1504 |
+
"grad_norm": 3.195039960104732,
|
| 1505 |
+
"learning_rate": 3.68112989544254e-06,
|
| 1506 |
+
"loss": 0.5569,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6294117647058823,
|
| 1511 |
+
"grad_norm": 3.0760143966442426,
|
| 1512 |
+
"learning_rate": 3.6316850496395863e-06,
|
| 1513 |
+
"loss": 0.5279,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6323529411764706,
|
| 1518 |
+
"grad_norm": 2.706150933189529,
|
| 1519 |
+
"learning_rate": 3.5823844283724464e-06,
|
| 1520 |
+
"loss": 0.5325,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6352941176470588,
|
| 1525 |
+
"grad_norm": 2.8909978900989683,
|
| 1526 |
+
"learning_rate": 3.5332332280757706e-06,
|
| 1527 |
+
"loss": 0.5275,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.638235294117647,
|
| 1532 |
+
"grad_norm": 2.4810301000022346,
|
| 1533 |
+
"learning_rate": 3.484236629434783e-06,
|
| 1534 |
+
"loss": 0.5449,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6411764705882353,
|
| 1539 |
+
"grad_norm": 2.6110146697831036,
|
| 1540 |
+
"learning_rate": 3.4353997968392295e-06,
|
| 1541 |
+
"loss": 0.5348,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6441176470588236,
|
| 1546 |
+
"grad_norm": 2.6207736282199208,
|
| 1547 |
+
"learning_rate": 3.386727877839027e-06,
|
| 1548 |
+
"loss": 0.5362,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6470588235294118,
|
| 1553 |
+
"grad_norm": 2.81407811382054,
|
| 1554 |
+
"learning_rate": 3.3382260026017027e-06,
|
| 1555 |
+
"loss": 0.5238,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.65,
|
| 1560 |
+
"grad_norm": 2.572013835784688,
|
| 1561 |
+
"learning_rate": 3.289899283371657e-06,
|
| 1562 |
+
"loss": 0.5251,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6529411764705882,
|
| 1567 |
+
"grad_norm": 2.587794908048521,
|
| 1568 |
+
"learning_rate": 3.241752813931316e-06,
|
| 1569 |
+
"loss": 0.5362,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6558823529411765,
|
| 1574 |
+
"grad_norm": 2.6632456475884423,
|
| 1575 |
+
"learning_rate": 3.1937916690642356e-06,
|
| 1576 |
+
"loss": 0.524,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6588235294117647,
|
| 1581 |
+
"grad_norm": 3.2994948697776407,
|
| 1582 |
+
"learning_rate": 3.1460209040201967e-06,
|
| 1583 |
+
"loss": 0.5246,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6617647058823529,
|
| 1588 |
+
"grad_norm": 2.5014355845754426,
|
| 1589 |
+
"learning_rate": 3.098445553982372e-06,
|
| 1590 |
+
"loss": 0.5201,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6647058823529411,
|
| 1595 |
+
"grad_norm": 3.615311051852331,
|
| 1596 |
+
"learning_rate": 3.0510706335366034e-06,
|
| 1597 |
+
"loss": 0.5328,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6676470588235294,
|
| 1602 |
+
"grad_norm": 2.8016358462604716,
|
| 1603 |
+
"learning_rate": 3.0039011361428466e-06,
|
| 1604 |
+
"loss": 0.5307,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6705882352941176,
|
| 1609 |
+
"grad_norm": 2.5797528822198896,
|
| 1610 |
+
"learning_rate": 2.956942033608843e-06,
|
| 1611 |
+
"loss": 0.5086,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6735294117647059,
|
| 1616 |
+
"grad_norm": 3.703697916863966,
|
| 1617 |
+
"learning_rate": 2.910198275566085e-06,
|
| 1618 |
+
"loss": 0.5119,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6764705882352942,
|
| 1623 |
+
"grad_norm": 2.4989182080083716,
|
| 1624 |
+
"learning_rate": 2.863674788948097e-06,
|
| 1625 |
+
"loss": 0.506,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6794117647058824,
|
| 1630 |
+
"grad_norm": 2.926008906071238,
|
| 1631 |
+
"learning_rate": 2.817376477471132e-06,
|
| 1632 |
+
"loss": 0.5061,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6823529411764706,
|
| 1637 |
+
"grad_norm": 2.5057586009739117,
|
| 1638 |
+
"learning_rate": 2.771308221117309e-06,
|
| 1639 |
+
"loss": 0.5158,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6852941176470588,
|
| 1644 |
+
"grad_norm": 2.556924814964562,
|
| 1645 |
+
"learning_rate": 2.725474875620228e-06,
|
| 1646 |
+
"loss": 0.5241,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6882352941176471,
|
| 1651 |
+
"grad_norm": 2.6209251376261586,
|
| 1652 |
+
"learning_rate": 2.6798812719531843e-06,
|
| 1653 |
+
"loss": 0.5139,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6911764705882353,
|
| 1658 |
+
"grad_norm": 3.565884155626241,
|
| 1659 |
+
"learning_rate": 2.6345322158199503e-06,
|
| 1660 |
+
"loss": 0.5196,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.6941176470588235,
|
| 1665 |
+
"grad_norm": 2.711017441488018,
|
| 1666 |
+
"learning_rate": 2.5894324871482557e-06,
|
| 1667 |
+
"loss": 0.4967,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6970588235294117,
|
| 1672 |
+
"grad_norm": 2.4420021545344355,
|
| 1673 |
+
"learning_rate": 2.544586839585961e-06,
|
| 1674 |
+
"loss": 0.4988,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7,
|
| 1679 |
+
"grad_norm": 3.318545670339647,
|
| 1680 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1681 |
+
"loss": 0.4935,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7029411764705882,
|
| 1686 |
+
"grad_norm": 2.7341852078487987,
|
| 1687 |
+
"learning_rate": 2.4556766679781763e-06,
|
| 1688 |
+
"loss": 0.4996,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7058823529411765,
|
| 1693 |
+
"grad_norm": 2.5448910575508994,
|
| 1694 |
+
"learning_rate": 2.411621515333788e-06,
|
| 1695 |
+
"loss": 0.5095,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7088235294117647,
|
| 1700 |
+
"grad_norm": 2.3716682194974448,
|
| 1701 |
+
"learning_rate": 2.3678391856132203e-06,
|
| 1702 |
+
"loss": 0.4869,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.711764705882353,
|
| 1707 |
+
"grad_norm": 2.5986125491232017,
|
| 1708 |
+
"learning_rate": 2.324334293606499e-06,
|
| 1709 |
+
"loss": 0.4964,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7147058823529412,
|
| 1714 |
+
"grad_norm": 2.6704869293015565,
|
| 1715 |
+
"learning_rate": 2.2811114248608675e-06,
|
| 1716 |
+
"loss": 0.4851,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7176470588235294,
|
| 1721 |
+
"grad_norm": 2.434069159222151,
|
| 1722 |
+
"learning_rate": 2.238175135197471e-06,
|
| 1723 |
+
"loss": 0.5144,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7205882352941176,
|
| 1728 |
+
"grad_norm": 2.8751039146064996,
|
| 1729 |
+
"learning_rate": 2.1955299502311523e-06,
|
| 1730 |
+
"loss": 0.4893,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7235294117647059,
|
| 1735 |
+
"grad_norm": 2.8716211792398103,
|
| 1736 |
+
"learning_rate": 2.1531803648934333e-06,
|
| 1737 |
+
"loss": 0.4998,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7264705882352941,
|
| 1742 |
+
"grad_norm": 2.5538647251200417,
|
| 1743 |
+
"learning_rate": 2.1111308429587446e-06,
|
| 1744 |
+
"loss": 0.4968,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7294117647058823,
|
| 1749 |
+
"grad_norm": 2.3552149327326406,
|
| 1750 |
+
"learning_rate": 2.069385816573928e-06,
|
| 1751 |
+
"loss": 0.4751,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7323529411764705,
|
| 1756 |
+
"grad_norm": 2.9092840839893994,
|
| 1757 |
+
"learning_rate": 2.0279496857910667e-06,
|
| 1758 |
+
"loss": 0.4892,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7352941176470589,
|
| 1763 |
+
"grad_norm": 2.7140627954551797,
|
| 1764 |
+
"learning_rate": 1.9868268181037186e-06,
|
| 1765 |
+
"loss": 0.5007,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7382352941176471,
|
| 1770 |
+
"grad_norm": 2.494839501975104,
|
| 1771 |
+
"learning_rate": 1.9460215479865613e-06,
|
| 1772 |
+
"loss": 0.4845,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7411764705882353,
|
| 1777 |
+
"grad_norm": 2.3723796356846556,
|
| 1778 |
+
"learning_rate": 1.9055381764385272e-06,
|
| 1779 |
+
"loss": 0.4829,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7441176470588236,
|
| 1784 |
+
"grad_norm": 2.3842886824345113,
|
| 1785 |
+
"learning_rate": 1.865380970529469e-06,
|
| 1786 |
+
"loss": 0.4853,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7470588235294118,
|
| 1791 |
+
"grad_norm": 3.0004523133733523,
|
| 1792 |
+
"learning_rate": 1.8255541629503865e-06,
|
| 1793 |
+
"loss": 0.4721,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.75,
|
| 1798 |
+
"grad_norm": 4.17555574871206,
|
| 1799 |
+
"learning_rate": 1.7860619515673034e-06,
|
| 1800 |
+
"loss": 0.4744,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7529411764705882,
|
| 1805 |
+
"grad_norm": 2.4034252951946233,
|
| 1806 |
+
"learning_rate": 1.746908498978791e-06,
|
| 1807 |
+
"loss": 0.4848,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7558823529411764,
|
| 1812 |
+
"grad_norm": 2.57638925801068,
|
| 1813 |
+
"learning_rate": 1.708097932077213e-06,
|
| 1814 |
+
"loss": 0.4771,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7588235294117647,
|
| 1819 |
+
"grad_norm": 2.648859686704048,
|
| 1820 |
+
"learning_rate": 1.6696343416137495e-06,
|
| 1821 |
+
"loss": 0.4977,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7617647058823529,
|
| 1826 |
+
"grad_norm": 2.695298184877126,
|
| 1827 |
+
"learning_rate": 1.6315217817672142e-06,
|
| 1828 |
+
"loss": 0.4785,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7647058823529411,
|
| 1833 |
+
"grad_norm": 2.5055321455184485,
|
| 1834 |
+
"learning_rate": 1.5937642697167288e-06,
|
| 1835 |
+
"loss": 0.4734,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7676470588235295,
|
| 1840 |
+
"grad_norm": 2.635547159269979,
|
| 1841 |
+
"learning_rate": 1.5563657852183072e-06,
|
| 1842 |
+
"loss": 0.4824,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7705882352941177,
|
| 1847 |
+
"grad_norm": 2.3664290913138757,
|
| 1848 |
+
"learning_rate": 1.5193302701853674e-06,
|
| 1849 |
+
"loss": 0.4622,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7735294117647059,
|
| 1854 |
+
"grad_norm": 3.119885061171242,
|
| 1855 |
+
"learning_rate": 1.4826616282732509e-06,
|
| 1856 |
+
"loss": 0.4718,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7764705882352941,
|
| 1861 |
+
"grad_norm": 2.490972462452176,
|
| 1862 |
+
"learning_rate": 1.4463637244677648e-06,
|
| 1863 |
+
"loss": 0.4669,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7794117647058824,
|
| 1868 |
+
"grad_norm": 2.458731892804391,
|
| 1869 |
+
"learning_rate": 1.410440384677791e-06,
|
| 1870 |
+
"loss": 0.4617,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7823529411764706,
|
| 1875 |
+
"grad_norm": 2.958025742420681,
|
| 1876 |
+
"learning_rate": 1.374895395332037e-06,
|
| 1877 |
+
"loss": 0.4696,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7852941176470588,
|
| 1882 |
+
"grad_norm": 2.9915004376321104,
|
| 1883 |
+
"learning_rate": 1.339732502979928e-06,
|
| 1884 |
+
"loss": 0.4636,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.788235294117647,
|
| 1889 |
+
"grad_norm": 2.966122147337875,
|
| 1890 |
+
"learning_rate": 1.3049554138967052e-06,
|
| 1891 |
+
"loss": 0.4506,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7911764705882353,
|
| 1896 |
+
"grad_norm": 2.37703783512199,
|
| 1897 |
+
"learning_rate": 1.2705677936927841e-06,
|
| 1898 |
+
"loss": 0.47,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.7941176470588235,
|
| 1903 |
+
"grad_norm": 2.378564585605087,
|
| 1904 |
+
"learning_rate": 1.2365732669273778e-06,
|
| 1905 |
+
"loss": 0.4494,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7970588235294118,
|
| 1910 |
+
"grad_norm": 2.9848625505080766,
|
| 1911 |
+
"learning_rate": 1.202975416726464e-06,
|
| 1912 |
+
"loss": 0.4545,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8,
|
| 1917 |
+
"grad_norm": 2.30742224730397,
|
| 1918 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1919 |
+
"loss": 0.4661,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8029411764705883,
|
| 1924 |
+
"grad_norm": 2.6109048720465386,
|
| 1925 |
+
"learning_rate": 1.1369838690942059e-06,
|
| 1926 |
+
"loss": 0.4533,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8058823529411765,
|
| 1931 |
+
"grad_norm": 2.9803797967996633,
|
| 1932 |
+
"learning_rate": 1.1045971273716476e-06,
|
| 1933 |
+
"loss": 0.4437,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8088235294117647,
|
| 1938 |
+
"grad_norm": 2.7563794002126416,
|
| 1939 |
+
"learning_rate": 1.072620972898007e-06,
|
| 1940 |
+
"loss": 0.4524,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8117647058823529,
|
| 1945 |
+
"grad_norm": 2.397027701415243,
|
| 1946 |
+
"learning_rate": 1.0410587760567104e-06,
|
| 1947 |
+
"loss": 0.4648,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8147058823529412,
|
| 1952 |
+
"grad_norm": 2.862646785242907,
|
| 1953 |
+
"learning_rate": 1.0099138635988026e-06,
|
| 1954 |
+
"loss": 0.4592,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8176470588235294,
|
| 1959 |
+
"grad_norm": 2.703458227745046,
|
| 1960 |
+
"learning_rate": 9.791895182922911e-07,
|
| 1961 |
+
"loss": 0.4656,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8205882352941176,
|
| 1966 |
+
"grad_norm": 2.502457085827656,
|
| 1967 |
+
"learning_rate": 9.488889785761324e-07,
|
| 1968 |
+
"loss": 0.4487,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8235294117647058,
|
| 1973 |
+
"grad_norm": 2.6423546848684167,
|
| 1974 |
+
"learning_rate": 9.190154382188921e-07,
|
| 1975 |
+
"loss": 0.4565,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8264705882352941,
|
| 1980 |
+
"grad_norm": 2.532810863987098,
|
| 1981 |
+
"learning_rate": 8.895720459821089e-07,
|
| 1982 |
+
"loss": 0.4603,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8294117647058824,
|
| 1987 |
+
"grad_norm": 2.24073062154865,
|
| 1988 |
+
"learning_rate": 8.605619052884106e-07,
|
| 1989 |
+
"loss": 0.455,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8323529411764706,
|
| 1994 |
+
"grad_norm": 2.3347312069958477,
|
| 1995 |
+
"learning_rate": 8.31988073894403e-07,
|
| 1996 |
+
"loss": 0.4542,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8352941176470589,
|
| 2001 |
+
"grad_norm": 2.2510986140412723,
|
| 2002 |
+
"learning_rate": 8.03853563568367e-07,
|
| 2003 |
+
"loss": 0.4422,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8382352941176471,
|
| 2008 |
+
"grad_norm": 2.466059819140582,
|
| 2009 |
+
"learning_rate": 7.761613397728174e-07,
|
| 2010 |
+
"loss": 0.4436,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8411764705882353,
|
| 2015 |
+
"grad_norm": 2.314552647197002,
|
| 2016 |
+
"learning_rate": 7.489143213519301e-07,
|
| 2017 |
+
"loss": 0.4526,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8441176470588235,
|
| 2022 |
+
"grad_norm": 2.2167187458342186,
|
| 2023 |
+
"learning_rate": 7.221153802238845e-07,
|
| 2024 |
+
"loss": 0.4396,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8470588235294118,
|
| 2029 |
+
"grad_norm": 2.4410281100307536,
|
| 2030 |
+
"learning_rate": 6.957673410781617e-07,
|
| 2031 |
+
"loss": 0.4493,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.85,
|
| 2036 |
+
"grad_norm": 2.391369587902256,
|
| 2037 |
+
"learning_rate": 6.698729810778065e-07,
|
| 2038 |
+
"loss": 0.4559,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8529411764705882,
|
| 2043 |
+
"grad_norm": 2.147240414992554,
|
| 2044 |
+
"learning_rate": 6.444350295667112e-07,
|
| 2045 |
+
"loss": 0.4423,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8558823529411764,
|
| 2050 |
+
"grad_norm": 2.425220228026074,
|
| 2051 |
+
"learning_rate": 6.194561677819327e-07,
|
| 2052 |
+
"loss": 0.454,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8588235294117647,
|
| 2057 |
+
"grad_norm": 2.478981866538809,
|
| 2058 |
+
"learning_rate": 5.949390285710777e-07,
|
| 2059 |
+
"loss": 0.4565,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.861764705882353,
|
| 2064 |
+
"grad_norm": 2.4324955757388484,
|
| 2065 |
+
"learning_rate": 5.708861961148004e-07,
|
| 2066 |
+
"loss": 0.4485,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8647058823529412,
|
| 2071 |
+
"grad_norm": 2.518242605502862,
|
| 2072 |
+
"learning_rate": 5.473002056544191e-07,
|
| 2073 |
+
"loss": 0.4465,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8676470588235294,
|
| 2078 |
+
"grad_norm": 2.4153603903758767,
|
| 2079 |
+
"learning_rate": 5.241835432246888e-07,
|
| 2080 |
+
"loss": 0.4449,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8705882352941177,
|
| 2085 |
+
"grad_norm": 2.5154960957655463,
|
| 2086 |
+
"learning_rate": 5.015386453917742e-07,
|
| 2087 |
+
"loss": 0.4416,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8735294117647059,
|
| 2092 |
+
"grad_norm": 2.5868546835610897,
|
| 2093 |
+
"learning_rate": 4.793678989964207e-07,
|
| 2094 |
+
"loss": 0.4343,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8764705882352941,
|
| 2099 |
+
"grad_norm": 2.483503936880007,
|
| 2100 |
+
"learning_rate": 4.576736409023813e-07,
|
| 2101 |
+
"loss": 0.4375,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8794117647058823,
|
| 2106 |
+
"grad_norm": 2.354484991161176,
|
| 2107 |
+
"learning_rate": 4.364581577500987e-07,
|
| 2108 |
+
"loss": 0.4396,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8823529411764706,
|
| 2113 |
+
"grad_norm": 2.32433065376712,
|
| 2114 |
+
"learning_rate": 4.15723685715686e-07,
|
| 2115 |
+
"loss": 0.4417,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8852941176470588,
|
| 2120 |
+
"grad_norm": 2.3785384367167173,
|
| 2121 |
+
"learning_rate": 3.9547241027523164e-07,
|
| 2122 |
+
"loss": 0.4372,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.888235294117647,
|
| 2127 |
+
"grad_norm": 2.2739855313885484,
|
| 2128 |
+
"learning_rate": 3.7570646597444196e-07,
|
| 2129 |
+
"loss": 0.4382,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8911764705882353,
|
| 2134 |
+
"grad_norm": 2.4248149464102324,
|
| 2135 |
+
"learning_rate": 3.564279362036488e-07,
|
| 2136 |
+
"loss": 0.4504,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8941176470588236,
|
| 2141 |
+
"grad_norm": 2.5206865046215623,
|
| 2142 |
+
"learning_rate": 3.3763885297822153e-07,
|
| 2143 |
+
"loss": 0.4425,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8970588235294118,
|
| 2148 |
+
"grad_norm": 2.321436598632796,
|
| 2149 |
+
"learning_rate": 3.1934119672438093e-07,
|
| 2150 |
+
"loss": 0.4508,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9,
|
| 2155 |
+
"grad_norm": 2.2956489839218253,
|
| 2156 |
+
"learning_rate": 3.015368960704584e-07,
|
| 2157 |
+
"loss": 0.4218,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9029411764705882,
|
| 2162 |
+
"grad_norm": 2.6380831075744045,
|
| 2163 |
+
"learning_rate": 2.842278276436128e-07,
|
| 2164 |
+
"loss": 0.4312,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9058823529411765,
|
| 2169 |
+
"grad_norm": 2.353068804554375,
|
| 2170 |
+
"learning_rate": 2.6741581587202747e-07,
|
| 2171 |
+
"loss": 0.4572,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9088235294117647,
|
| 2176 |
+
"grad_norm": 2.3504938166434286,
|
| 2177 |
+
"learning_rate": 2.511026327926114e-07,
|
| 2178 |
+
"loss": 0.4415,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9117647058823529,
|
| 2183 |
+
"grad_norm": 2.4033670936853864,
|
| 2184 |
+
"learning_rate": 2.3528999786421758e-07,
|
| 2185 |
+
"loss": 0.4383,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9147058823529411,
|
| 2190 |
+
"grad_norm": 2.5018773276624557,
|
| 2191 |
+
"learning_rate": 2.1997957778641166e-07,
|
| 2192 |
+
"loss": 0.448,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9176470588235294,
|
| 2197 |
+
"grad_norm": 2.3367055632905753,
|
| 2198 |
+
"learning_rate": 2.0517298632379445e-07,
|
| 2199 |
+
"loss": 0.4497,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9205882352941176,
|
| 2204 |
+
"grad_norm": 2.765125262874024,
|
| 2205 |
+
"learning_rate": 1.908717841359048e-07,
|
| 2206 |
+
"loss": 0.4507,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9235294117647059,
|
| 2211 |
+
"grad_norm": 2.376902313703198,
|
| 2212 |
+
"learning_rate": 1.770774786127244e-07,
|
| 2213 |
+
"loss": 0.4183,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9264705882352942,
|
| 2218 |
+
"grad_norm": 2.413442304678235,
|
| 2219 |
+
"learning_rate": 1.6379152371579277e-07,
|
| 2220 |
+
"loss": 0.4334,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9294117647058824,
|
| 2225 |
+
"grad_norm": 2.3143531531697596,
|
| 2226 |
+
"learning_rate": 1.510153198249531e-07,
|
| 2227 |
+
"loss": 0.436,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9323529411764706,
|
| 2232 |
+
"grad_norm": 2.34608810626235,
|
| 2233 |
+
"learning_rate": 1.3875021359075257e-07,
|
| 2234 |
+
"loss": 0.44,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9352941176470588,
|
| 2239 |
+
"grad_norm": 2.35355433201976,
|
| 2240 |
+
"learning_rate": 1.2699749779249926e-07,
|
| 2241 |
+
"loss": 0.4314,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9382352941176471,
|
| 2246 |
+
"grad_norm": 2.218100203321306,
|
| 2247 |
+
"learning_rate": 1.157584112019966e-07,
|
| 2248 |
+
"loss": 0.4384,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9411764705882353,
|
| 2253 |
+
"grad_norm": 2.3877266166874676,
|
| 2254 |
+
"learning_rate": 1.0503413845297739e-07,
|
| 2255 |
+
"loss": 0.4357,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9441176470588235,
|
| 2260 |
+
"grad_norm": 2.3187529416036936,
|
| 2261 |
+
"learning_rate": 9.482580991623747e-08,
|
| 2262 |
+
"loss": 0.4323,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9470588235294117,
|
| 2267 |
+
"grad_norm": 2.2887581674114648,
|
| 2268 |
+
"learning_rate": 8.513450158049109e-08,
|
| 2269 |
+
"loss": 0.4448,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.95,
|
| 2274 |
+
"grad_norm": 2.3836470257770945,
|
| 2275 |
+
"learning_rate": 7.59612349389599e-08,
|
| 2276 |
+
"loss": 0.4269,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9529411764705882,
|
| 2281 |
+
"grad_norm": 2.233426183677112,
|
| 2282 |
+
"learning_rate": 6.730697688170251e-08,
|
| 2283 |
+
"loss": 0.4351,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9558823529411765,
|
| 2288 |
+
"grad_norm": 2.2576790941626284,
|
| 2289 |
+
"learning_rate": 5.917263959370312e-08,
|
| 2290 |
+
"loss": 0.4355,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9588235294117647,
|
| 2295 |
+
"grad_norm": 2.2984294403136163,
|
| 2296 |
+
"learning_rate": 5.155908045872349e-08,
|
| 2297 |
+
"loss": 0.4406,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.961764705882353,
|
| 2302 |
+
"grad_norm": 2.3707792344113945,
|
| 2303 |
+
"learning_rate": 4.446710196893245e-08,
|
| 2304 |
+
"loss": 0.4385,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9647058823529412,
|
| 2309 |
+
"grad_norm": 2.2246371108164262,
|
| 2310 |
+
"learning_rate": 3.7897451640321326e-08,
|
| 2311 |
+
"loss": 0.4246,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9676470588235294,
|
| 2316 |
+
"grad_norm": 2.622284137214347,
|
| 2317 |
+
"learning_rate": 3.185082193391143e-08,
|
| 2318 |
+
"loss": 0.4279,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9705882352941176,
|
| 2323 |
+
"grad_norm": 2.3946023382142383,
|
| 2324 |
+
"learning_rate": 2.6327850182769065e-08,
|
| 2325 |
+
"loss": 0.4573,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9735294117647059,
|
| 2330 |
+
"grad_norm": 2.354547522725208,
|
| 2331 |
+
"learning_rate": 2.1329118524827662e-08,
|
| 2332 |
+
"loss": 0.4405,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9764705882352941,
|
| 2337 |
+
"grad_norm": 2.3346382733247815,
|
| 2338 |
+
"learning_rate": 1.6855153841527915e-08,
|
| 2339 |
+
"loss": 0.4386,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.9794117647058823,
|
| 2344 |
+
"grad_norm": 2.3205008178624764,
|
| 2345 |
+
"learning_rate": 1.2906427702284452e-08,
|
| 2346 |
+
"loss": 0.4304,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.9823529411764705,
|
| 2351 |
+
"grad_norm": 2.4270314026839714,
|
| 2352 |
+
"learning_rate": 9.48335631477948e-09,
|
| 2353 |
+
"loss": 0.4415,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9852941176470589,
|
| 2358 |
+
"grad_norm": 2.6174654383893214,
|
| 2359 |
+
"learning_rate": 6.586300481095098e-09,
|
| 2360 |
+
"loss": 0.4272,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9882352941176471,
|
| 2365 |
+
"grad_norm": 2.2960002654994973,
|
| 2366 |
+
"learning_rate": 4.2155655596809455e-09,
|
| 2367 |
+
"loss": 0.4334,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9911764705882353,
|
| 2372 |
+
"grad_norm": 2.443491425156529,
|
| 2373 |
+
"learning_rate": 2.371401433170495e-09,
|
| 2374 |
+
"loss": 0.4249,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9941176470588236,
|
| 2379 |
+
"grad_norm": 2.2442264144591926,
|
| 2380 |
+
"learning_rate": 1.054002482043237e-09,
|
| 2381 |
+
"loss": 0.428,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9970588235294118,
|
| 2386 |
+
"grad_norm": 2.4769821745544953,
|
| 2387 |
+
"learning_rate": 2.6350756413440203e-10,
|
| 2388 |
+
"loss": 0.4334,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 1.0,
|
| 2393 |
+
"grad_norm": 2.3764816774266806,
|
| 2394 |
+
"learning_rate": 0.0,
|
| 2395 |
+
"loss": 0.4299,
|
| 2396 |
+
"step": 1700
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 1.0,
|
| 2400 |
+
"eval_runtime": 4.4513,
|
| 2401 |
+
"eval_samples_per_second": 2.247,
|
| 2402 |
+
"eval_steps_per_second": 0.674,
|
| 2403 |
+
"step": 1700
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 1.0,
|
| 2407 |
+
"step": 1700,
|
| 2408 |
+
"total_flos": 117051280588800.0,
|
| 2409 |
+
"train_loss": 0.691351400122923,
|
| 2410 |
+
"train_runtime": 20676.98,
|
| 2411 |
+
"train_samples_per_second": 1.315,
|
| 2412 |
+
"train_steps_per_second": 0.082
|
| 2413 |
+
}
|
| 2414 |
+
],
|
| 2415 |
+
"logging_steps": 5,
|
| 2416 |
+
"max_steps": 1700,
|
| 2417 |
+
"num_input_tokens_seen": 0,
|
| 2418 |
+
"num_train_epochs": 1,
|
| 2419 |
+
"save_steps": 100,
|
| 2420 |
+
"stateful_callbacks": {
|
| 2421 |
+
"TrainerControl": {
|
| 2422 |
+
"args": {
|
| 2423 |
+
"should_epoch_stop": false,
|
| 2424 |
+
"should_evaluate": false,
|
| 2425 |
+
"should_log": false,
|
| 2426 |
+
"should_save": true,
|
| 2427 |
+
"should_training_stop": true
|
| 2428 |
+
},
|
| 2429 |
+
"attributes": {}
|
| 2430 |
+
}
|
| 2431 |
+
},
|
| 2432 |
+
"total_flos": 117051280588800.0,
|
| 2433 |
+
"train_batch_size": 4,
|
| 2434 |
+
"trial_name": null,
|
| 2435 |
+
"trial_params": null
|
| 2436 |
+
}
|