Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- runs/Sep19_17-38-55_COE-CS-sv003/events.out.tfevents.1726767828.COE-CS-sv003.388702.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2429 -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_dirty
|
| 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_dirty
|
| 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": 116982426894336.0,
|
| 4 |
+
"train_loss": 0.6940072515278861,
|
| 5 |
+
"train_runtime": 20640.2326,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.317,
|
| 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_17-38-55_COE-CS-sv003/events.out.tfevents.1726767828.COE-CS-sv003.388702.0
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 sha256:7c6d67b324eca9c14a811ea9bc3602e4b4e97eb265bb0cee1ea183f32f032ba1
|
| 3 |
+
size 77405
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 116982426894336.0,
|
| 4 |
+
"train_loss": 0.6940072515278861,
|
| 5 |
+
"train_runtime": 20640.2326,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.317,
|
| 8 |
+
"train_steps_per_second": 0.082
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2429 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1699,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005885815185403178,
|
| 13 |
+
"grad_norm": 81.18565670358198,
|
| 14 |
+
"learning_rate": 5.882352941176471e-08,
|
| 15 |
+
"loss": 4.1356,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.002942907592701589,
|
| 20 |
+
"grad_norm": 77.04540875822474,
|
| 21 |
+
"learning_rate": 2.9411764705882356e-07,
|
| 22 |
+
"loss": 4.1935,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005885815185403178,
|
| 27 |
+
"grad_norm": 35.44829095248985,
|
| 28 |
+
"learning_rate": 5.882352941176471e-07,
|
| 29 |
+
"loss": 3.9438,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008828722778104767,
|
| 34 |
+
"grad_norm": 22.5439620439223,
|
| 35 |
+
"learning_rate": 8.823529411764707e-07,
|
| 36 |
+
"loss": 3.4429,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011771630370806356,
|
| 41 |
+
"grad_norm": 13.2019843638693,
|
| 42 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 43 |
+
"loss": 3.1438,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.014714537963507945,
|
| 48 |
+
"grad_norm": 14.772036939700145,
|
| 49 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 50 |
+
"loss": 2.9141,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.017657445556209534,
|
| 55 |
+
"grad_norm": 20.5779186199068,
|
| 56 |
+
"learning_rate": 1.7647058823529414e-06,
|
| 57 |
+
"loss": 2.6074,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.020600353148911125,
|
| 62 |
+
"grad_norm": 27.382526732783585,
|
| 63 |
+
"learning_rate": 2.058823529411765e-06,
|
| 64 |
+
"loss": 2.2919,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.023543260741612712,
|
| 69 |
+
"grad_norm": 20.764560265022364,
|
| 70 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 71 |
+
"loss": 1.7801,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.026486168334314303,
|
| 76 |
+
"grad_norm": 11.788637809965795,
|
| 77 |
+
"learning_rate": 2.647058823529412e-06,
|
| 78 |
+
"loss": 1.3328,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02942907592701589,
|
| 83 |
+
"grad_norm": 22.53245322685116,
|
| 84 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 85 |
+
"loss": 1.1855,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.032371983519717484,
|
| 90 |
+
"grad_norm": 6.015914744919097,
|
| 91 |
+
"learning_rate": 3.2352941176470594e-06,
|
| 92 |
+
"loss": 1.1366,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03531489111241907,
|
| 97 |
+
"grad_norm": 5.217837662899846,
|
| 98 |
+
"learning_rate": 3.529411764705883e-06,
|
| 99 |
+
"loss": 1.1055,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.03825779870512066,
|
| 104 |
+
"grad_norm": 3.357783317857939,
|
| 105 |
+
"learning_rate": 3.8235294117647055e-06,
|
| 106 |
+
"loss": 1.0953,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04120070629782225,
|
| 111 |
+
"grad_norm": 4.996001471413591,
|
| 112 |
+
"learning_rate": 4.11764705882353e-06,
|
| 113 |
+
"loss": 1.0496,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04414361389052384,
|
| 118 |
+
"grad_norm": 8.621809656222444,
|
| 119 |
+
"learning_rate": 4.411764705882353e-06,
|
| 120 |
+
"loss": 1.0292,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.047086521483225424,
|
| 125 |
+
"grad_norm": 5.339001901811776,
|
| 126 |
+
"learning_rate": 4.705882352941177e-06,
|
| 127 |
+
"loss": 1.0423,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.050029429075927015,
|
| 132 |
+
"grad_norm": 3.493074689207993,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 1.026,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.052972336668628606,
|
| 139 |
+
"grad_norm": 9.685062179270181,
|
| 140 |
+
"learning_rate": 5.294117647058824e-06,
|
| 141 |
+
"loss": 1.005,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.055915244261330196,
|
| 146 |
+
"grad_norm": 6.933487463420427,
|
| 147 |
+
"learning_rate": 5.588235294117647e-06,
|
| 148 |
+
"loss": 0.9885,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.05885815185403178,
|
| 153 |
+
"grad_norm": 8.28144947566066,
|
| 154 |
+
"learning_rate": 5.882352941176471e-06,
|
| 155 |
+
"loss": 0.9873,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06180105944673337,
|
| 160 |
+
"grad_norm": 17.077053218905366,
|
| 161 |
+
"learning_rate": 6.176470588235295e-06,
|
| 162 |
+
"loss": 0.9694,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06474396703943497,
|
| 167 |
+
"grad_norm": 19.687199094462986,
|
| 168 |
+
"learning_rate": 6.470588235294119e-06,
|
| 169 |
+
"loss": 0.9622,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06768687463213655,
|
| 174 |
+
"grad_norm": 13.804537783692052,
|
| 175 |
+
"learning_rate": 6.764705882352942e-06,
|
| 176 |
+
"loss": 0.9563,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07062978222483814,
|
| 181 |
+
"grad_norm": 10.94725198202626,
|
| 182 |
+
"learning_rate": 7.058823529411766e-06,
|
| 183 |
+
"loss": 0.9414,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07357268981753973,
|
| 188 |
+
"grad_norm": 18.46306700568472,
|
| 189 |
+
"learning_rate": 7.352941176470589e-06,
|
| 190 |
+
"loss": 0.9289,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07651559741024132,
|
| 195 |
+
"grad_norm": 14.894213258364779,
|
| 196 |
+
"learning_rate": 7.647058823529411e-06,
|
| 197 |
+
"loss": 0.9238,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.07945850500294291,
|
| 202 |
+
"grad_norm": 7.569487604501164,
|
| 203 |
+
"learning_rate": 7.941176470588236e-06,
|
| 204 |
+
"loss": 0.9183,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.0824014125956445,
|
| 209 |
+
"grad_norm": 21.99231321792487,
|
| 210 |
+
"learning_rate": 8.23529411764706e-06,
|
| 211 |
+
"loss": 0.911,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08534432018834609,
|
| 216 |
+
"grad_norm": 18.115769647379153,
|
| 217 |
+
"learning_rate": 8.529411764705883e-06,
|
| 218 |
+
"loss": 0.9047,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08828722778104768,
|
| 223 |
+
"grad_norm": 10.705484744227235,
|
| 224 |
+
"learning_rate": 8.823529411764707e-06,
|
| 225 |
+
"loss": 0.8932,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09123013537374926,
|
| 230 |
+
"grad_norm": 16.313242301856516,
|
| 231 |
+
"learning_rate": 9.11764705882353e-06,
|
| 232 |
+
"loss": 0.8899,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09417304296645085,
|
| 237 |
+
"grad_norm": 12.978636094048102,
|
| 238 |
+
"learning_rate": 9.411764705882354e-06,
|
| 239 |
+
"loss": 0.8916,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09711595055915244,
|
| 244 |
+
"grad_norm": 11.506756600035605,
|
| 245 |
+
"learning_rate": 9.705882352941177e-06,
|
| 246 |
+
"loss": 0.8789,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10005885815185403,
|
| 251 |
+
"grad_norm": 24.989193177477567,
|
| 252 |
+
"learning_rate": 1e-05,
|
| 253 |
+
"loss": 0.8768,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10300176574455562,
|
| 258 |
+
"grad_norm": 13.397711202568765,
|
| 259 |
+
"learning_rate": 9.99973614764657e-06,
|
| 260 |
+
"loss": 0.8801,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10594467333725721,
|
| 265 |
+
"grad_norm": 12.037309022627504,
|
| 266 |
+
"learning_rate": 9.99894461843351e-06,
|
| 267 |
+
"loss": 0.8731,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1088875809299588,
|
| 272 |
+
"grad_norm": 22.40769728661026,
|
| 273 |
+
"learning_rate": 9.997625495899555e-06,
|
| 274 |
+
"loss": 0.8712,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11183048852266039,
|
| 279 |
+
"grad_norm": 14.584492756570954,
|
| 280 |
+
"learning_rate": 9.99577891926614e-06,
|
| 281 |
+
"loss": 0.8491,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11477339611536198,
|
| 286 |
+
"grad_norm": 9.426683321327127,
|
| 287 |
+
"learning_rate": 9.9934050834227e-06,
|
| 288 |
+
"loss": 0.8674,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.11771630370806356,
|
| 293 |
+
"grad_norm": 25.968161707891955,
|
| 294 |
+
"learning_rate": 9.990504238906107e-06,
|
| 295 |
+
"loss": 0.8573,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12065921130076515,
|
| 300 |
+
"grad_norm": 13.072895533976467,
|
| 301 |
+
"learning_rate": 9.98707669187422e-06,
|
| 302 |
+
"loss": 0.8525,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12360211889346674,
|
| 307 |
+
"grad_norm": 11.095395155610879,
|
| 308 |
+
"learning_rate": 9.98312280407358e-06,
|
| 309 |
+
"loss": 0.8553,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12654502648616833,
|
| 314 |
+
"grad_norm": 16.418159232638153,
|
| 315 |
+
"learning_rate": 9.978642992801229e-06,
|
| 316 |
+
"loss": 0.8555,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.12948793407886994,
|
| 321 |
+
"grad_norm": 4.450231202820506,
|
| 322 |
+
"learning_rate": 9.973637730860664e-06,
|
| 323 |
+
"loss": 0.8413,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.13243084167157151,
|
| 328 |
+
"grad_norm": 4.111542494606179,
|
| 329 |
+
"learning_rate": 9.968107546511942e-06,
|
| 330 |
+
"loss": 0.8424,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1353737492642731,
|
| 335 |
+
"grad_norm": 13.693095667132473,
|
| 336 |
+
"learning_rate": 9.962053023415926e-06,
|
| 337 |
+
"loss": 0.8561,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1383166568569747,
|
| 342 |
+
"grad_norm": 5.646518368088086,
|
| 343 |
+
"learning_rate": 9.955474800572684e-06,
|
| 344 |
+
"loss": 0.8636,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.14125956444967627,
|
| 349 |
+
"grad_norm": 5.108991758643153,
|
| 350 |
+
"learning_rate": 9.948373572254044e-06,
|
| 351 |
+
"loss": 0.8503,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.14420247204237788,
|
| 356 |
+
"grad_norm": 9.81020252703176,
|
| 357 |
+
"learning_rate": 9.94075008793033e-06,
|
| 358 |
+
"loss": 0.873,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.14714537963507945,
|
| 363 |
+
"grad_norm": 7.935285563014131,
|
| 364 |
+
"learning_rate": 9.932605152191255e-06,
|
| 365 |
+
"loss": 0.8433,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15008828722778106,
|
| 370 |
+
"grad_norm": 3.3393405756912484,
|
| 371 |
+
"learning_rate": 9.923939624661005e-06,
|
| 372 |
+
"loss": 0.8296,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.15303119482048264,
|
| 377 |
+
"grad_norm": 3.830858966207542,
|
| 378 |
+
"learning_rate": 9.914754419907508e-06,
|
| 379 |
+
"loss": 0.8317,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1559741024131842,
|
| 384 |
+
"grad_norm": 16.404195583675744,
|
| 385 |
+
"learning_rate": 9.905050507345926e-06,
|
| 386 |
+
"loss": 0.8419,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.15891701000588582,
|
| 391 |
+
"grad_norm": 9.685851257998209,
|
| 392 |
+
"learning_rate": 9.894828911136323e-06,
|
| 393 |
+
"loss": 0.8649,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1618599175985874,
|
| 398 |
+
"grad_norm": 5.518903601787907,
|
| 399 |
+
"learning_rate": 9.884090710075585e-06,
|
| 400 |
+
"loss": 0.8508,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.164802825191289,
|
| 405 |
+
"grad_norm": 6.557040852323014,
|
| 406 |
+
"learning_rate": 9.87283703748356e-06,
|
| 407 |
+
"loss": 0.8537,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.16774573278399058,
|
| 412 |
+
"grad_norm": 6.51479177962607,
|
| 413 |
+
"learning_rate": 9.861069081083452e-06,
|
| 414 |
+
"loss": 0.8454,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17068864037669218,
|
| 419 |
+
"grad_norm": 4.48759740362246,
|
| 420 |
+
"learning_rate": 9.84878808287645e-06,
|
| 421 |
+
"loss": 0.8494,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17363154796939376,
|
| 426 |
+
"grad_norm": 3.077914439009112,
|
| 427 |
+
"learning_rate": 9.835995339010673e-06,
|
| 428 |
+
"loss": 0.8371,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.17657445556209536,
|
| 433 |
+
"grad_norm": 7.741990389634322,
|
| 434 |
+
"learning_rate": 9.822692199644346e-06,
|
| 435 |
+
"loss": 0.8489,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.17951736315479694,
|
| 440 |
+
"grad_norm": 7.12110354743661,
|
| 441 |
+
"learning_rate": 9.808880068803323e-06,
|
| 442 |
+
"loss": 0.8386,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.18246027074749852,
|
| 447 |
+
"grad_norm": 4.874821068849924,
|
| 448 |
+
"learning_rate": 9.794560404232898e-06,
|
| 449 |
+
"loss": 0.8399,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.18540317834020012,
|
| 454 |
+
"grad_norm": 4.563419733002005,
|
| 455 |
+
"learning_rate": 9.779734717243946e-06,
|
| 456 |
+
"loss": 0.8271,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1883460859329017,
|
| 461 |
+
"grad_norm": 10.93234206126768,
|
| 462 |
+
"learning_rate": 9.764404572553427e-06,
|
| 463 |
+
"loss": 0.8123,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1912889935256033,
|
| 468 |
+
"grad_norm": 9.419206687007401,
|
| 469 |
+
"learning_rate": 9.748571588119247e-06,
|
| 470 |
+
"loss": 0.8188,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.19423190111830488,
|
| 475 |
+
"grad_norm": 6.977872680215194,
|
| 476 |
+
"learning_rate": 9.732237434969487e-06,
|
| 477 |
+
"loss": 0.8255,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.19717480871100648,
|
| 482 |
+
"grad_norm": 5.908304115359454,
|
| 483 |
+
"learning_rate": 9.715403837026046e-06,
|
| 484 |
+
"loss": 0.8265,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.20011771630370806,
|
| 489 |
+
"grad_norm": 6.797809972888078,
|
| 490 |
+
"learning_rate": 9.698072570922698e-06,
|
| 491 |
+
"loss": 0.8118,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.20306062389640966,
|
| 496 |
+
"grad_norm": 6.704121135386284,
|
| 497 |
+
"learning_rate": 9.680245465817583e-06,
|
| 498 |
+
"loss": 0.8096,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.20600353148911124,
|
| 503 |
+
"grad_norm": 8.645898444887452,
|
| 504 |
+
"learning_rate": 9.661924403200155e-06,
|
| 505 |
+
"loss": 0.8038,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.20894643908181282,
|
| 510 |
+
"grad_norm": 6.272395534856629,
|
| 511 |
+
"learning_rate": 9.64311131669261e-06,
|
| 512 |
+
"loss": 0.7848,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.21188934667451442,
|
| 517 |
+
"grad_norm": 7.0264316123358155,
|
| 518 |
+
"learning_rate": 9.623808191845806e-06,
|
| 519 |
+
"loss": 0.7885,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.214832254267216,
|
| 524 |
+
"grad_norm": 6.754946681714514,
|
| 525 |
+
"learning_rate": 9.604017065929715e-06,
|
| 526 |
+
"loss": 0.7937,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2177751618599176,
|
| 531 |
+
"grad_norm": 6.611299682141483,
|
| 532 |
+
"learning_rate": 9.583740027718396e-06,
|
| 533 |
+
"loss": 0.8036,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22071806945261918,
|
| 538 |
+
"grad_norm": 4.018544136474155,
|
| 539 |
+
"learning_rate": 9.562979217269547e-06,
|
| 540 |
+
"loss": 0.7969,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.22366097704532079,
|
| 545 |
+
"grad_norm": 7.101373596601596,
|
| 546 |
+
"learning_rate": 9.54173682569865e-06,
|
| 547 |
+
"loss": 0.796,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.22660388463802236,
|
| 552 |
+
"grad_norm": 6.399510185421723,
|
| 553 |
+
"learning_rate": 9.520015094947704e-06,
|
| 554 |
+
"loss": 0.8075,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.22954679223072397,
|
| 559 |
+
"grad_norm": 9.296984628579494,
|
| 560 |
+
"learning_rate": 9.497816317548625e-06,
|
| 561 |
+
"loss": 0.7799,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.23248969982342554,
|
| 566 |
+
"grad_norm": 4.786255899468576,
|
| 567 |
+
"learning_rate": 9.475142836381274e-06,
|
| 568 |
+
"loss": 0.7695,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.23543260741612712,
|
| 573 |
+
"grad_norm": 4.559456259891477,
|
| 574 |
+
"learning_rate": 9.4519970444262e-06,
|
| 575 |
+
"loss": 0.7741,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.23837551500882873,
|
| 580 |
+
"grad_norm": 7.466092853647954,
|
| 581 |
+
"learning_rate": 9.428381384512073e-06,
|
| 582 |
+
"loss": 0.7619,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2413184226015303,
|
| 587 |
+
"grad_norm": 6.623530673850783,
|
| 588 |
+
"learning_rate": 9.404298349057873e-06,
|
| 589 |
+
"loss": 0.7731,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2442613301942319,
|
| 594 |
+
"grad_norm": 6.068028651768666,
|
| 595 |
+
"learning_rate": 9.379750479809832e-06,
|
| 596 |
+
"loss": 0.7574,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.24720423778693348,
|
| 601 |
+
"grad_norm": 8.419516596990583,
|
| 602 |
+
"learning_rate": 9.354740367573178e-06,
|
| 603 |
+
"loss": 0.7821,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25014714537963506,
|
| 608 |
+
"grad_norm": 4.198177608767893,
|
| 609 |
+
"learning_rate": 9.329270651938702e-06,
|
| 610 |
+
"loss": 0.7547,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.25309005297233667,
|
| 615 |
+
"grad_norm": 4.566575750438902,
|
| 616 |
+
"learning_rate": 9.303344021004169e-06,
|
| 617 |
+
"loss": 0.7664,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.25603296056503827,
|
| 622 |
+
"grad_norm": 8.680313490795726,
|
| 623 |
+
"learning_rate": 9.27696321109061e-06,
|
| 624 |
+
"loss": 0.7668,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2589758681577399,
|
| 629 |
+
"grad_norm": 9.12784954370401,
|
| 630 |
+
"learning_rate": 9.250131006453544e-06,
|
| 631 |
+
"loss": 0.7678,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2619187757504414,
|
| 636 |
+
"grad_norm": 10.144210993180703,
|
| 637 |
+
"learning_rate": 9.222850238989104e-06,
|
| 638 |
+
"loss": 0.765,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.26486168334314303,
|
| 643 |
+
"grad_norm": 3.0429868552146964,
|
| 644 |
+
"learning_rate": 9.195123787935168e-06,
|
| 645 |
+
"loss": 0.7445,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.26780459093584463,
|
| 650 |
+
"grad_norm": 4.691784082396774,
|
| 651 |
+
"learning_rate": 9.166954579567483e-06,
|
| 652 |
+
"loss": 0.7649,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2707474985285462,
|
| 657 |
+
"grad_norm": 4.7416354261595615,
|
| 658 |
+
"learning_rate": 9.138345586890818e-06,
|
| 659 |
+
"loss": 0.7496,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2736904061212478,
|
| 664 |
+
"grad_norm": 6.223772502898528,
|
| 665 |
+
"learning_rate": 9.109299829325191e-06,
|
| 666 |
+
"loss": 0.7439,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2766333137139494,
|
| 671 |
+
"grad_norm": 5.069169897647759,
|
| 672 |
+
"learning_rate": 9.0798203723872e-06,
|
| 673 |
+
"loss": 0.7552,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.279576221306651,
|
| 678 |
+
"grad_norm": 4.067678548778218,
|
| 679 |
+
"learning_rate": 9.049910327366481e-06,
|
| 680 |
+
"loss": 0.7479,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.28251912889935255,
|
| 685 |
+
"grad_norm": 7.216787820352084,
|
| 686 |
+
"learning_rate": 9.019572850997339e-06,
|
| 687 |
+
"loss": 0.7456,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.28546203649205415,
|
| 692 |
+
"grad_norm": 6.118517874215981,
|
| 693 |
+
"learning_rate": 8.988811145125592e-06,
|
| 694 |
+
"loss": 0.7373,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.28840494408475575,
|
| 699 |
+
"grad_norm": 6.974203549951666,
|
| 700 |
+
"learning_rate": 8.957628456370636e-06,
|
| 701 |
+
"loss": 0.7544,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2913478516774573,
|
| 706 |
+
"grad_norm": 6.659489098506463,
|
| 707 |
+
"learning_rate": 8.926028075782791e-06,
|
| 708 |
+
"loss": 0.7485,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2942907592701589,
|
| 713 |
+
"grad_norm": 6.963976313909436,
|
| 714 |
+
"learning_rate": 8.89401333849598e-06,
|
| 715 |
+
"loss": 0.7364,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.2972336668628605,
|
| 720 |
+
"grad_norm": 5.060368338459147,
|
| 721 |
+
"learning_rate": 8.861587623375711e-06,
|
| 722 |
+
"loss": 0.7335,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3001765744555621,
|
| 727 |
+
"grad_norm": 11.693154707714637,
|
| 728 |
+
"learning_rate": 8.828754352662483e-06,
|
| 729 |
+
"loss": 0.7469,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.30311948204826367,
|
| 734 |
+
"grad_norm": 8.708261644979073,
|
| 735 |
+
"learning_rate": 8.795516991610592e-06,
|
| 736 |
+
"loss": 0.7296,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.30606238964096527,
|
| 741 |
+
"grad_norm": 5.926416635055216,
|
| 742 |
+
"learning_rate": 8.761879048122414e-06,
|
| 743 |
+
"loss": 0.7443,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3090052972336669,
|
| 748 |
+
"grad_norm": 12.904699986898022,
|
| 749 |
+
"learning_rate": 8.727844072378173e-06,
|
| 750 |
+
"loss": 0.7182,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3119482048263684,
|
| 755 |
+
"grad_norm": 6.368508385097698,
|
| 756 |
+
"learning_rate": 8.693415656461244e-06,
|
| 757 |
+
"loss": 0.734,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.31489111241907003,
|
| 762 |
+
"grad_norm": 3.3588114427301607,
|
| 763 |
+
"learning_rate": 8.658597433979057e-06,
|
| 764 |
+
"loss": 0.7248,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.31783402001177163,
|
| 769 |
+
"grad_norm": 12.331947606606729,
|
| 770 |
+
"learning_rate": 8.623393079679588e-06,
|
| 771 |
+
"loss": 0.7344,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.32077692760447324,
|
| 776 |
+
"grad_norm": 12.166702155420914,
|
| 777 |
+
"learning_rate": 8.587806309063531e-06,
|
| 778 |
+
"loss": 0.726,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3237198351971748,
|
| 783 |
+
"grad_norm": 4.612832603972562,
|
| 784 |
+
"learning_rate": 8.551840877992155e-06,
|
| 785 |
+
"loss": 0.7283,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3266627427898764,
|
| 790 |
+
"grad_norm": 5.77424459107182,
|
| 791 |
+
"learning_rate": 8.515500582290914e-06,
|
| 792 |
+
"loss": 0.7155,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.329605650382578,
|
| 797 |
+
"grad_norm": 3.6459134260142214,
|
| 798 |
+
"learning_rate": 8.478789257348827e-06,
|
| 799 |
+
"loss": 0.7116,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3325485579752796,
|
| 804 |
+
"grad_norm": 3.0011507986582733,
|
| 805 |
+
"learning_rate": 8.441710777713683e-06,
|
| 806 |
+
"loss": 0.7037,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.33549146556798115,
|
| 811 |
+
"grad_norm": 5.360195180614321,
|
| 812 |
+
"learning_rate": 8.404269056683133e-06,
|
| 813 |
+
"loss": 0.7169,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.33843437316068276,
|
| 818 |
+
"grad_norm": 5.707195645216086,
|
| 819 |
+
"learning_rate": 8.366468045891658e-06,
|
| 820 |
+
"loss": 0.7102,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.34137728075338436,
|
| 825 |
+
"grad_norm": 3.5974238955778457,
|
| 826 |
+
"learning_rate": 8.32831173489352e-06,
|
| 827 |
+
"loss": 0.7115,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3443201883460859,
|
| 832 |
+
"grad_norm": 3.2599056325905367,
|
| 833 |
+
"learning_rate": 8.289804150741704e-06,
|
| 834 |
+
"loss": 0.7154,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3472630959387875,
|
| 839 |
+
"grad_norm": 4.229649245193178,
|
| 840 |
+
"learning_rate": 8.250949357562892e-06,
|
| 841 |
+
"loss": 0.7057,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3502060035314891,
|
| 846 |
+
"grad_norm": 5.194627560952757,
|
| 847 |
+
"learning_rate": 8.211751456128532e-06,
|
| 848 |
+
"loss": 0.7251,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3531489111241907,
|
| 853 |
+
"grad_norm": 3.0868739009823583,
|
| 854 |
+
"learning_rate": 8.172214583422038e-06,
|
| 855 |
+
"loss": 0.7067,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.35609181871689227,
|
| 860 |
+
"grad_norm": 4.2974502097872405,
|
| 861 |
+
"learning_rate": 8.132342912202178e-06,
|
| 862 |
+
"loss": 0.7144,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3590347263095939,
|
| 867 |
+
"grad_norm": 3.271991629082987,
|
| 868 |
+
"learning_rate": 8.092140650562665e-06,
|
| 869 |
+
"loss": 0.7103,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3619776339022955,
|
| 874 |
+
"grad_norm": 4.2083604726976915,
|
| 875 |
+
"learning_rate": 8.051612041488038e-06,
|
| 876 |
+
"loss": 0.7211,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.36492054149499703,
|
| 881 |
+
"grad_norm": 3.6340200304717114,
|
| 882 |
+
"learning_rate": 8.010761362405851e-06,
|
| 883 |
+
"loss": 0.7272,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.36786344908769864,
|
| 888 |
+
"grad_norm": 5.783094842613372,
|
| 889 |
+
"learning_rate": 7.969592924735234e-06,
|
| 890 |
+
"loss": 0.7054,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.37080635668040024,
|
| 895 |
+
"grad_norm": 5.0424325252776745,
|
| 896 |
+
"learning_rate": 7.928111073431847e-06,
|
| 897 |
+
"loss": 0.7122,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.37374926427310184,
|
| 902 |
+
"grad_norm": 5.196701737520978,
|
| 903 |
+
"learning_rate": 7.886320186529327e-06,
|
| 904 |
+
"loss": 0.6953,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3766921718658034,
|
| 909 |
+
"grad_norm": 3.608090265928746,
|
| 910 |
+
"learning_rate": 7.844224674677225e-06,
|
| 911 |
+
"loss": 0.6978,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.379635079458505,
|
| 916 |
+
"grad_norm": 4.731749503563869,
|
| 917 |
+
"learning_rate": 7.801828980675485e-06,
|
| 918 |
+
"loss": 0.6953,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3825779870512066,
|
| 923 |
+
"grad_norm": 3.0237351828264547,
|
| 924 |
+
"learning_rate": 7.75913757900556e-06,
|
| 925 |
+
"loss": 0.6738,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3855208946439082,
|
| 930 |
+
"grad_norm": 3.0717508997327365,
|
| 931 |
+
"learning_rate": 7.716154975358174e-06,
|
| 932 |
+
"loss": 0.7046,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.38846380223660976,
|
| 937 |
+
"grad_norm": 3.4394705043914477,
|
| 938 |
+
"learning_rate": 7.672885706157778e-06,
|
| 939 |
+
"loss": 0.6982,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.39140670982931136,
|
| 944 |
+
"grad_norm": 3.354889681681399,
|
| 945 |
+
"learning_rate": 7.629334338083774e-06,
|
| 946 |
+
"loss": 0.665,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.39434961742201297,
|
| 951 |
+
"grad_norm": 4.968268480091777,
|
| 952 |
+
"learning_rate": 7.585505467588547e-06,
|
| 953 |
+
"loss": 0.6836,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.3972925250147145,
|
| 958 |
+
"grad_norm": 3.4164646148678917,
|
| 959 |
+
"learning_rate": 7.541403720412353e-06,
|
| 960 |
+
"loss": 0.6951,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4002354326074161,
|
| 965 |
+
"grad_norm": 5.683831574907592,
|
| 966 |
+
"learning_rate": 7.497033751095099e-06,
|
| 967 |
+
"loss": 0.6928,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4031783402001177,
|
| 972 |
+
"grad_norm": 4.472151489983979,
|
| 973 |
+
"learning_rate": 7.4524002424851185e-06,
|
| 974 |
+
"loss": 0.6797,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.40612124779281933,
|
| 979 |
+
"grad_norm": 6.186377434937082,
|
| 980 |
+
"learning_rate": 7.407507905244924e-06,
|
| 981 |
+
"loss": 0.672,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4090641553855209,
|
| 986 |
+
"grad_norm": 2.9651290505497045,
|
| 987 |
+
"learning_rate": 7.362361477354052e-06,
|
| 988 |
+
"loss": 0.6791,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4120070629782225,
|
| 993 |
+
"grad_norm": 5.390553555502008,
|
| 994 |
+
"learning_rate": 7.3169657236090004e-06,
|
| 995 |
+
"loss": 0.6689,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4149499705709241,
|
| 1000 |
+
"grad_norm": 4.10914126631982,
|
| 1001 |
+
"learning_rate": 7.271325435120353e-06,
|
| 1002 |
+
"loss": 0.6845,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.41789287816362564,
|
| 1007 |
+
"grad_norm": 8.644264069486146,
|
| 1008 |
+
"learning_rate": 7.225445428807121e-06,
|
| 1009 |
+
"loss": 0.6791,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.42083578575632724,
|
| 1014 |
+
"grad_norm": 4.677705564927546,
|
| 1015 |
+
"learning_rate": 7.179330546888364e-06,
|
| 1016 |
+
"loss": 0.6713,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.42377869334902885,
|
| 1021 |
+
"grad_norm": 6.154537296560613,
|
| 1022 |
+
"learning_rate": 7.132985656372126e-06,
|
| 1023 |
+
"loss": 0.6579,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.42672160094173045,
|
| 1028 |
+
"grad_norm": 8.540942755538525,
|
| 1029 |
+
"learning_rate": 7.086415648541782e-06,
|
| 1030 |
+
"loss": 0.6576,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.429664508534432,
|
| 1035 |
+
"grad_norm": 5.046772189845541,
|
| 1036 |
+
"learning_rate": 7.0396254384398e-06,
|
| 1037 |
+
"loss": 0.6435,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4326074161271336,
|
| 1042 |
+
"grad_norm": 3.4905163179341967,
|
| 1043 |
+
"learning_rate": 6.9926199643489985e-06,
|
| 1044 |
+
"loss": 0.6405,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4355503237198352,
|
| 1049 |
+
"grad_norm": 3.005636207711306,
|
| 1050 |
+
"learning_rate": 6.945404187271365e-06,
|
| 1051 |
+
"loss": 0.6461,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.4384932313125368,
|
| 1056 |
+
"grad_norm": 3.4654929507322803,
|
| 1057 |
+
"learning_rate": 6.897983090404457e-06,
|
| 1058 |
+
"loss": 0.6239,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.44143613890523836,
|
| 1063 |
+
"grad_norm": 3.6846119188733666,
|
| 1064 |
+
"learning_rate": 6.850361678615481e-06,
|
| 1065 |
+
"loss": 0.6548,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.44437904649793997,
|
| 1070 |
+
"grad_norm": 3.2356789677548106,
|
| 1071 |
+
"learning_rate": 6.802544977913067e-06,
|
| 1072 |
+
"loss": 0.6741,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.44732195409064157,
|
| 1077 |
+
"grad_norm": 4.998250903563615,
|
| 1078 |
+
"learning_rate": 6.754538034916819e-06,
|
| 1079 |
+
"loss": 0.6631,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4502648616833431,
|
| 1084 |
+
"grad_norm": 5.140876271152121,
|
| 1085 |
+
"learning_rate": 6.706345916324692e-06,
|
| 1086 |
+
"loss": 0.6458,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.4532077692760447,
|
| 1091 |
+
"grad_norm": 3.486582775324299,
|
| 1092 |
+
"learning_rate": 6.657973708378251e-06,
|
| 1093 |
+
"loss": 0.6468,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.45615067686874633,
|
| 1098 |
+
"grad_norm": 3.2392766779187543,
|
| 1099 |
+
"learning_rate": 6.609426516325859e-06,
|
| 1100 |
+
"loss": 0.6453,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.45909358446144793,
|
| 1105 |
+
"grad_norm": 8.066053393796732,
|
| 1106 |
+
"learning_rate": 6.560709463883865e-06,
|
| 1107 |
+
"loss": 0.629,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4620364920541495,
|
| 1112 |
+
"grad_norm": 4.508373615955478,
|
| 1113 |
+
"learning_rate": 6.511827692695847e-06,
|
| 1114 |
+
"loss": 0.6364,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4649793996468511,
|
| 1119 |
+
"grad_norm": 3.7473795272990755,
|
| 1120 |
+
"learning_rate": 6.462786361789948e-06,
|
| 1121 |
+
"loss": 0.6499,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4679223072395527,
|
| 1126 |
+
"grad_norm": 2.9615724505608543,
|
| 1127 |
+
"learning_rate": 6.413590647034401e-06,
|
| 1128 |
+
"loss": 0.6346,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.47086521483225424,
|
| 1133 |
+
"grad_norm": 3.04892457656011,
|
| 1134 |
+
"learning_rate": 6.364245740591251e-06,
|
| 1135 |
+
"loss": 0.6322,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.47380812242495585,
|
| 1140 |
+
"grad_norm": 2.918336925557892,
|
| 1141 |
+
"learning_rate": 6.314756850368377e-06,
|
| 1142 |
+
"loss": 0.648,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.47675103001765745,
|
| 1147 |
+
"grad_norm": 3.7501594414028516,
|
| 1148 |
+
"learning_rate": 6.2651291994698425e-06,
|
| 1149 |
+
"loss": 0.6372,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.47969393761035906,
|
| 1154 |
+
"grad_norm": 2.6175973358406597,
|
| 1155 |
+
"learning_rate": 6.215368025644637e-06,
|
| 1156 |
+
"loss": 0.6233,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4826368452030606,
|
| 1161 |
+
"grad_norm": 3.216759692623588,
|
| 1162 |
+
"learning_rate": 6.1654785807338925e-06,
|
| 1163 |
+
"loss": 0.6261,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4855797527957622,
|
| 1168 |
+
"grad_norm": 4.2663028236726,
|
| 1169 |
+
"learning_rate": 6.11546613011659e-06,
|
| 1170 |
+
"loss": 0.6402,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.4885226603884638,
|
| 1175 |
+
"grad_norm": 3.0098989981042683,
|
| 1176 |
+
"learning_rate": 6.065335952153846e-06,
|
| 1177 |
+
"loss": 0.6283,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.4914655679811654,
|
| 1182 |
+
"grad_norm": 2.9805289997816997,
|
| 1183 |
+
"learning_rate": 6.015093337631833e-06,
|
| 1184 |
+
"loss": 0.625,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.49440847557386697,
|
| 1189 |
+
"grad_norm": 2.828079466519479,
|
| 1190 |
+
"learning_rate": 5.964743589203388e-06,
|
| 1191 |
+
"loss": 0.6271,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.4973513831665686,
|
| 1196 |
+
"grad_norm": 3.056832170036487,
|
| 1197 |
+
"learning_rate": 5.9142920208283564e-06,
|
| 1198 |
+
"loss": 0.6163,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5002942907592701,
|
| 1203 |
+
"grad_norm": 4.967318743335985,
|
| 1204 |
+
"learning_rate": 5.863743957212759e-06,
|
| 1205 |
+
"loss": 0.6231,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5032371983519718,
|
| 1210 |
+
"grad_norm": 7.691396927048123,
|
| 1211 |
+
"learning_rate": 5.8131047332468106e-06,
|
| 1212 |
+
"loss": 0.6279,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5061801059446733,
|
| 1217 |
+
"grad_norm": 10.485146930542516,
|
| 1218 |
+
"learning_rate": 5.762379693441883e-06,
|
| 1219 |
+
"loss": 0.617,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5091230135373749,
|
| 1224 |
+
"grad_norm": 5.793807677023011,
|
| 1225 |
+
"learning_rate": 5.711574191366427e-06,
|
| 1226 |
+
"loss": 0.6117,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5120659211300765,
|
| 1231 |
+
"grad_norm": 3.75783456933447,
|
| 1232 |
+
"learning_rate": 5.660693589080958e-06,
|
| 1233 |
+
"loss": 0.6083,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5150088287227781,
|
| 1238 |
+
"grad_norm": 4.787442679225073,
|
| 1239 |
+
"learning_rate": 5.60974325657214e-06,
|
| 1240 |
+
"loss": 0.6062,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5179517363154797,
|
| 1245 |
+
"grad_norm": 4.424520409909319,
|
| 1246 |
+
"learning_rate": 5.558728571186027e-06,
|
| 1247 |
+
"loss": 0.6087,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5208946439081813,
|
| 1252 |
+
"grad_norm": 4.60877448542516,
|
| 1253 |
+
"learning_rate": 5.507654917060541e-06,
|
| 1254 |
+
"loss": 0.6144,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5238375515008828,
|
| 1259 |
+
"grad_norm": 5.463543352263803,
|
| 1260 |
+
"learning_rate": 5.456527684557217e-06,
|
| 1261 |
+
"loss": 0.6234,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5267804590935845,
|
| 1266 |
+
"grad_norm": 4.713832072636091,
|
| 1267 |
+
"learning_rate": 5.405352269692301e-06,
|
| 1268 |
+
"loss": 0.6036,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5297233666862861,
|
| 1273 |
+
"grad_norm": 3.074792807909731,
|
| 1274 |
+
"learning_rate": 5.354134073567255e-06,
|
| 1275 |
+
"loss": 0.5953,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5326662742789876,
|
| 1280 |
+
"grad_norm": 2.624507780176144,
|
| 1281 |
+
"learning_rate": 5.302878501798714e-06,
|
| 1282 |
+
"loss": 0.6024,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5356091818716893,
|
| 1287 |
+
"grad_norm": 2.4586044761464505,
|
| 1288 |
+
"learning_rate": 5.2515909639479715e-06,
|
| 1289 |
+
"loss": 0.583,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5385520894643908,
|
| 1294 |
+
"grad_norm": 3.1643013838767318,
|
| 1295 |
+
"learning_rate": 5.200276872950052e-06,
|
| 1296 |
+
"loss": 0.5848,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5414949970570924,
|
| 1301 |
+
"grad_norm": 5.524616335650943,
|
| 1302 |
+
"learning_rate": 5.148941644542428e-06,
|
| 1303 |
+
"loss": 0.6066,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.544437904649794,
|
| 1308 |
+
"grad_norm": 2.9626617937595383,
|
| 1309 |
+
"learning_rate": 5.09759069669343e-06,
|
| 1310 |
+
"loss": 0.5868,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5473808122424956,
|
| 1315 |
+
"grad_norm": 4.179026591658555,
|
| 1316 |
+
"learning_rate": 5.046229449030432e-06,
|
| 1317 |
+
"loss": 0.5938,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5503237198351971,
|
| 1322 |
+
"grad_norm": 2.9736230937939436,
|
| 1323 |
+
"learning_rate": 4.994863322267865e-06,
|
| 1324 |
+
"loss": 0.5854,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5532666274278988,
|
| 1329 |
+
"grad_norm": 2.6674395025105384,
|
| 1330 |
+
"learning_rate": 4.943497737635103e-06,
|
| 1331 |
+
"loss": 0.5773,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5562095350206003,
|
| 1336 |
+
"grad_norm": 3.3999264932445965,
|
| 1337 |
+
"learning_rate": 4.892138116304297e-06,
|
| 1338 |
+
"loss": 0.5899,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.559152442613302,
|
| 1343 |
+
"grad_norm": 3.7946695534722954,
|
| 1344 |
+
"learning_rate": 4.840789878818238e-06,
|
| 1345 |
+
"loss": 0.5909,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5620953502060035,
|
| 1350 |
+
"grad_norm": 2.9170374244463777,
|
| 1351 |
+
"learning_rate": 4.789458444518243e-06,
|
| 1352 |
+
"loss": 0.585,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5650382577987051,
|
| 1357 |
+
"grad_norm": 3.804085500882333,
|
| 1358 |
+
"learning_rate": 4.73814923097221e-06,
|
| 1359 |
+
"loss": 0.5811,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5679811653914068,
|
| 1364 |
+
"grad_norm": 2.627248184612479,
|
| 1365 |
+
"learning_rate": 4.6868676534028415e-06,
|
| 1366 |
+
"loss": 0.5741,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5709240729841083,
|
| 1371 |
+
"grad_norm": 2.758710046147963,
|
| 1372 |
+
"learning_rate": 4.635619124116105e-06,
|
| 1373 |
+
"loss": 0.5751,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5738669805768098,
|
| 1378 |
+
"grad_norm": 2.533940367785718,
|
| 1379 |
+
"learning_rate": 4.584409051930027e-06,
|
| 1380 |
+
"loss": 0.5727,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5768098881695115,
|
| 1385 |
+
"grad_norm": 4.7034871762477755,
|
| 1386 |
+
"learning_rate": 4.533242841603836e-06,
|
| 1387 |
+
"loss": 0.5958,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5797527957622131,
|
| 1392 |
+
"grad_norm": 3.9130158274657205,
|
| 1393 |
+
"learning_rate": 4.4821258932675324e-06,
|
| 1394 |
+
"loss": 0.5752,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5826957033549146,
|
| 1399 |
+
"grad_norm": 2.923581564704699,
|
| 1400 |
+
"learning_rate": 4.431063601851965e-06,
|
| 1401 |
+
"loss": 0.5855,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5856386109476163,
|
| 1406 |
+
"grad_norm": 2.68100418754764,
|
| 1407 |
+
"learning_rate": 4.38006135651944e-06,
|
| 1408 |
+
"loss": 0.5592,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5885815185403178,
|
| 1413 |
+
"grad_norm": 2.5427064423534054,
|
| 1414 |
+
"learning_rate": 4.329124540094939e-06,
|
| 1415 |
+
"loss": 0.5738,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.5915244261330195,
|
| 1420 |
+
"grad_norm": 2.56361328953547,
|
| 1421 |
+
"learning_rate": 4.278258528498023e-06,
|
| 1422 |
+
"loss": 0.5518,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.594467333725721,
|
| 1427 |
+
"grad_norm": 3.570826652815698,
|
| 1428 |
+
"learning_rate": 4.227468690175433e-06,
|
| 1429 |
+
"loss": 0.5565,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.5974102413184226,
|
| 1434 |
+
"grad_norm": 3.144931194434484,
|
| 1435 |
+
"learning_rate": 4.17676038553452e-06,
|
| 1436 |
+
"loss": 0.5542,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6003531489111242,
|
| 1441 |
+
"grad_norm": 3.0428538279770487,
|
| 1442 |
+
"learning_rate": 4.126138966377495e-06,
|
| 1443 |
+
"loss": 0.5724,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6032960565038258,
|
| 1448 |
+
"grad_norm": 2.9781764362393863,
|
| 1449 |
+
"learning_rate": 4.075609775336585e-06,
|
| 1450 |
+
"loss": 0.5651,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6062389640965273,
|
| 1455 |
+
"grad_norm": 2.8792302508241896,
|
| 1456 |
+
"learning_rate": 4.02517814531018e-06,
|
| 1457 |
+
"loss": 0.5538,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.609181871689229,
|
| 1462 |
+
"grad_norm": 2.852533794505324,
|
| 1463 |
+
"learning_rate": 3.974849398899991e-06,
|
| 1464 |
+
"loss": 0.5488,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6121247792819305,
|
| 1469 |
+
"grad_norm": 3.645969861195433,
|
| 1470 |
+
"learning_rate": 3.924628847849287e-06,
|
| 1471 |
+
"loss": 0.5534,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6150676868746321,
|
| 1476 |
+
"grad_norm": 2.4948233282423002,
|
| 1477 |
+
"learning_rate": 3.874521792482305e-06,
|
| 1478 |
+
"loss": 0.5661,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6180105944673338,
|
| 1483 |
+
"grad_norm": 2.6660019183479577,
|
| 1484 |
+
"learning_rate": 3.8245335211448404e-06,
|
| 1485 |
+
"loss": 0.5601,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6209535020600353,
|
| 1490 |
+
"grad_norm": 3.7571638678391346,
|
| 1491 |
+
"learning_rate": 3.7746693096461027e-06,
|
| 1492 |
+
"loss": 0.5522,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6238964096527368,
|
| 1497 |
+
"grad_norm": 2.891024598849757,
|
| 1498 |
+
"learning_rate": 3.7249344207019163e-06,
|
| 1499 |
+
"loss": 0.5384,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6268393172454385,
|
| 1504 |
+
"grad_norm": 3.117981111832482,
|
| 1505 |
+
"learning_rate": 3.675334103379282e-06,
|
| 1506 |
+
"loss": 0.5297,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6297822248381401,
|
| 1511 |
+
"grad_norm": 2.471751054844093,
|
| 1512 |
+
"learning_rate": 3.625873592542377e-06,
|
| 1513 |
+
"loss": 0.5488,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6327251324308417,
|
| 1518 |
+
"grad_norm": 2.9406065713758545,
|
| 1519 |
+
"learning_rate": 3.576558108300081e-06,
|
| 1520 |
+
"loss": 0.5406,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6356680400235433,
|
| 1525 |
+
"grad_norm": 2.5304758647440466,
|
| 1526 |
+
"learning_rate": 3.5273928554550218e-06,
|
| 1527 |
+
"loss": 0.5377,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6386109476162448,
|
| 1532 |
+
"grad_norm": 4.358193416602383,
|
| 1533 |
+
"learning_rate": 3.478383022954267e-06,
|
| 1534 |
+
"loss": 0.5262,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6415538552089465,
|
| 1539 |
+
"grad_norm": 3.751043254388853,
|
| 1540 |
+
"learning_rate": 3.4295337833416797e-06,
|
| 1541 |
+
"loss": 0.5347,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.644496762801648,
|
| 1546 |
+
"grad_norm": 2.8478083940715058,
|
| 1547 |
+
"learning_rate": 3.3808502922119884e-06,
|
| 1548 |
+
"loss": 0.5517,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6474396703943496,
|
| 1553 |
+
"grad_norm": 3.592053051641115,
|
| 1554 |
+
"learning_rate": 3.3323376876666787e-06,
|
| 1555 |
+
"loss": 0.5185,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6503825779870512,
|
| 1560 |
+
"grad_norm": 3.2832783558616523,
|
| 1561 |
+
"learning_rate": 3.2840010897717045e-06,
|
| 1562 |
+
"loss": 0.5283,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6533254855797528,
|
| 1567 |
+
"grad_norm": 3.7958128452191313,
|
| 1568 |
+
"learning_rate": 3.235845600017107e-06,
|
| 1569 |
+
"loss": 0.5304,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6562683931724543,
|
| 1574 |
+
"grad_norm": 2.458980007684118,
|
| 1575 |
+
"learning_rate": 3.1878763007786075e-06,
|
| 1576 |
+
"loss": 0.5119,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.659211300765156,
|
| 1581 |
+
"grad_norm": 5.356928896927627,
|
| 1582 |
+
"learning_rate": 3.1400982547812088e-06,
|
| 1583 |
+
"loss": 0.5356,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6621542083578575,
|
| 1588 |
+
"grad_norm": 4.006321789010636,
|
| 1589 |
+
"learning_rate": 3.0925165045648564e-06,
|
| 1590 |
+
"loss": 0.5144,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6650971159505592,
|
| 1595 |
+
"grad_norm": 3.1366794171510164,
|
| 1596 |
+
"learning_rate": 3.0451360719522653e-06,
|
| 1597 |
+
"loss": 0.5259,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6680400235432608,
|
| 1602 |
+
"grad_norm": 3.075050005395039,
|
| 1603 |
+
"learning_rate": 2.997961957518892e-06,
|
| 1604 |
+
"loss": 0.5126,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6709829311359623,
|
| 1609 |
+
"grad_norm": 2.614210901095361,
|
| 1610 |
+
"learning_rate": 2.950999140065184e-06,
|
| 1611 |
+
"loss": 0.526,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.673925838728664,
|
| 1616 |
+
"grad_norm": 2.668099567229874,
|
| 1617 |
+
"learning_rate": 2.904252576091105e-06,
|
| 1618 |
+
"loss": 0.5207,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6768687463213655,
|
| 1623 |
+
"grad_norm": 2.822258046228147,
|
| 1624 |
+
"learning_rate": 2.8577271992730223e-06,
|
| 1625 |
+
"loss": 0.5179,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6798116539140671,
|
| 1630 |
+
"grad_norm": 2.7498849578418842,
|
| 1631 |
+
"learning_rate": 2.8114279199430027e-06,
|
| 1632 |
+
"loss": 0.5208,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6827545615067687,
|
| 1637 |
+
"grad_norm": 3.7794257172644086,
|
| 1638 |
+
"learning_rate": 2.765359624570574e-06,
|
| 1639 |
+
"loss": 0.5156,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6856974690994703,
|
| 1644 |
+
"grad_norm": 2.9809135197781527,
|
| 1645 |
+
"learning_rate": 2.719527175246993e-06,
|
| 1646 |
+
"loss": 0.5233,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.6886403766921718,
|
| 1651 |
+
"grad_norm": 2.600094892048898,
|
| 1652 |
+
"learning_rate": 2.6739354091721094e-06,
|
| 1653 |
+
"loss": 0.5223,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.6915832842848735,
|
| 1658 |
+
"grad_norm": 3.392788220792135,
|
| 1659 |
+
"learning_rate": 2.6285891381438346e-06,
|
| 1660 |
+
"loss": 0.5343,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.694526191877575,
|
| 1665 |
+
"grad_norm": 2.620537806287387,
|
| 1666 |
+
"learning_rate": 2.583493148050297e-06,
|
| 1667 |
+
"loss": 0.5113,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.6974690994702767,
|
| 1672 |
+
"grad_norm": 2.4542848921872817,
|
| 1673 |
+
"learning_rate": 2.5386521983647468e-06,
|
| 1674 |
+
"loss": 0.511,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7004120070629782,
|
| 1679 |
+
"grad_norm": 2.5882709829545556,
|
| 1680 |
+
"learning_rate": 2.494071021643222e-06,
|
| 1681 |
+
"loss": 0.5061,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7033549146556798,
|
| 1686 |
+
"grad_norm": 2.2977723019021306,
|
| 1687 |
+
"learning_rate": 2.4497543230250843e-06,
|
| 1688 |
+
"loss": 0.5041,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7062978222483814,
|
| 1693 |
+
"grad_norm": 2.706988899429715,
|
| 1694 |
+
"learning_rate": 2.4057067797364247e-06,
|
| 1695 |
+
"loss": 0.5034,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.709240729841083,
|
| 1700 |
+
"grad_norm": 2.8986602480661716,
|
| 1701 |
+
"learning_rate": 2.361933040596424e-06,
|
| 1702 |
+
"loss": 0.5081,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7121836374337845,
|
| 1707 |
+
"grad_norm": 3.069283780524911,
|
| 1708 |
+
"learning_rate": 2.31843772552672e-06,
|
| 1709 |
+
"loss": 0.4812,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7151265450264862,
|
| 1714 |
+
"grad_norm": 2.942823198350691,
|
| 1715 |
+
"learning_rate": 2.275225425063813e-06,
|
| 1716 |
+
"loss": 0.4862,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7180694526191878,
|
| 1721 |
+
"grad_norm": 4.0835750894481,
|
| 1722 |
+
"learning_rate": 2.2323006998745673e-06,
|
| 1723 |
+
"loss": 0.5002,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7210123602118893,
|
| 1728 |
+
"grad_norm": 2.6814535983918764,
|
| 1729 |
+
"learning_rate": 2.189668080274891e-06,
|
| 1730 |
+
"loss": 0.4844,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.723955267804591,
|
| 1735 |
+
"grad_norm": 2.476076418297489,
|
| 1736 |
+
"learning_rate": 2.1473320657515913e-06,
|
| 1737 |
+
"loss": 0.4882,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7268981753972925,
|
| 1742 |
+
"grad_norm": 2.9506625115965073,
|
| 1743 |
+
"learning_rate": 2.1052971244874905e-06,
|
| 1744 |
+
"loss": 0.496,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7298410829899941,
|
| 1749 |
+
"grad_norm": 2.4718878265701214,
|
| 1750 |
+
"learning_rate": 2.063567692889864e-06,
|
| 1751 |
+
"loss": 0.496,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7327839905826957,
|
| 1756 |
+
"grad_norm": 2.5394148181117724,
|
| 1757 |
+
"learning_rate": 2.0221481751222027e-06,
|
| 1758 |
+
"loss": 0.4918,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7357268981753973,
|
| 1763 |
+
"grad_norm": 2.4239900265691436,
|
| 1764 |
+
"learning_rate": 1.981042942639404e-06,
|
| 1765 |
+
"loss": 0.4825,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7386698057680989,
|
| 1770 |
+
"grad_norm": 2.582847306611647,
|
| 1771 |
+
"learning_rate": 1.9402563337264012e-06,
|
| 1772 |
+
"loss": 0.479,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7416127133608005,
|
| 1777 |
+
"grad_norm": 2.8447951582688344,
|
| 1778 |
+
"learning_rate": 1.899792653040291e-06,
|
| 1779 |
+
"loss": 0.4856,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.744555620953502,
|
| 1784 |
+
"grad_norm": 2.7091883817883504,
|
| 1785 |
+
"learning_rate": 1.8596561711560267e-06,
|
| 1786 |
+
"loss": 0.4785,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7474985285462037,
|
| 1791 |
+
"grad_norm": 2.5437072619561443,
|
| 1792 |
+
"learning_rate": 1.8198511241156902e-06,
|
| 1793 |
+
"loss": 0.4846,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7504414361389052,
|
| 1798 |
+
"grad_norm": 3.156459227921938,
|
| 1799 |
+
"learning_rate": 1.780381712981415e-06,
|
| 1800 |
+
"loss": 0.4765,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7533843437316068,
|
| 1805 |
+
"grad_norm": 2.606788537147239,
|
| 1806 |
+
"learning_rate": 1.7412521033920087e-06,
|
| 1807 |
+
"loss": 0.4945,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7563272513243084,
|
| 1812 |
+
"grad_norm": 2.7549835249891057,
|
| 1813 |
+
"learning_rate": 1.7024664251233052e-06,
|
| 1814 |
+
"loss": 0.4795,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.75927015891701,
|
| 1819 |
+
"grad_norm": 2.499144800895581,
|
| 1820 |
+
"learning_rate": 1.6640287716522975e-06,
|
| 1821 |
+
"loss": 0.4691,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7622130665097115,
|
| 1826 |
+
"grad_norm": 2.4799314569355677,
|
| 1827 |
+
"learning_rate": 1.6259431997251191e-06,
|
| 1828 |
+
"loss": 0.4972,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7651559741024132,
|
| 1833 |
+
"grad_norm": 2.5207515424299034,
|
| 1834 |
+
"learning_rate": 1.5882137289288812e-06,
|
| 1835 |
+
"loss": 0.4734,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7680988816951148,
|
| 1840 |
+
"grad_norm": 2.383436407772927,
|
| 1841 |
+
"learning_rate": 1.5508443412674518e-06,
|
| 1842 |
+
"loss": 0.4944,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7710417892878164,
|
| 1847 |
+
"grad_norm": 2.7793788321817656,
|
| 1848 |
+
"learning_rate": 1.513838980741183e-06,
|
| 1849 |
+
"loss": 0.4749,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.773984696880518,
|
| 1854 |
+
"grad_norm": 2.301895560667882,
|
| 1855 |
+
"learning_rate": 1.4772015529306587e-06,
|
| 1856 |
+
"loss": 0.4619,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7769276044732195,
|
| 1861 |
+
"grad_norm": 2.446229173062197,
|
| 1862 |
+
"learning_rate": 1.4409359245845005e-06,
|
| 1863 |
+
"loss": 0.4671,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7798705120659212,
|
| 1868 |
+
"grad_norm": 2.998324656500335,
|
| 1869 |
+
"learning_rate": 1.4050459232112652e-06,
|
| 1870 |
+
"loss": 0.4708,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7828134196586227,
|
| 1875 |
+
"grad_norm": 2.287094207368545,
|
| 1876 |
+
"learning_rate": 1.3695353366754798e-06,
|
| 1877 |
+
"loss": 0.4666,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7857563272513243,
|
| 1882 |
+
"grad_norm": 2.652209930668726,
|
| 1883 |
+
"learning_rate": 1.3344079127978788e-06,
|
| 1884 |
+
"loss": 0.458,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.7886992348440259,
|
| 1889 |
+
"grad_norm": 2.6373384490448872,
|
| 1890 |
+
"learning_rate": 1.299667358959847e-06,
|
| 1891 |
+
"loss": 0.4802,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.7916421424367275,
|
| 1896 |
+
"grad_norm": 2.6241240764044034,
|
| 1897 |
+
"learning_rate": 1.2653173417121367e-06,
|
| 1898 |
+
"loss": 0.458,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.794585050029429,
|
| 1903 |
+
"grad_norm": 2.596475069136379,
|
| 1904 |
+
"learning_rate": 1.2313614863879074e-06,
|
| 1905 |
+
"loss": 0.4589,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7975279576221307,
|
| 1910 |
+
"grad_norm": 2.6042310296871856,
|
| 1911 |
+
"learning_rate": 1.1978033767200936e-06,
|
| 1912 |
+
"loss": 0.4613,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8004708652148322,
|
| 1917 |
+
"grad_norm": 2.4907632128373614,
|
| 1918 |
+
"learning_rate": 1.1646465544631802e-06,
|
| 1919 |
+
"loss": 0.4672,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8034137728075338,
|
| 1924 |
+
"grad_norm": 2.8404270982027624,
|
| 1925 |
+
"learning_rate": 1.131894519019403e-06,
|
| 1926 |
+
"loss": 0.4647,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8063566804002354,
|
| 1931 |
+
"grad_norm": 2.511899627431363,
|
| 1932 |
+
"learning_rate": 1.0995507270694121e-06,
|
| 1933 |
+
"loss": 0.4568,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.809299587992937,
|
| 1938 |
+
"grad_norm": 2.49013220750732,
|
| 1939 |
+
"learning_rate": 1.0676185922074583e-06,
|
| 1940 |
+
"loss": 0.4657,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8122424955856387,
|
| 1945 |
+
"grad_norm": 2.4156192880535747,
|
| 1946 |
+
"learning_rate": 1.036101484581117e-06,
|
| 1947 |
+
"loss": 0.4616,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8151854031783402,
|
| 1952 |
+
"grad_norm": 2.4647230171771364,
|
| 1953 |
+
"learning_rate": 1.0050027305355926e-06,
|
| 1954 |
+
"loss": 0.4651,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8181283107710418,
|
| 1959 |
+
"grad_norm": 2.90805484751002,
|
| 1960 |
+
"learning_rate": 9.743256122626644e-07,
|
| 1961 |
+
"loss": 0.4582,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8210712183637434,
|
| 1966 |
+
"grad_norm": 2.770355546856387,
|
| 1967 |
+
"learning_rate": 9.440733674542751e-07,
|
| 1968 |
+
"loss": 0.47,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.824014125956445,
|
| 1973 |
+
"grad_norm": 2.894362207485248,
|
| 1974 |
+
"learning_rate": 9.142491889608179e-07,
|
| 1975 |
+
"loss": 0.4525,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8269570335491465,
|
| 1980 |
+
"grad_norm": 2.404589039415564,
|
| 1981 |
+
"learning_rate": 8.848562244541675e-07,
|
| 1982 |
+
"loss": 0.4682,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8298999411418482,
|
| 1987 |
+
"grad_norm": 2.492758443409389,
|
| 1988 |
+
"learning_rate": 8.558975760954685e-07,
|
| 1989 |
+
"loss": 0.4595,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8328428487345497,
|
| 1994 |
+
"grad_norm": 3.0988070143588202,
|
| 1995 |
+
"learning_rate": 8.273763002077284e-07,
|
| 1996 |
+
"loss": 0.4568,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8357857563272513,
|
| 2001 |
+
"grad_norm": 3.440047173962592,
|
| 2002 |
+
"learning_rate": 7.992954069532543e-07,
|
| 2003 |
+
"loss": 0.461,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8387286639199529,
|
| 2008 |
+
"grad_norm": 2.657598459534566,
|
| 2009 |
+
"learning_rate": 7.716578600159535e-07,
|
| 2010 |
+
"loss": 0.4503,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8416715715126545,
|
| 2015 |
+
"grad_norm": 2.572563487365978,
|
| 2016 |
+
"learning_rate": 7.444665762885484e-07,
|
| 2017 |
+
"loss": 0.4615,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8446144791053561,
|
| 2022 |
+
"grad_norm": 2.439187441026128,
|
| 2023 |
+
"learning_rate": 7.177244255647209e-07,
|
| 2024 |
+
"loss": 0.4446,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8475573866980577,
|
| 2029 |
+
"grad_norm": 2.233969483468557,
|
| 2030 |
+
"learning_rate": 6.914342302362309e-07,
|
| 2031 |
+
"loss": 0.4511,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8505002942907592,
|
| 2036 |
+
"grad_norm": 2.471799044846796,
|
| 2037 |
+
"learning_rate": 6.65598764995043e-07,
|
| 2038 |
+
"loss": 0.4497,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8534432018834609,
|
| 2043 |
+
"grad_norm": 2.755442497972038,
|
| 2044 |
+
"learning_rate": 6.402207565404811e-07,
|
| 2045 |
+
"loss": 0.4487,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8563861094761624,
|
| 2050 |
+
"grad_norm": 2.4539325525236326,
|
| 2051 |
+
"learning_rate": 6.153028832914453e-07,
|
| 2052 |
+
"loss": 0.4541,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.859329017068864,
|
| 2057 |
+
"grad_norm": 2.2940758992178014,
|
| 2058 |
+
"learning_rate": 5.908477751037361e-07,
|
| 2059 |
+
"loss": 0.4425,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8622719246615657,
|
| 2064 |
+
"grad_norm": 2.5143891555656435,
|
| 2065 |
+
"learning_rate": 5.668580129924949e-07,
|
| 2066 |
+
"loss": 0.4475,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8652148322542672,
|
| 2071 |
+
"grad_norm": 2.643528212719313,
|
| 2072 |
+
"learning_rate": 5.433361288597949e-07,
|
| 2073 |
+
"loss": 0.4374,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8681577398469688,
|
| 2078 |
+
"grad_norm": 2.3103830994227774,
|
| 2079 |
+
"learning_rate": 5.20284605227433e-07,
|
| 2080 |
+
"loss": 0.4434,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8711006474396704,
|
| 2085 |
+
"grad_norm": 2.1890313025355668,
|
| 2086 |
+
"learning_rate": 4.977058749749115e-07,
|
| 2087 |
+
"loss": 0.4464,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.874043555032372,
|
| 2092 |
+
"grad_norm": 2.311388455333076,
|
| 2093 |
+
"learning_rate": 4.7560232108267714e-07,
|
| 2094 |
+
"loss": 0.4501,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8769864626250736,
|
| 2099 |
+
"grad_norm": 2.2768695736500817,
|
| 2100 |
+
"learning_rate": 4.5397627638061604e-07,
|
| 2101 |
+
"loss": 0.4519,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8799293702177752,
|
| 2106 |
+
"grad_norm": 2.4901928210573305,
|
| 2107 |
+
"learning_rate": 4.3283002330184233e-07,
|
| 2108 |
+
"loss": 0.4426,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8828722778104767,
|
| 2113 |
+
"grad_norm": 2.2512157372682213,
|
| 2114 |
+
"learning_rate": 4.1216579364181333e-07,
|
| 2115 |
+
"loss": 0.4465,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.8858151854031784,
|
| 2120 |
+
"grad_norm": 2.2396613255090947,
|
| 2121 |
+
"learning_rate": 3.919857683227818e-07,
|
| 2122 |
+
"loss": 0.447,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.8887580929958799,
|
| 2127 |
+
"grad_norm": 2.4441734845394096,
|
| 2128 |
+
"learning_rate": 3.72292077163614e-07,
|
| 2129 |
+
"loss": 0.4455,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8917010005885815,
|
| 2134 |
+
"grad_norm": 2.3390391914462128,
|
| 2135 |
+
"learning_rate": 3.5308679865501516e-07,
|
| 2136 |
+
"loss": 0.4329,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.8946439081812831,
|
| 2141 |
+
"grad_norm": 2.467951011914622,
|
| 2142 |
+
"learning_rate": 3.343719597401596e-07,
|
| 2143 |
+
"loss": 0.4373,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.8975868157739847,
|
| 2148 |
+
"grad_norm": 2.506469216208831,
|
| 2149 |
+
"learning_rate": 3.1614953560076213e-07,
|
| 2150 |
+
"loss": 0.4413,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9005297233666862,
|
| 2155 |
+
"grad_norm": 2.3383168243956174,
|
| 2156 |
+
"learning_rate": 2.984214494486215e-07,
|
| 2157 |
+
"loss": 0.4548,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9034726309593879,
|
| 2162 |
+
"grad_norm": 2.630425109995007,
|
| 2163 |
+
"learning_rate": 2.811895723226382e-07,
|
| 2164 |
+
"loss": 0.4524,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9064155385520895,
|
| 2169 |
+
"grad_norm": 2.4286937188874775,
|
| 2170 |
+
"learning_rate": 2.6445572289134634e-07,
|
| 2171 |
+
"loss": 0.4376,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.909358446144791,
|
| 2176 |
+
"grad_norm": 2.216575869780935,
|
| 2177 |
+
"learning_rate": 2.482216672609677e-07,
|
| 2178 |
+
"loss": 0.4351,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9123013537374927,
|
| 2183 |
+
"grad_norm": 2.3654422420764973,
|
| 2184 |
+
"learning_rate": 2.3248911878901547e-07,
|
| 2185 |
+
"loss": 0.4355,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9152442613301942,
|
| 2190 |
+
"grad_norm": 2.3549052063334495,
|
| 2191 |
+
"learning_rate": 2.1725973790346522e-07,
|
| 2192 |
+
"loss": 0.4346,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9181871689228959,
|
| 2197 |
+
"grad_norm": 2.3113103859994535,
|
| 2198 |
+
"learning_rate": 2.0253513192751374e-07,
|
| 2199 |
+
"loss": 0.4367,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9211300765155974,
|
| 2204 |
+
"grad_norm": 2.7576756319444553,
|
| 2205 |
+
"learning_rate": 1.8831685490993457e-07,
|
| 2206 |
+
"loss": 0.4381,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.924072984108299,
|
| 2211 |
+
"grad_norm": 2.369997826369225,
|
| 2212 |
+
"learning_rate": 1.7460640746107094e-07,
|
| 2213 |
+
"loss": 0.4454,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9270158917010006,
|
| 2218 |
+
"grad_norm": 2.6393554236756,
|
| 2219 |
+
"learning_rate": 1.614052365944524e-07,
|
| 2220 |
+
"loss": 0.4481,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9299587992937022,
|
| 2225 |
+
"grad_norm": 2.4054005953759128,
|
| 2226 |
+
"learning_rate": 1.4871473557407924e-07,
|
| 2227 |
+
"loss": 0.4412,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9329017068864037,
|
| 2232 |
+
"grad_norm": 2.491831338754737,
|
| 2233 |
+
"learning_rate": 1.3653624376737562e-07,
|
| 2234 |
+
"loss": 0.4451,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9358446144791054,
|
| 2239 |
+
"grad_norm": 3.0815694783925305,
|
| 2240 |
+
"learning_rate": 1.2487104650383165e-07,
|
| 2241 |
+
"loss": 0.4283,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9387875220718069,
|
| 2246 |
+
"grad_norm": 2.2641401566584545,
|
| 2247 |
+
"learning_rate": 1.1372037493934785e-07,
|
| 2248 |
+
"loss": 0.4237,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9417304296645085,
|
| 2253 |
+
"grad_norm": 2.7762443739962057,
|
| 2254 |
+
"learning_rate": 1.0308540592629756e-07,
|
| 2255 |
+
"loss": 0.4404,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9446733372572101,
|
| 2260 |
+
"grad_norm": 2.6178876501865145,
|
| 2261 |
+
"learning_rate": 9.296726188932293e-08,
|
| 2262 |
+
"loss": 0.4306,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9476162448499117,
|
| 2267 |
+
"grad_norm": 2.6239109904141604,
|
| 2268 |
+
"learning_rate": 8.336701070686904e-08,
|
| 2269 |
+
"loss": 0.4461,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9505591524426134,
|
| 2274 |
+
"grad_norm": 2.2437383637665715,
|
| 2275 |
+
"learning_rate": 7.428566559848416e-08,
|
| 2276 |
+
"loss": 0.4345,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9535020600353149,
|
| 2281 |
+
"grad_norm": 2.1833908010974605,
|
| 2282 |
+
"learning_rate": 6.572418501788025e-08,
|
| 2283 |
+
"loss": 0.4359,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9564449676280165,
|
| 2288 |
+
"grad_norm": 2.3432337307633215,
|
| 2289 |
+
"learning_rate": 5.768347255177609e-08,
|
| 2290 |
+
"loss": 0.4169,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9593878752207181,
|
| 2295 |
+
"grad_norm": 2.809678686382748,
|
| 2296 |
+
"learning_rate": 5.0164376824535235e-08,
|
| 2297 |
+
"loss": 0.4353,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9623307828134197,
|
| 2302 |
+
"grad_norm": 2.36827229347058,
|
| 2303 |
+
"learning_rate": 4.3167691408599334e-08,
|
| 2304 |
+
"loss": 0.4284,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9652736904061212,
|
| 2309 |
+
"grad_norm": 2.29034432840639,
|
| 2310 |
+
"learning_rate": 3.6694154740733456e-08,
|
| 2311 |
+
"loss": 0.4443,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9682165979988229,
|
| 2316 |
+
"grad_norm": 2.313570657915266,
|
| 2317 |
+
"learning_rate": 3.074445004409121e-08,
|
| 2318 |
+
"loss": 0.4382,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9711595055915244,
|
| 2323 |
+
"grad_norm": 2.3743788836695776,
|
| 2324 |
+
"learning_rate": 2.5319205256107425e-08,
|
| 2325 |
+
"loss": 0.418,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.974102413184226,
|
| 2330 |
+
"grad_norm": 2.3747057502664854,
|
| 2331 |
+
"learning_rate": 2.0418992962224495e-08,
|
| 2332 |
+
"loss": 0.4463,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9770453207769276,
|
| 2337 |
+
"grad_norm": 2.2646005029150618,
|
| 2338 |
+
"learning_rate": 1.6044330335460712e-08,
|
| 2339 |
+
"loss": 0.4517,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.9799882283696292,
|
| 2344 |
+
"grad_norm": 2.3395141439258302,
|
| 2345 |
+
"learning_rate": 1.2195679081827283e-08,
|
| 2346 |
+
"loss": 0.447,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.9829311359623308,
|
| 2351 |
+
"grad_norm": 2.7310828433799377,
|
| 2352 |
+
"learning_rate": 8.873445391601176e-09,
|
| 2353 |
+
"loss": 0.4445,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9858740435550324,
|
| 2358 |
+
"grad_norm": 2.3333374487486203,
|
| 2359 |
+
"learning_rate": 6.077979896453312e-09,
|
| 2360 |
+
"loss": 0.4354,
|
| 2361 |
+
"step": 1675
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.9888169511477339,
|
| 2365 |
+
"grad_norm": 2.3497335141047513,
|
| 2366 |
+
"learning_rate": 3.809577632443717e-09,
|
| 2367 |
+
"loss": 0.4321,
|
| 2368 |
+
"step": 1680
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.9917598587404356,
|
| 2372 |
+
"grad_norm": 2.3914591270933983,
|
| 2373 |
+
"learning_rate": 2.0684780088825463e-09,
|
| 2374 |
+
"loss": 0.4372,
|
| 2375 |
+
"step": 1685
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.9947027663331371,
|
| 2379 |
+
"grad_norm": 2.3017358226327653,
|
| 2380 |
+
"learning_rate": 8.548647830630696e-10,
|
| 2381 |
+
"loss": 0.4339,
|
| 2382 |
+
"step": 1690
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.9976456739258387,
|
| 2386 |
+
"grad_norm": 2.621035259676665,
|
| 2387 |
+
"learning_rate": 1.6886604086774337e-10,
|
| 2388 |
+
"loss": 0.4319,
|
| 2389 |
+
"step": 1695
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 1.0,
|
| 2393 |
+
"eval_runtime": 4.2193,
|
| 2394 |
+
"eval_samples_per_second": 2.37,
|
| 2395 |
+
"eval_steps_per_second": 0.711,
|
| 2396 |
+
"step": 1699
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 1.0,
|
| 2400 |
+
"step": 1699,
|
| 2401 |
+
"total_flos": 116982426894336.0,
|
| 2402 |
+
"train_loss": 0.6940072515278861,
|
| 2403 |
+
"train_runtime": 20640.2326,
|
| 2404 |
+
"train_samples_per_second": 1.317,
|
| 2405 |
+
"train_steps_per_second": 0.082
|
| 2406 |
+
}
|
| 2407 |
+
],
|
| 2408 |
+
"logging_steps": 5,
|
| 2409 |
+
"max_steps": 1699,
|
| 2410 |
+
"num_input_tokens_seen": 0,
|
| 2411 |
+
"num_train_epochs": 1,
|
| 2412 |
+
"save_steps": 100,
|
| 2413 |
+
"stateful_callbacks": {
|
| 2414 |
+
"TrainerControl": {
|
| 2415 |
+
"args": {
|
| 2416 |
+
"should_epoch_stop": false,
|
| 2417 |
+
"should_evaluate": false,
|
| 2418 |
+
"should_log": false,
|
| 2419 |
+
"should_save": true,
|
| 2420 |
+
"should_training_stop": true
|
| 2421 |
+
},
|
| 2422 |
+
"attributes": {}
|
| 2423 |
+
}
|
| 2424 |
+
},
|
| 2425 |
+
"total_flos": 116982426894336.0,
|
| 2426 |
+
"train_batch_size": 4,
|
| 2427 |
+
"trial_name": null,
|
| 2428 |
+
"trial_params": null
|
| 2429 |
+
}
|