Model save
Browse files- README.md +60 -0
- all_results.json +9 -0
- generation_config.json +9 -0
- runs/Sep29_17-51-15_COE-CS-sv003/events.out.tfevents.1727632736.COE-CS-sv003.1214200.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2373 -0
README.md
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3
|
| 4 |
+
base_model: meta-llama/Meta-Llama-3-8B
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: llama3_feedback
|
| 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 |
+
# llama3_feedback
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 1e-05
|
| 39 |
+
- train_batch_size: 4
|
| 40 |
+
- eval_batch_size: 1
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 4
|
| 44 |
+
- total_train_batch_size: 16
|
| 45 |
+
- total_eval_batch_size: 4
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
### Framework versions
|
| 56 |
+
|
| 57 |
+
- Transformers 4.45.1
|
| 58 |
+
- Pytorch 2.4.1+cu121
|
| 59 |
+
- Datasets 3.0.1
|
| 60 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 173366354903040.0,
|
| 4 |
+
"train_loss": 0.6223726071021407,
|
| 5 |
+
"train_runtime": 17321.1955,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.529,
|
| 8 |
+
"train_steps_per_second": 0.096
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": 128001,
|
| 5 |
+
"max_length": 4096,
|
| 6 |
+
"temperature": 0.6,
|
| 7 |
+
"top_p": 0.9,
|
| 8 |
+
"transformers_version": "4.45.1"
|
| 9 |
+
}
|
runs/Sep29_17-51-15_COE-CS-sv003/events.out.tfevents.1727632736.COE-CS-sv003.1214200.0
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 sha256:2777ded145d5f5d1f5c39e9ae0d6211c705d4a62a8621f5423daa5f0f6dbb848
|
| 3 |
+
size 76286
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 173366354903040.0,
|
| 4 |
+
"train_loss": 0.6223726071021407,
|
| 5 |
+
"train_runtime": 17321.1955,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.529,
|
| 8 |
+
"train_steps_per_second": 0.096
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2373 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1656,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006038647342995169,
|
| 13 |
+
"grad_norm": 11.647146166653753,
|
| 14 |
+
"learning_rate": 6.02409638554217e-08,
|
| 15 |
+
"loss": 1.4319,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0030193236714975845,
|
| 20 |
+
"grad_norm": 10.76596246435424,
|
| 21 |
+
"learning_rate": 3.0120481927710845e-07,
|
| 22 |
+
"loss": 1.4423,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006038647342995169,
|
| 27 |
+
"grad_norm": 8.345396159148025,
|
| 28 |
+
"learning_rate": 6.024096385542169e-07,
|
| 29 |
+
"loss": 1.4123,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009057971014492754,
|
| 34 |
+
"grad_norm": 5.728658238969367,
|
| 35 |
+
"learning_rate": 9.036144578313254e-07,
|
| 36 |
+
"loss": 1.3779,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012077294685990338,
|
| 41 |
+
"grad_norm": 8.858123210879867,
|
| 42 |
+
"learning_rate": 1.2048192771084338e-06,
|
| 43 |
+
"loss": 1.307,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.015096618357487922,
|
| 48 |
+
"grad_norm": 5.956632109318297,
|
| 49 |
+
"learning_rate": 1.5060240963855425e-06,
|
| 50 |
+
"loss": 1.2575,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.018115942028985508,
|
| 55 |
+
"grad_norm": 2.691768583064308,
|
| 56 |
+
"learning_rate": 1.8072289156626508e-06,
|
| 57 |
+
"loss": 1.1412,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.021135265700483092,
|
| 62 |
+
"grad_norm": 5.521729994682871,
|
| 63 |
+
"learning_rate": 2.1084337349397595e-06,
|
| 64 |
+
"loss": 1.0901,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.024154589371980676,
|
| 69 |
+
"grad_norm": 1.950978681738152,
|
| 70 |
+
"learning_rate": 2.4096385542168676e-06,
|
| 71 |
+
"loss": 1.0489,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02717391304347826,
|
| 76 |
+
"grad_norm": 1.6004592652392855,
|
| 77 |
+
"learning_rate": 2.710843373493976e-06,
|
| 78 |
+
"loss": 1.0294,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.030193236714975844,
|
| 83 |
+
"grad_norm": 1.8752688704592362,
|
| 84 |
+
"learning_rate": 3.012048192771085e-06,
|
| 85 |
+
"loss": 1.0009,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03321256038647343,
|
| 90 |
+
"grad_norm": 1.730463603619638,
|
| 91 |
+
"learning_rate": 3.313253012048193e-06,
|
| 92 |
+
"loss": 0.9798,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.036231884057971016,
|
| 97 |
+
"grad_norm": 1.5964460177303024,
|
| 98 |
+
"learning_rate": 3.6144578313253016e-06,
|
| 99 |
+
"loss": 0.9684,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0392512077294686,
|
| 104 |
+
"grad_norm": 1.6451870176416807,
|
| 105 |
+
"learning_rate": 3.91566265060241e-06,
|
| 106 |
+
"loss": 0.9598,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.042270531400966184,
|
| 111 |
+
"grad_norm": 1.7104520603402644,
|
| 112 |
+
"learning_rate": 4.216867469879519e-06,
|
| 113 |
+
"loss": 0.9603,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04528985507246377,
|
| 118 |
+
"grad_norm": 1.5767703064322158,
|
| 119 |
+
"learning_rate": 4.518072289156627e-06,
|
| 120 |
+
"loss": 0.9292,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04830917874396135,
|
| 125 |
+
"grad_norm": 1.451795891571848,
|
| 126 |
+
"learning_rate": 4.819277108433735e-06,
|
| 127 |
+
"loss": 0.9191,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.051328502415458936,
|
| 132 |
+
"grad_norm": 1.5778301532020658,
|
| 133 |
+
"learning_rate": 5.120481927710844e-06,
|
| 134 |
+
"loss": 0.9131,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05434782608695652,
|
| 139 |
+
"grad_norm": 1.5490469939143814,
|
| 140 |
+
"learning_rate": 5.421686746987952e-06,
|
| 141 |
+
"loss": 0.9147,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.057367149758454104,
|
| 146 |
+
"grad_norm": 1.6958988602467524,
|
| 147 |
+
"learning_rate": 5.722891566265061e-06,
|
| 148 |
+
"loss": 0.9031,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06038647342995169,
|
| 153 |
+
"grad_norm": 1.695663023457705,
|
| 154 |
+
"learning_rate": 6.02409638554217e-06,
|
| 155 |
+
"loss": 0.8903,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06340579710144928,
|
| 160 |
+
"grad_norm": 1.696466007675861,
|
| 161 |
+
"learning_rate": 6.325301204819277e-06,
|
| 162 |
+
"loss": 0.8852,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06642512077294686,
|
| 167 |
+
"grad_norm": 1.607366527826758,
|
| 168 |
+
"learning_rate": 6.626506024096386e-06,
|
| 169 |
+
"loss": 0.8729,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06944444444444445,
|
| 174 |
+
"grad_norm": 1.592578389550239,
|
| 175 |
+
"learning_rate": 6.927710843373494e-06,
|
| 176 |
+
"loss": 0.8685,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07246376811594203,
|
| 181 |
+
"grad_norm": 1.5599550038753036,
|
| 182 |
+
"learning_rate": 7.228915662650603e-06,
|
| 183 |
+
"loss": 0.8643,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07548309178743962,
|
| 188 |
+
"grad_norm": 1.7526452577051699,
|
| 189 |
+
"learning_rate": 7.530120481927712e-06,
|
| 190 |
+
"loss": 0.8472,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.0785024154589372,
|
| 195 |
+
"grad_norm": 1.570734587958465,
|
| 196 |
+
"learning_rate": 7.83132530120482e-06,
|
| 197 |
+
"loss": 0.8457,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08152173913043478,
|
| 202 |
+
"grad_norm": 1.3578392369216397,
|
| 203 |
+
"learning_rate": 8.132530120481928e-06,
|
| 204 |
+
"loss": 0.8419,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08454106280193237,
|
| 209 |
+
"grad_norm": 1.5345711918538654,
|
| 210 |
+
"learning_rate": 8.433734939759038e-06,
|
| 211 |
+
"loss": 0.8293,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08756038647342995,
|
| 216 |
+
"grad_norm": 1.5199887652875657,
|
| 217 |
+
"learning_rate": 8.734939759036145e-06,
|
| 218 |
+
"loss": 0.8116,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09057971014492754,
|
| 223 |
+
"grad_norm": 1.4329178750495808,
|
| 224 |
+
"learning_rate": 9.036144578313254e-06,
|
| 225 |
+
"loss": 0.8131,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09359903381642512,
|
| 230 |
+
"grad_norm": 1.5357940171741657,
|
| 231 |
+
"learning_rate": 9.337349397590362e-06,
|
| 232 |
+
"loss": 0.8109,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0966183574879227,
|
| 237 |
+
"grad_norm": 1.4127066136547346,
|
| 238 |
+
"learning_rate": 9.63855421686747e-06,
|
| 239 |
+
"loss": 0.8116,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09963768115942029,
|
| 244 |
+
"grad_norm": 1.5153723323216768,
|
| 245 |
+
"learning_rate": 9.93975903614458e-06,
|
| 246 |
+
"loss": 0.8067,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10265700483091787,
|
| 251 |
+
"grad_norm": 1.3541934052887161,
|
| 252 |
+
"learning_rate": 9.999822178354131e-06,
|
| 253 |
+
"loss": 0.7957,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10567632850241546,
|
| 258 |
+
"grad_norm": 1.3822332075059867,
|
| 259 |
+
"learning_rate": 9.999099799595088e-06,
|
| 260 |
+
"loss": 0.7915,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10869565217391304,
|
| 265 |
+
"grad_norm": 1.4372310169992935,
|
| 266 |
+
"learning_rate": 9.997821830092095e-06,
|
| 267 |
+
"loss": 0.7999,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11171497584541062,
|
| 272 |
+
"grad_norm": 1.6716440887834658,
|
| 273 |
+
"learning_rate": 9.995988411876328e-06,
|
| 274 |
+
"loss": 0.7992,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11473429951690821,
|
| 279 |
+
"grad_norm": 1.5723808182751033,
|
| 280 |
+
"learning_rate": 9.993599748710505e-06,
|
| 281 |
+
"loss": 0.783,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11775362318840579,
|
| 286 |
+
"grad_norm": 1.4610043673248276,
|
| 287 |
+
"learning_rate": 9.990656106066257e-06,
|
| 288 |
+
"loss": 0.7827,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12077294685990338,
|
| 293 |
+
"grad_norm": 1.4846567788944018,
|
| 294 |
+
"learning_rate": 9.9871578110946e-06,
|
| 295 |
+
"loss": 0.7773,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12379227053140096,
|
| 300 |
+
"grad_norm": 1.4132792684184563,
|
| 301 |
+
"learning_rate": 9.983105252589599e-06,
|
| 302 |
+
"loss": 0.7662,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12681159420289856,
|
| 307 |
+
"grad_norm": 1.3392248870506875,
|
| 308 |
+
"learning_rate": 9.978498880945138e-06,
|
| 309 |
+
"loss": 0.7643,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12983091787439613,
|
| 314 |
+
"grad_norm": 1.4340057279445353,
|
| 315 |
+
"learning_rate": 9.97333920810488e-06,
|
| 316 |
+
"loss": 0.7791,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.13285024154589373,
|
| 321 |
+
"grad_norm": 1.2837053197488815,
|
| 322 |
+
"learning_rate": 9.967626807505359e-06,
|
| 323 |
+
"loss": 0.7647,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1358695652173913,
|
| 328 |
+
"grad_norm": 1.4133054167156096,
|
| 329 |
+
"learning_rate": 9.961362314012258e-06,
|
| 330 |
+
"loss": 0.7851,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1388888888888889,
|
| 335 |
+
"grad_norm": 1.4609903990792592,
|
| 336 |
+
"learning_rate": 9.954546423849842e-06,
|
| 337 |
+
"loss": 0.7689,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14190821256038647,
|
| 342 |
+
"grad_norm": 1.4430590677898403,
|
| 343 |
+
"learning_rate": 9.947179894523594e-06,
|
| 344 |
+
"loss": 0.7483,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.14492753623188406,
|
| 349 |
+
"grad_norm": 1.1927964292128603,
|
| 350 |
+
"learning_rate": 9.93926354473601e-06,
|
| 351 |
+
"loss": 0.7549,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.14794685990338163,
|
| 356 |
+
"grad_norm": 1.267141348725353,
|
| 357 |
+
"learning_rate": 9.930798254295628e-06,
|
| 358 |
+
"loss": 0.7528,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.15096618357487923,
|
| 363 |
+
"grad_norm": 1.2738788294787875,
|
| 364 |
+
"learning_rate": 9.921784964019234e-06,
|
| 365 |
+
"loss": 0.7588,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1539855072463768,
|
| 370 |
+
"grad_norm": 1.3665251025923502,
|
| 371 |
+
"learning_rate": 9.91222467562731e-06,
|
| 372 |
+
"loss": 0.752,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1570048309178744,
|
| 377 |
+
"grad_norm": 1.2895638701049197,
|
| 378 |
+
"learning_rate": 9.902118451632694e-06,
|
| 379 |
+
"loss": 0.7443,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16002415458937197,
|
| 384 |
+
"grad_norm": 1.2792248819171141,
|
| 385 |
+
"learning_rate": 9.891467415222511e-06,
|
| 386 |
+
"loss": 0.7389,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16304347826086957,
|
| 391 |
+
"grad_norm": 1.2577426141477723,
|
| 392 |
+
"learning_rate": 9.880272750133328e-06,
|
| 393 |
+
"loss": 0.7495,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16606280193236714,
|
| 398 |
+
"grad_norm": 1.2581092756100691,
|
| 399 |
+
"learning_rate": 9.868535700519605e-06,
|
| 400 |
+
"loss": 0.7574,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.16908212560386474,
|
| 405 |
+
"grad_norm": 1.3915721437755781,
|
| 406 |
+
"learning_rate": 9.856257570815415e-06,
|
| 407 |
+
"loss": 0.7451,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.1721014492753623,
|
| 412 |
+
"grad_norm": 1.4182529730794977,
|
| 413 |
+
"learning_rate": 9.843439725589481e-06,
|
| 414 |
+
"loss": 0.7392,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1751207729468599,
|
| 419 |
+
"grad_norm": 1.4007115307336426,
|
| 420 |
+
"learning_rate": 9.83008358939351e-06,
|
| 421 |
+
"loss": 0.7488,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17814009661835747,
|
| 426 |
+
"grad_norm": 1.2602901130922803,
|
| 427 |
+
"learning_rate": 9.81619064660388e-06,
|
| 428 |
+
"loss": 0.7481,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.18115942028985507,
|
| 433 |
+
"grad_norm": 1.230299206050831,
|
| 434 |
+
"learning_rate": 9.801762441256663e-06,
|
| 435 |
+
"loss": 0.7277,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18417874396135267,
|
| 440 |
+
"grad_norm": 1.1834363014701523,
|
| 441 |
+
"learning_rate": 9.786800576876026e-06,
|
| 442 |
+
"loss": 0.7458,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.18719806763285024,
|
| 447 |
+
"grad_norm": 1.257138541703525,
|
| 448 |
+
"learning_rate": 9.77130671629602e-06,
|
| 449 |
+
"loss": 0.7338,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.19021739130434784,
|
| 454 |
+
"grad_norm": 1.316511247899504,
|
| 455 |
+
"learning_rate": 9.755282581475769e-06,
|
| 456 |
+
"loss": 0.7382,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.1932367149758454,
|
| 461 |
+
"grad_norm": 1.2036218489959571,
|
| 462 |
+
"learning_rate": 9.738729953308104e-06,
|
| 463 |
+
"loss": 0.734,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.196256038647343,
|
| 468 |
+
"grad_norm": 1.3147689783592325,
|
| 469 |
+
"learning_rate": 9.72165067142163e-06,
|
| 470 |
+
"loss": 0.7133,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.19927536231884058,
|
| 475 |
+
"grad_norm": 1.2043088743921089,
|
| 476 |
+
"learning_rate": 9.70404663397628e-06,
|
| 477 |
+
"loss": 0.724,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.20229468599033817,
|
| 482 |
+
"grad_norm": 1.184487132016919,
|
| 483 |
+
"learning_rate": 9.68591979745235e-06,
|
| 484 |
+
"loss": 0.7227,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.20531400966183574,
|
| 489 |
+
"grad_norm": 1.1868803234306302,
|
| 490 |
+
"learning_rate": 9.667272176433063e-06,
|
| 491 |
+
"loss": 0.7211,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.20833333333333334,
|
| 496 |
+
"grad_norm": 1.354442450720458,
|
| 497 |
+
"learning_rate": 9.648105843380674e-06,
|
| 498 |
+
"loss": 0.7246,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2113526570048309,
|
| 503 |
+
"grad_norm": 1.3221102565499465,
|
| 504 |
+
"learning_rate": 9.628422928406133e-06,
|
| 505 |
+
"loss": 0.723,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2143719806763285,
|
| 510 |
+
"grad_norm": 1.298626503464365,
|
| 511 |
+
"learning_rate": 9.608225619032361e-06,
|
| 512 |
+
"loss": 0.7204,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.21739130434782608,
|
| 517 |
+
"grad_norm": 1.2235482139383798,
|
| 518 |
+
"learning_rate": 9.587516159951118e-06,
|
| 519 |
+
"loss": 0.7157,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.22041062801932368,
|
| 524 |
+
"grad_norm": 1.4044097057246043,
|
| 525 |
+
"learning_rate": 9.566296852773541e-06,
|
| 526 |
+
"loss": 0.7444,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.22342995169082125,
|
| 531 |
+
"grad_norm": 1.204069094083874,
|
| 532 |
+
"learning_rate": 9.544570055774348e-06,
|
| 533 |
+
"loss": 0.7068,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22644927536231885,
|
| 538 |
+
"grad_norm": 1.433962682853488,
|
| 539 |
+
"learning_rate": 9.522338183629737e-06,
|
| 540 |
+
"loss": 0.7136,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.22946859903381642,
|
| 545 |
+
"grad_norm": 1.3644548503760257,
|
| 546 |
+
"learning_rate": 9.499603707149035e-06,
|
| 547 |
+
"loss": 0.7121,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23248792270531402,
|
| 552 |
+
"grad_norm": 1.3126699782875395,
|
| 553 |
+
"learning_rate": 9.476369153000076e-06,
|
| 554 |
+
"loss": 0.7184,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.23550724637681159,
|
| 559 |
+
"grad_norm": 1.1868983270075697,
|
| 560 |
+
"learning_rate": 9.45263710342842e-06,
|
| 561 |
+
"loss": 0.6992,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.23852657004830918,
|
| 566 |
+
"grad_norm": 1.4139162160182768,
|
| 567 |
+
"learning_rate": 9.428410195970337e-06,
|
| 568 |
+
"loss": 0.6926,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.24154589371980675,
|
| 573 |
+
"grad_norm": 1.1181104001140345,
|
| 574 |
+
"learning_rate": 9.403691123159707e-06,
|
| 575 |
+
"loss": 0.7041,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24456521739130435,
|
| 580 |
+
"grad_norm": 1.4069308087516696,
|
| 581 |
+
"learning_rate": 9.378482632228745e-06,
|
| 582 |
+
"loss": 0.6956,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.24758454106280192,
|
| 587 |
+
"grad_norm": 1.230108931004278,
|
| 588 |
+
"learning_rate": 9.352787524802707e-06,
|
| 589 |
+
"loss": 0.7121,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2506038647342995,
|
| 594 |
+
"grad_norm": 1.2509235012582307,
|
| 595 |
+
"learning_rate": 9.326608656588502e-06,
|
| 596 |
+
"loss": 0.7013,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2536231884057971,
|
| 601 |
+
"grad_norm": 1.36604045335286,
|
| 602 |
+
"learning_rate": 9.299948937057325e-06,
|
| 603 |
+
"loss": 0.6953,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25664251207729466,
|
| 608 |
+
"grad_norm": 1.1730575669226555,
|
| 609 |
+
"learning_rate": 9.272811329121305e-06,
|
| 610 |
+
"loss": 0.7144,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.25966183574879226,
|
| 615 |
+
"grad_norm": 1.1622429028280914,
|
| 616 |
+
"learning_rate": 9.245198848804197e-06,
|
| 617 |
+
"loss": 0.6991,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.26268115942028986,
|
| 622 |
+
"grad_norm": 1.264307120570429,
|
| 623 |
+
"learning_rate": 9.217114564906208e-06,
|
| 624 |
+
"loss": 0.6867,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.26570048309178745,
|
| 629 |
+
"grad_norm": 1.1848039169987097,
|
| 630 |
+
"learning_rate": 9.188561598662921e-06,
|
| 631 |
+
"loss": 0.691,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.26871980676328505,
|
| 636 |
+
"grad_norm": 1.2376999426889066,
|
| 637 |
+
"learning_rate": 9.159543123398416e-06,
|
| 638 |
+
"loss": 0.7033,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2717391304347826,
|
| 643 |
+
"grad_norm": 1.2317420665829562,
|
| 644 |
+
"learning_rate": 9.130062364172582e-06,
|
| 645 |
+
"loss": 0.6907,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2747584541062802,
|
| 650 |
+
"grad_norm": 1.2724956183147942,
|
| 651 |
+
"learning_rate": 9.1001225974227e-06,
|
| 652 |
+
"loss": 0.6805,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2777777777777778,
|
| 657 |
+
"grad_norm": 1.1495012873362647,
|
| 658 |
+
"learning_rate": 9.0697271505993e-06,
|
| 659 |
+
"loss": 0.6813,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2807971014492754,
|
| 664 |
+
"grad_norm": 1.267845133752999,
|
| 665 |
+
"learning_rate": 9.038879401796358e-06,
|
| 666 |
+
"loss": 0.6947,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.28381642512077293,
|
| 671 |
+
"grad_norm": 1.2127192185018594,
|
| 672 |
+
"learning_rate": 9.00758277937586e-06,
|
| 673 |
+
"loss": 0.6966,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.28683574879227053,
|
| 678 |
+
"grad_norm": 1.2447551629635505,
|
| 679 |
+
"learning_rate": 8.975840761586772e-06,
|
| 680 |
+
"loss": 0.7047,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2898550724637681,
|
| 685 |
+
"grad_norm": 1.1928908039430874,
|
| 686 |
+
"learning_rate": 8.94365687617849e-06,
|
| 687 |
+
"loss": 0.6935,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2928743961352657,
|
| 692 |
+
"grad_norm": 1.2672585178249127,
|
| 693 |
+
"learning_rate": 8.911034700008757e-06,
|
| 694 |
+
"loss": 0.6739,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.29589371980676327,
|
| 699 |
+
"grad_norm": 1.3798934886196015,
|
| 700 |
+
"learning_rate": 8.87797785864615e-06,
|
| 701 |
+
"loss": 0.6877,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.29891304347826086,
|
| 706 |
+
"grad_norm": 1.2248493689194777,
|
| 707 |
+
"learning_rate": 8.844490025967126e-06,
|
| 708 |
+
"loss": 0.6758,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.30193236714975846,
|
| 713 |
+
"grad_norm": 1.1804949407776295,
|
| 714 |
+
"learning_rate": 8.810574923747729e-06,
|
| 715 |
+
"loss": 0.6798,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.30495169082125606,
|
| 720 |
+
"grad_norm": 1.2545058603527033,
|
| 721 |
+
"learning_rate": 8.776236321249955e-06,
|
| 722 |
+
"loss": 0.6891,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3079710144927536,
|
| 727 |
+
"grad_norm": 1.2224106907939065,
|
| 728 |
+
"learning_rate": 8.741478034802835e-06,
|
| 729 |
+
"loss": 0.6866,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3109903381642512,
|
| 734 |
+
"grad_norm": 1.2220324268766434,
|
| 735 |
+
"learning_rate": 8.706303927378306e-06,
|
| 736 |
+
"loss": 0.6537,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3140096618357488,
|
| 741 |
+
"grad_norm": 1.2102281315861285,
|
| 742 |
+
"learning_rate": 8.670717908161878e-06,
|
| 743 |
+
"loss": 0.6656,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3170289855072464,
|
| 748 |
+
"grad_norm": 1.3602250350216378,
|
| 749 |
+
"learning_rate": 8.634723932118184e-06,
|
| 750 |
+
"loss": 0.6738,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.32004830917874394,
|
| 755 |
+
"grad_norm": 1.289145913095343,
|
| 756 |
+
"learning_rate": 8.598325999551425e-06,
|
| 757 |
+
"loss": 0.6717,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.32306763285024154,
|
| 762 |
+
"grad_norm": 1.1804864914238673,
|
| 763 |
+
"learning_rate": 8.56152815566078e-06,
|
| 764 |
+
"loss": 0.6579,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.32608695652173914,
|
| 769 |
+
"grad_norm": 1.2188640624977634,
|
| 770 |
+
"learning_rate": 8.524334490090848e-06,
|
| 771 |
+
"loss": 0.6802,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.32910628019323673,
|
| 776 |
+
"grad_norm": 1.277685319237274,
|
| 777 |
+
"learning_rate": 8.486749136477112e-06,
|
| 778 |
+
"loss": 0.6693,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3321256038647343,
|
| 783 |
+
"grad_norm": 1.300215594029609,
|
| 784 |
+
"learning_rate": 8.448776271986542e-06,
|
| 785 |
+
"loss": 0.6681,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3351449275362319,
|
| 790 |
+
"grad_norm": 1.1966744788724837,
|
| 791 |
+
"learning_rate": 8.41042011685336e-06,
|
| 792 |
+
"loss": 0.6698,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.33816425120772947,
|
| 797 |
+
"grad_norm": 1.354369397725386,
|
| 798 |
+
"learning_rate": 8.371684933909996e-06,
|
| 799 |
+
"loss": 0.6662,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.34118357487922707,
|
| 804 |
+
"grad_norm": 1.2209167981844395,
|
| 805 |
+
"learning_rate": 8.33257502811334e-06,
|
| 806 |
+
"loss": 0.6714,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3442028985507246,
|
| 811 |
+
"grad_norm": 1.3116820491493628,
|
| 812 |
+
"learning_rate": 8.293094746066283e-06,
|
| 813 |
+
"loss": 0.6686,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3472222222222222,
|
| 818 |
+
"grad_norm": 1.1920221462379554,
|
| 819 |
+
"learning_rate": 8.253248475534656e-06,
|
| 820 |
+
"loss": 0.666,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3502415458937198,
|
| 825 |
+
"grad_norm": 1.2619018537421245,
|
| 826 |
+
"learning_rate": 8.213040644959572e-06,
|
| 827 |
+
"loss": 0.6376,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3532608695652174,
|
| 832 |
+
"grad_norm": 1.3255977529395933,
|
| 833 |
+
"learning_rate": 8.172475722965263e-06,
|
| 834 |
+
"loss": 0.6485,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.35628019323671495,
|
| 839 |
+
"grad_norm": 1.1572256141420745,
|
| 840 |
+
"learning_rate": 8.131558217862444e-06,
|
| 841 |
+
"loss": 0.6265,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.35929951690821255,
|
| 846 |
+
"grad_norm": 1.2055573426722468,
|
| 847 |
+
"learning_rate": 8.090292677147268e-06,
|
| 848 |
+
"loss": 0.6488,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.36231884057971014,
|
| 853 |
+
"grad_norm": 1.2861724281724545,
|
| 854 |
+
"learning_rate": 8.048683686995921e-06,
|
| 855 |
+
"loss": 0.6577,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.36533816425120774,
|
| 860 |
+
"grad_norm": 1.1614784125218114,
|
| 861 |
+
"learning_rate": 8.006735871754932e-06,
|
| 862 |
+
"loss": 0.6639,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.36835748792270534,
|
| 867 |
+
"grad_norm": 1.2651442426402355,
|
| 868 |
+
"learning_rate": 7.96445389342722e-06,
|
| 869 |
+
"loss": 0.6429,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3713768115942029,
|
| 874 |
+
"grad_norm": 1.1763415581175696,
|
| 875 |
+
"learning_rate": 7.921842451153982e-06,
|
| 876 |
+
"loss": 0.642,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3743961352657005,
|
| 881 |
+
"grad_norm": 1.2194809852155362,
|
| 882 |
+
"learning_rate": 7.878906280692424e-06,
|
| 883 |
+
"loss": 0.658,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3774154589371981,
|
| 888 |
+
"grad_norm": 1.234621233448178,
|
| 889 |
+
"learning_rate": 7.835650153889449e-06,
|
| 890 |
+
"loss": 0.6455,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3804347826086957,
|
| 895 |
+
"grad_norm": 1.2755378740652312,
|
| 896 |
+
"learning_rate": 7.792078878151318e-06,
|
| 897 |
+
"loss": 0.6475,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3834541062801932,
|
| 902 |
+
"grad_norm": 1.2958591755788766,
|
| 903 |
+
"learning_rate": 7.748197295909359e-06,
|
| 904 |
+
"loss": 0.6387,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.3864734299516908,
|
| 909 |
+
"grad_norm": 1.2667202622939782,
|
| 910 |
+
"learning_rate": 7.704010284081801e-06,
|
| 911 |
+
"loss": 0.6403,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.3894927536231884,
|
| 916 |
+
"grad_norm": 1.5195261481113882,
|
| 917 |
+
"learning_rate": 7.65952275353175e-06,
|
| 918 |
+
"loss": 0.6303,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.392512077294686,
|
| 923 |
+
"grad_norm": 1.2745676764475649,
|
| 924 |
+
"learning_rate": 7.614739648521412e-06,
|
| 925 |
+
"loss": 0.6427,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.39553140096618356,
|
| 930 |
+
"grad_norm": 1.249145405044073,
|
| 931 |
+
"learning_rate": 7.56966594616259e-06,
|
| 932 |
+
"loss": 0.6324,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.39855072463768115,
|
| 937 |
+
"grad_norm": 1.3163147586943797,
|
| 938 |
+
"learning_rate": 7.524306655863544e-06,
|
| 939 |
+
"loss": 0.6467,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.40157004830917875,
|
| 944 |
+
"grad_norm": 1.2102246889732293,
|
| 945 |
+
"learning_rate": 7.478666818772252e-06,
|
| 946 |
+
"loss": 0.637,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.40458937198067635,
|
| 951 |
+
"grad_norm": 1.270031191494333,
|
| 952 |
+
"learning_rate": 7.432751507216146e-06,
|
| 953 |
+
"loss": 0.6375,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4076086956521739,
|
| 958 |
+
"grad_norm": 1.2566934217617964,
|
| 959 |
+
"learning_rate": 7.386565824138378e-06,
|
| 960 |
+
"loss": 0.6315,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4106280193236715,
|
| 965 |
+
"grad_norm": 1.2617759238489816,
|
| 966 |
+
"learning_rate": 7.3401149025306995e-06,
|
| 967 |
+
"loss": 0.6243,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4136473429951691,
|
| 972 |
+
"grad_norm": 1.2024909559860102,
|
| 973 |
+
"learning_rate": 7.293403904862981e-06,
|
| 974 |
+
"loss": 0.6369,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4166666666666667,
|
| 979 |
+
"grad_norm": 1.3149707857289934,
|
| 980 |
+
"learning_rate": 7.246438022509465e-06,
|
| 981 |
+
"loss": 0.6272,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4196859903381642,
|
| 986 |
+
"grad_norm": 1.3641187440427756,
|
| 987 |
+
"learning_rate": 7.199222475171812e-06,
|
| 988 |
+
"loss": 0.6286,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.4227053140096618,
|
| 993 |
+
"grad_norm": 1.2521464928430228,
|
| 994 |
+
"learning_rate": 7.151762510298985e-06,
|
| 995 |
+
"loss": 0.6292,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4257246376811594,
|
| 1000 |
+
"grad_norm": 1.4667183264262524,
|
| 1001 |
+
"learning_rate": 7.104063402504065e-06,
|
| 1002 |
+
"loss": 0.6159,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.428743961352657,
|
| 1007 |
+
"grad_norm": 1.2306489207096507,
|
| 1008 |
+
"learning_rate": 7.056130452978039e-06,
|
| 1009 |
+
"loss": 0.6184,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.43176328502415456,
|
| 1014 |
+
"grad_norm": 1.2724217204325428,
|
| 1015 |
+
"learning_rate": 7.0079689889006275e-06,
|
| 1016 |
+
"loss": 0.6166,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.43478260869565216,
|
| 1021 |
+
"grad_norm": 1.363470409728175,
|
| 1022 |
+
"learning_rate": 6.959584362848239e-06,
|
| 1023 |
+
"loss": 0.6115,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.43780193236714976,
|
| 1028 |
+
"grad_norm": 1.2907389058316032,
|
| 1029 |
+
"learning_rate": 6.910981952199097e-06,
|
| 1030 |
+
"loss": 0.6184,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.44082125603864736,
|
| 1035 |
+
"grad_norm": 1.2785918958945928,
|
| 1036 |
+
"learning_rate": 6.862167158535599e-06,
|
| 1037 |
+
"loss": 0.6075,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4438405797101449,
|
| 1042 |
+
"grad_norm": 1.3363100906946352,
|
| 1043 |
+
"learning_rate": 6.813145407044003e-06,
|
| 1044 |
+
"loss": 0.6062,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4468599033816425,
|
| 1049 |
+
"grad_norm": 1.2872953040326178,
|
| 1050 |
+
"learning_rate": 6.763922145911474e-06,
|
| 1051 |
+
"loss": 0.6133,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.4498792270531401,
|
| 1056 |
+
"grad_norm": 1.2195163567464784,
|
| 1057 |
+
"learning_rate": 6.714502845720595e-06,
|
| 1058 |
+
"loss": 0.6081,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4528985507246377,
|
| 1063 |
+
"grad_norm": 1.2421351910574,
|
| 1064 |
+
"learning_rate": 6.664892998841361e-06,
|
| 1065 |
+
"loss": 0.6114,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.4559178743961353,
|
| 1070 |
+
"grad_norm": 1.199322165560994,
|
| 1071 |
+
"learning_rate": 6.61509811882078e-06,
|
| 1072 |
+
"loss": 0.5977,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.45893719806763283,
|
| 1077 |
+
"grad_norm": 1.3225200571161657,
|
| 1078 |
+
"learning_rate": 6.565123739770102e-06,
|
| 1079 |
+
"loss": 0.6035,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.46195652173913043,
|
| 1084 |
+
"grad_norm": 1.2787776232265113,
|
| 1085 |
+
"learning_rate": 6.5149754157497645e-06,
|
| 1086 |
+
"loss": 0.5973,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.46497584541062803,
|
| 1091 |
+
"grad_norm": 1.254549030567658,
|
| 1092 |
+
"learning_rate": 6.464658720152135e-06,
|
| 1093 |
+
"loss": 0.5986,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.46799516908212563,
|
| 1098 |
+
"grad_norm": 1.2314288543177414,
|
| 1099 |
+
"learning_rate": 6.41417924508208e-06,
|
| 1100 |
+
"loss": 0.5894,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.47101449275362317,
|
| 1105 |
+
"grad_norm": 1.2493709608575305,
|
| 1106 |
+
"learning_rate": 6.363542600735486e-06,
|
| 1107 |
+
"loss": 0.602,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.47403381642512077,
|
| 1112 |
+
"grad_norm": 1.2411573705254966,
|
| 1113 |
+
"learning_rate": 6.312754414775737e-06,
|
| 1114 |
+
"loss": 0.5984,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.47705314009661837,
|
| 1119 |
+
"grad_norm": 1.5223815896896797,
|
| 1120 |
+
"learning_rate": 6.261820331708275e-06,
|
| 1121 |
+
"loss": 0.5917,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.48007246376811596,
|
| 1126 |
+
"grad_norm": 1.3324419027195697,
|
| 1127 |
+
"learning_rate": 6.210746012253277e-06,
|
| 1128 |
+
"loss": 0.5894,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.4830917874396135,
|
| 1133 |
+
"grad_norm": 1.44284253649467,
|
| 1134 |
+
"learning_rate": 6.159537132716532e-06,
|
| 1135 |
+
"loss": 0.608,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4861111111111111,
|
| 1140 |
+
"grad_norm": 1.2847962576743663,
|
| 1141 |
+
"learning_rate": 6.108199384358595e-06,
|
| 1142 |
+
"loss": 0.5699,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4891304347826087,
|
| 1147 |
+
"grad_norm": 1.2281985675073568,
|
| 1148 |
+
"learning_rate": 6.0567384727622566e-06,
|
| 1149 |
+
"loss": 0.5978,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4921497584541063,
|
| 1154 |
+
"grad_norm": 1.2429721815217631,
|
| 1155 |
+
"learning_rate": 6.005160117198448e-06,
|
| 1156 |
+
"loss": 0.5655,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.49516908212560384,
|
| 1161 |
+
"grad_norm": 1.2553916316159304,
|
| 1162 |
+
"learning_rate": 5.953470049990605e-06,
|
| 1163 |
+
"loss": 0.5824,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.49818840579710144,
|
| 1168 |
+
"grad_norm": 1.2732162189081664,
|
| 1169 |
+
"learning_rate": 5.90167401587759e-06,
|
| 1170 |
+
"loss": 0.582,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.501207729468599,
|
| 1175 |
+
"grad_norm": 1.2868094593799089,
|
| 1176 |
+
"learning_rate": 5.84977777137523e-06,
|
| 1177 |
+
"loss": 0.5755,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5042270531400966,
|
| 1182 |
+
"grad_norm": 1.2418575696704444,
|
| 1183 |
+
"learning_rate": 5.797787084136556e-06,
|
| 1184 |
+
"loss": 0.5887,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5072463768115942,
|
| 1189 |
+
"grad_norm": 1.346918393906291,
|
| 1190 |
+
"learning_rate": 5.745707732310781e-06,
|
| 1191 |
+
"loss": 0.5955,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5102657004830918,
|
| 1196 |
+
"grad_norm": 1.34120555394378,
|
| 1197 |
+
"learning_rate": 5.693545503901149e-06,
|
| 1198 |
+
"loss": 0.5855,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5132850241545893,
|
| 1203 |
+
"grad_norm": 1.2970124804625551,
|
| 1204 |
+
"learning_rate": 5.641306196121643e-06,
|
| 1205 |
+
"loss": 0.5703,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5163043478260869,
|
| 1210 |
+
"grad_norm": 1.3089147536533883,
|
| 1211 |
+
"learning_rate": 5.5889956147527156e-06,
|
| 1212 |
+
"loss": 0.581,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5193236714975845,
|
| 1217 |
+
"grad_norm": 1.317588086682751,
|
| 1218 |
+
"learning_rate": 5.536619573496027e-06,
|
| 1219 |
+
"loss": 0.5827,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5223429951690821,
|
| 1224 |
+
"grad_norm": 1.3414854828076408,
|
| 1225 |
+
"learning_rate": 5.484183893328332e-06,
|
| 1226 |
+
"loss": 0.5832,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5253623188405797,
|
| 1231 |
+
"grad_norm": 1.3014560301267408,
|
| 1232 |
+
"learning_rate": 5.431694401854545e-06,
|
| 1233 |
+
"loss": 0.576,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5283816425120773,
|
| 1238 |
+
"grad_norm": 1.3095457716443242,
|
| 1239 |
+
"learning_rate": 5.379156932660067e-06,
|
| 1240 |
+
"loss": 0.575,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5314009661835749,
|
| 1245 |
+
"grad_norm": 1.2198501596042746,
|
| 1246 |
+
"learning_rate": 5.326577324662459e-06,
|
| 1247 |
+
"loss": 0.5731,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5344202898550725,
|
| 1252 |
+
"grad_norm": 1.6231967828184084,
|
| 1253 |
+
"learning_rate": 5.273961421462505e-06,
|
| 1254 |
+
"loss": 0.5641,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5374396135265701,
|
| 1259 |
+
"grad_norm": 1.308363933960681,
|
| 1260 |
+
"learning_rate": 5.221315070694775e-06,
|
| 1261 |
+
"loss": 0.5597,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5404589371980676,
|
| 1266 |
+
"grad_norm": 1.2349082122280453,
|
| 1267 |
+
"learning_rate": 5.168644123377725e-06,
|
| 1268 |
+
"loss": 0.5665,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5434782608695652,
|
| 1273 |
+
"grad_norm": 1.360538833482079,
|
| 1274 |
+
"learning_rate": 5.1159544332634256e-06,
|
| 1275 |
+
"loss": 0.5584,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5464975845410628,
|
| 1280 |
+
"grad_norm": 1.3836112606752502,
|
| 1281 |
+
"learning_rate": 5.063251856186991e-06,
|
| 1282 |
+
"loss": 0.5613,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5495169082125604,
|
| 1287 |
+
"grad_norm": 1.3886831951922487,
|
| 1288 |
+
"learning_rate": 5.010542249415761e-06,
|
| 1289 |
+
"loss": 0.5709,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.552536231884058,
|
| 1294 |
+
"grad_norm": 1.4704931160983408,
|
| 1295 |
+
"learning_rate": 4.95783147099835e-06,
|
| 1296 |
+
"loss": 0.5594,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5555555555555556,
|
| 1301 |
+
"grad_norm": 1.2850129485673656,
|
| 1302 |
+
"learning_rate": 4.90512537911358e-06,
|
| 1303 |
+
"loss": 0.5483,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5585748792270532,
|
| 1308 |
+
"grad_norm": 1.3661715593612251,
|
| 1309 |
+
"learning_rate": 4.852429831419428e-06,
|
| 1310 |
+
"loss": 0.5619,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5615942028985508,
|
| 1315 |
+
"grad_norm": 1.3522573537703417,
|
| 1316 |
+
"learning_rate": 4.799750684402006e-06,
|
| 1317 |
+
"loss": 0.5415,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5646135265700483,
|
| 1322 |
+
"grad_norm": 1.2608087316548307,
|
| 1323 |
+
"learning_rate": 4.747093792724679e-06,
|
| 1324 |
+
"loss": 0.552,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5676328502415459,
|
| 1329 |
+
"grad_norm": 1.2839946310245753,
|
| 1330 |
+
"learning_rate": 4.6944650085774095e-06,
|
| 1331 |
+
"loss": 0.5632,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5706521739130435,
|
| 1336 |
+
"grad_norm": 1.3007085793096578,
|
| 1337 |
+
"learning_rate": 4.641870181026322e-06,
|
| 1338 |
+
"loss": 0.5625,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5736714975845411,
|
| 1343 |
+
"grad_norm": 1.3004895083183792,
|
| 1344 |
+
"learning_rate": 4.589315155363683e-06,
|
| 1345 |
+
"loss": 0.5507,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5766908212560387,
|
| 1350 |
+
"grad_norm": 1.325648774927767,
|
| 1351 |
+
"learning_rate": 4.53680577245824e-06,
|
| 1352 |
+
"loss": 0.5596,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5797101449275363,
|
| 1357 |
+
"grad_norm": 1.3082395762281465,
|
| 1358 |
+
"learning_rate": 4.484347868106097e-06,
|
| 1359 |
+
"loss": 0.5478,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5827294685990339,
|
| 1364 |
+
"grad_norm": 1.3393641276484884,
|
| 1365 |
+
"learning_rate": 4.431947272382118e-06,
|
| 1366 |
+
"loss": 0.5441,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5857487922705314,
|
| 1371 |
+
"grad_norm": 1.312719486472206,
|
| 1372 |
+
"learning_rate": 4.379609808992e-06,
|
| 1373 |
+
"loss": 0.5497,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5887681159420289,
|
| 1378 |
+
"grad_norm": 1.3204903978926055,
|
| 1379 |
+
"learning_rate": 4.327341294625019e-06,
|
| 1380 |
+
"loss": 0.5402,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5917874396135265,
|
| 1385 |
+
"grad_norm": 1.3284223155848323,
|
| 1386 |
+
"learning_rate": 4.275147538307594e-06,
|
| 1387 |
+
"loss": 0.5454,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5948067632850241,
|
| 1392 |
+
"grad_norm": 1.3534667187989002,
|
| 1393 |
+
"learning_rate": 4.223034340757666e-06,
|
| 1394 |
+
"loss": 0.5449,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5978260869565217,
|
| 1399 |
+
"grad_norm": 1.4006362214343153,
|
| 1400 |
+
"learning_rate": 4.171007493740023e-06,
|
| 1401 |
+
"loss": 0.5488,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6008454106280193,
|
| 1406 |
+
"grad_norm": 1.344509170501366,
|
| 1407 |
+
"learning_rate": 4.1190727794226175e-06,
|
| 1408 |
+
"loss": 0.5548,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6038647342995169,
|
| 1413 |
+
"grad_norm": 1.3164908246322924,
|
| 1414 |
+
"learning_rate": 4.067235969733937e-06,
|
| 1415 |
+
"loss": 0.5457,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6068840579710145,
|
| 1420 |
+
"grad_norm": 1.303498686637293,
|
| 1421 |
+
"learning_rate": 4.015502825721537e-06,
|
| 1422 |
+
"loss": 0.5409,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6099033816425121,
|
| 1427 |
+
"grad_norm": 1.2724683031487374,
|
| 1428 |
+
"learning_rate": 3.963879096911751e-06,
|
| 1429 |
+
"loss": 0.5312,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6129227053140096,
|
| 1434 |
+
"grad_norm": 1.2726486865020912,
|
| 1435 |
+
"learning_rate": 3.91237052067072e-06,
|
| 1436 |
+
"loss": 0.5401,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6159420289855072,
|
| 1441 |
+
"grad_norm": 1.343440037009517,
|
| 1442 |
+
"learning_rate": 3.860982821566729e-06,
|
| 1443 |
+
"loss": 0.5301,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6189613526570048,
|
| 1448 |
+
"grad_norm": 1.353619163212925,
|
| 1449 |
+
"learning_rate": 3.8097217107340107e-06,
|
| 1450 |
+
"loss": 0.5365,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6219806763285024,
|
| 1455 |
+
"grad_norm": 1.3477813018813276,
|
| 1456 |
+
"learning_rate": 3.7585928852380025e-06,
|
| 1457 |
+
"loss": 0.5393,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.625,
|
| 1462 |
+
"grad_norm": 1.3731933102965312,
|
| 1463 |
+
"learning_rate": 3.7076020274421996e-06,
|
| 1464 |
+
"loss": 0.5464,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6280193236714976,
|
| 1469 |
+
"grad_norm": 1.3390473233403632,
|
| 1470 |
+
"learning_rate": 3.6567548043766157e-06,
|
| 1471 |
+
"loss": 0.5444,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6310386473429952,
|
| 1476 |
+
"grad_norm": 1.3046493242835504,
|
| 1477 |
+
"learning_rate": 3.6060568671079658e-06,
|
| 1478 |
+
"loss": 0.5353,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6340579710144928,
|
| 1483 |
+
"grad_norm": 1.4419256695572908,
|
| 1484 |
+
"learning_rate": 3.5555138501116247e-06,
|
| 1485 |
+
"loss": 0.5244,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6370772946859904,
|
| 1490 |
+
"grad_norm": 1.2976760979714546,
|
| 1491 |
+
"learning_rate": 3.5051313706453995e-06,
|
| 1492 |
+
"loss": 0.5404,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6400966183574879,
|
| 1497 |
+
"grad_norm": 1.2919651906622562,
|
| 1498 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1499 |
+
"loss": 0.5183,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6431159420289855,
|
| 1504 |
+
"grad_norm": 1.3937729306554194,
|
| 1505 |
+
"learning_rate": 3.40487040350303e-06,
|
| 1506 |
+
"loss": 0.5279,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6461352657004831,
|
| 1511 |
+
"grad_norm": 1.4754894489154289,
|
| 1512 |
+
"learning_rate": 3.355003058646105e-06,
|
| 1513 |
+
"loss": 0.5262,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6491545893719807,
|
| 1518 |
+
"grad_norm": 1.39657778630638,
|
| 1519 |
+
"learning_rate": 3.305318535719343e-06,
|
| 1520 |
+
"loss": 0.5375,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6521739130434783,
|
| 1525 |
+
"grad_norm": 1.350395780922992,
|
| 1526 |
+
"learning_rate": 3.2558223565691104e-06,
|
| 1527 |
+
"loss": 0.5238,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6551932367149759,
|
| 1532 |
+
"grad_norm": 1.3685872873162936,
|
| 1533 |
+
"learning_rate": 3.2065200221095905e-06,
|
| 1534 |
+
"loss": 0.5298,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6582125603864735,
|
| 1539 |
+
"grad_norm": 1.3233219201374211,
|
| 1540 |
+
"learning_rate": 3.1574170117114293e-06,
|
| 1541 |
+
"loss": 0.5184,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6612318840579711,
|
| 1546 |
+
"grad_norm": 1.3288039422625464,
|
| 1547 |
+
"learning_rate": 3.1085187825927555e-06,
|
| 1548 |
+
"loss": 0.5278,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6642512077294686,
|
| 1553 |
+
"grad_norm": 1.3601324138928483,
|
| 1554 |
+
"learning_rate": 3.0598307692126904e-06,
|
| 1555 |
+
"loss": 0.5216,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6672705314009661,
|
| 1560 |
+
"grad_norm": 1.2974323886442631,
|
| 1561 |
+
"learning_rate": 3.0113583826673655e-06,
|
| 1562 |
+
"loss": 0.5281,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6702898550724637,
|
| 1567 |
+
"grad_norm": 1.3817579947889416,
|
| 1568 |
+
"learning_rate": 2.9631070100885373e-06,
|
| 1569 |
+
"loss": 0.5344,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6733091787439613,
|
| 1574 |
+
"grad_norm": 1.373246896186375,
|
| 1575 |
+
"learning_rate": 2.915082014044883e-06,
|
| 1576 |
+
"loss": 0.5118,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6763285024154589,
|
| 1581 |
+
"grad_norm": 1.3914109961429577,
|
| 1582 |
+
"learning_rate": 2.867288731946004e-06,
|
| 1583 |
+
"loss": 0.5219,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6793478260869565,
|
| 1588 |
+
"grad_norm": 1.3456129482520864,
|
| 1589 |
+
"learning_rate": 2.8197324754492456e-06,
|
| 1590 |
+
"loss": 0.5123,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6823671497584541,
|
| 1595 |
+
"grad_norm": 1.3516053089371904,
|
| 1596 |
+
"learning_rate": 2.7724185298693596e-06,
|
| 1597 |
+
"loss": 0.5211,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6853864734299517,
|
| 1602 |
+
"grad_norm": 1.6512789595808264,
|
| 1603 |
+
"learning_rate": 2.7253521535911144e-06,
|
| 1604 |
+
"loss": 0.5069,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6884057971014492,
|
| 1609 |
+
"grad_norm": 1.3963436494331172,
|
| 1610 |
+
"learning_rate": 2.678538577484871e-06,
|
| 1611 |
+
"loss": 0.5163,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6914251207729468,
|
| 1616 |
+
"grad_norm": 1.327652930753795,
|
| 1617 |
+
"learning_rate": 2.6319830043252616e-06,
|
| 1618 |
+
"loss": 0.5062,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6944444444444444,
|
| 1623 |
+
"grad_norm": 1.3883407663503802,
|
| 1624 |
+
"learning_rate": 2.5856906082129313e-06,
|
| 1625 |
+
"loss": 0.5099,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.697463768115942,
|
| 1630 |
+
"grad_norm": 1.3541895548105385,
|
| 1631 |
+
"learning_rate": 2.53966653399952e-06,
|
| 1632 |
+
"loss": 0.5151,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7004830917874396,
|
| 1637 |
+
"grad_norm": 1.6671674084618924,
|
| 1638 |
+
"learning_rate": 2.4939158967158657e-06,
|
| 1639 |
+
"loss": 0.5234,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7035024154589372,
|
| 1644 |
+
"grad_norm": 1.4194726508864435,
|
| 1645 |
+
"learning_rate": 2.448443781003527e-06,
|
| 1646 |
+
"loss": 0.5009,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7065217391304348,
|
| 1651 |
+
"grad_norm": 1.4328105129938522,
|
| 1652 |
+
"learning_rate": 2.403255240549693e-06,
|
| 1653 |
+
"loss": 0.5055,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7095410628019324,
|
| 1658 |
+
"grad_norm": 1.4936167581096147,
|
| 1659 |
+
"learning_rate": 2.3583552975255108e-06,
|
| 1660 |
+
"loss": 0.5163,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7125603864734299,
|
| 1665 |
+
"grad_norm": 1.366182685953503,
|
| 1666 |
+
"learning_rate": 2.313748942027956e-06,
|
| 1667 |
+
"loss": 0.5014,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7155797101449275,
|
| 1672 |
+
"grad_norm": 1.4219678396595035,
|
| 1673 |
+
"learning_rate": 2.269441131525213e-06,
|
| 1674 |
+
"loss": 0.5043,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7185990338164251,
|
| 1679 |
+
"grad_norm": 1.3494187919079106,
|
| 1680 |
+
"learning_rate": 2.225436790305733e-06,
|
| 1681 |
+
"loss": 0.5017,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7216183574879227,
|
| 1686 |
+
"grad_norm": 1.3118718901779507,
|
| 1687 |
+
"learning_rate": 2.181740808930947e-06,
|
| 1688 |
+
"loss": 0.4854,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7246376811594203,
|
| 1693 |
+
"grad_norm": 1.410018165296287,
|
| 1694 |
+
"learning_rate": 2.1383580436917452e-06,
|
| 1695 |
+
"loss": 0.4977,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7276570048309179,
|
| 1700 |
+
"grad_norm": 1.4006801036455434,
|
| 1701 |
+
"learning_rate": 2.0952933160687456e-06,
|
| 1702 |
+
"loss": 0.4985,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7306763285024155,
|
| 1707 |
+
"grad_norm": 1.4335252946653403,
|
| 1708 |
+
"learning_rate": 2.052551412196456e-06,
|
| 1709 |
+
"loss": 0.4939,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7336956521739131,
|
| 1714 |
+
"grad_norm": 1.3979757625658358,
|
| 1715 |
+
"learning_rate": 2.010137082331354e-06,
|
| 1716 |
+
"loss": 0.5048,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7367149758454107,
|
| 1721 |
+
"grad_norm": 1.2854697344016206,
|
| 1722 |
+
"learning_rate": 1.96805504032393e-06,
|
| 1723 |
+
"loss": 0.5047,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7397342995169082,
|
| 1728 |
+
"grad_norm": 1.3683323463695072,
|
| 1729 |
+
"learning_rate": 1.9263099630948274e-06,
|
| 1730 |
+
"loss": 0.5042,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7427536231884058,
|
| 1735 |
+
"grad_norm": 1.398514848998387,
|
| 1736 |
+
"learning_rate": 1.8849064901150372e-06,
|
| 1737 |
+
"loss": 0.473,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7457729468599034,
|
| 1742 |
+
"grad_norm": 1.4022273688712774,
|
| 1743 |
+
"learning_rate": 1.8438492228902893e-06,
|
| 1744 |
+
"loss": 0.4995,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.748792270531401,
|
| 1749 |
+
"grad_norm": 1.293867549241767,
|
| 1750 |
+
"learning_rate": 1.8031427244496357e-06,
|
| 1751 |
+
"loss": 0.5009,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7518115942028986,
|
| 1756 |
+
"grad_norm": 1.2978900181373894,
|
| 1757 |
+
"learning_rate": 1.7627915188383382e-06,
|
| 1758 |
+
"loss": 0.4901,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7548309178743962,
|
| 1763 |
+
"grad_norm": 1.4448114494500728,
|
| 1764 |
+
"learning_rate": 1.7228000906150672e-06,
|
| 1765 |
+
"loss": 0.5034,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7578502415458938,
|
| 1770 |
+
"grad_norm": 1.3817066267485774,
|
| 1771 |
+
"learning_rate": 1.6831728843534962e-06,
|
| 1772 |
+
"loss": 0.4928,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7608695652173914,
|
| 1777 |
+
"grad_norm": 1.5474646288550162,
|
| 1778 |
+
"learning_rate": 1.6439143041483352e-06,
|
| 1779 |
+
"loss": 0.4753,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7638888888888888,
|
| 1784 |
+
"grad_norm": 1.3836879466510739,
|
| 1785 |
+
"learning_rate": 1.6050287131258862e-06,
|
| 1786 |
+
"loss": 0.4929,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7669082125603864,
|
| 1791 |
+
"grad_norm": 1.352247909907531,
|
| 1792 |
+
"learning_rate": 1.5665204329591066e-06,
|
| 1793 |
+
"loss": 0.4915,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.769927536231884,
|
| 1798 |
+
"grad_norm": 1.3795806073347745,
|
| 1799 |
+
"learning_rate": 1.528393743387328e-06,
|
| 1800 |
+
"loss": 0.486,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7729468599033816,
|
| 1805 |
+
"grad_norm": 1.3399181762266046,
|
| 1806 |
+
"learning_rate": 1.4906528817406052e-06,
|
| 1807 |
+
"loss": 0.4765,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7759661835748792,
|
| 1812 |
+
"grad_norm": 1.432793399552329,
|
| 1813 |
+
"learning_rate": 1.453302042468786e-06,
|
| 1814 |
+
"loss": 0.4793,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7789855072463768,
|
| 1819 |
+
"grad_norm": 1.440038491030672,
|
| 1820 |
+
"learning_rate": 1.4163453766753537e-06,
|
| 1821 |
+
"loss": 0.489,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7820048309178744,
|
| 1826 |
+
"grad_norm": 1.3868984310668784,
|
| 1827 |
+
"learning_rate": 1.3797869916560692e-06,
|
| 1828 |
+
"loss": 0.4933,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.785024154589372,
|
| 1833 |
+
"grad_norm": 1.382680178173945,
|
| 1834 |
+
"learning_rate": 1.3436309504425137e-06,
|
| 1835 |
+
"loss": 0.4816,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7880434782608695,
|
| 1840 |
+
"grad_norm": 1.3699907284486668,
|
| 1841 |
+
"learning_rate": 1.3078812713505079e-06,
|
| 1842 |
+
"loss": 0.4836,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7910628019323671,
|
| 1847 |
+
"grad_norm": 1.3748734244080174,
|
| 1848 |
+
"learning_rate": 1.2725419275335404e-06,
|
| 1849 |
+
"loss": 0.4787,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7940821256038647,
|
| 1854 |
+
"grad_norm": 1.4416418877555286,
|
| 1855 |
+
"learning_rate": 1.237616846541192e-06,
|
| 1856 |
+
"loss": 0.4861,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7971014492753623,
|
| 1861 |
+
"grad_norm": 1.4582733607311282,
|
| 1862 |
+
"learning_rate": 1.2031099098826376e-06,
|
| 1863 |
+
"loss": 0.4902,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8001207729468599,
|
| 1868 |
+
"grad_norm": 1.3951099319990903,
|
| 1869 |
+
"learning_rate": 1.1690249525952569e-06,
|
| 1870 |
+
"loss": 0.4698,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8031400966183575,
|
| 1875 |
+
"grad_norm": 1.3211083162574102,
|
| 1876 |
+
"learning_rate": 1.1353657628184217e-06,
|
| 1877 |
+
"loss": 0.4654,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8061594202898551,
|
| 1882 |
+
"grad_norm": 1.39861267444744,
|
| 1883 |
+
"learning_rate": 1.1021360813724924e-06,
|
| 1884 |
+
"loss": 0.4814,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8091787439613527,
|
| 1889 |
+
"grad_norm": 1.3855723067060435,
|
| 1890 |
+
"learning_rate": 1.0693396013430552e-06,
|
| 1891 |
+
"loss": 0.4744,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8121980676328503,
|
| 1896 |
+
"grad_norm": 1.4445315702507702,
|
| 1897 |
+
"learning_rate": 1.036979967670494e-06,
|
| 1898 |
+
"loss": 0.4724,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8152173913043478,
|
| 1903 |
+
"grad_norm": 1.4005662044580711,
|
| 1904 |
+
"learning_rate": 1.0050607767448928e-06,
|
| 1905 |
+
"loss": 0.4692,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8182367149758454,
|
| 1910 |
+
"grad_norm": 1.3423573516670364,
|
| 1911 |
+
"learning_rate": 9.735855760063412e-07,
|
| 1912 |
+
"loss": 0.4769,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.821256038647343,
|
| 1917 |
+
"grad_norm": 1.3388950541234161,
|
| 1918 |
+
"learning_rate": 9.425578635506721e-07,
|
| 1919 |
+
"loss": 0.4801,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8242753623188406,
|
| 1924 |
+
"grad_norm": 1.3520391195043011,
|
| 1925 |
+
"learning_rate": 9.119810877406998e-07,
|
| 1926 |
+
"loss": 0.4812,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8272946859903382,
|
| 1931 |
+
"grad_norm": 1.364798918561803,
|
| 1932 |
+
"learning_rate": 8.818586468229695e-07,
|
| 1933 |
+
"loss": 0.4694,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8303140096618358,
|
| 1938 |
+
"grad_norm": 1.3718140961001366,
|
| 1939 |
+
"learning_rate": 8.521938885500825e-07,
|
| 1940 |
+
"loss": 0.4695,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8333333333333334,
|
| 1945 |
+
"grad_norm": 1.3996474606181322,
|
| 1946 |
+
"learning_rate": 8.229901098086335e-07,
|
| 1947 |
+
"loss": 0.4873,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.836352657004831,
|
| 1952 |
+
"grad_norm": 1.3941499294551503,
|
| 1953 |
+
"learning_rate": 7.942505562528024e-07,
|
| 1954 |
+
"loss": 0.4812,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8393719806763285,
|
| 1959 |
+
"grad_norm": 1.4317232059058598,
|
| 1960 |
+
"learning_rate": 7.659784219436373e-07,
|
| 1961 |
+
"loss": 0.4775,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.842391304347826,
|
| 1966 |
+
"grad_norm": 1.3368338252764984,
|
| 1967 |
+
"learning_rate": 7.381768489940678e-07,
|
| 1968 |
+
"loss": 0.4834,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8454106280193237,
|
| 1973 |
+
"grad_norm": 1.376767316724929,
|
| 1974 |
+
"learning_rate": 7.108489272197089e-07,
|
| 1975 |
+
"loss": 0.4777,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8484299516908212,
|
| 1980 |
+
"grad_norm": 1.4416146692158918,
|
| 1981 |
+
"learning_rate": 6.839976937954479e-07,
|
| 1982 |
+
"loss": 0.4805,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8514492753623188,
|
| 1987 |
+
"grad_norm": 1.3884346338383682,
|
| 1988 |
+
"learning_rate": 6.576261329179123e-07,
|
| 1989 |
+
"loss": 0.4659,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8544685990338164,
|
| 1994 |
+
"grad_norm": 1.461661357035372,
|
| 1995 |
+
"learning_rate": 6.317371754738044e-07,
|
| 1996 |
+
"loss": 0.4835,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.857487922705314,
|
| 2001 |
+
"grad_norm": 1.3155537677854425,
|
| 2002 |
+
"learning_rate": 6.06333698714171e-07,
|
| 2003 |
+
"loss": 0.4612,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8605072463768116,
|
| 2008 |
+
"grad_norm": 1.430549233517389,
|
| 2009 |
+
"learning_rate": 5.814185259346267e-07,
|
| 2010 |
+
"loss": 0.4652,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8635265700483091,
|
| 2015 |
+
"grad_norm": 1.5438182616910798,
|
| 2016 |
+
"learning_rate": 5.56994426161584e-07,
|
| 2017 |
+
"loss": 0.4816,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8665458937198067,
|
| 2022 |
+
"grad_norm": 1.3944382700425997,
|
| 2023 |
+
"learning_rate": 5.330641138445064e-07,
|
| 2024 |
+
"loss": 0.4668,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8695652173913043,
|
| 2029 |
+
"grad_norm": 1.3524996972490821,
|
| 2030 |
+
"learning_rate": 5.096302485542265e-07,
|
| 2031 |
+
"loss": 0.4626,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8725845410628019,
|
| 2036 |
+
"grad_norm": 1.3677343406038267,
|
| 2037 |
+
"learning_rate": 4.866954346873715e-07,
|
| 2038 |
+
"loss": 0.4658,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8756038647342995,
|
| 2043 |
+
"grad_norm": 1.3464641098418646,
|
| 2044 |
+
"learning_rate": 4.642622211769099e-07,
|
| 2045 |
+
"loss": 0.4685,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8786231884057971,
|
| 2050 |
+
"grad_norm": 1.390326156036712,
|
| 2051 |
+
"learning_rate": 4.4233310120887387e-07,
|
| 2052 |
+
"loss": 0.4591,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8816425120772947,
|
| 2057 |
+
"grad_norm": 1.4052855328615665,
|
| 2058 |
+
"learning_rate": 4.209105119452628e-07,
|
| 2059 |
+
"loss": 0.4581,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8846618357487923,
|
| 2064 |
+
"grad_norm": 1.4105848251200876,
|
| 2065 |
+
"learning_rate": 3.999968342531918e-07,
|
| 2066 |
+
"loss": 0.4707,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8876811594202898,
|
| 2071 |
+
"grad_norm": 1.333713097837263,
|
| 2072 |
+
"learning_rate": 3.7959439244027727e-07,
|
| 2073 |
+
"loss": 0.4647,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8907004830917874,
|
| 2078 |
+
"grad_norm": 1.356947580984905,
|
| 2079 |
+
"learning_rate": 3.5970545399632574e-07,
|
| 2080 |
+
"loss": 0.4572,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.893719806763285,
|
| 2085 |
+
"grad_norm": 1.3949650199499017,
|
| 2086 |
+
"learning_rate": 3.4033222934131914e-07,
|
| 2087 |
+
"loss": 0.466,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8967391304347826,
|
| 2092 |
+
"grad_norm": 1.485033631723171,
|
| 2093 |
+
"learning_rate": 3.214768715797656e-07,
|
| 2094 |
+
"loss": 0.4735,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8997584541062802,
|
| 2099 |
+
"grad_norm": 1.3806955768204767,
|
| 2100 |
+
"learning_rate": 3.0314147626139543e-07,
|
| 2101 |
+
"loss": 0.4671,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9027777777777778,
|
| 2106 |
+
"grad_norm": 1.4448739661661896,
|
| 2107 |
+
"learning_rate": 2.853280811482734e-07,
|
| 2108 |
+
"loss": 0.4746,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9057971014492754,
|
| 2113 |
+
"grad_norm": 1.4190366783020854,
|
| 2114 |
+
"learning_rate": 2.6803866598832216e-07,
|
| 2115 |
+
"loss": 0.4544,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.908816425120773,
|
| 2120 |
+
"grad_norm": 1.3988884827045063,
|
| 2121 |
+
"learning_rate": 2.5127515229529665e-07,
|
| 2122 |
+
"loss": 0.4659,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9118357487922706,
|
| 2127 |
+
"grad_norm": 1.3539792184833292,
|
| 2128 |
+
"learning_rate": 2.350394031352343e-07,
|
| 2129 |
+
"loss": 0.4683,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9148550724637681,
|
| 2134 |
+
"grad_norm": 1.3390472231534103,
|
| 2135 |
+
"learning_rate": 2.1933322291938897e-07,
|
| 2136 |
+
"loss": 0.4587,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9178743961352657,
|
| 2141 |
+
"grad_norm": 1.2982303956915597,
|
| 2142 |
+
"learning_rate": 2.041583572037037e-07,
|
| 2143 |
+
"loss": 0.4705,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9208937198067633,
|
| 2148 |
+
"grad_norm": 1.398922259288625,
|
| 2149 |
+
"learning_rate": 1.8951649249480287e-07,
|
| 2150 |
+
"loss": 0.4674,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9239130434782609,
|
| 2155 |
+
"grad_norm": 1.3706043106328762,
|
| 2156 |
+
"learning_rate": 1.7540925606256088e-07,
|
| 2157 |
+
"loss": 0.4738,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9269323671497585,
|
| 2162 |
+
"grad_norm": 1.4051091915658231,
|
| 2163 |
+
"learning_rate": 1.6183821575925186e-07,
|
| 2164 |
+
"loss": 0.4678,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9299516908212561,
|
| 2169 |
+
"grad_norm": 1.3953334052871644,
|
| 2170 |
+
"learning_rate": 1.4880487984529846e-07,
|
| 2171 |
+
"loss": 0.4789,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9329710144927537,
|
| 2176 |
+
"grad_norm": 1.4473194738633308,
|
| 2177 |
+
"learning_rate": 1.363106968216482e-07,
|
| 2178 |
+
"loss": 0.474,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9359903381642513,
|
| 2183 |
+
"grad_norm": 1.3696328617667175,
|
| 2184 |
+
"learning_rate": 1.243570552687895e-07,
|
| 2185 |
+
"loss": 0.4549,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9390096618357487,
|
| 2190 |
+
"grad_norm": 1.4005643653184559,
|
| 2191 |
+
"learning_rate": 1.1294528369242663e-07,
|
| 2192 |
+
"loss": 0.4545,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9420289855072463,
|
| 2197 |
+
"grad_norm": 1.3929824476134263,
|
| 2198 |
+
"learning_rate": 1.020766503758347e-07,
|
| 2199 |
+
"loss": 0.4793,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9450483091787439,
|
| 2204 |
+
"grad_norm": 1.3824037261570252,
|
| 2205 |
+
"learning_rate": 9.175236323890058e-08,
|
| 2206 |
+
"loss": 0.4773,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9480676328502415,
|
| 2211 |
+
"grad_norm": 1.4006805147153165,
|
| 2212 |
+
"learning_rate": 8.197356970388148e-08,
|
| 2213 |
+
"loss": 0.4672,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9510869565217391,
|
| 2218 |
+
"grad_norm": 1.3886975460206281,
|
| 2219 |
+
"learning_rate": 7.274135656787917e-08,
|
| 2220 |
+
"loss": 0.4526,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9541062801932367,
|
| 2225 |
+
"grad_norm": 1.4838070116928936,
|
| 2226 |
+
"learning_rate": 6.405674988205602e-08,
|
| 2227 |
+
"loss": 0.4572,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9571256038647343,
|
| 2232 |
+
"grad_norm": 1.4235233786161414,
|
| 2233 |
+
"learning_rate": 5.592071483760397e-08,
|
| 2234 |
+
"loss": 0.4599,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9601449275362319,
|
| 2239 |
+
"grad_norm": 1.402808009230018,
|
| 2240 |
+
"learning_rate": 4.833415565847155e-08,
|
| 2241 |
+
"loss": 0.4634,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9631642512077294,
|
| 2246 |
+
"grad_norm": 1.397718470953625,
|
| 2247 |
+
"learning_rate": 4.1297915500873034e-08,
|
| 2248 |
+
"loss": 0.4605,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.966183574879227,
|
| 2253 |
+
"grad_norm": 1.3622090829571152,
|
| 2254 |
+
"learning_rate": 3.481277635957903e-08,
|
| 2255 |
+
"loss": 0.4644,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9692028985507246,
|
| 2260 |
+
"grad_norm": 1.4192128536229283,
|
| 2261 |
+
"learning_rate": 2.88794589810093e-08,
|
| 2262 |
+
"loss": 0.4513,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9722222222222222,
|
| 2267 |
+
"grad_norm": 1.4607543198292594,
|
| 2268 |
+
"learning_rate": 2.3498622783128533e-08,
|
| 2269 |
+
"loss": 0.4643,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9752415458937198,
|
| 2274 |
+
"grad_norm": 1.4519572105646263,
|
| 2275 |
+
"learning_rate": 1.8670865782161042e-08,
|
| 2276 |
+
"loss": 0.4737,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9782608695652174,
|
| 2281 |
+
"grad_norm": 1.5587138432284067,
|
| 2282 |
+
"learning_rate": 1.4396724526127282e-08,
|
| 2283 |
+
"loss": 0.4719,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.981280193236715,
|
| 2288 |
+
"grad_norm": 1.3504712276503201,
|
| 2289 |
+
"learning_rate": 1.067667403521433e-08,
|
| 2290 |
+
"loss": 0.4597,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9842995169082126,
|
| 2295 |
+
"grad_norm": 1.3482863134408452,
|
| 2296 |
+
"learning_rate": 7.51112774898144e-09,
|
| 2297 |
+
"loss": 0.4678,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9873188405797102,
|
| 2302 |
+
"grad_norm": 1.3970409533706654,
|
| 2303 |
+
"learning_rate": 4.900437480413467e-09,
|
| 2304 |
+
"loss": 0.4718,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9903381642512077,
|
| 2309 |
+
"grad_norm": 1.4616532167767295,
|
| 2310 |
+
"learning_rate": 2.844893376816593e-09,
|
| 2311 |
+
"loss": 0.4653,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9933574879227053,
|
| 2316 |
+
"grad_norm": 1.3667351204068807,
|
| 2317 |
+
"learning_rate": 1.3447238875774482e-09,
|
| 2318 |
+
"loss": 0.468,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9963768115942029,
|
| 2323 |
+
"grad_norm": 1.3873085164877705,
|
| 2324 |
+
"learning_rate": 4.000957387700899e-10,
|
| 2325 |
+
"loss": 0.4735,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9993961352657005,
|
| 2330 |
+
"grad_norm": 1.369426340332071,
|
| 2331 |
+
"learning_rate": 1.1113914626381672e-11,
|
| 2332 |
+
"loss": 0.4693,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 1.0,
|
| 2337 |
+
"eval_runtime": 3.6076,
|
| 2338 |
+
"eval_samples_per_second": 2.772,
|
| 2339 |
+
"eval_steps_per_second": 0.832,
|
| 2340 |
+
"step": 1656
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 1.0,
|
| 2344 |
+
"step": 1656,
|
| 2345 |
+
"total_flos": 173366354903040.0,
|
| 2346 |
+
"train_loss": 0.6223726071021407,
|
| 2347 |
+
"train_runtime": 17321.1955,
|
| 2348 |
+
"train_samples_per_second": 1.529,
|
| 2349 |
+
"train_steps_per_second": 0.096
|
| 2350 |
+
}
|
| 2351 |
+
],
|
| 2352 |
+
"logging_steps": 5,
|
| 2353 |
+
"max_steps": 1656,
|
| 2354 |
+
"num_input_tokens_seen": 0,
|
| 2355 |
+
"num_train_epochs": 1,
|
| 2356 |
+
"save_steps": 100,
|
| 2357 |
+
"stateful_callbacks": {
|
| 2358 |
+
"TrainerControl": {
|
| 2359 |
+
"args": {
|
| 2360 |
+
"should_epoch_stop": false,
|
| 2361 |
+
"should_evaluate": false,
|
| 2362 |
+
"should_log": false,
|
| 2363 |
+
"should_save": true,
|
| 2364 |
+
"should_training_stop": true
|
| 2365 |
+
},
|
| 2366 |
+
"attributes": {}
|
| 2367 |
+
}
|
| 2368 |
+
},
|
| 2369 |
+
"total_flos": 173366354903040.0,
|
| 2370 |
+
"train_batch_size": 4,
|
| 2371 |
+
"trial_name": null,
|
| 2372 |
+
"trial_params": null
|
| 2373 |
+
}
|