Model save
Browse files- README.md +62 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- train_results.json +9 -0
- trainer_state.json +2387 -0
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
+
base_model: gpt2
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gpt2-cpt-dutch
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# gpt2-cpt-dutch
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 16
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- total_train_batch_size: 64
|
| 47 |
+
- total_eval_batch_size: 32
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
### Framework versions
|
| 58 |
+
|
| 59 |
+
- Transformers 4.44.2
|
| 60 |
+
- Pytorch 2.1.2
|
| 61 |
+
- Datasets 3.0.0
|
| 62 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.396344619008e+16,
|
| 4 |
+
"train_loss": 2.9739012581145694,
|
| 5 |
+
"train_runtime": 9187.7213,
|
| 6 |
+
"train_samples": 125938,
|
| 7 |
+
"train_samples_per_second": 11.629,
|
| 8 |
+
"train_steps_per_second": 0.182
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.44.2"
|
| 6 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.396344619008e+16,
|
| 4 |
+
"train_loss": 2.9739012581145694,
|
| 5 |
+
"train_runtime": 9187.7213,
|
| 6 |
+
"train_samples": 125938,
|
| 7 |
+
"train_samples_per_second": 11.629,
|
| 8 |
+
"train_steps_per_second": 0.182
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2387 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1670,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0005988023952095808,
|
| 13 |
+
"grad_norm": 5.398237705230713,
|
| 14 |
+
"learning_rate": 1.1976047904191619e-06,
|
| 15 |
+
"loss": 5.2288,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0029940119760479044,
|
| 20 |
+
"grad_norm": 4.768432140350342,
|
| 21 |
+
"learning_rate": 5.9880239520958085e-06,
|
| 22 |
+
"loss": 5.1682,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.005988023952095809,
|
| 27 |
+
"grad_norm": 4.144229412078857,
|
| 28 |
+
"learning_rate": 1.1976047904191617e-05,
|
| 29 |
+
"loss": 5.1486,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008982035928143712,
|
| 34 |
+
"grad_norm": 3.3602042198181152,
|
| 35 |
+
"learning_rate": 1.7964071856287426e-05,
|
| 36 |
+
"loss": 5.0451,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011976047904191617,
|
| 41 |
+
"grad_norm": 3.0493969917297363,
|
| 42 |
+
"learning_rate": 2.3952095808383234e-05,
|
| 43 |
+
"loss": 4.9569,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.014970059880239521,
|
| 48 |
+
"grad_norm": 2.564634323120117,
|
| 49 |
+
"learning_rate": 2.994011976047904e-05,
|
| 50 |
+
"loss": 4.8593,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.017964071856287425,
|
| 55 |
+
"grad_norm": 2.1028244495391846,
|
| 56 |
+
"learning_rate": 3.592814371257485e-05,
|
| 57 |
+
"loss": 4.7325,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.020958083832335328,
|
| 62 |
+
"grad_norm": 2.0943355560302734,
|
| 63 |
+
"learning_rate": 4.191616766467066e-05,
|
| 64 |
+
"loss": 4.6032,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.023952095808383235,
|
| 69 |
+
"grad_norm": 2.1831929683685303,
|
| 70 |
+
"learning_rate": 4.790419161676647e-05,
|
| 71 |
+
"loss": 4.5291,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02694610778443114,
|
| 76 |
+
"grad_norm": 2.2486143112182617,
|
| 77 |
+
"learning_rate": 5.389221556886228e-05,
|
| 78 |
+
"loss": 4.4485,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.029940119760479042,
|
| 83 |
+
"grad_norm": 1.977081537246704,
|
| 84 |
+
"learning_rate": 5.988023952095808e-05,
|
| 85 |
+
"loss": 4.3569,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03293413173652695,
|
| 90 |
+
"grad_norm": 2.1440491676330566,
|
| 91 |
+
"learning_rate": 6.58682634730539e-05,
|
| 92 |
+
"loss": 4.2673,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03592814371257485,
|
| 97 |
+
"grad_norm": 1.845775842666626,
|
| 98 |
+
"learning_rate": 7.18562874251497e-05,
|
| 99 |
+
"loss": 4.1614,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.038922155688622756,
|
| 104 |
+
"grad_norm": 2.0755035877227783,
|
| 105 |
+
"learning_rate": 7.784431137724552e-05,
|
| 106 |
+
"loss": 4.1784,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.041916167664670656,
|
| 111 |
+
"grad_norm": 2.0788209438323975,
|
| 112 |
+
"learning_rate": 8.383233532934131e-05,
|
| 113 |
+
"loss": 4.0587,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04491017964071856,
|
| 118 |
+
"grad_norm": 2.0177366733551025,
|
| 119 |
+
"learning_rate": 8.982035928143712e-05,
|
| 120 |
+
"loss": 4.0414,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04790419161676647,
|
| 125 |
+
"grad_norm": 1.9424874782562256,
|
| 126 |
+
"learning_rate": 9.580838323353294e-05,
|
| 127 |
+
"loss": 3.9817,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05089820359281437,
|
| 132 |
+
"grad_norm": 2.1245999336242676,
|
| 133 |
+
"learning_rate": 0.00010179640718562875,
|
| 134 |
+
"loss": 3.8902,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05389221556886228,
|
| 139 |
+
"grad_norm": 1.7750043869018555,
|
| 140 |
+
"learning_rate": 0.00010778443113772456,
|
| 141 |
+
"loss": 3.8991,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.05688622754491018,
|
| 146 |
+
"grad_norm": 1.888543725013733,
|
| 147 |
+
"learning_rate": 0.00011377245508982037,
|
| 148 |
+
"loss": 3.8088,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.059880239520958084,
|
| 153 |
+
"grad_norm": 1.696926236152649,
|
| 154 |
+
"learning_rate": 0.00011976047904191617,
|
| 155 |
+
"loss": 3.814,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06287425149700598,
|
| 160 |
+
"grad_norm": 1.7164498567581177,
|
| 161 |
+
"learning_rate": 0.00012574850299401196,
|
| 162 |
+
"loss": 3.7896,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0658682634730539,
|
| 167 |
+
"grad_norm": 1.924892544746399,
|
| 168 |
+
"learning_rate": 0.0001317365269461078,
|
| 169 |
+
"loss": 3.733,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0688622754491018,
|
| 174 |
+
"grad_norm": 2.0697178840637207,
|
| 175 |
+
"learning_rate": 0.00013772455089820359,
|
| 176 |
+
"loss": 3.7365,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0718562874251497,
|
| 181 |
+
"grad_norm": 1.8498303890228271,
|
| 182 |
+
"learning_rate": 0.0001437125748502994,
|
| 183 |
+
"loss": 3.6889,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.0748502994011976,
|
| 188 |
+
"grad_norm": 1.799263596534729,
|
| 189 |
+
"learning_rate": 0.0001497005988023952,
|
| 190 |
+
"loss": 3.6785,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.07784431137724551,
|
| 195 |
+
"grad_norm": 1.8041021823883057,
|
| 196 |
+
"learning_rate": 0.00015568862275449103,
|
| 197 |
+
"loss": 3.6193,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08083832335329341,
|
| 202 |
+
"grad_norm": 2.032904863357544,
|
| 203 |
+
"learning_rate": 0.00016167664670658683,
|
| 204 |
+
"loss": 3.5986,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08383233532934131,
|
| 209 |
+
"grad_norm": 2.044524908065796,
|
| 210 |
+
"learning_rate": 0.00016766467065868263,
|
| 211 |
+
"loss": 3.549,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.08682634730538923,
|
| 216 |
+
"grad_norm": 2.168320655822754,
|
| 217 |
+
"learning_rate": 0.00017365269461077845,
|
| 218 |
+
"loss": 3.5249,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.08982035928143713,
|
| 223 |
+
"grad_norm": 1.8781085014343262,
|
| 224 |
+
"learning_rate": 0.00017964071856287425,
|
| 225 |
+
"loss": 3.5475,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09281437125748503,
|
| 230 |
+
"grad_norm": 1.6960855722427368,
|
| 231 |
+
"learning_rate": 0.00018562874251497007,
|
| 232 |
+
"loss": 3.5204,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.09580838323353294,
|
| 237 |
+
"grad_norm": 1.9243298768997192,
|
| 238 |
+
"learning_rate": 0.00019161676646706587,
|
| 239 |
+
"loss": 3.5206,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.09880239520958084,
|
| 244 |
+
"grad_norm": 2.000701904296875,
|
| 245 |
+
"learning_rate": 0.0001976047904191617,
|
| 246 |
+
"loss": 3.5163,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10179640718562874,
|
| 251 |
+
"grad_norm": 1.6321606636047363,
|
| 252 |
+
"learning_rate": 0.00019999803395762152,
|
| 253 |
+
"loss": 3.4382,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10479041916167664,
|
| 258 |
+
"grad_norm": 1.5973315238952637,
|
| 259 |
+
"learning_rate": 0.00019998601953415373,
|
| 260 |
+
"loss": 3.4415,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.10778443113772455,
|
| 265 |
+
"grad_norm": 1.6465246677398682,
|
| 266 |
+
"learning_rate": 0.00019996308424365594,
|
| 267 |
+
"loss": 3.4091,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11077844311377245,
|
| 272 |
+
"grad_norm": 1.6946362257003784,
|
| 273 |
+
"learning_rate": 0.00019992923059121106,
|
| 274 |
+
"loss": 3.3752,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11377245508982035,
|
| 279 |
+
"grad_norm": 1.602065086364746,
|
| 280 |
+
"learning_rate": 0.0001998844622744483,
|
| 281 |
+
"loss": 3.415,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.11676646706586827,
|
| 286 |
+
"grad_norm": 1.8944158554077148,
|
| 287 |
+
"learning_rate": 0.0001998287841831396,
|
| 288 |
+
"loss": 3.3535,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.11976047904191617,
|
| 293 |
+
"grad_norm": 2.0829012393951416,
|
| 294 |
+
"learning_rate": 0.00019976220239866562,
|
| 295 |
+
"loss": 3.3339,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12275449101796407,
|
| 300 |
+
"grad_norm": 1.8381649255752563,
|
| 301 |
+
"learning_rate": 0.00019968472419335106,
|
| 302 |
+
"loss": 3.3609,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12574850299401197,
|
| 307 |
+
"grad_norm": 1.864918828010559,
|
| 308 |
+
"learning_rate": 0.00019959635802967087,
|
| 309 |
+
"loss": 3.3285,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.12874251497005987,
|
| 314 |
+
"grad_norm": 2.0552523136138916,
|
| 315 |
+
"learning_rate": 0.00019949711355932566,
|
| 316 |
+
"loss": 3.3141,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.1317365269461078,
|
| 321 |
+
"grad_norm": 1.4688948392868042,
|
| 322 |
+
"learning_rate": 0.0001993870016221875,
|
| 323 |
+
"loss": 3.3043,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1347305389221557,
|
| 328 |
+
"grad_norm": 1.7277065515518188,
|
| 329 |
+
"learning_rate": 0.000199266034245116,
|
| 330 |
+
"loss": 3.3307,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1377245508982036,
|
| 335 |
+
"grad_norm": 1.6926337480545044,
|
| 336 |
+
"learning_rate": 0.0001991342246406448,
|
| 337 |
+
"loss": 3.302,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1407185628742515,
|
| 342 |
+
"grad_norm": 1.7383888959884644,
|
| 343 |
+
"learning_rate": 0.00019899158720553824,
|
| 344 |
+
"loss": 3.2949,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1437125748502994,
|
| 349 |
+
"grad_norm": 1.4267903566360474,
|
| 350 |
+
"learning_rate": 0.00019883813751921903,
|
| 351 |
+
"loss": 3.2932,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.1467065868263473,
|
| 356 |
+
"grad_norm": 1.4597861766815186,
|
| 357 |
+
"learning_rate": 0.00019867389234206654,
|
| 358 |
+
"loss": 3.2629,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1497005988023952,
|
| 363 |
+
"grad_norm": 1.6646181344985962,
|
| 364 |
+
"learning_rate": 0.00019849886961358621,
|
| 365 |
+
"loss": 3.2334,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15269461077844312,
|
| 370 |
+
"grad_norm": 1.46786367893219,
|
| 371 |
+
"learning_rate": 0.0001983130884504501,
|
| 372 |
+
"loss": 3.2368,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.15568862275449102,
|
| 377 |
+
"grad_norm": 1.3595575094223022,
|
| 378 |
+
"learning_rate": 0.00019811656914440885,
|
| 379 |
+
"loss": 3.2124,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.15868263473053892,
|
| 384 |
+
"grad_norm": 1.7106595039367676,
|
| 385 |
+
"learning_rate": 0.0001979093331600754,
|
| 386 |
+
"loss": 3.2399,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16167664670658682,
|
| 391 |
+
"grad_norm": 1.7180885076522827,
|
| 392 |
+
"learning_rate": 0.0001976914031325806,
|
| 393 |
+
"loss": 3.1848,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16467065868263472,
|
| 398 |
+
"grad_norm": 1.5397311449050903,
|
| 399 |
+
"learning_rate": 0.0001974628028651007,
|
| 400 |
+
"loss": 3.1918,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.16766467065868262,
|
| 405 |
+
"grad_norm": 1.2814067602157593,
|
| 406 |
+
"learning_rate": 0.00019722355732625774,
|
| 407 |
+
"loss": 3.1661,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.17065868263473055,
|
| 412 |
+
"grad_norm": 1.3572536706924438,
|
| 413 |
+
"learning_rate": 0.0001969736926473921,
|
| 414 |
+
"loss": 3.2363,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17365269461077845,
|
| 419 |
+
"grad_norm": 1.1298632621765137,
|
| 420 |
+
"learning_rate": 0.0001967132361197086,
|
| 421 |
+
"loss": 3.1441,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.17664670658682635,
|
| 426 |
+
"grad_norm": 1.3632957935333252,
|
| 427 |
+
"learning_rate": 0.00019644221619129548,
|
| 428 |
+
"loss": 3.1402,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.17964071856287425,
|
| 433 |
+
"grad_norm": 1.4946106672286987,
|
| 434 |
+
"learning_rate": 0.00019616066246401717,
|
| 435 |
+
"loss": 3.1352,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18263473053892215,
|
| 440 |
+
"grad_norm": 1.3740743398666382,
|
| 441 |
+
"learning_rate": 0.00019586860569028124,
|
| 442 |
+
"loss": 3.1609,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.18562874251497005,
|
| 447 |
+
"grad_norm": 1.359784483909607,
|
| 448 |
+
"learning_rate": 0.0001955660777696793,
|
| 449 |
+
"loss": 3.1191,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.18862275449101795,
|
| 454 |
+
"grad_norm": 1.0663578510284424,
|
| 455 |
+
"learning_rate": 0.00019525311174550285,
|
| 456 |
+
"loss": 3.1508,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.19161676646706588,
|
| 461 |
+
"grad_norm": 1.2853511571884155,
|
| 462 |
+
"learning_rate": 0.00019492974180113426,
|
| 463 |
+
"loss": 3.1197,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.19461077844311378,
|
| 468 |
+
"grad_norm": 1.2358075380325317,
|
| 469 |
+
"learning_rate": 0.00019459600325631303,
|
| 470 |
+
"loss": 3.0737,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.19760479041916168,
|
| 475 |
+
"grad_norm": 1.2823344469070435,
|
| 476 |
+
"learning_rate": 0.0001942519325632781,
|
| 477 |
+
"loss": 3.1486,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.20059880239520958,
|
| 482 |
+
"grad_norm": 1.0905367136001587,
|
| 483 |
+
"learning_rate": 0.00019389756730278627,
|
| 484 |
+
"loss": 3.1254,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.20359281437125748,
|
| 489 |
+
"grad_norm": 1.1666399240493774,
|
| 490 |
+
"learning_rate": 0.00019353294618000758,
|
| 491 |
+
"loss": 3.0982,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.20658682634730538,
|
| 496 |
+
"grad_norm": 1.352597951889038,
|
| 497 |
+
"learning_rate": 0.00019315810902029786,
|
| 498 |
+
"loss": 3.1079,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.20958083832335328,
|
| 503 |
+
"grad_norm": 1.5166900157928467,
|
| 504 |
+
"learning_rate": 0.00019277309676484858,
|
| 505 |
+
"loss": 3.1142,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2125748502994012,
|
| 510 |
+
"grad_norm": 1.1076050996780396,
|
| 511 |
+
"learning_rate": 0.0001923779514662154,
|
| 512 |
+
"loss": 3.0817,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2155688622754491,
|
| 517 |
+
"grad_norm": 1.1077985763549805,
|
| 518 |
+
"learning_rate": 0.00019197271628372482,
|
| 519 |
+
"loss": 3.0804,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.218562874251497,
|
| 524 |
+
"grad_norm": 1.0608789920806885,
|
| 525 |
+
"learning_rate": 0.00019155743547876023,
|
| 526 |
+
"loss": 3.0823,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2215568862275449,
|
| 531 |
+
"grad_norm": 1.2678635120391846,
|
| 532 |
+
"learning_rate": 0.00019113215440992752,
|
| 533 |
+
"loss": 3.0845,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2245508982035928,
|
| 538 |
+
"grad_norm": 1.1210081577301025,
|
| 539 |
+
"learning_rate": 0.0001906969195281007,
|
| 540 |
+
"loss": 3.066,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2275449101796407,
|
| 545 |
+
"grad_norm": 1.3343095779418945,
|
| 546 |
+
"learning_rate": 0.00019025177837134858,
|
| 547 |
+
"loss": 3.0512,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23053892215568864,
|
| 552 |
+
"grad_norm": 1.1427243947982788,
|
| 553 |
+
"learning_rate": 0.00018979677955974228,
|
| 554 |
+
"loss": 3.073,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.23353293413173654,
|
| 559 |
+
"grad_norm": 1.2000013589859009,
|
| 560 |
+
"learning_rate": 0.0001893319727900448,
|
| 561 |
+
"loss": 3.0551,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.23652694610778444,
|
| 566 |
+
"grad_norm": 1.0942010879516602,
|
| 567 |
+
"learning_rate": 0.0001888574088302831,
|
| 568 |
+
"loss": 3.0414,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.23952095808383234,
|
| 573 |
+
"grad_norm": 0.9435099363327026,
|
| 574 |
+
"learning_rate": 0.00018837313951420272,
|
| 575 |
+
"loss": 3.0231,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24251497005988024,
|
| 580 |
+
"grad_norm": 1.0394734144210815,
|
| 581 |
+
"learning_rate": 0.00018787921773560657,
|
| 582 |
+
"loss": 3.0175,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.24550898203592814,
|
| 587 |
+
"grad_norm": 0.9441319704055786,
|
| 588 |
+
"learning_rate": 0.00018737569744257756,
|
| 589 |
+
"loss": 3.0385,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.24850299401197604,
|
| 594 |
+
"grad_norm": 1.020687460899353,
|
| 595 |
+
"learning_rate": 0.00018686263363158602,
|
| 596 |
+
"loss": 3.0505,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.25149700598802394,
|
| 601 |
+
"grad_norm": 0.9998067617416382,
|
| 602 |
+
"learning_rate": 0.0001863400823414831,
|
| 603 |
+
"loss": 3.0358,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25449101796407186,
|
| 608 |
+
"grad_norm": 1.0903741121292114,
|
| 609 |
+
"learning_rate": 0.00018580810064737965,
|
| 610 |
+
"loss": 3.0195,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.25748502994011974,
|
| 615 |
+
"grad_norm": 1.0015404224395752,
|
| 616 |
+
"learning_rate": 0.00018526674665441257,
|
| 617 |
+
"loss": 3.0362,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.26047904191616766,
|
| 622 |
+
"grad_norm": 0.9371241927146912,
|
| 623 |
+
"learning_rate": 0.00018471607949139803,
|
| 624 |
+
"loss": 3.0123,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2634730538922156,
|
| 629 |
+
"grad_norm": 0.9184331893920898,
|
| 630 |
+
"learning_rate": 0.00018415615930437337,
|
| 631 |
+
"loss": 3.0075,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.26646706586826346,
|
| 636 |
+
"grad_norm": 1.0484634637832642,
|
| 637 |
+
"learning_rate": 0.00018358704725002768,
|
| 638 |
+
"loss": 3.0029,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2694610778443114,
|
| 643 |
+
"grad_norm": 1.0423493385314941,
|
| 644 |
+
"learning_rate": 0.000183008805489022,
|
| 645 |
+
"loss": 2.9913,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.27245508982035926,
|
| 650 |
+
"grad_norm": 1.2088091373443604,
|
| 651 |
+
"learning_rate": 0.00018242149717919993,
|
| 652 |
+
"loss": 2.9834,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2754491017964072,
|
| 657 |
+
"grad_norm": 0.8996514678001404,
|
| 658 |
+
"learning_rate": 0.0001818251864686893,
|
| 659 |
+
"loss": 2.972,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.27844311377245506,
|
| 664 |
+
"grad_norm": 1.014440894126892,
|
| 665 |
+
"learning_rate": 0.00018121993848889552,
|
| 666 |
+
"loss": 2.9687,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.281437125748503,
|
| 671 |
+
"grad_norm": 0.8995974063873291,
|
| 672 |
+
"learning_rate": 0.00018060581934738784,
|
| 673 |
+
"loss": 2.9317,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2844311377245509,
|
| 678 |
+
"grad_norm": 1.0253899097442627,
|
| 679 |
+
"learning_rate": 0.00017998289612067864,
|
| 680 |
+
"loss": 2.9695,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2874251497005988,
|
| 685 |
+
"grad_norm": 0.954319179058075,
|
| 686 |
+
"learning_rate": 0.00017935123684689733,
|
| 687 |
+
"loss": 2.9884,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2904191616766467,
|
| 692 |
+
"grad_norm": 0.8713873028755188,
|
| 693 |
+
"learning_rate": 0.00017871091051835874,
|
| 694 |
+
"loss": 2.9689,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2934131736526946,
|
| 699 |
+
"grad_norm": 0.9465665817260742,
|
| 700 |
+
"learning_rate": 0.00017806198707402752,
|
| 701 |
+
"loss": 2.9073,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.2964071856287425,
|
| 706 |
+
"grad_norm": 0.9693202972412109,
|
| 707 |
+
"learning_rate": 0.00017740453739187922,
|
| 708 |
+
"loss": 2.953,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2994011976047904,
|
| 713 |
+
"grad_norm": 0.9308465719223022,
|
| 714 |
+
"learning_rate": 0.0001767386332811587,
|
| 715 |
+
"loss": 2.9439,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3023952095808383,
|
| 720 |
+
"grad_norm": 0.8614184260368347,
|
| 721 |
+
"learning_rate": 0.0001760643474745368,
|
| 722 |
+
"loss": 2.9567,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.30538922155688625,
|
| 727 |
+
"grad_norm": 0.913870632648468,
|
| 728 |
+
"learning_rate": 0.00017538175362016622,
|
| 729 |
+
"loss": 2.9447,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3083832335329341,
|
| 734 |
+
"grad_norm": 0.8058397769927979,
|
| 735 |
+
"learning_rate": 0.00017469092627363738,
|
| 736 |
+
"loss": 2.9677,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.31137724550898205,
|
| 741 |
+
"grad_norm": 0.7762190699577332,
|
| 742 |
+
"learning_rate": 0.00017399194088983511,
|
| 743 |
+
"loss": 2.9242,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3143712574850299,
|
| 748 |
+
"grad_norm": 0.8444433808326721,
|
| 749 |
+
"learning_rate": 0.0001732848738146973,
|
| 750 |
+
"loss": 2.9612,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.31736526946107785,
|
| 755 |
+
"grad_norm": 0.7365703582763672,
|
| 756 |
+
"learning_rate": 0.00017256980227687595,
|
| 757 |
+
"loss": 2.9629,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3203592814371258,
|
| 762 |
+
"grad_norm": 0.841096818447113,
|
| 763 |
+
"learning_rate": 0.00017184680437930198,
|
| 764 |
+
"loss": 2.9349,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.32335329341317365,
|
| 769 |
+
"grad_norm": 0.7778092622756958,
|
| 770 |
+
"learning_rate": 0.00017111595909065466,
|
| 771 |
+
"loss": 2.9147,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3263473053892216,
|
| 776 |
+
"grad_norm": 0.8155597448348999,
|
| 777 |
+
"learning_rate": 0.00017037734623673615,
|
| 778 |
+
"loss": 2.9255,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.32934131736526945,
|
| 783 |
+
"grad_norm": 0.693092942237854,
|
| 784 |
+
"learning_rate": 0.00016963104649175272,
|
| 785 |
+
"loss": 2.902,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3323353293413174,
|
| 790 |
+
"grad_norm": 0.8276283144950867,
|
| 791 |
+
"learning_rate": 0.0001688771413695032,
|
| 792 |
+
"loss": 2.9414,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.33532934131736525,
|
| 797 |
+
"grad_norm": 0.8656878471374512,
|
| 798 |
+
"learning_rate": 0.00016811571321447566,
|
| 799 |
+
"loss": 2.9105,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3383233532934132,
|
| 804 |
+
"grad_norm": 0.713503897190094,
|
| 805 |
+
"learning_rate": 0.00016734684519285344,
|
| 806 |
+
"loss": 2.9138,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3413173652694611,
|
| 811 |
+
"grad_norm": 0.7443877458572388,
|
| 812 |
+
"learning_rate": 0.00016657062128343144,
|
| 813 |
+
"loss": 2.8801,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.344311377245509,
|
| 818 |
+
"grad_norm": 0.8257678151130676,
|
| 819 |
+
"learning_rate": 0.00016578712626844365,
|
| 820 |
+
"loss": 2.8803,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3473053892215569,
|
| 825 |
+
"grad_norm": 0.8291301131248474,
|
| 826 |
+
"learning_rate": 0.00016499644572430278,
|
| 827 |
+
"loss": 2.9076,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3502994011976048,
|
| 832 |
+
"grad_norm": 0.738528311252594,
|
| 833 |
+
"learning_rate": 0.0001641986660122534,
|
| 834 |
+
"loss": 2.8978,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3532934131736527,
|
| 839 |
+
"grad_norm": 0.7673987150192261,
|
| 840 |
+
"learning_rate": 0.00016339387426893918,
|
| 841 |
+
"loss": 2.9111,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.3562874251497006,
|
| 846 |
+
"grad_norm": 0.8196045756340027,
|
| 847 |
+
"learning_rate": 0.0001625821583968855,
|
| 848 |
+
"loss": 2.8467,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3592814371257485,
|
| 853 |
+
"grad_norm": 0.7220514416694641,
|
| 854 |
+
"learning_rate": 0.00016176360705489823,
|
| 855 |
+
"loss": 2.8808,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.36227544910179643,
|
| 860 |
+
"grad_norm": 0.8424475193023682,
|
| 861 |
+
"learning_rate": 0.00016093830964838035,
|
| 862 |
+
"loss": 2.858,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.3652694610778443,
|
| 867 |
+
"grad_norm": 0.7498836517333984,
|
| 868 |
+
"learning_rate": 0.00016010635631956652,
|
| 869 |
+
"loss": 2.8987,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.36826347305389223,
|
| 874 |
+
"grad_norm": 0.7145126461982727,
|
| 875 |
+
"learning_rate": 0.0001592678379376775,
|
| 876 |
+
"loss": 2.8769,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.3712574850299401,
|
| 881 |
+
"grad_norm": 0.6803230047225952,
|
| 882 |
+
"learning_rate": 0.0001584228460889949,
|
| 883 |
+
"loss": 2.8812,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.37425149700598803,
|
| 888 |
+
"grad_norm": 0.6997132897377014,
|
| 889 |
+
"learning_rate": 0.00015757147306685808,
|
| 890 |
+
"loss": 2.8886,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3772455089820359,
|
| 895 |
+
"grad_norm": 0.7119357585906982,
|
| 896 |
+
"learning_rate": 0.00015671381186158312,
|
| 897 |
+
"loss": 2.9011,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.38023952095808383,
|
| 902 |
+
"grad_norm": 0.8835166096687317,
|
| 903 |
+
"learning_rate": 0.00015584995615030634,
|
| 904 |
+
"loss": 2.8876,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.38323353293413176,
|
| 909 |
+
"grad_norm": 0.7940571904182434,
|
| 910 |
+
"learning_rate": 0.0001549800002867524,
|
| 911 |
+
"loss": 2.8739,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.38622754491017963,
|
| 916 |
+
"grad_norm": 0.7138169407844543,
|
| 917 |
+
"learning_rate": 0.00015410403929092857,
|
| 918 |
+
"loss": 2.869,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.38922155688622756,
|
| 923 |
+
"grad_norm": 0.711258053779602,
|
| 924 |
+
"learning_rate": 0.00015322216883874643,
|
| 925 |
+
"loss": 2.878,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.39221556886227543,
|
| 930 |
+
"grad_norm": 0.730831503868103,
|
| 931 |
+
"learning_rate": 0.0001523344852515716,
|
| 932 |
+
"loss": 2.8573,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.39520958083832336,
|
| 937 |
+
"grad_norm": 0.7164433598518372,
|
| 938 |
+
"learning_rate": 0.00015144108548570322,
|
| 939 |
+
"loss": 2.8375,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.39820359281437123,
|
| 944 |
+
"grad_norm": 0.7521623969078064,
|
| 945 |
+
"learning_rate": 0.000150542067121784,
|
| 946 |
+
"loss": 2.8714,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.40119760479041916,
|
| 951 |
+
"grad_norm": 0.6829126477241516,
|
| 952 |
+
"learning_rate": 0.00014963752835414203,
|
| 953 |
+
"loss": 2.8614,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4041916167664671,
|
| 958 |
+
"grad_norm": 0.6588369011878967,
|
| 959 |
+
"learning_rate": 0.00014872756798006576,
|
| 960 |
+
"loss": 2.8498,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.40718562874251496,
|
| 965 |
+
"grad_norm": 0.670486569404602,
|
| 966 |
+
"learning_rate": 0.00014781228538901267,
|
| 967 |
+
"loss": 2.853,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4101796407185629,
|
| 972 |
+
"grad_norm": 0.6251630783081055,
|
| 973 |
+
"learning_rate": 0.00014689178055175394,
|
| 974 |
+
"loss": 2.8709,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.41317365269461076,
|
| 979 |
+
"grad_norm": 0.6421228647232056,
|
| 980 |
+
"learning_rate": 0.00014596615400945496,
|
| 981 |
+
"loss": 2.8585,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4161676646706587,
|
| 986 |
+
"grad_norm": 0.7128710746765137,
|
| 987 |
+
"learning_rate": 0.0001450355068626939,
|
| 988 |
+
"loss": 2.8333,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.41916167664670656,
|
| 993 |
+
"grad_norm": 0.7639265060424805,
|
| 994 |
+
"learning_rate": 0.0001440999407604192,
|
| 995 |
+
"loss": 2.8495,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4221556886227545,
|
| 1000 |
+
"grad_norm": 0.7937913537025452,
|
| 1001 |
+
"learning_rate": 0.00014315955788884698,
|
| 1002 |
+
"loss": 2.8667,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4251497005988024,
|
| 1007 |
+
"grad_norm": 0.7219135761260986,
|
| 1008 |
+
"learning_rate": 0.00014221446096029992,
|
| 1009 |
+
"loss": 2.8137,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4281437125748503,
|
| 1014 |
+
"grad_norm": 0.6893067359924316,
|
| 1015 |
+
"learning_rate": 0.00014126475320198843,
|
| 1016 |
+
"loss": 2.8147,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4311377245508982,
|
| 1021 |
+
"grad_norm": 0.7299759387969971,
|
| 1022 |
+
"learning_rate": 0.00014031053834473613,
|
| 1023 |
+
"loss": 2.8157,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4341317365269461,
|
| 1028 |
+
"grad_norm": 0.6203035116195679,
|
| 1029 |
+
"learning_rate": 0.00013935192061164956,
|
| 1030 |
+
"loss": 2.842,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.437125748502994,
|
| 1035 |
+
"grad_norm": 0.67593914270401,
|
| 1036 |
+
"learning_rate": 0.0001383890047067348,
|
| 1037 |
+
"loss": 2.8281,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.44011976047904194,
|
| 1042 |
+
"grad_norm": 0.629625678062439,
|
| 1043 |
+
"learning_rate": 0.0001374218958034612,
|
| 1044 |
+
"loss": 2.8321,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.4431137724550898,
|
| 1049 |
+
"grad_norm": 0.7435698509216309,
|
| 1050 |
+
"learning_rate": 0.0001364506995332739,
|
| 1051 |
+
"loss": 2.8069,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.44610778443113774,
|
| 1056 |
+
"grad_norm": 0.7014979124069214,
|
| 1057 |
+
"learning_rate": 0.00013547552197405632,
|
| 1058 |
+
"loss": 2.8271,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4491017964071856,
|
| 1063 |
+
"grad_norm": 0.6280860900878906,
|
| 1064 |
+
"learning_rate": 0.00013449646963854396,
|
| 1065 |
+
"loss": 2.7851,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.45209580838323354,
|
| 1070 |
+
"grad_norm": 0.6429287195205688,
|
| 1071 |
+
"learning_rate": 0.00013351364946269072,
|
| 1072 |
+
"loss": 2.8342,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4550898203592814,
|
| 1077 |
+
"grad_norm": 0.6879150867462158,
|
| 1078 |
+
"learning_rate": 0.00013252716879398884,
|
| 1079 |
+
"loss": 2.8017,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.45808383233532934,
|
| 1084 |
+
"grad_norm": 0.6519134044647217,
|
| 1085 |
+
"learning_rate": 0.00013153713537974394,
|
| 1086 |
+
"loss": 2.8272,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.46107784431137727,
|
| 1091 |
+
"grad_norm": 0.7050421833992004,
|
| 1092 |
+
"learning_rate": 0.00013054365735530664,
|
| 1093 |
+
"loss": 2.8046,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.46407185628742514,
|
| 1098 |
+
"grad_norm": 0.6754117012023926,
|
| 1099 |
+
"learning_rate": 0.00012954684323226136,
|
| 1100 |
+
"loss": 2.7996,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.46706586826347307,
|
| 1105 |
+
"grad_norm": 0.6119791269302368,
|
| 1106 |
+
"learning_rate": 0.00012854680188657437,
|
| 1107 |
+
"loss": 2.825,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.47005988023952094,
|
| 1112 |
+
"grad_norm": 0.7737504839897156,
|
| 1113 |
+
"learning_rate": 0.00012754364254670192,
|
| 1114 |
+
"loss": 2.8229,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.47305389221556887,
|
| 1119 |
+
"grad_norm": 0.6356379985809326,
|
| 1120 |
+
"learning_rate": 0.00012653747478165987,
|
| 1121 |
+
"loss": 2.7901,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.47604790419161674,
|
| 1126 |
+
"grad_norm": 0.679063081741333,
|
| 1127 |
+
"learning_rate": 0.0001255284084890562,
|
| 1128 |
+
"loss": 2.7847,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.47904191616766467,
|
| 1133 |
+
"grad_norm": 0.6137480139732361,
|
| 1134 |
+
"learning_rate": 0.0001245165538830873,
|
| 1135 |
+
"loss": 2.7921,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4820359281437126,
|
| 1140 |
+
"grad_norm": 0.6218224167823792,
|
| 1141 |
+
"learning_rate": 0.00012350202148250037,
|
| 1142 |
+
"loss": 2.8368,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.48502994011976047,
|
| 1147 |
+
"grad_norm": 0.6735543012619019,
|
| 1148 |
+
"learning_rate": 0.0001224849220985218,
|
| 1149 |
+
"loss": 2.7808,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4880239520958084,
|
| 1154 |
+
"grad_norm": 0.5928522944450378,
|
| 1155 |
+
"learning_rate": 0.00012146536682275387,
|
| 1156 |
+
"loss": 2.7922,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.49101796407185627,
|
| 1161 |
+
"grad_norm": 0.640355110168457,
|
| 1162 |
+
"learning_rate": 0.00012044346701504128,
|
| 1163 |
+
"loss": 2.8388,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4940119760479042,
|
| 1168 |
+
"grad_norm": 0.5893163084983826,
|
| 1169 |
+
"learning_rate": 0.00011941933429130758,
|
| 1170 |
+
"loss": 2.8005,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.49700598802395207,
|
| 1175 |
+
"grad_norm": 0.5506909489631653,
|
| 1176 |
+
"learning_rate": 0.0001183930805113643,
|
| 1177 |
+
"loss": 2.8018,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5,
|
| 1182 |
+
"grad_norm": 0.6417153477668762,
|
| 1183 |
+
"learning_rate": 0.00011736481776669306,
|
| 1184 |
+
"loss": 2.7829,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5029940119760479,
|
| 1189 |
+
"grad_norm": 0.6764044761657715,
|
| 1190 |
+
"learning_rate": 0.00011633465836820243,
|
| 1191 |
+
"loss": 2.7853,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5059880239520959,
|
| 1196 |
+
"grad_norm": 0.6103654503822327,
|
| 1197 |
+
"learning_rate": 0.00011530271483396115,
|
| 1198 |
+
"loss": 2.7922,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5089820359281437,
|
| 1203 |
+
"grad_norm": 0.6200462579727173,
|
| 1204 |
+
"learning_rate": 0.00011426909987690819,
|
| 1205 |
+
"loss": 2.798,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5119760479041916,
|
| 1210 |
+
"grad_norm": 0.556036114692688,
|
| 1211 |
+
"learning_rate": 0.00011323392639254193,
|
| 1212 |
+
"loss": 2.823,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5149700598802395,
|
| 1217 |
+
"grad_norm": 0.5561334490776062,
|
| 1218 |
+
"learning_rate": 0.00011219730744658921,
|
| 1219 |
+
"loss": 2.7946,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5179640718562875,
|
| 1224 |
+
"grad_norm": 0.6291623711585999,
|
| 1225 |
+
"learning_rate": 0.00011115935626265594,
|
| 1226 |
+
"loss": 2.769,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5209580838323353,
|
| 1231 |
+
"grad_norm": 0.622279703617096,
|
| 1232 |
+
"learning_rate": 0.00011012018620986028,
|
| 1233 |
+
"loss": 2.7905,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5239520958083832,
|
| 1238 |
+
"grad_norm": 0.5651618242263794,
|
| 1239 |
+
"learning_rate": 0.00010907991079045006,
|
| 1240 |
+
"loss": 2.7611,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5269461077844312,
|
| 1245 |
+
"grad_norm": 0.5883045792579651,
|
| 1246 |
+
"learning_rate": 0.00010803864362740562,
|
| 1247 |
+
"loss": 2.7853,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5299401197604791,
|
| 1252 |
+
"grad_norm": 0.5984832048416138,
|
| 1253 |
+
"learning_rate": 0.00010699649845202934,
|
| 1254 |
+
"loss": 2.7686,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5329341317365269,
|
| 1259 |
+
"grad_norm": 0.631034791469574,
|
| 1260 |
+
"learning_rate": 0.00010595358909152378,
|
| 1261 |
+
"loss": 2.7649,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5359281437125748,
|
| 1266 |
+
"grad_norm": 0.5855838656425476,
|
| 1267 |
+
"learning_rate": 0.00010491002945655861,
|
| 1268 |
+
"loss": 2.7972,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5389221556886228,
|
| 1273 |
+
"grad_norm": 0.644307017326355,
|
| 1274 |
+
"learning_rate": 0.00010386593352882909,
|
| 1275 |
+
"loss": 2.7894,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5419161676646707,
|
| 1280 |
+
"grad_norm": 0.6458478569984436,
|
| 1281 |
+
"learning_rate": 0.0001028214153486066,
|
| 1282 |
+
"loss": 2.7875,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5449101796407185,
|
| 1287 |
+
"grad_norm": 0.6464332342147827,
|
| 1288 |
+
"learning_rate": 0.00010177658900228249,
|
| 1289 |
+
"loss": 2.7967,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5479041916167665,
|
| 1294 |
+
"grad_norm": 0.564663827419281,
|
| 1295 |
+
"learning_rate": 0.0001007315686099072,
|
| 1296 |
+
"loss": 2.8149,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5508982035928144,
|
| 1301 |
+
"grad_norm": 0.669620931148529,
|
| 1302 |
+
"learning_rate": 9.96864683127257e-05,
|
| 1303 |
+
"loss": 2.773,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5538922155688623,
|
| 1308 |
+
"grad_norm": 0.6294126510620117,
|
| 1309 |
+
"learning_rate": 9.864140226071053e-05,
|
| 1310 |
+
"loss": 2.7909,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5568862275449101,
|
| 1315 |
+
"grad_norm": 0.5996381044387817,
|
| 1316 |
+
"learning_rate": 9.759648460009376e-05,
|
| 1317 |
+
"loss": 2.7736,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5598802395209581,
|
| 1322 |
+
"grad_norm": 0.6062589287757874,
|
| 1323 |
+
"learning_rate": 9.655182946089956e-05,
|
| 1324 |
+
"loss": 2.7693,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.562874251497006,
|
| 1329 |
+
"grad_norm": 0.6130212545394897,
|
| 1330 |
+
"learning_rate": 9.550755094447848e-05,
|
| 1331 |
+
"loss": 2.7422,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5658682634730539,
|
| 1336 |
+
"grad_norm": 0.6118180155754089,
|
| 1337 |
+
"learning_rate": 9.446376311104494e-05,
|
| 1338 |
+
"loss": 2.7847,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5688622754491018,
|
| 1343 |
+
"grad_norm": 0.6395862698554993,
|
| 1344 |
+
"learning_rate": 9.342057996721894e-05,
|
| 1345 |
+
"loss": 2.7557,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5718562874251497,
|
| 1350 |
+
"grad_norm": 0.5906466841697693,
|
| 1351 |
+
"learning_rate": 9.237811545357392e-05,
|
| 1352 |
+
"loss": 2.7821,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5748502994011976,
|
| 1357 |
+
"grad_norm": 0.6405777335166931,
|
| 1358 |
+
"learning_rate": 9.133648343219168e-05,
|
| 1359 |
+
"loss": 2.7561,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5778443113772455,
|
| 1364 |
+
"grad_norm": 0.537714421749115,
|
| 1365 |
+
"learning_rate": 9.029579767422592e-05,
|
| 1366 |
+
"loss": 2.7624,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5808383233532934,
|
| 1371 |
+
"grad_norm": 0.5861092805862427,
|
| 1372 |
+
"learning_rate": 8.925617184747584e-05,
|
| 1373 |
+
"loss": 2.7885,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5838323353293413,
|
| 1378 |
+
"grad_norm": 0.5366087555885315,
|
| 1379 |
+
"learning_rate": 8.821771950397066e-05,
|
| 1380 |
+
"loss": 2.7717,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5868263473053892,
|
| 1385 |
+
"grad_norm": 0.5149749517440796,
|
| 1386 |
+
"learning_rate": 8.718055406756714e-05,
|
| 1387 |
+
"loss": 2.7722,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5898203592814372,
|
| 1392 |
+
"grad_norm": 0.6070750951766968,
|
| 1393 |
+
"learning_rate": 8.614478882156103e-05,
|
| 1394 |
+
"loss": 2.7747,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.592814371257485,
|
| 1399 |
+
"grad_norm": 0.5566183924674988,
|
| 1400 |
+
"learning_rate": 8.51105368963137e-05,
|
| 1401 |
+
"loss": 2.7363,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5958083832335329,
|
| 1406 |
+
"grad_norm": 0.5821384191513062,
|
| 1407 |
+
"learning_rate": 8.407791125689578e-05,
|
| 1408 |
+
"loss": 2.7286,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.5988023952095808,
|
| 1413 |
+
"grad_norm": 0.5972596406936646,
|
| 1414 |
+
"learning_rate": 8.30470246907484e-05,
|
| 1415 |
+
"loss": 2.8008,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6017964071856288,
|
| 1420 |
+
"grad_norm": 0.5653948187828064,
|
| 1421 |
+
"learning_rate": 8.201798979536437e-05,
|
| 1422 |
+
"loss": 2.7381,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6047904191616766,
|
| 1427 |
+
"grad_norm": 0.5987522602081299,
|
| 1428 |
+
"learning_rate": 8.099091896598964e-05,
|
| 1429 |
+
"loss": 2.7171,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6077844311377245,
|
| 1434 |
+
"grad_norm": 0.6162002086639404,
|
| 1435 |
+
"learning_rate": 7.996592438334728e-05,
|
| 1436 |
+
"loss": 2.7595,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6107784431137725,
|
| 1441 |
+
"grad_norm": 0.5301627516746521,
|
| 1442 |
+
"learning_rate": 7.894311800138432e-05,
|
| 1443 |
+
"loss": 2.7199,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6137724550898204,
|
| 1448 |
+
"grad_norm": 0.5902780890464783,
|
| 1449 |
+
"learning_rate": 7.792261153504402e-05,
|
| 1450 |
+
"loss": 2.7465,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6167664670658682,
|
| 1455 |
+
"grad_norm": 0.5662907958030701,
|
| 1456 |
+
"learning_rate": 7.690451644806372e-05,
|
| 1457 |
+
"loss": 2.7463,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6197604790419161,
|
| 1462 |
+
"grad_norm": 0.5153141617774963,
|
| 1463 |
+
"learning_rate": 7.588894394080045e-05,
|
| 1464 |
+
"loss": 2.7497,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6227544910179641,
|
| 1469 |
+
"grad_norm": 0.5608941316604614,
|
| 1470 |
+
"learning_rate": 7.487600493808513e-05,
|
| 1471 |
+
"loss": 2.7587,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.625748502994012,
|
| 1476 |
+
"grad_norm": 0.5900591015815735,
|
| 1477 |
+
"learning_rate": 7.386581007710693e-05,
|
| 1478 |
+
"loss": 2.721,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6287425149700598,
|
| 1483 |
+
"grad_norm": 0.5923062562942505,
|
| 1484 |
+
"learning_rate": 7.285846969532907e-05,
|
| 1485 |
+
"loss": 2.7245,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6317365269461078,
|
| 1490 |
+
"grad_norm": 0.5622250437736511,
|
| 1491 |
+
"learning_rate": 7.185409381843727e-05,
|
| 1492 |
+
"loss": 2.7514,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6347305389221557,
|
| 1497 |
+
"grad_norm": 0.5878713130950928,
|
| 1498 |
+
"learning_rate": 7.085279214832233e-05,
|
| 1499 |
+
"loss": 2.7361,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6377245508982036,
|
| 1504 |
+
"grad_norm": 0.5391808748245239,
|
| 1505 |
+
"learning_rate": 6.985467405109815e-05,
|
| 1506 |
+
"loss": 2.7708,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6407185628742516,
|
| 1511 |
+
"grad_norm": 0.5266147255897522,
|
| 1512 |
+
"learning_rate": 6.885984854515623e-05,
|
| 1513 |
+
"loss": 2.739,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6437125748502994,
|
| 1518 |
+
"grad_norm": 0.5293008089065552,
|
| 1519 |
+
"learning_rate": 6.786842428925821e-05,
|
| 1520 |
+
"loss": 2.7338,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6467065868263473,
|
| 1525 |
+
"grad_norm": 0.5374533534049988,
|
| 1526 |
+
"learning_rate": 6.688050957066787e-05,
|
| 1527 |
+
"loss": 2.7521,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6497005988023952,
|
| 1532 |
+
"grad_norm": 0.519822895526886,
|
| 1533 |
+
"learning_rate": 6.58962122933234e-05,
|
| 1534 |
+
"loss": 2.7509,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6526946107784432,
|
| 1539 |
+
"grad_norm": 0.4745498597621918,
|
| 1540 |
+
"learning_rate": 6.491563996605198e-05,
|
| 1541 |
+
"loss": 2.6944,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.655688622754491,
|
| 1546 |
+
"grad_norm": 0.5428311228752136,
|
| 1547 |
+
"learning_rate": 6.393889969082691e-05,
|
| 1548 |
+
"loss": 2.746,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6586826347305389,
|
| 1553 |
+
"grad_norm": 0.6130102276802063,
|
| 1554 |
+
"learning_rate": 6.29660981510697e-05,
|
| 1555 |
+
"loss": 2.7142,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6616766467065869,
|
| 1560 |
+
"grad_norm": 0.565981388092041,
|
| 1561 |
+
"learning_rate": 6.199734159999769e-05,
|
| 1562 |
+
"loss": 2.7377,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6646706586826348,
|
| 1567 |
+
"grad_norm": 0.5873706340789795,
|
| 1568 |
+
"learning_rate": 6.103273584901856e-05,
|
| 1569 |
+
"loss": 2.7707,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6676646706586826,
|
| 1574 |
+
"grad_norm": 0.6199204325675964,
|
| 1575 |
+
"learning_rate": 6.007238625617333e-05,
|
| 1576 |
+
"loss": 2.7288,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6706586826347305,
|
| 1581 |
+
"grad_norm": 0.5575565695762634,
|
| 1582 |
+
"learning_rate": 5.911639771462858e-05,
|
| 1583 |
+
"loss": 2.7454,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6736526946107785,
|
| 1588 |
+
"grad_norm": 0.5518234372138977,
|
| 1589 |
+
"learning_rate": 5.8164874641219735e-05,
|
| 1590 |
+
"loss": 2.7345,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6766467065868264,
|
| 1595 |
+
"grad_norm": 0.5617517828941345,
|
| 1596 |
+
"learning_rate": 5.721792096504611e-05,
|
| 1597 |
+
"loss": 2.756,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6796407185628742,
|
| 1602 |
+
"grad_norm": 0.5457236170768738,
|
| 1603 |
+
"learning_rate": 5.627564011611961e-05,
|
| 1604 |
+
"loss": 2.7492,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6826347305389222,
|
| 1609 |
+
"grad_norm": 0.5588122606277466,
|
| 1610 |
+
"learning_rate": 5.5338135014067395e-05,
|
| 1611 |
+
"loss": 2.7385,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6856287425149701,
|
| 1616 |
+
"grad_norm": 0.5922512412071228,
|
| 1617 |
+
"learning_rate": 5.440550805689075e-05,
|
| 1618 |
+
"loss": 2.7128,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.688622754491018,
|
| 1623 |
+
"grad_norm": 0.5327529311180115,
|
| 1624 |
+
"learning_rate": 5.3477861109780835e-05,
|
| 1625 |
+
"loss": 2.7507,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6916167664670658,
|
| 1630 |
+
"grad_norm": 0.5234887003898621,
|
| 1631 |
+
"learning_rate": 5.255529549399234e-05,
|
| 1632 |
+
"loss": 2.77,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6946107784431138,
|
| 1637 |
+
"grad_norm": 0.5305109024047852,
|
| 1638 |
+
"learning_rate": 5.163791197577714e-05,
|
| 1639 |
+
"loss": 2.7765,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.6976047904191617,
|
| 1644 |
+
"grad_norm": 0.5282033085823059,
|
| 1645 |
+
"learning_rate": 5.0725810755377825e-05,
|
| 1646 |
+
"loss": 2.7518,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7005988023952096,
|
| 1651 |
+
"grad_norm": 0.5210415124893188,
|
| 1652 |
+
"learning_rate": 4.9819091456083644e-05,
|
| 1653 |
+
"loss": 2.7493,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7035928143712575,
|
| 1658 |
+
"grad_norm": 0.560530424118042,
|
| 1659 |
+
"learning_rate": 4.891785311334923e-05,
|
| 1660 |
+
"loss": 2.7429,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7065868263473054,
|
| 1665 |
+
"grad_norm": 0.5484561324119568,
|
| 1666 |
+
"learning_rate": 4.8022194163977494e-05,
|
| 1667 |
+
"loss": 2.7131,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7095808383233533,
|
| 1672 |
+
"grad_norm": 0.5590568780899048,
|
| 1673 |
+
"learning_rate": 4.713221243536816e-05,
|
| 1674 |
+
"loss": 2.715,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7125748502994012,
|
| 1679 |
+
"grad_norm": 0.5286269783973694,
|
| 1680 |
+
"learning_rate": 4.6248005134832394e-05,
|
| 1681 |
+
"loss": 2.7006,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7155688622754491,
|
| 1686 |
+
"grad_norm": 0.5053289532661438,
|
| 1687 |
+
"learning_rate": 4.5369668838975597e-05,
|
| 1688 |
+
"loss": 2.7375,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.718562874251497,
|
| 1693 |
+
"grad_norm": 0.5476385354995728,
|
| 1694 |
+
"learning_rate": 4.449729948314894e-05,
|
| 1695 |
+
"loss": 2.7402,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7215568862275449,
|
| 1700 |
+
"grad_norm": 0.5178956985473633,
|
| 1701 |
+
"learning_rate": 4.363099235097087e-05,
|
| 1702 |
+
"loss": 2.7,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7245508982035929,
|
| 1707 |
+
"grad_norm": 0.5388510227203369,
|
| 1708 |
+
"learning_rate": 4.277084206391989e-05,
|
| 1709 |
+
"loss": 2.7014,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7275449101796407,
|
| 1714 |
+
"grad_norm": 0.510368824005127,
|
| 1715 |
+
"learning_rate": 4.191694257099962e-05,
|
| 1716 |
+
"loss": 2.7267,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7305389221556886,
|
| 1721 |
+
"grad_norm": 0.563225269317627,
|
| 1722 |
+
"learning_rate": 4.10693871384773e-05,
|
| 1723 |
+
"loss": 2.7277,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7335329341317365,
|
| 1728 |
+
"grad_norm": 0.5411152243614197,
|
| 1729 |
+
"learning_rate": 4.022826833969692e-05,
|
| 1730 |
+
"loss": 2.7442,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7365269461077845,
|
| 1735 |
+
"grad_norm": 0.5123066902160645,
|
| 1736 |
+
"learning_rate": 3.93936780449679e-05,
|
| 1737 |
+
"loss": 2.7042,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7395209580838323,
|
| 1742 |
+
"grad_norm": 0.5124099254608154,
|
| 1743 |
+
"learning_rate": 3.856570741153087e-05,
|
| 1744 |
+
"loss": 2.7288,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7425149700598802,
|
| 1749 |
+
"grad_norm": 0.5201666355133057,
|
| 1750 |
+
"learning_rate": 3.774444687360082e-05,
|
| 1751 |
+
"loss": 2.7266,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7455089820359282,
|
| 1756 |
+
"grad_norm": 0.5414671301841736,
|
| 1757 |
+
"learning_rate": 3.692998613248977e-05,
|
| 1758 |
+
"loss": 2.7237,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7485029940119761,
|
| 1763 |
+
"grad_norm": 0.5279539823532104,
|
| 1764 |
+
"learning_rate": 3.6122414146809014e-05,
|
| 1765 |
+
"loss": 2.7114,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7514970059880239,
|
| 1770 |
+
"grad_norm": 0.5688903331756592,
|
| 1771 |
+
"learning_rate": 3.532181912275301e-05,
|
| 1772 |
+
"loss": 2.7434,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7544910179640718,
|
| 1777 |
+
"grad_norm": 0.5434580445289612,
|
| 1778 |
+
"learning_rate": 3.4528288504464844e-05,
|
| 1779 |
+
"loss": 2.7369,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7574850299401198,
|
| 1784 |
+
"grad_norm": 0.5412757992744446,
|
| 1785 |
+
"learning_rate": 3.3741908964485414e-05,
|
| 1786 |
+
"loss": 2.7672,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7604790419161677,
|
| 1791 |
+
"grad_norm": 0.5109795331954956,
|
| 1792 |
+
"learning_rate": 3.296276639428665e-05,
|
| 1793 |
+
"loss": 2.6753,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7634730538922155,
|
| 1798 |
+
"grad_norm": 0.5179950594902039,
|
| 1799 |
+
"learning_rate": 3.21909458948901e-05,
|
| 1800 |
+
"loss": 2.7301,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7664670658682635,
|
| 1805 |
+
"grad_norm": 0.5099706649780273,
|
| 1806 |
+
"learning_rate": 3.1426531767572e-05,
|
| 1807 |
+
"loss": 2.7018,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7694610778443114,
|
| 1812 |
+
"grad_norm": 0.5369354486465454,
|
| 1813 |
+
"learning_rate": 3.0669607504655326e-05,
|
| 1814 |
+
"loss": 2.7473,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7724550898203593,
|
| 1819 |
+
"grad_norm": 0.45971599221229553,
|
| 1820 |
+
"learning_rate": 2.9920255780390617e-05,
|
| 1821 |
+
"loss": 2.7035,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7754491017964071,
|
| 1826 |
+
"grad_norm": 0.49725160002708435,
|
| 1827 |
+
"learning_rate": 2.917855844192584e-05,
|
| 1828 |
+
"loss": 2.7258,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7784431137724551,
|
| 1833 |
+
"grad_norm": 0.48002126812934875,
|
| 1834 |
+
"learning_rate": 2.8444596500366825e-05,
|
| 1835 |
+
"loss": 2.6843,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.781437125748503,
|
| 1840 |
+
"grad_norm": 0.47459498047828674,
|
| 1841 |
+
"learning_rate": 2.7718450121928918e-05,
|
| 1842 |
+
"loss": 2.7133,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7844311377245509,
|
| 1847 |
+
"grad_norm": 0.5074918270111084,
|
| 1848 |
+
"learning_rate": 2.7000198619180794e-05,
|
| 1849 |
+
"loss": 2.7191,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7874251497005988,
|
| 1854 |
+
"grad_norm": 0.48583555221557617,
|
| 1855 |
+
"learning_rate": 2.6289920442381722e-05,
|
| 1856 |
+
"loss": 2.7248,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7904191616766467,
|
| 1861 |
+
"grad_norm": 0.5015809535980225,
|
| 1862 |
+
"learning_rate": 2.5587693170912875e-05,
|
| 1863 |
+
"loss": 2.7373,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7934131736526946,
|
| 1868 |
+
"grad_norm": 0.49757909774780273,
|
| 1869 |
+
"learning_rate": 2.4893593504803826e-05,
|
| 1870 |
+
"loss": 2.7156,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.7964071856287425,
|
| 1875 |
+
"grad_norm": 0.48436880111694336,
|
| 1876 |
+
"learning_rate": 2.4207697256355145e-05,
|
| 1877 |
+
"loss": 2.7378,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.7994011976047904,
|
| 1882 |
+
"grad_norm": 0.5281265377998352,
|
| 1883 |
+
"learning_rate": 2.353007934185768e-05,
|
| 1884 |
+
"loss": 2.7332,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8023952095808383,
|
| 1889 |
+
"grad_norm": 0.5300716161727905,
|
| 1890 |
+
"learning_rate": 2.2860813773410106e-05,
|
| 1891 |
+
"loss": 2.663,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8053892215568862,
|
| 1896 |
+
"grad_norm": 0.429849237203598,
|
| 1897 |
+
"learning_rate": 2.2199973650834906e-05,
|
| 1898 |
+
"loss": 2.6993,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8083832335329342,
|
| 1903 |
+
"grad_norm": 0.5136377811431885,
|
| 1904 |
+
"learning_rate": 2.154763115369419e-05,
|
| 1905 |
+
"loss": 2.7297,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.811377245508982,
|
| 1910 |
+
"grad_norm": 0.523758053779602,
|
| 1911 |
+
"learning_rate": 2.0903857533405958e-05,
|
| 1912 |
+
"loss": 2.7227,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8143712574850299,
|
| 1917 |
+
"grad_norm": 0.5091609954833984,
|
| 1918 |
+
"learning_rate": 2.026872310546165e-05,
|
| 1919 |
+
"loss": 2.7174,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8173652694610778,
|
| 1924 |
+
"grad_norm": 0.5340681672096252,
|
| 1925 |
+
"learning_rate": 1.9642297241746142e-05,
|
| 1926 |
+
"loss": 2.706,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8203592814371258,
|
| 1931 |
+
"grad_norm": 0.4736381471157074,
|
| 1932 |
+
"learning_rate": 1.902464836296054e-05,
|
| 1933 |
+
"loss": 2.7244,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8233532934131736,
|
| 1938 |
+
"grad_norm": 0.5413190126419067,
|
| 1939 |
+
"learning_rate": 1.841584393114919e-05,
|
| 1940 |
+
"loss": 2.682,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8263473053892215,
|
| 1945 |
+
"grad_norm": 0.45962581038475037,
|
| 1946 |
+
"learning_rate": 1.7815950442330963e-05,
|
| 1947 |
+
"loss": 2.6946,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8293413173652695,
|
| 1952 |
+
"grad_norm": 0.5146113634109497,
|
| 1953 |
+
"learning_rate": 1.7225033419236503e-05,
|
| 1954 |
+
"loss": 2.6919,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8323353293413174,
|
| 1959 |
+
"grad_norm": 0.48548585176467896,
|
| 1960 |
+
"learning_rate": 1.6643157404151467e-05,
|
| 1961 |
+
"loss": 2.7278,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8353293413173652,
|
| 1966 |
+
"grad_norm": 0.46142202615737915,
|
| 1967 |
+
"learning_rate": 1.6070385951866952e-05,
|
| 1968 |
+
"loss": 2.7241,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8383233532934131,
|
| 1973 |
+
"grad_norm": 0.5457370281219482,
|
| 1974 |
+
"learning_rate": 1.5506781622737942e-05,
|
| 1975 |
+
"loss": 2.7047,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8413173652694611,
|
| 1980 |
+
"grad_norm": 0.48171794414520264,
|
| 1981 |
+
"learning_rate": 1.4952405975850026e-05,
|
| 1982 |
+
"loss": 2.6901,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.844311377245509,
|
| 1987 |
+
"grad_norm": 0.5125330090522766,
|
| 1988 |
+
"learning_rate": 1.4407319562295762e-05,
|
| 1989 |
+
"loss": 2.7052,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8473053892215568,
|
| 1994 |
+
"grad_norm": 0.4971305727958679,
|
| 1995 |
+
"learning_rate": 1.387158191856105e-05,
|
| 1996 |
+
"loss": 2.7405,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8502994011976048,
|
| 2001 |
+
"grad_norm": 0.539444625377655,
|
| 2002 |
+
"learning_rate": 1.3345251560022288e-05,
|
| 2003 |
+
"loss": 2.6869,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8532934131736527,
|
| 2008 |
+
"grad_norm": 0.5079681873321533,
|
| 2009 |
+
"learning_rate": 1.2828385974555202e-05,
|
| 2010 |
+
"loss": 2.7132,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8562874251497006,
|
| 2015 |
+
"grad_norm": 0.5335781574249268,
|
| 2016 |
+
"learning_rate": 1.2321041616255614e-05,
|
| 2017 |
+
"loss": 2.7378,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8592814371257484,
|
| 2022 |
+
"grad_norm": 0.4930953085422516,
|
| 2023 |
+
"learning_rate": 1.1823273899273435e-05,
|
| 2024 |
+
"loss": 2.678,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8622754491017964,
|
| 2029 |
+
"grad_norm": 0.4556470513343811,
|
| 2030 |
+
"learning_rate": 1.1335137191760093e-05,
|
| 2031 |
+
"loss": 2.6918,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8652694610778443,
|
| 2036 |
+
"grad_norm": 0.49691587686538696,
|
| 2037 |
+
"learning_rate": 1.0856684809930151e-05,
|
| 2038 |
+
"loss": 2.6948,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8682634730538922,
|
| 2043 |
+
"grad_norm": 0.48159509897232056,
|
| 2044 |
+
"learning_rate": 1.0387969012238064e-05,
|
| 2045 |
+
"loss": 2.694,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8712574850299402,
|
| 2050 |
+
"grad_norm": 0.4907771050930023,
|
| 2051 |
+
"learning_rate": 9.929040993670114e-06,
|
| 2052 |
+
"loss": 2.724,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.874251497005988,
|
| 2057 |
+
"grad_norm": 0.4428238570690155,
|
| 2058 |
+
"learning_rate": 9.47995088015281e-06,
|
| 2059 |
+
"loss": 2.6853,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8772455089820359,
|
| 2064 |
+
"grad_norm": 0.4720156192779541,
|
| 2065 |
+
"learning_rate": 9.040747723077902e-06,
|
| 2066 |
+
"loss": 2.7465,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8802395209580839,
|
| 2071 |
+
"grad_norm": 0.4877772629261017,
|
| 2072 |
+
"learning_rate": 8.61147949394483e-06,
|
| 2073 |
+
"loss": 2.7284,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8832335329341318,
|
| 2078 |
+
"grad_norm": 0.48317137360572815,
|
| 2079 |
+
"learning_rate": 8.192193079121002e-06,
|
| 2080 |
+
"loss": 2.7261,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8862275449101796,
|
| 2085 |
+
"grad_norm": 0.48045286536216736,
|
| 2086 |
+
"learning_rate": 7.782934274720777e-06,
|
| 2087 |
+
"loss": 2.6818,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8892215568862275,
|
| 2092 |
+
"grad_norm": 0.47768479585647583,
|
| 2093 |
+
"learning_rate": 7.3837477816033896e-06,
|
| 2094 |
+
"loss": 2.692,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8922155688622755,
|
| 2099 |
+
"grad_norm": 0.5294510722160339,
|
| 2100 |
+
"learning_rate": 6.994677200490507e-06,
|
| 2101 |
+
"loss": 2.6831,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.8952095808383234,
|
| 2106 |
+
"grad_norm": 0.48493650555610657,
|
| 2107 |
+
"learning_rate": 6.615765027204102e-06,
|
| 2108 |
+
"loss": 2.6695,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.8982035928143712,
|
| 2113 |
+
"grad_norm": 0.5145286917686462,
|
| 2114 |
+
"learning_rate": 6.247052648024765e-06,
|
| 2115 |
+
"loss": 2.6966,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9011976047904192,
|
| 2120 |
+
"grad_norm": 0.4654744565486908,
|
| 2121 |
+
"learning_rate": 5.888580335171368e-06,
|
| 2122 |
+
"loss": 2.7247,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9041916167664671,
|
| 2127 |
+
"grad_norm": 0.4812209904193878,
|
| 2128 |
+
"learning_rate": 5.540387242402434e-06,
|
| 2129 |
+
"loss": 2.7261,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.907185628742515,
|
| 2134 |
+
"grad_norm": 0.4804396331310272,
|
| 2135 |
+
"learning_rate": 5.20251140073953e-06,
|
| 2136 |
+
"loss": 2.692,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9101796407185628,
|
| 2141 |
+
"grad_norm": 0.4617965519428253,
|
| 2142 |
+
"learning_rate": 4.874989714313449e-06,
|
| 2143 |
+
"loss": 2.7015,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9131736526946108,
|
| 2148 |
+
"grad_norm": 0.4632161259651184,
|
| 2149 |
+
"learning_rate": 4.5578579563333e-06,
|
| 2150 |
+
"loss": 2.6971,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9161676646706587,
|
| 2155 |
+
"grad_norm": 0.530784010887146,
|
| 2156 |
+
"learning_rate": 4.251150765179291e-06,
|
| 2157 |
+
"loss": 2.7031,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9191616766467066,
|
| 2162 |
+
"grad_norm": 0.49502164125442505,
|
| 2163 |
+
"learning_rate": 3.954901640619368e-06,
|
| 2164 |
+
"loss": 2.6704,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9221556886227545,
|
| 2169 |
+
"grad_norm": 0.4807766079902649,
|
| 2170 |
+
"learning_rate": 3.6691429401502053e-06,
|
| 2171 |
+
"loss": 2.6889,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9251497005988024,
|
| 2176 |
+
"grad_norm": 0.49187591671943665,
|
| 2177 |
+
"learning_rate": 3.3939058754630882e-06,
|
| 2178 |
+
"loss": 2.7056,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.9281437125748503,
|
| 2183 |
+
"grad_norm": 0.5204874277114868,
|
| 2184 |
+
"learning_rate": 3.1292205090347248e-06,
|
| 2185 |
+
"loss": 2.7579,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9311377245508982,
|
| 2190 |
+
"grad_norm": 0.5218049883842468,
|
| 2191 |
+
"learning_rate": 2.875115750843771e-06,
|
| 2192 |
+
"loss": 2.722,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9341317365269461,
|
| 2197 |
+
"grad_norm": 0.4557659327983856,
|
| 2198 |
+
"learning_rate": 2.6316193552131884e-06,
|
| 2199 |
+
"loss": 2.7076,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.937125748502994,
|
| 2204 |
+
"grad_norm": 0.4934738874435425,
|
| 2205 |
+
"learning_rate": 2.398757917778727e-06,
|
| 2206 |
+
"loss": 2.7231,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9401197604790419,
|
| 2211 |
+
"grad_norm": 0.48129838705062866,
|
| 2212 |
+
"learning_rate": 2.176556872584168e-06,
|
| 2213 |
+
"loss": 2.7231,
|
| 2214 |
+
"step": 1570
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.9431137724550899,
|
| 2218 |
+
"grad_norm": 0.4986419975757599,
|
| 2219 |
+
"learning_rate": 1.965040489303194e-06,
|
| 2220 |
+
"loss": 2.7142,
|
| 2221 |
+
"step": 1575
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9461077844311377,
|
| 2225 |
+
"grad_norm": 0.48590248823165894,
|
| 2226 |
+
"learning_rate": 1.7642318705886286e-06,
|
| 2227 |
+
"loss": 2.7517,
|
| 2228 |
+
"step": 1580
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.9491017964071856,
|
| 2232 |
+
"grad_norm": 0.4665396809577942,
|
| 2233 |
+
"learning_rate": 1.574152949549057e-06,
|
| 2234 |
+
"loss": 2.7028,
|
| 2235 |
+
"step": 1585
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9520958083832335,
|
| 2239 |
+
"grad_norm": 0.4755544364452362,
|
| 2240 |
+
"learning_rate": 1.3948244873532078e-06,
|
| 2241 |
+
"loss": 2.6893,
|
| 2242 |
+
"step": 1590
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9550898203592815,
|
| 2246 |
+
"grad_norm": 0.44289329648017883,
|
| 2247 |
+
"learning_rate": 1.226266070962323e-06,
|
| 2248 |
+
"loss": 2.7409,
|
| 2249 |
+
"step": 1595
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.9580838323353293,
|
| 2253 |
+
"grad_norm": 0.5039628744125366,
|
| 2254 |
+
"learning_rate": 1.0684961109908353e-06,
|
| 2255 |
+
"loss": 2.6975,
|
| 2256 |
+
"step": 1600
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9610778443113772,
|
| 2260 |
+
"grad_norm": 0.47264382243156433,
|
| 2261 |
+
"learning_rate": 9.21531839695411e-07,
|
| 2262 |
+
"loss": 2.6629,
|
| 2263 |
+
"step": 1605
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.9640718562874252,
|
| 2267 |
+
"grad_norm": 0.4993440806865692,
|
| 2268 |
+
"learning_rate": 7.853893090928654e-07,
|
| 2269 |
+
"loss": 2.7854,
|
| 2270 |
+
"step": 1610
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.9670658682634731,
|
| 2274 |
+
"grad_norm": 0.4866315722465515,
|
| 2275 |
+
"learning_rate": 6.600833892068336e-07,
|
| 2276 |
+
"loss": 2.7192,
|
| 2277 |
+
"step": 1615
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.9700598802395209,
|
| 2281 |
+
"grad_norm": 0.42745715379714966,
|
| 2282 |
+
"learning_rate": 5.456277664436127e-07,
|
| 2283 |
+
"loss": 2.7148,
|
| 2284 |
+
"step": 1620
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.9730538922155688,
|
| 2288 |
+
"grad_norm": 0.49950963258743286,
|
| 2289 |
+
"learning_rate": 4.4203494209733576e-07,
|
| 2290 |
+
"loss": 2.743,
|
| 2291 |
+
"step": 1625
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9760479041916168,
|
| 2295 |
+
"grad_norm": 0.45195409655570984,
|
| 2296 |
+
"learning_rate": 3.4931623098445334e-07,
|
| 2297 |
+
"loss": 2.7275,
|
| 2298 |
+
"step": 1630
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.9790419161676647,
|
| 2302 |
+
"grad_norm": 0.47945547103881836,
|
| 2303 |
+
"learning_rate": 2.674817602079327e-07,
|
| 2304 |
+
"loss": 2.725,
|
| 2305 |
+
"step": 1635
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.9820359281437125,
|
| 2309 |
+
"grad_norm": 0.48270925879478455,
|
| 2310 |
+
"learning_rate": 1.965404680511207e-07,
|
| 2311 |
+
"loss": 2.6793,
|
| 2312 |
+
"step": 1640
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.9850299401197605,
|
| 2316 |
+
"grad_norm": 0.4643472731113434,
|
| 2317 |
+
"learning_rate": 1.3650010300150228e-07,
|
| 2318 |
+
"loss": 2.7078,
|
| 2319 |
+
"step": 1645
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.9880239520958084,
|
| 2323 |
+
"grad_norm": 0.4867922365665436,
|
| 2324 |
+
"learning_rate": 8.736722290429988e-08,
|
| 2325 |
+
"loss": 2.7148,
|
| 2326 |
+
"step": 1650
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.9910179640718563,
|
| 2330 |
+
"grad_norm": 0.5043019652366638,
|
| 2331 |
+
"learning_rate": 4.9147194246290664e-08,
|
| 2332 |
+
"loss": 2.7255,
|
| 2333 |
+
"step": 1655
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.9940119760479041,
|
| 2337 |
+
"grad_norm": 0.4517767131328583,
|
| 2338 |
+
"learning_rate": 2.1844191569597716e-08,
|
| 2339 |
+
"loss": 2.6949,
|
| 2340 |
+
"step": 1660
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.9970059880239521,
|
| 2344 |
+
"grad_norm": 0.4739466905593872,
|
| 2345 |
+
"learning_rate": 5.461197015765862e-09,
|
| 2346 |
+
"loss": 2.7152,
|
| 2347 |
+
"step": 1665
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 1.0,
|
| 2351 |
+
"grad_norm": 0.46331822872161865,
|
| 2352 |
+
"learning_rate": 0.0,
|
| 2353 |
+
"loss": 2.7246,
|
| 2354 |
+
"step": 1670
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 1.0,
|
| 2358 |
+
"step": 1670,
|
| 2359 |
+
"total_flos": 1.396344619008e+16,
|
| 2360 |
+
"train_loss": 2.9739012581145694,
|
| 2361 |
+
"train_runtime": 9187.7213,
|
| 2362 |
+
"train_samples_per_second": 11.629,
|
| 2363 |
+
"train_steps_per_second": 0.182
|
| 2364 |
+
}
|
| 2365 |
+
],
|
| 2366 |
+
"logging_steps": 5,
|
| 2367 |
+
"max_steps": 1670,
|
| 2368 |
+
"num_input_tokens_seen": 0,
|
| 2369 |
+
"num_train_epochs": 1,
|
| 2370 |
+
"save_steps": 100,
|
| 2371 |
+
"stateful_callbacks": {
|
| 2372 |
+
"TrainerControl": {
|
| 2373 |
+
"args": {
|
| 2374 |
+
"should_epoch_stop": false,
|
| 2375 |
+
"should_evaluate": false,
|
| 2376 |
+
"should_log": false,
|
| 2377 |
+
"should_save": true,
|
| 2378 |
+
"should_training_stop": true
|
| 2379 |
+
},
|
| 2380 |
+
"attributes": {}
|
| 2381 |
+
}
|
| 2382 |
+
},
|
| 2383 |
+
"total_flos": 1.396344619008e+16,
|
| 2384 |
+
"train_batch_size": 16,
|
| 2385 |
+
"trial_name": null,
|
| 2386 |
+
"trial_params": null
|
| 2387 |
+
}
|